Вы находитесь на странице: 1из 115

Think Twice: How the Gut's "Second Brain" Inuences Mood...

http://www.scienticamerican.com/article/gut-second-brain/?p...

ADVERTISEMENT

Permanent Address: http://www.scientificamerican.com/article/gut-second-brain/


Health Features
This article is from the In-Depth Report Science at the Winter Olympics

Think Twice: How the Gut's "Second Brain" Inuences


Mood and Well-Being
The emerging and surprising view of how the enteric nervous system in our bellies goes far beyond just processing the food we eat
By Adam Hadhazy | February 12, 2010 |

As Olympians go for the gold in Vancouver, even the steeliest are likely to experience
that familiar feeling of "butterflies" in the stomach. Underlying this sensation is an
often-overlooked network of neurons lining our guts that is so extensive some
scientists have nicknamed it our "second brain".
A deeper understanding of this mass of neural tissue, filled with important
neurotransmitters, is revealing that it does much more than merely handle digestion
or inflict the occasional nervous pang. The little brain in our innards, in connection
with the big one in our skulls, partly determines our mental state and plays key roles
in certain diseases throughout the body.
Although its influence is far-reaching, the second brain is not the seat of any
conscious thoughts or decision-making.
"The second brain doesn't help with the great thought processesreligion,
philosophy and poetry is left to the brain in the head," says Michael Gershon,

ISTOCKPHOTO/ERAXION

ADVERTISEMENT

chairman of the Department of Anatomy and Cell Biology at New YorkPresbyterian


Hospital/Columbia University Medical Center, an expert in the nascent field of neurogastroenterology and author of the 1998 book The
Second Brain (HarperCollins).
Technically known as the enteric nervous system, the second brain consists of sheaths of neurons embedded in the walls of the long
tube of our gut, or alimentary canal, which measures about nine meters end to end from the esophagus to the anus. The second brain
contains some 100 million neurons, more than in either the spinal cord or the peripheral nervous system, Gershon says.
This multitude of neurons in the enteric nervous system enables us to "feel" the inner world of our gut and its contents. Much of this
neural firepower comes to bear in the elaborate daily grind of digestion. Breaking down food, absorbing nutrients, and expelling of
waste requires chemical processing, mechanical mixing and rhythmic muscle contractions that move everything on down the line.
Thus equipped with its own reflexes and senses, the second brain can control gut behavior independently of the brain, Gershon says. We
likely evolved this intricate web of nerves to perform digestion and excretion "on site," rather than remotely from our brains through the
middleman of the spinal cord. "The brain in the head doesn't need to get its hands dirty with the messy business of digestion, which is
delegated to the brain in the gut," Gershon says. He and other researchers explain, however, that the second brain's complexity likely
cannot be interpreted through this process alone.
"The system is way too complicated to have evolved only to make sure things move out of your colon," says Emeran Mayer, professor of

1 of 3

20/07/2015 2:32 pm

Think Twice: How the Gut's "Second Brain" Inuences Mood...

http://www.scienticamerican.com/article/gut-second-brain/?p...

physiology, psychiatry and biobehavioral sciences at the David Geffen School of Medicine at the University of California, Los Angeles
(U.C.L.A.). For example, scientists were shocked to learn that about 90 percent of the fibers in the primary visceral nerve, the vagus,
carry information from the gut to the brain and not the other way around. "Some of that info is decidedly unpleasant," Gershon says.
The second brain informs our state of mind in other more obscure ways, as well. "A big part of our emotions are probably influenced by
the nerves in our gut," Mayer says. Butterflies in the stomachsignaling in the gut as part of our physiological stress response, Gershon
saysis but one example. Although gastrointestinal (GI) turmoil can sour one's moods, everyday emotional well-being may rely on
messages from the brain below to the brain above. For example, electrical stimulation of the vagus nervea useful treatment for
depressionmay mimic these signals, Gershon says.
Given the two brains' commonalities, other depression treatments that target the mind can unintentionally impact the gut. The enteric
nervous system uses more than 30 neurotransmitters, just like the brain, and in fact 95 percent of the body's serotonin is found in the
bowels. Because antidepressant medications called selective serotonin reuptake inhibitors (SSRIs) increase serotonin levels, it's little
wonder that meds meant to cause chemical changes in the mind often provoke GI issues as a side effect. Irritable bowel syndrome
which afflicts more than two million Americansalso arises in part from too much serotonin in our entrails, and could perhaps be
regarded as a "mental illness" of the second brain.
Scientists are learning that the serotonin made by the enteric nervous system might also play a role in more surprising diseases: In a
new Nature Medicine study published online February 7, a drug that inhibited the release of serotonin from the gut counteracted the
bone-deteriorating disease osteoporosis in postmenopausal rodents. (Scientific American is part of Nature Publishing Group.) "It was
totally unexpected that the gut would regulate bone mass to the extent that one could use this regulation to cureat least in rodents
osteoporosis," says Gerard Karsenty, lead author of the study and chair of the Department of Genetics and Development at Columbia
University Medical Center.
Serotonin seeping from the second brain might even play some part in autism, the developmental disorder often first noticed in early
childhood. Gershon has discovered that the same genes involved in synapse formation between neurons in the brain are involved in the
alimentary synapse formation. "If these genes are affected in autism," he says, "it could explain why so many kids with autism have GI
motor abnormalities" in addition to elevated levels of gut-produced serotonin in their blood.
Down the road, the blossoming field of neurogastroenterology will likely offer some new insight into the workings of the second
brainand its impact on the body and mind. "We have never systematically looked at [the enteric nervous system] in relating lesions in
it to diseases like they have for the" central nervous system, Gershon says. One day, perhaps there will be well-known connections
between diseases and lesions in the gut's nervous system as some in the brain and spinal cord today indicate multiple sclerosis.
Cutting-edge research is currently investigating how the second brain mediates the body's immune response; after all, at least 70
percent of our immune system is aimed at the gut to expel and kill foreign invaders.
U.C.L.A.'s Mayer is doing work on how the trillions of bacteria in the gut "communicate" with enteric nervous system cells (which they
greatly outnumber). His work with the gut's nervous system has led him to think that in coming years psychiatry will need to expand to
treat the second brain in addition to the one atop the shoulders.
So for those physically skilled and mentally strong enough to compete in the Olympic Gamesas well as those watching at homeit
may well behoove us all to pay more heed to our so-called "gut feelings" in the future.
SEE ALSO:

Energy & Sustainability: 5 Steps to Feed the World and Sustain the Planet | Evolution: New Fossil Reveals Velociraptor Sported
Feathers | Mind & Brain: Nail Biting May Arise from Perfectionism | Space: Europa's "Brown Gunk" Suggests a Briny
Sea | Technology: Timeline: The Amazing Multimillion-Year History of Processed Food | More Science: The Flavor Connection

2 of 3

20/07/2015 2:32 pm

Think Twice: How the Gut's "Second Brain" Inuences Mood...

http://www.scienticamerican.com/article/gut-second-brain/?p...

Recommended For You


1.

The Problem with Female Superheroes 3 weeks ago scienticamerican.com ScienticAmerican.com Mind & Brain
2.

Metacognition Is the Forgotten Secret to Success 11 months ago scienticamerican.com ScienticAmerican.com Features

What Kind of Introvert Are You? 9 months ago blogs.scienticamerican.com ScienticAmerican.com Jennifer Odessa Grimes

2015 Scientific American, a Division of Nature America, Inc.


All Rights Reserved.
YYEESS!! Send me a free issue of Scientific
American with no obligation to continue
the subscription. If I like it, I will be billed
for the one-year subscription.

Subscribe Now

3 of 3

20/07/2015 2:32 pm

Psychological Medicine, 2001, 31, 761767.


" 2001 Cambridge University Press

Printed in the United Kingdom

EDITORIAL

Acute sickness behaviour : an immune system-to-brain


communication ?"
Over the past 20 years, psychoneuroimmunological research has produced a large body of evidence
that challenges the historically dominant view that the immune system operates in an autonomous
manner independent of other physiological systems. Today, there is little doubt that the brain and
the immune system are intimately linked and capable of reciprocal communication (Ader et al.
1991). Despite the acknowledged bi-directional nature of the brainimmune system connection, the
predominant focus of study has been on the effects of psychological and behavioural events (e.g.
stress) on immune responses and disease processes, and the mechanisms underlying such effects (see
Kusnekov & Rabin, 1994 ; Maier et al. 1994 ; Rozlog et al. 1999). However, considerable interest
in the possibilities of immune-system-to-brain communication was initiated by a seminal paper
considering the biological basis of behaviour in sick animals (Hart, 1988). Subsequently, the
immunological determinants of the behavioural, cognitive and emotional changes associated with
acute illness, as well as with more chronic psychopathological states (e.g. depression) have become
the subject of rapidly expanding areas of research (e.g. Kent et al. 1992 ; Lloyd et al. 1992 ; Hickie
& Lloyd, 1995 ; Maes et al. 1995 a ; Rothwell & Hopkins, 1995 ; Dantzer et al. 1996 ; Maier &
Watkins, 1998 ; Vollmer-Conna et al. 1998 ; Maes, 1999).
The main objective of this editorial is to provide a succinct overview of current knowledge of the
normal behavioural correlates of acute infective illness, their adaptive function and underlying
mechanisms. Elucidation of the processes involved in the appearance, maintenance and inhibition
of normal sickness behaviour is important if extrapolations from this phenomenon to more
chronic psychopathological conditions are to provide more than a new label for poorly understood
non-specific symptom clusters.
A ROLE FOR SICKNESS BEHAVIOUR IN THE HOST DEFENCE AGAINST INFECTION
AND INFLAMMATION
Acute infective illnesses, both in animals and humans, are typically accompanied by a cluster of
non-specific symptoms such as fever, an increased need to sleep, hyperalgesia, anorexia, loss of
interest in usual activities, decreased social interaction and body care, depression and impaired
concentration (Hart, 1988 ; Dantzer et al. 1996). Perhaps because they are prevalent and nonspecific concomitants of infective illnesses, these phenomena are commonly dismissed or relatively
ignored by physicians. Yet, it has recently been argued that sickness behaviours constitute a highly
organized and evolved strategy to combat infection and injury. Specifically, it has been suggested
that the behavioural changes function to conserve energy and, thus, facilitate the role of fever in
stimulating immune function and inhibiting the proliferation of thermo-sensitive pathogens (Hart,
1988).
The adaptive nature of the fever response is well documented and apparent from numerous
demonstrations that inhibition of fever (e.g. by placing infected animals in a cold environment, or
treating them with antipyretic drugs) is detrimental to survival (Kluger, 1979). Interference with
sickness behaviours, by force-feeding or sleep-depriving animals with acute infections, similarly
produces a marked reduction in survival rates (Hart, 1988). Moreover, there is evidence to suggest
" Addressforcorrespondence :DrUte! Vollmer-Conna,DepartmentofHumanBehaviour,SchoolofPsychiatry,UniversityofNewSouth
Wales, UNSW Sydney 2052, Australia.

761

762

Editorial : Acute sickness behaviour

that sickness behaviour is not merely an automatic, inflexible reaction to illness, but rather the
expression of a central motivational state. That is, animals will actively learn responses to be able
to engage in these behaviours when they are infected and will choose conditions under which they
can occur (Dantzer et al. 1996 ; Maier & Watkins, 1999).
ACUTE SICKNESS BEHAVIOUR IS IMMUNOLOGICALLY MEDIATED
The production of the pro-inflammatory cytokines such as tumour necrosis factor (TNF),
interleukin-1 (IL-1) and IL-6 by activated immune cells (monocyte}macrophages, lymphocytes) is
an integral part of the host response to infection. These cytokines act as messenger molecules and
play a pivotal role in the orchestration of the acute phase response (Dinarello, 1997 ; Papanicolaou,
1998). In recent years, it has also become clear that the necessary synchrony between metabolic,
physiological and behavioural aspects of the individuals response to infection depends on the
activities of these same cytokines. It has long been known that fever is not caused directly by
invading pathogens but rather by the action of soluble immunological products. These endogenous
pyrogens were subsequently identified as being pro-inflammatory cytokines, with IL-1 the most
potent pyrogenic agent (Dinarello et al. 1977). The first indication of a role for cytokines in the
induction of acute sickness behaviour came from clinical trials with purified or recombinant
cytokines. Administration of cytokines in the treatment of cancer and chronic viral infections such
as hepatitis B and C produced a syndrome of adverse effects, including fever, fatigue, malaise,
headaches, anorexia, depression and, at high doses, delirium (Renault & Hoofnagel, 1989 ;
Dinarello, 1997). However, as these clinical observations were made predominantly on patients with
significant medical illnesses, generalization from these findings to argue for a role for cytokines
in normal sickness behaviour is problematical.
Animal experiments have confirmed that most aspects of sickness behaviour can be induced in
a dose-dependent fashion by systemic or intracerebral injections of pro-inflammatory cytokines,
particularly IL-1 (Dantzer et al. 1996). Systemic injections with lipopolysaccharide (LPS), a strong
inducer of the synthesis and release of most pro-inflammatory cytokines, also produces the full
spectrum of sickness behaviour. Conversely, administration of specific antagonists (e.g. IL-1
receptor antagonist (IL-1Ra)) inhibits many of the central effects of this cytokine (Rothwell &
Hopkins, 1995). Given the pleiotropism and redundancy of the cytokine network, it has been
difficult to determine the precise contribution of individual cytokines to sickness behaviour.
However, there is evidence to suggests that TNF- and IL-1-, which are synthesized very early in
the immune response, are more potent than those that are induced later (e.g. IL-6) and act in
synergy to produce sickness behaviour (Dantzer et al. 1996). IL-6 appears to require the presence
of other pro-inflammatory cytokines (e.g. IL-1) to produce behavioural symptoms of sickness
(Bluthe et al. 1998). The role of IL-6 in sickness behaviour is likely to be complex, as recent evidence
suggests that IL-6 has both pro- and anti-inflammatory functions, the latter including the induction
of natural antagonists to TNF and IL-1 and stimulation of the hypothalamicpituitaryadrenal
(HPA) axis (Papanicolaou, 1998).
LINKS BETWEEN ACUTE SICKNESS BEHAVIOUR AND THE STRESS RESPONSE
Experimental administration of pro-inflammatory cytokines or LPS to animals has revealed that,
in addition to acute sickness behaviour, a classic stress response is produced. This is characterized
by activation of the sympathetic nervous system and release of plasma catecholamines as well as
activation of the HPA system leading to the release of adrenocorticotrophic hormone (ACTH) and
glucocorticoids (Dunn, 1995 ; Maier & Watkins, 1999). In humans, IL-6 appears to be a particularly
potent stimulator of the HPA axis. For example, daily administration of recombinant IL-6 over a
week was found to produce a remarkable activation and enlargement of the adrenal glands, similar
to that seen in patients with Cushings disease (Mastorakos et al. 1993). Because glucocorticoids,
in turn, exert negative feedback on the secretion of IL-6, it has been argued that this cytokine
displays the traditional characteristics of a hormone (Papanicolaou, 1998).

Editorial : Acute sickness behaviour

763

It is well established that exposure to stressors may result in the suppression of cell-mediated
immune responses (Kusnekov & Rabin, 1994). However, new evidence suggests that the sequelae
of exposure to acute stressors are considerably more complex than previously thought and include
aspects of immune activation not unlike the host response to acute infections (Maier & Watkins,
1998 ; Maes, 1999 ; Tringali et al. 2000, for reviews). Briefly, in both humans and animals, diverse
physical and psychosocial stressors were found to produce leucocytosis, a shift in liver metabolism
toward production of acute phase proteins, secretion of pro-inflammatory cytokines and fever.
Moreover, behavioural changes including anorexia, decreases in activation, exploration, social
interaction and aggression, depressed affect and cognitive impairment, bear a strong resemblance
to cytokine-induced sickness behaviour and appear more adaptive in the context of sickness then
as a component of the frightfight response.
The similarities between aspects of the stress response and acute sickness behaviour point to a
common mediator. Indeed, a direct role for pro-inflammatory cytokines in the stress response was
established by experiments blocking specific cytokine receptors in the brain during exposure to a
stressor. For example, injections of the IL-1 receptor antagonist (IL-1Ra) abolished the learned
helplessness response and exaggerated fear typically induced by inescapable shock (Maier &
Watkins, 1995), as well as blunting the pituitaryadrenal response and inhibiting the release of
hypothalamic monoamines normally associated with immobilization stress (Shintani et al. 1995).
There is also evidence that the stress-induced secretion of cytokines in hypothalamic structures and
the periphery are mediated by catecholaminergic mechanisms such as the activation of -adrenergic
receptors (Takaki et al. 1994 ; Papanicolaou, 1998 ; Tringali et al. 2000).
There remains little doubt that the systems mediating the organisms defence against infection
and injury on the one hand, and against psychosocial and environmental stressors on the other, are
closely connected. Indeed, it has been proposed that environmental (i.e. external ) stressors
essentially activate the same neuroendorineimmune circuitry as immunogenic agents (i.e.
internal stressors), although they may enter the circuit at different sites (Maier & Watkins, 1998 ;
Maes, 1999). The end product of this activation (in terms of neuroendocrine, behavioural, cognitive,
emotional, or immunological consequences which are clearly different for different trigger stimuli)
presumably depends on the specific characteristics of the trigger (Maier &Watkins, 1998 ; Tringali
et al. 2000).
From an evolutionary perspective, the similarities in the response to stress and acute illness may
be understood by studying the systems subserving adaptation and defence in primitive organisms.
Even in the most primitive organisms such as molluscs and sponges, which are incapable of
elaborate defence against distal threats, some form of host response against infection and injury can
be identified (Beck et al. 1994). Moreover, there is evidence that as early in evolution as the molluscs,
defence against infection and injury involved pro-inflammatory cytokines in bi-directional
communication between immunological and neural structures, as well as the release peptides
traditionally viewed as stress hormones (Clatworthy, 1996 ; Maier & Watkins, 1998, for reviews).
When the frightflight response evolved later in more complex organisms, existing mechanisms such
as cytokine-based communication networks appear to have been incorporated in this new
adaptation (Maier & Watkins, 1998). This account, although speculative, offers an explanation for
the mobilization of inflammatory processes during acute stress, and provides a basis for the
manifestation of seemingly maladaptive behavioural changes (i.e. sickness behaviour) after
exposure to a significant stressor.
POSSIBLE MECHANISMS UNDERLYING THE PRODUCTION OF ACUTE SICKNESS
BEHAVIOUR
In view of their documented involvement in diverse centrally-mediated phenomena, there is little
doubt that cytokines are capable of providing signals to the brain to alter neural activity. The
mechanism through which these peripherally produced molecules might act on the brain, however,
has been the subject of much debate. Cytokines are not thought to cross the bloodbrain barrier,

764

Editorial : Acute sickness behaviour

owing to their large molecular weight and hydrophilic properties. A number of specialized
mechanisms have been identified, however, that would allow blood-borne cytokines to signal the
brain. These include an active transport system across the barrier (Banks et al. 1991), and cytokine
entry at regions of the brain (e.g. circumventricular organs) where the barrier is weak or absent to
trigger the production of second messengers (e.g. prostaglandins) to neural targets (e.g.
hypothalamic regions) or to induce local cytokine production (Saper & Breder, 1994). As an
alternative to the bloodborne route, several authors have pursued the possibility of direct neural
signalling via the vagus nerve (Dantzer et al. 1996 ; Maier & Watkins, 1999). Whatever the route
used to signal the brain, most authors agree that local production of cytokines by glial cells plays
an important role in the induction of sickness behaviour. This offers the best explanation of the welldocumented increase in brain levels of cytokines after peripheral administration, as well as
accounting for the presence of cytokine-producing cells (glial cells) and specific binding sites
throughout the brain (Vollmer-Conna et al. 1998 ; Maier & Watkins, 1999). The exact mechanism
responsible to translate the immune signal into a neural signal is still unclear, but appears to involve
alterations in a variety of neuropeptide (e.g. corticotropin-releasing hormone (CRH), substance P,
opioids) and neurotransmitter systems (nonadrenaline, serotonin, gamma-aminobutyric acid
(GABA)) (Rothwell & Hopkins, 1995).
HUMAN STUDIES OF ACUTE SICKNESS BEHAVIOUR
To date, very few studies have systematically examined the potential relationships between proinflammatory cytokines and mental and behavioural symptoms in sick humans. Although animal
studies have contributed much to our understanding of acute sickness behaviour, they are poorly
suited to examine complex behaviours and}or more subtle changes in brain function (e.g. cognitive
deficits, mood alterations) that are nonetheless an integral part of sickness behaviour. Smith and
colleagues (1987, 1988) studied cognitive performance in healthy volunteers with experimentally
induced common cold (rhinovirus) or influenza, or an infusion of interferon-. Both influenza
infections and interferon- administration produced impairments in stimulus detection tasks,
whereas common colds were associated with impaired handeye coordination. Interestingly, these
cognitive performance deficits were also demonstrated in subjects with subclinical infections.
Interpretation of these findings is somewhat limited by the artificial context in which they were
obtained, and the use of a cytokine not generally considered one of the prototypic inflammatory
cytokines. They do suggest, however, that even minor infections with agents that do not directly
infect the brain, may be associated with significant impairment in cognitive performance. Moreover,
such impairment appears to be related to the action of cytokines.
Significant neurocognitive deficits have also been demonstrated in a clinical sample of patients
with acute infections (both EpsteinBarr virus (EBV), and influenza-like infections ; Vollmer-Conna
et al. 1997). In comparison to well-matched healthy controls, these patients reported a significant
increase in negative affect and fatigue, and showed deficits in selective attention, response speed,
pursuit tracking, and performance accuracy. A similar study recently demonstrated that patients
with influenza-like illnesses were impaired on aspects of everyday memory (Capuron et al. 1999).
These findings are suggestive only, as immunological correlates were not assessed and not all
infections were serologically documented. What is clearly needed is the unambiguous identification
of cognitive deficits associated with specific acute infections. In addition, the relationship between
such deficits and the action of cytokines must be examined.
CLINICAL AND PRACTICAL IMPLICATIONS OF THE BEHAVIOURAL AND MENTAL
EFFECTS OF CYTOKINES
An accumulation of evidence suggests that behavioural}mental changes typically associated with
acute infections (i.e. acute sickness behaviours) are immunologically mediated and form part of the
host defence. The efficacy of these behavioural disease-fighting strategies is demonstrated by the fact

Editorial : Acute sickness behaviour

765

that both animals and man have survived infections and injury through evolutionary history. A
better understanding of acute sickness behaviour and its underlying mechanism(s) will provide new
insights into how sickness and recovery processes are organized in the brain. This may lead to the
development of management practices that complement the innate disease-coping strategies, and
may be particularly beneficial when dealing with viruses or drug resistant bacteria (Hart, 1988).
Several reports suggest that acute infective illness is associated with significant cognitive
impairment (Smith et al. 1987 ; Volmer-Conna et al. 1997 ; Capuron et al. 1999). Yet, many patients
continue with their usual daily routine throughout an illness such as influenza, glandular fever or
Q fever. There is some indication from an ongoing large, prospective study of infective cohorts
(Bennett et al. 1998) that accidents in the workplace are more prevalent in the context of acute Q
fever (Lloyd, A. unpublished data). Elucidation of the full extent of behavioural and mental changes
during acute infections may thus have important implications for road and work place safety.
POTENTIALLY ABNORMAL MANIFESTATIONS OF SICKNESS BEHAVIOUR
Sickness behaviour, similar to the frightflight response, is viewed as a highly organized strategy
critical to survival. However, in the same way as pathological fear and anxiety are debilitating,
excessive sickness behaviour can be detrimental. Over the past decade there has been much
speculation on a possible role for cytokines in the pathogenesis of neuropsychiatric syndromes,
notably post-infective fatigue syndromes (e.g. Chao et al. 1991 ; Hickie & Lloyd, 1995 ; VollmerConna et al. 1998 ;) and, more recently, major depression (e.g. Maes et al. 1995 a ; Maes 1999 ;
Charlton, 2000).
Clinical and scientific interest in exploring the possibility of an immunological basis for
neuropsychiatric disorders was initially fuelled by the striking similarity between acute sickness
behaviours and key symptoms reported in depression and fatigue syndromes (i.e. loss of appetite,
malaise, psychomotor slowing, altered sleep, fatigue, anhedonia, depressed affect and cognitive
impairment). In addition, the discovery that psychosocial stressors can activate the inflammatory
response system has lent more credence to a proposed role for pro-inflammatory cytokines in the
pathogenesis of stress-related disorders such as depression (Maes, 1999).
Research examining immunological correlates of depression has produced evidence consistent
with the notion of immune activation, including elevated leukocyte counts and activation markers,
increased production of pro-inflammatory cytokines and acute phase proteins. Moreover, tricyclic
antidepressants and serotonin reuptake inhibitors appear to suppress these inflammatory responses
(Maes et al. 1995 a, for review). On the other hand, there is substantial evidence documenting
immunosuppression and increased susceptibility to disease in patients with major depression
(Herbert & Cohen, 1993). It has been suggested that the simultaneous signs of immune activation
and suppression in depression may be reconciled by reference to the action of multiple feedback
systems generated to contain the immune response or T cell exhaustion (Maes et al. 1995 b).
The emergence of immunological hypotheses for neuropsychiatric disorders, such as depression,
challenges traditional pathophysiological views, and provides a fresh perspective on symptomatology (e.g. depression as a variant of sickness behaviour). However, the data to date are not
sufficient to determine whether altered levels of pro-inflammatory cytokines play an aetiological
role, are the consequence of depression, or merely reflect an epiphenomenon. Interpretation of the
available evidence is limited by a variety of methodological problems including the predominant
reliance on cross-sectional, correlational designs, one-off sampling and the inevitable heterogeneity
of subject samples. An additional complication inherent in this type of research lies in the attempt
to establish associations between centrally-mediated symptoms and measurements obtained from
peripheral blood samples. Although such endeavours are justified, in principle, by the bi-directional
nature of the connection between the brain and the immune system, it is unlikely that clear and
definitive answers about the role of specific circulating cytokines in the production of specific
psychopathological symptoms can be established in this way.
Understanding potentially abnormal variants of a natural phenomenon clearly requires an in-

766

Editorial : Acute sickness behaviour

depth knowledge of the phenomenon in question. Therefore, a systematic study of normal sickness
behaviour in humans (examining the full spectrum, development and immunological mediators of
symptoms during acute infections) is needed to construct a sound knowledge-base enabling
identification and evaluation of inappropriate or excessive manifestations of this phenomenon.
Moreover, such knowledge is essential in light of the fact that the manipulation of cytokine systems
is rapidly developing into a new area of therapeutics (Dinarello, 1997), and which may have
unanticipated neurobehavioural consequences.
! -
I wish to thank Professor Gordon Parker for invaluable advice and encouragement throughout the preparation
of this paper.

REFERENCES
Ader, R., Felten, D. L. & Cohen, N. (1991). Psychoneuroimmunology,
2nd edn. Academic Press : San Diego, CA.
Banks, W. A., Oritz, L., Plotkin, S. R. & Kastin, A. J. (1991).
Human interleukin (IL) 1 alpha, murine IL-1 alpha, and murine
IL-1 beta are transported from blood to brain in the mouse by a
shared saturable mechanism. Journal of Pharmacology and Experimental Therapeutics 259, 988996.
Beck, G., Cooper, E. L., Hobicht, G. S. & Marchalonis, J. J. (1994).
Primordial immunity : foundations for the vertebrate immune
system. Annals of the New York Academy of Sciences 712,
206212.
Bennett, B. K., Hickie, I. B., Vollmer-Conna, U. S., Quigley, B.,
Brennan, C. M., Wakefield, D., Douglas, M. P., Hansen, G. R.,
Tahmindjis, A. J. & Lloyd, A. R. (1998). The relationship between
fatigue, psychological and immunological variables in acute
infectious illness. Australian and New Zealand Journal of Psychiatry
32, 180186.
Bluthe, R. M., Michaud, B., Poli, V., Bernay, F., Parnet, P. &
Dantzer, R. (1998). Interleukin-6 is active only in presence of other
proinflammatory cytokines to induce sickness behaviour. Neuroimmunomodulation 5, 7.
Capuron, L., Lamarque, D., Dantzer, R. & Goodall, G. (1999).
Attentional and mnemonic deficits associated with infectious
diseases in humans. Psychological Medicine 29, 291297.
Chao, C. C., Janoff, E. N., Hu, S., Thomas, K., Gallagher, M.,
Tsang, M. & Peterson, P. K. (1991). Altered cytokine release in
peripheral blood mononuclear cell cultures from patients with
chronic fatigue syndrome. Cytokine 3, 292298.
Charlton, B. G. (2000). The malaise theory of depression : major
depressive disorder is sickness behaviour and antidepressants are
analgesics. Medical Hypotheses 54, 126130.
Clatworthy, A. L. (1996). A simple system approach to neuralimmune communication. Comparative Biochemistry and Physiology
115A, 110.
Dantzer, R., Bluthe, R.-M., Aubert, A., Goodall, G., Bret-Dibat, JL., Kent, S., Goujon, E., Laye, S., Parnet, P. & Kelley, K. W.
(1996). Cytokine actions on behavior. In Cytokines in the Nervous
System (ed. N. J. Rothwell), pp. 117144. R. G. Landes Co.: New
York.
Dinarello, C. A. (1997). Proinflammatory and anti-inflammatory
cytokines as mediators in the pathogenesis of septic shock. Chest
112, 312S329S.
Dinarello, C. A., Renfer, L. & Wolff, S. M. (1977). Human leukocytic
pyrogen : purification and development of a radioummunoassay.
Proceedings of the National Academy of Science USA 74, 4624.
Dunn, A. J. (1995). Interactions between the nervous system and the
immune system : implications for psychopharmacology. In Psychopharmacology : The fourth Generation of Progress (ed. F. E. Bloom
and D. J. Kupfer), pp. 719733. Raven Press : New York.
Hart, B. L. (1988). Biological basis of the behavior of sick animals.
Neuroscience and Biobehavioral Reviews 12, 123137.
Herbert, I. & Cohen, S. (1993). Depression and immunity : a metaanalytic review. Psychological Bulletin 113, 472486.

Hickie, I. & Lloyd, A. (1995). Are cytokines associated with


neuropsychiatric syndromes in humans ? International Journal of
Immunopharmacology 17, 677683.
Kent, S., Bluthe, R. M. & Kelley, K. W. (1992). Sickness behaviour
as a new target for drug development. Trends in Pharmacological
Sciences 13, 2428.
Kluger, M. J. (1979). Fever : Its Biology, Evolution and Function.
Princeton University Press : Princeton, NJ.
Kusnekov, A. W. & Rabin, B. S. (1994). Stressor-induced alterations
of immune function : mechanisms and issues. International Archives
of Allergy and Immunology 105, 107121.
Lloyd, A., Hickie, I., Hickie, C., Dwyer, J. & Wakefield, D. (1992).
Cell-mediated immunity in patients with chronic fatigue syndrome,
healthy control subjects and patients with major depression.
Clinical and Experimental Immunology 87, 7679.
Maes, M. (1999). Psychological stress, cytokines and the inflammatory response system. Current Opinion in Psychiatry 12, 695700.
Maes, M., Smith, R. & Scharpe, S. (1995 a). The monocyteT
lymphocyte hypothesis of major depression. Psychoneuroendocrinology 20, 111116.
Maes, M., Meltzer, H. Y., Bosmans, E., Bergmans, R.,
Vandoolaeghe, E., Ranjan, R. & Desynder, R. (1995 b). Increased
plasma concentrations of interleukin-6, soluble interleukin-6,
soluble interleukin-2 and transferrin receptor in major depression.
Journal of Affective Disorders 34, 301309.
Maier, S. F. & Watkins, L. R. (1995). Intracerebroventricular
intereukin-1 receptor antagonist blocks the enhancement of fear
conditioning and interference with escape learning produced by
inescabable shock. Brain Research 695, 279286.
Maier, S. F. & Watkins, L. R. (1998). Cytokines for psychologists :
implications of bi-directional immune-to-brain communication for
understanding behavior, mood, and cognition. Psychological
Review 105, 83107.
Maier, S. F. & Watkins, L. R. (1999). Bidirectional communication
between the brain and the immune system : implications for
behaviour. Animal Behaviour 57, 741751.
Maier, S. F., Watkins, L. R. & Fleshner, M. (1994). Psychoneuroimmunology : the interface between behaviour, brain, and immunity. American Psychologist 49, 10041017.
Mastorakos, G., Chrousos, G. P. & Weber, J. S. (1993). Recombinant
interleukin-6 activates the hypothalamicpituitaryadrenal axis in
humans. Journal of Clinical Endocrinology and Metabolism 77,
16901694.
Papanicolaou, D. A. (1998). The pathophysiologic roles of
interleukin-6 in human disease. Annals of Internal Medicine 128,
127137.
Renault, N. J. & Hoofnagel, J. H. (1989). Side effects of alpha
interferon. Seminars in Liver Disease 9, 273277.
Rothwell, N. J. & Hopkins, S. J. (1995). Cytokines and the nervous
system: actions and mechanisms of action. Trends in Neuroscience
18, 130136.
Rozlog, L. A., Kiecolt-Glaser, J. K., Marucha, P. T., Sheridan, J. F.
& Glaser, R. (1999). Stress and immunity : implications for viral
disease and would healing. Journal of Periodontology 70, 786792.

Editorial : Acute sickness behaviour


Saper, C. B. & Breder, C. D. (1994). The neurologic basis of fever.
New England Journal of Medicine 330, 18801886.
Shintani, F., Nakaki, T., Kanba, S., Sato, K., Yagi, G., Shiozawa,
M., Aiso, S., Kato, R. & Asai, M. (1995). Involvement of
interleukin-1 in immobilization stress-induced increase in plasma
adrenocorticotropic hormone and in the release of hypothalamic
monoamines in the rat. Journal of Neuroscience 15, 19611970.
Smith, A. P., Tyrrell, D. A. J., Al-Nakib, W., Coyle, K. B., Donovan,
C. B., Higgins, P. G. & Willman, J. S. (1987). Effects of experimentally induced respiratory virus infections and illness on
psychomotor performance. Neuropsychology 18, 144148.
Smith, A., Tyrrell, D., Coyle, K. & Higgins, P. (1988). Effects of
interferon alpha on performance in man : a preliminary report.
Psychopharmacology 96, 414416.

767

Takaki, A., Huang, Q. H., Somogyvari-Vigh, A. & Arimura, A.


(1994). Immobilization stress may increase plasma IL-6 activity via
central and peripheral catecholamines. Neuroimmunomodulation 1,
335342.
Tringali, G., Dello Russo, C., Preziosi, P. & Navarra, P. (2000).
Interleukin-1 in the central nervous system : from physiology to
pathology. TheU rapie 55, 171175.
Vollmer-Conna, U., Wakefield, D., Lloyd, A., Hickie, I., Lemon, J.,
Bird, K. D. & Westbrook, R. F. (1997). Cognitive deficits in
patients suffering from chronic fatigue syndrome, acute infective
illness or depression. British Journal of Psychiatry 171, 377381.
Vollmer-Conna, U., Lloyd, A., Hickie, I. & Wakefield, D. (1998).
Chronic fatigue syndrome : an immunological perspective.
Australian and New Zealand Journal of Psychiatry 32, 523527.

30 Scientiic American, February 2013

Photograph by Tktk Tktk

Rodrigo Quian Quiroga, a native of Argentina, is


professor and head of the Bioengineering Research
Group at the University of Leicester in England. He is
author of the recently published Borges and Memory:
Encounters with the Human Brain (MIT Press, 2012).
Itzhak Fried is a professor of neurosurgery and director
of the Epilepsy Surgery Program at the U.C.L.A. David
Geen School of Medicine. He is also a professor at the
Tel Aviv Sourasky Medical Center and Tel Aviv University.
Christof Koch is professor of cognitive and
behavioral biology at the California Institute of
Technology and chief scientic ocer at the
Allen Institute for Brain Science in Seattle.

NEUROSCIENCE

Brain Cells for

Grandmother
Each concepteach person or thing in our
everyday experiencemay have a set of
corresponding neurons assigned to it
By Rodrigo Quian Quiroga,
Itzhak Fried and Christof Koch

IN BRIEF

For decades neuroscientists have debated how


memories are stored. That debate continues today,
with competing theoriesone of which suggests
that single neurons hold the recollection, say, of your
grandmother or of a famous movie star.

Photograph by Dan Saelinger

The alternative theory asserts that each memory is


distributed across many millions of neurons. A number
of recent experiments during brain surgeries provide
evidence that relatively small sets of neurons in specific regions are involved with the encoding of memories.

At the same time, these small groupings of cells may


represent many instances of one thing; a visual image of Grandmas face or her entire bodyeven a
front and side view or the voice of a Hollywood star
such as Jennifer Aniston.

February 2013, ScientiicAmerican.com 31

The story, of course, is iction. The late neuroscientist Jerry


Lettvin (who, unlike Akakhievitch, was real) told it to a crowd of
students at the Massachusetts Institute of Technology in 1969 to
illustrate the provocative idea that as few as about 18,000 neurons could form the basis of any particular conscious experience,
thought or memory of a relative or any other person or object we
might come across. Lettvin never proved or disproved his audacious hypothesis, and for more than 40 years scientists have debated, mostly in jest, the idea of grandmother cells.
The idea of neurons that store memories in such a highly
speciic manner goes all the way back to William James, who in
the late 19th century conceived of pontiicial cells to which
our consciousness is attached. The existence of these cells,
though, runs counter to the dominant view that the perception
of any speciic individual or object is accomplished by the collective activity of many millions if not billions of nerve cells,
what Nobel laureate Charles Sherrington in 1940 called a millionfold democracy. In this case, the activity of any one individual nerve cell is meaningless. Only the collaboration of very
large populations of neurons creates meaning.
Neuroscientists continue to argue about whether it takes
relatively few neuronson the order of thousands or lessto
serve as repositories for a particular concept or whether it takes
hundreds of millions distributed widely throughout the brain.
Attempts to resolve this dispute are leading to new understanding of the workings of memory and conscious thoughtwith a
little help from Hollywood.
JENNIFER ANISTON NEURONS

some years agotogether with Gabriel Kreiman, now a faculty


member at Harvard Medical School, and Leila Reddy, now a researcher at the Brain and Cognition Research Center in Toulouse,
Francewe performed experiments that led to the discovery of a
neuron in the hippocampus of one patient, a brain region known
to be involved in memory processes, that responded very strongly
to diferent photographs of actress Jennifer Aniston but not to
dozens of other actors, celebrities, places and animals. In another

32 Scientiic American, February 2013

patient, a neuron in the hippocampus lit


up at the sight of pictures of actress Halle
Berry and even to her name written on the
computer screen but responded to nothing else. Another neuron ired selectively
to pictures of Oprah Winfrey and to her
name written on the screen and spoken by
a computer-synthesized voice. Yet another
ired to pictures of Luke Skywalker and to
his written and spoken name, and so on.
This kind of observation is made possible by the direct recording of the activity of individual neurons. Other more common techniques, such as functional brain imaging, can pinpoint activity throughout the brain when a volunteer performs
a given task. Yet although functional imaging can track the
overall power consumption of typically a few million cells, it
cannot identify small groups of neurons, let alone individual
cells. To record the electrical pulses emitted by individual neurons, microelectrodes thinner than a human hair need to be
implanted in the brain. This technique is used less commonly
than functional imaging, and only special medical circumstances warrant implantation of these electrodes in humans.
One of those rare circumstances occurs when treating patients with epilepsy. When seizures cannot be controlled with
medication, these patients may be candidates for remedial surgery. The medical team examines clinical evidence that can
pinpoint the location of the area where seizures start, the epileptic focus, which can potentially be surgically removed to
cure the patient. Initially this evaluation involves noninvasive
procedures, such as brain imaging, consideration of clinical evidence and the study of pathological electrical activitya multitude of epileptic discharges that all occur in lockstepwith
EEG recordings made from the patients scalp. But when it is
not possible to accurately determine the location of the epileptic focus with these methods, neurosurgeons may implant electrodes deep inside the skull to continuously monitor in the
hospital brain activity over several days and then analyze the
seizures observed.
Scientists sometimes ask patients to volunteer for research
studies during the monitoring period, studies in which a variety of cognitive tasks are performed as brain activity is recorded. At the University of California, Los Angeles, we have employed a unique technique to record within the skull using
lexible electrodes with tiny microwires; the technology was developed by one of us (Fried), who heads the Epilepsy Surgery
Program at U.C.L.A. and collaborates with other scientists from

PRECEDING PAGES: DOMINIQUE BAYNES (prop styling)

nce a brilliant russian neurosurgeon named


Akakhi Akakhievitch had a patient who wanted
to forget his overbearing, impossible mother.
Eager to oblige, Akakhievitch opened up the
patients brain and, one by one, ablated several
thousand neurons, each of which related to the
concept of his mother. When the patient woke
up from anesthesia, he had lost all notion of his mother. All memories of her, good and bad, were gone. Jubilant with his success,
Akakhievitch turned his attention to the next endeavorthe search
for cells linked to the memory of grandmother.

around the world, including Kochs group at the California Institute of Technology and Quian Quirogas laboratory at the University of Leicester in England. This technique furnishes an extraordinary opportunity to record directly from single neurons for
days at a time in awake patients and provides the ability to study
the iring of neurons during various tasksmonitoring the incessant chattering that occurs while patients look at images on a
laptop, recall memories or perform other tasks. That is how we
discovered the Jennifer Aniston neurons and unwittingly revived the debate ignited by Lettvins parable.
GRANDMOTHER CELLS REVISITED

are nerve cells such as the Jennifer Aniston neuron the longdebated grandmother cells? To answer that question, we have to
be more precise about what we mean by grandmother cells. One
extreme way of thinking about the grandmother cell hypothesis
is that only one neuron responds to one concept. But if we could
ind one single neuron that ired to Jennifer Aniston, it strongly
suggests that there must be morethe chance of inding the one
and only one among billions is minuscule. Moreover, if only a
single neuron would be responsible for a persons entire concept
of Jennifer Aniston, and it were damaged or destroyed by disease or accident, all trace of Jennifer Aniston would disappear
from memory, an extremely unlikely prospect.
A less extreme deinition of
grandmother cells postulates
that many more than a solitary
neuron respond to any one concept. This hypothesis is plausible but very diicult, if not impossible, to prove. We cannot
try every possible concept to
prove that the neuron ires only
to Jennifer Aniston. In fact, the
opposite is often the case: we
often ind neurons that respond
to more than one concept. Thus,
if a neuron ires only to one person during an experiment, we
cannot rule out that it could have also ired to some other stimuli that we did not happen to show.
For example, the day after inding the Jennifer Aniston neuron we repeated the experiment, now using many more pictures
related to her, and found that the neuron also ired to Lisa Kudrow, a costar in the TV series Friends that catapulted both to
fame. The neuron that responded to Luke Skywalker also ired to
Yoda, another Jedi from Star Wars; another neuron ired to two
basketball players; another to one of the authors (Quian Quiroga) of this article and other colleagues who interacted with the
patient at U.C.L.A., and so on. Even then, one can still argue that
these neurons are grandmother cells that are responding to
broader concepts, namely, the two blond women from Friends,
the Jedis from Star Wars, the basketball players, or the scientists
doing experiments with the patient. This expanded deinition
turns the discussion of whether these neurons should be considered grandmother cells into a semantic issue.
Let us leave semantics aside for now and focus instead on a
few critical aspects of these so-called Jennifer Aniston neurons.
First, we found that the responses of each cell are quite selec-

A single neuron
that responded to
Luke Skywalker
and his written
and spoken name
also ired to the
image of Yoda.

tiveeach ires to a small fraction of the pictures of celebrities,


politicians, relatives, landmarks, and so on, presented to the patient. Second, each cell responds to multiple representations of a
particular individual or place, regardless of speciic visual features of the picture used. Indeed, a cell ires in a similar manner
in response to diferent pictures of the same person and even to
his or her written or spoken name. It is as if the neuron in its iring patterns tells us: I know it is Jennifer Aniston, and it does
not matter how you present her to me, whether in a red dress, in
proile, as a written name or even when you call her name out
loud. The neuron, then, seems to respond to the conceptto any
representation of the thing itself. Thus, these neurons may be
more appropriately called concept cells instead of grandmother
cells. Concept cells may sometimes ire to more than one concept, but if they do, these concepts tend to be closely related.
A CODE FOR CONCEPTS

to understand the way a small number of cells become attached to a particular concept such as Jennifer Aniston, it helps
to know something about the brains complex processes for
capturing and storing images of the myriad of objects and people encountered in the world around us. The information taken
in by the eyes irst goesvia the optic nerve leaving the eyeballto the primary visual cortex at the back of the head. Neurons there ire in response to a tiny portion of the minute details that compose an image, as if each were lighting up like a
pixel in a digital image or as if they were the colored dots in a
pointillist painting by Georges Seurat.
One neuron does not suice to tell whether the detail is part
of a face, a cup of tea or the Eifel Tower. Each cell forms part of
an ensemble, a combination that generates a composite image
presented, say, as A Sunday Afternoon on the Island of La
Grande Jatte. If the picture changes slightly, some of the details
will vary, and the iring of the corresponding set of neurons will
change as well.
The brain needs to process sensory information so that it
captures more than a photographit must recognize an object
and integrate it with what is already known. From the primary
visual cortex, the neuronal activation triggered by an image
moves through a series of cortical regions toward more frontal
areas. Individual neurons in these higher visual areas respond
to entire faces or whole objects and not to local details. Just one
of these high-level neurons can tell us that the image is a face
and not the Eifel Tower. If we slightly vary the picture, move it
about or change the lighting illuminating it, it will change some
features, but these neurons do not care much about small differences in detail, and their iring will remain more or less the
samea property known as visual invariance.
Neurons in high-level visual areas send their information to
the medial temporal lobethe hippocampus and surrounding
cortexwhich is involved in memory functions and is where we
found the Jennifer Aniston neurons. The responses of neurons in
the hippocampus are much more speciic than in the higher visual
cortex. Each of these neurons responds to a particular person or,
more precisely, to the concept of that person: not only to the face
and other facets of appearance but also to closely associated attributes such as the persons name.
In our research, we have tried to explore how many individual neurons ire to represent a given concept. We had to ask

February 2013, ScientiicAmerican.com 33

whether it is just one, dozens, thousands or perhaps millions. In


other words, how sparse is the representation of concepts?
Clearly, we cannot measure this number directly, because we
cannot record the activity of all neurons in a given area. Using
statistical methods, Stephen Waydo, at the time a doctoral student with one of us (Koch) at Caltech, estimated that a particular
concept triggers the iring of no more than a million or so neurons, out of about a billion in the medial temporal lobe. But be-

cause we use pictures of things that are very familiar to the patients in our researchwhich tend to trigger more responsesthis
number should be taken strictly as an upper bound; the number
of cells representing a concept may be 10 or 100 times as small,
perhaps close to Lettvins guess of 18,000 neurons per concept.
Contrary to this argument, one reason to think that the brain
does not code concepts sparsely, but rather distributes them
across very large neuronal populations, is that we may not have
enough neurons to represent all possible
concepts and their variations. Do we, for
CONCEPT CELLS
instance, have a big enough store of brain
cells to picture Grandma smiling, weaving, drinking tea or waiting at the bus
stop, as well as the Queen of England
Neuroscientists ardently debate two alternative theories of how memories are encoded
greeting the crowds, Luke Skywalker as a
in the brain. One theory contends that the representation of a single memorythe
child on Tatooine or ighting Darth Vader,
image of Luke Skywalker, for instanceis stored as bits and pieces distributed across
and so on?
millions or perhaps billions of neurons. The alternative view, which has gained more
To answer this question, we should
scientic credibility in recent years, holds that a relatively few neurons, numbering in the
irst consider that, in fact, a typical person
thousands or perhaps even less, constitute a sparse representation of an image. Each of
remembers no more than 10,000 conthose neurons will switch on to the image of Luke, whether from a distance or close-up.
cepts. And this is not a lot in comparison
Some but not all of the same group of neurons will also re to the related image of Yoda.
to the billion nerve cells that make up the
Similarly, a separate set of specic neurons activates when perceiving Jennifer Aniston.
medial temporal lobe. Furthermore, we
have good reason to think that concepts
may be coded and stored very efficiently in
Sparse
Distributed
a sparse way. Neurons in the medial temImage of
poral lobe just do not care about different
Luke Skywalker
instances of the same conceptthey do
not care if Luke is sitting or standing; they
only care if a stimulus has something to do
with Luke. They re to the concept itself
no matter how it is presented. Making the
concept more abstractring to all inMedial temporal lobe
stances of Lukereduces the information
that a neuron needs to encode and allows
Dierent image of
Luke Skywalker
it to become highly selective, responding
to Luke but not to Jennifer.
Simulation studies by Waydo underscore this view even further. Drawing on
a detailed model of visual processing,
Waydo built a software-based neural network that learned to recognize many unlabeled pictures of airplanes, cars, motorImage of
bikes and human faces. The software did
Yoda
so without supervision from a teacher. It
was not told this is a plane and that a
car. It had to gure this out by itself, using the assumption that the immense variety of possible images is in reality based
on a small number of people or things
and that each is represented by a small
Image of
subset of neurons, just as we found in the
Jennifer Aniston
medial temporal lobe. By incorporating
this sparse representation in the software
simulation, the network learned to distinguish the same persons or objects even
when shown in myriad different ways, a
nding similar to our observations from
human brain recordings.

To Code a Memory

34 Scientiic American, February 2013

Illustration by Stephen Savage

WHY CONCEPT CELLS?

our research is closely related to the question of how the brain


interprets the outside world and translates perceptions into
memories. Consider the famous 1953 case of patient H.M., who
sufered from intractable epilepsy. As a desperate approach to
try to stop his seizures, a neurosurgeon removed his hippocam
pus and adjoining regions in both sides of the brain. After the
surgery, H.M. could still recognize people and objects and re
member events that he had known before the surgery, but the
unexpected result was that he could no longer make new long
lasting memories. Without the hippocampus, everything that
happened to him quickly fell into oblivion. The 2000 movie
Memento revolves around a character who has a similar neuro
logical condition.
H.Ms case demonstrates that the hippocampus, and the me
dial temporal lobe in general, is not necessary for perception but
is critical for transferring shortterm memories (things we re
member for a short while) into longterm memories (things re
membered for hours, days or years). In line with this evidence, we
argue that concept cells, which reside in these areas, are critical
for translating what is in our awarenesswhatever is triggered
by sensory inputs or internal recallinto longterm memories
that will later be stored in other areas in the cerebral cortex. We
believe that the Jennifer Aniston neuron we found was not neces
sary for the patient to recognize the actress or to remember who
she was, but it was critical to bring Aniston into awareness for
forging new links and memories related to her, such as later re
membering seeing her picture.
Our brains may use a small number of concept cells to repre
sent many instances of one thing as a unique concepta sparse
and invariant representation. The workings of concept cells go a
long way toward explaining the way we remember: we recall Jen
nifer and Luke in all guises instead of remembering every pore on
their faces. We neither need (nor want) to remember every detail
of whatever happens to us.
What is important is to grasp the gist of particular situations
involving persons and concepts that are relevant to us, rather
than remembering an overwhelming myriad of meaningless de
tails. If we run into somebody we know in a caf, it is more impor
tant to remember a few salient events at this encounter than
what exactly the person was wearing, every single word he used
or what the other strangers relaxing in the caf looked like. Con
cept cells tend to ire to personally relevant things because we
typically remember events involving people and things that are
familiar to us and we do not invest in making memories of things
that have no particular relevance.
Memories are much more than single isolated concepts. A
memory of Jennifer Aniston involves a series of events in which
sheor her character in Friends for that mattertakes part.
The full recollection of a single memory episode requires links
between diferent but associated concepts: Jennifer Aniston
linked to the concept of your sitting on a sofa while spooning
ice cream and watching Friends.
If two concepts are related, some of the neurons encoding
one concept may also ire to the other one. This hypothesis
gives a physiological explanation for how neurons in the brain
encode associations. The tendency for cells to ire to related
concepts may indeed be the basis for the creation of episodic
memories (such as the particular sequence of events during the

caf encounter) or the low of consciousness, moving spontane


ously from one concept to the other. We see Jennifer Aniston,
and this perception evokes the memory of the TV, the sofa and
ice creamrelated concepts that underlie the memory of
watching an episode of Friends. A similar process may also cre
ate the links between aspects of the same concept stored in dif
ferent cortical areas, bringing together the smell, shape, color
and texture of a roseor Jennifers appearance and voice.
Given the obvious advantages of storing highlevel memories
as abstract concepts, we can also ask why the representation of
these concepts has to be sparsely distributed in the medial tem
poral lobe. One answer is provided by modeling studies, which
have consistently shown that sparse representations are neces
sary for creating rapid associations.
The technical details are complex, but the general idea is quite
simple. Imagine a distributedas opposite of sparserepresen
tation for the person we met in the caf, with neurons coding for
each minute feature of that person. Imagine another distributed
representation for the caf itself. Making a connection between
the person and the caf would require creating links among the
diferent details representing each concept but without mixing
them up with others, because the caf looks like a comfortable
bookstore and our friend looks like somebody else we know.
Creating such links with distributed networks is very slow
and leads to the mixing of memories. Establishing such connec
tions with sparse networks is, in contrast, fast and easy. It just re
quires creating a few links between the groups of cells represent
ing each concept, by getting a few neurons to start iring to both
concepts. Another advantage of a sparse representation is that
something new can be added without profoundly afecting every
thing else in the network. This separation is much more diicult
to achieve with distributed networks, where adding a new con
cept shifts boundaries for the entire network.
Concept cells link perception to memory; they give an ab
stract and sparse representation of semantic knowledgethe
people, places, objects, all the meaningful concepts that make
up our individual worlds. They constitute the building blocks
for the memories of facts and events of our lives. Their elegant
coding scheme allows our minds to leave aside countless unim
portant details and extract meaning that can be used to make
new associations and memories. They encode what is critical to
retain from our experiences.
Concept cells are not quite like the grandmother cells that
Lettvin envisioned, but they may be an important physical ba
sis of human cognitive abilities, the hardware components of
thought and memory.
MORE TO ExPlORE

Sparse but Not Grandmother-Cell Coding in the Medial Temporal Lobe. R. Quian
Quiroga, G. Kreiman, C. Koch and I. Fried in Trends in Cognitive Sciences, Vol. 12, No. 3, pages
8791; March 2008.
Percepts to Recollections: Insights from Single Neuron Recordings in the Human
Brain. Nanthia Suthana and Itzhak Fried in Trends in Cognitive Sciences, Vol. 16, No. 8, pages
427436; July 16, 2012.
Concept Cells: The Building Blocks of Declarative Memory Functions. Rodrigo Quian
Quiroga in Nature Reviews Neuroscience, Vol. 13, pages 587597; August 2012.
SCIENTIFIC AMERICAN ONLINE
Read an excerpt of Quian Quirogas book on memory at
ScientiicAmerican.com/feb2013/brain-cells

February 2013, ScientiicAmerican.com 35

Neuron

Review
Experimental and Theoretical Approaches
to Conscious Processing
Stanislas Dehaene1,2,3,4,* and Jean-Pierre Changeux4,5,*
1INSERM,

Cognitive Neuroimaging Unit, Gif sur Yvette, 91191 France


DSV, I2BM, Neurospin center, Gif sur Yvette, 91191 France
3University Paris 11, Orsay 91401, France
4Colle
` ge de France, 11 Place Marcelin Berthelot, 75005 Paris, France
5Institut Pasteur CNRS URA 2182, Institut Pasteur, 75015 Paris, France
*Correspondence: stanislas.dehaene@gmail.com (S.D.), changeux@noos.fr (J.-P.C.)
DOI 10.1016/j.neuron.2011.03.018
2CEA,

Recent experimental studies and theoretical models have begun to address the challenge of establishing
a causal link between subjective conscious experience and measurable neuronal activity. The present
review focuses on the well-delimited issue of how an external or internal piece of information goes
beyond nonconscious processing and gains access to conscious processing, a transition characterized
by the existence of a reportable subjective experience. Converging neuroimaging and neurophysiological
data, acquired during minimal experimental contrasts between conscious and nonconscious processing,
point to objective neural measures of conscious access: late amplification of relevant sensory activity,
long-distance cortico-cortical synchronization at beta and gamma frequencies, and ignition of
a large-scale prefronto-parietal network. We compare these findings to current theoretical models of
conscious processing, including the Global Neuronal Workspace (GNW) model according to which
conscious access occurs when incoming information is made globally available to multiple brain systems
through a network of neurons with long-range axons densely distributed in prefrontal, parieto-temporal,
and cingulate cortices. The clinical implications of these results for general anesthesia, coma, vegetative
state, and schizophrenia are discussed.
Introduction
Understanding the neuronal architectures that give rise to
conscious experience is one of the central unsolved problems
of todays neuroscience, despite its major clinical implications
for general anesthesia, coma, vegetative-state, or minimally
conscious patients. The difficulties are numerous. Notably, the
term consciousness has multiple meanings, most of which
are difficult to precisely define in a manner amenable to experimentation. In this review, we outline recent advances made in
understanding the delimited issue of conscious access: how
does an external or internal piece of information gain access
to conscious processing, defined as a reportable subjective
experience?
We start with a brief overview of the relevant vocabulary and
theoretical concepts. We then examine the experimental studies
that have attempted to delineate the objective physiological
mechanisms of conscious sensory perception by contrasting it
with minimally different, yet nonconscious processing conditions, using a variety of methods: behavior, neuroimaging,
time-resolved electro- and magneto-encephalography, and
finally single-cell electrophysiology and pharmacology. We critically examine how the present evidence fits or argues against
existing models of conscious processing, including the Global
Neuronal Workspace (GNW) model. We end by examining
possible consequences of these advances for pathological brain
states, including general anesthesia, coma, and vegetative
states.

200 Neuron 70, April 28, 2011 2011 Elsevier Inc.

I. Vocabulary and Major Experimental Paradigms


Conscious State versus Conscious Contents
Conscious is an ambiguous word. In its intransitive use (e.g.,
the patient was still conscious), it refers to the state of
consciousness, also called wakefulness or vigilance, which is
thought to vary almost continuously from coma and slow-wave
sleep to full vigilance. In its transitive use (e.g., I was not
conscious of the red light), it refers to conscious access to
and/or conscious processing of a specific piece of information.
The latter meaning is the primary focus of this review. At any
given moment, only a limited amount of information is
consciously accessed and defines the current conscious
content, which is reportable verbally or by an intended gesture.
At the same time, many other processing streams co-occur
but remain nonconscious.
Major Experimental Paradigms
A broad variety of paradigms (reviewed in Kim and Blake, 2005)
are now available to create a minimal contrast between
conscious and nonconscious stimuli (Baars, 1989) and thus
isolate the moment and the physiological properties of
conscious access. A basic distinction is whether the nonconscious stimulus is subliminal or preconscious (Dehaene et al.,
2006; Kanai et al., 2010). A subliminal stimulus is one in which
the bottom-up, stimulus-driven information is so reduced as to
make it undetectable, even with focused attention. A preconscious stimulus, by contrast, is one that is potentially visible (its
energy and duration are such that it could be seen), but which,

Neuron

Review
on a given trial, is not consciously perceived due to temporary
distraction or inattention.
Subliminal presentation is often achieved by masking,
a method whereby the subjective visibility of a stimulus is
reduced or eliminated by the presentation, in close spatial and
temporal contiguity, of other stimuli acting as masks (Breitmeyer, 2006). For instance, a word flashed for 33 ms is visible
when presented in isolation but becomes fully invisible when
preceded and followed by geometrical shapes. Masked stimuli
are frequently used to induce subliminal priming, the facilitation
of the processing of a visible target by the prior presentation of
an identical or related subliminal prime (for review, see Kouider
and Dehaene, 2007). Subliminal presentation can also be
achieved with threshold stimuli, where the contrast or energy
of a stimulus is progressively reduced until its presence is unnoticeable. Binocular rivalry is another common paradigm whereby
the image in one eye becomes subliminal by competition with
a rivaling image presented in the other eye. Participants typically
report temporal alternations in the image that is consciously
perceived. However, a variant of binocular rivalry, the continuous
flash suppression paradigm allows an image to be made permanently invisible by presenting continuously flashing shapes in the
other eye (Tsuchiya and Koch, 2005).
An equally large range of techniques allows for preconscious
presentation. In inattentional blindness, a potentially visible but
unexpected stimulus remains unreported when the participants
attention is focused on another task (Mack and Rock, 1998;
Simons and Ambinder, 2005). The attentional blink (AB) is
a short-term variant of this effect where a brief distraction by
a first stimulus T1 prevents the conscious perception of a second
stimulus T2 briefly presented within a few hundreds of milliseconds of T1 (Raymond et al., 1992). In the related psychological
refractory period (PRP) effect (Pashler, 1994; Welford, 1952),
T2 is unmasked and is therefore eventually perceived and processed, but only after a delay during which it remains nonconscious (Corallo et al., 2008; Marti et al., 2010). The distracting
event T1 can be a surprise event that merely captures attention
(Asplund et al., 2010). The minimum requirement, in order to
induce AB, appears to be that T1 is consciously perceived
(Nieuwenstein et al., 2009). Thus, PRP and AB are closely related
phenomena that point to a serial limit or bottleneck in conscous access (Jolicoeur, 1999; Marti et al., 2010; Wong, 2002)
and can be used to contrast the neural fate of two identical
stimuli, only one of which is consciously perceived (Sergent
et al., 2005).
Objective versus Subjective Criteria for Conscious
Access
How can an experimenter decide whether his experimental
subject was or was not conscious of a stimulus? According to
a long psychophysical tradition, grounded in signal-detection
theory, a stimulus should be accepted as nonconscious only if
subjects are unable to perform above chance on some direct
task of stimulus detection or classification. This strict objective
criterion raises problems, however (Persaud et al., 2007;
Schurger and Sher, 2008). First, it tends to overestimate
conscious perception: there are many conditions in which
subjects perform better than chance, yet still deny perceiving
the stimulus. Second, performance can be at chance level for

some tasks, but not others, raising the issue of which tasks count
as evidence of conscious perception or merely of subliminal processing. Third, the approach requires accepting the null hypothesis of chance-level performance, yet performance never really
falls down to zero, and whether it is significant or not often
depends on arbitrary choices such as the number of trials dedicated to its measurement.
For these reasons, recent alternative approaches emphasize
either pure subjective reports, such as ratings of stimulus visibility
(Sergent and Dehaene, 2004), or second-order commentaries
such as postdecision wagering (e.g., would you bet that your
response was correct?; Persaud et al., 2007). The wagering
method and related confidence judgements provide a high motivation to respond truthfully and in an unbiased manner (Schurger
and Sher, 2008). Furthemore, they can be adapted to nonhuman
subjects (Kiani and Shadlen, 2009; Terrace and Son, 2009).
However, they can sometimes exceed chance level even when
subjects deny seeing the stimulus (Kanai et al., 2010).
Conversely, subjective report is arguably the primary data of
interest in consciousness research. Furthermore, reports of stimulus visibility can be finely quantified, leading to the discovery that
conscious perception can be all-or-none in some paradigms
(Del Cul et al., 2007; Del Cul et al., 2006; Sergent and Dehaene,
2004). Subjective reports also present the advantage of assessing conscious access immediately and on every trial, thus permitting postexperiment sorting of conscious versus nonconscious
trials with identical stimuli (e.g., Del Cul et al., 2007; Lamy et al.,
2009; Pins and Ffytche, 2003; Sergent et al., 2005; Wyart and
Tallon-Baudry, 2008).
Although the debate about optimal measures of conscious
perception continues, it is important to acknowledge that objective assessments, wagering indices and subjective reports are
generally in excellent agreement (Del Cul et al., 2006; Del Cul
et al., 2009; Persaud et al., 2007). For instance, in visual masking,
the conscious perception thresholds derived from objective and
subjective data are essentially identical across subjects (r2 =
0.96, slope z 1) (Del Cul et al., 2006). Those data suggest that
conscious access causes a major change in the global availability of information, whether queried by objective or by subjective means, whose mechanism is the focus of the present review.
Selective Attention versus Conscious Access
Conscious access must be distinguished from the related
concept of attention. William James (1890) provided a wellknown definition of attention as the taking possession by the
mind, in clear and vivid form, of one out of what seem several
simultaneously possible objects or trains of thought. The
problem with this definition is that it conflates two processes
that are now clearly separated in cognitive psychology and cognitive neuroscience (e.g., Huang, 2010; Posner and Dehaene,
1994): selection and access. Selection, also called selective
attention, refers to the separation of relevant versus irrelevant
information, isolation of an object or spatial location, based on
its saliency or relevance to current goals, and amplification of
its sensory attributes. Access refers to its conscious taking
possession of the mindthe subject of the present review.
Empirical evidence indicates that selection can occur without
conscious processing (Koch and Tsuchiya, 2007). For instance,
selective spatial attention can be attracted to the location of

Neuron 70, April 28, 2011 2011 Elsevier Inc. 201

Neuron

Review
a target stimulus that remains invisible (Bressan and Pizzighello,
2008; McCormick, 1997; Robitaille and Jolicoeur, 2006;
Woodman and Luck, 2003). Selective attention can also amplify
the processing of stimuli that remain nonconscious (Kentridge
et al., 2008; Kiefer and Brendel, 2006; Naccache et al., 2002).
Finally, in simple displays with a single target, conscious access
can occur independently of selection (Wyart and Tallon-Baudry,
2008). In cluttered displays, however, selection appears to be
a prerequisite of conscious access: when faced with several
competing stimuli, we need attentional selection in order to
gain conscious access to just one of them (Dehaene and Naccache, 2001; Mack and Rock, 1998). These findings indicate that
selective attention and conscious access are related but dissociable concepts that should be carefully separated, attention
frequently serving as a gateway that regulates which information reaches conscious processing.
II. Experimental Studies of the Brain Mechanisms
of Conscious Access
With this vocabulary at hand, we turn to empirical studies of
conscious access. The simplest experiments consist in presenting a brief sensory stimulus that is sometimes consciously accessible, sometimes not, and using behavior, neuroimaging, and
neurophysiological recording to monitor the depth of its processing and how it differs as a function of conscious reportability.
Experiments Contrasting Visible and Invisible Stimuli
Behavioral evidence. A visual stimulus that is masked and
remains invisible can nevertheless affect behavior and brain
activity at multiple levels (for review, see Kouider and Dehaene,
2007; Van den Bussche et al., 2009b). Subliminal priming has
now been convincingly demonstrated at visual, semantic, and
even motor levels. For instance, when a visible target image is
preceded by a subliminal presentation of the same image, simple
decisions, such as judging whether it refers to an object or
animal, are accelerated compared to when the image is not
repeated. Crucially, this repetition effect resists major changes
in the physical stimulus, such as presenting the same word in
upper case versus lower case (Dehaene et al., 2001) or presenting the same face in two different orientations (Kouider et al.,
2009), suggesting that invariant visual recognition can be
achieved without awareness. At the semantic level, subliminal
extraction of the meaning of words has now been demonstrated
for a variety of word categories (e.g., Gaillard et al., 2006; Naccache and Dehaene, 2001; Van den Bussche et al., 2009a). At
even more advanced levels, a subliminal stimulus can bias motor
responses (Dehaene et al., 1998b; Leuthold and Kopp, 1998).
Subliminal monetary incentives enhance subjects motivation
in a demanding force task, indicating that motivation is modulated by nonconscious signals (Pessiglione et al., 2007). So is
task setting: masked shapes can act as cues for task switching
and lead to detectable changes in task set (Lau and Passingham,
2007). Even inhibitory control can be partially launched nonconsciously, as when a nonconscious stop signal slows down or
interrupts motor responses (van Gaal et al., 2008) (see Figure 1).
The above list suggests that entire chains of specialized
processors can be subject to nonconscious influences. Nevertheless, three potential limits to subliminal processing have
been identified (Dehaene and Naccache, 2001). First, subliminal

202 Neuron 70, April 28, 2011 2011 Elsevier Inc.

priming quickly decreases with processing depth, such that only


small influences are detectable at higher cognitive and decision
levels (Dehaene, 2008; van Gaal et al., 2008). For instance,
a subliminal number can enter into a single numerical operation,
but not a series of two arbitrary operations (Sackur and Dehaene,
2009). Second, subliminal priming decreases with elapsed time,
and therefore typically ceases to be detectable after 500 ms
(Dupoux et al., 2008; Greenwald et al., 1996; Mattler, 2005).
For instance, classical conditioning across a temporal gap only
obtains when participants report being aware of the relations
among the stimuli (Clark et al., 2002) (although see Bekinschtein
et al., 2009b). Third, subliminal stimuli typically fail to yield lasting
and flexible modifications in executive control. Human subjects
generally excel in identifying strategies that exploit virtually any
statistical relation among stimuli, but such strategic control
appears to require consciousness (Posner et al., 1975/2004)
and is not deployed when the stimuli are masked or unattended
and therefore are not consciously detected (Heinemann et al.,
2009; Kinoshita et al., 2008; Merikle and Joordens, 1997; Van
den Bussche et al., 2008). For instance, under conscious conditions, subjects typically slow down after a conflict or error trial
but may not do so when the error or conflict is nonconscious
(Kunde, 2003; Nieuwenhuis et al., 2001) (for two interesting
exceptions, see Logan and Crump, 2010; van Gaal et al., 2010).
Brain-scale neuroimaging. Functional magnetic resonance
imaging (fMRI) can provide a global image of the brain activity
evoked by a visible or invisible stimulus, integrated over a few
seconds. Grill-Spector et al. (2000) first used fMRI to measure
visual activity evoked by masked pictures presented below or
above the visibility threshold. Activation of the primary visual
area V1 was largely unaffected by masking, but the amount of
activation in more anterior regions of lateral occipital and fusiform cortex strong correlated with perceptual reports. A year
later (Dehaene et al., 2001), a similar contrast between masked
and unmasked words, now at the whole-brain level, again revealed a strong correlation of conscious perception with fusiform
activity, but also demonstrated extended areas of activation
uniquely evoked by conscious words, including inferior prefrontal, mesial frontal, and parietal sites (Figure 1). In more recent
fMRI work, using a masking paradigm where conscious reports
followed a characteristic U-shaped curve as a function of the
target-mask delay, fusiform and midline prefrontal and inferior
parietal regions again closely tracked conscious perception
(Haynes et al., 2005b). An important control was recently added:
participants objective performance could be equated while
subjective visibility was manipulated (Lau and Passingham,
2006). In this case, a correlate of visibility could only be detected
in left dorsolateral prefrontal cortex.
Some authors have found correlations of fMRI activation with
visibility of masked versus unmasked stimuli exclusively in
posterior visual areas (e.g., Tse et al., 2005). However, in their
paradigm, even the unmasked stimuli were probably not seen
because they were unattended and irrelevant, which can prevent
conscious access (Dehaene et al., 2006; Kouider et al., 2007;
Mack and Rock, 1998). Overall, fMRI evidence suggests two
convergent correlates of conscious access: (1) amplification of
activity in visual cortex, clearest in higher-visual areas such as
the fusiform gyrus, but possibly including earlier visual areas

Neuron

Review
A

Visible word

Figure 1. fMRI Measures of Conscious


Access

Invisible word

0.3

Visible words
percent signal change

0.2

0.1

Masked
words
0.0
-5

10

15

time (s)
-0.1

Left visual word form area


(-48, -60, -12)

Detected sound

Non-detected sound
Heard
Not heard

Inhibitory control
by visible cue

Inhibitory control
by invisible cue
Visible go/nogo signals

right IFC/anterior insula

Masked go/nogo signals

(e.g., Haynes et al., 2005a; Polonsky et al., 2000; Williams et al.,


2008); (2) emergence of a correlated distributed set of areas,
virtually always including bilateral parietal and prefrontal cortices
(see Figure 1).
Time-resolved imaging methods. Event-related potentials
(ERPs) and magneto-encephalography (MEG) are noninvasive
methods for monitoring at a millisecond scale, respectively, the
electrical and magnetic fields evoked by cortical and subcortical
sources in the human brain. Both techniques have been used to
track the processing of a masked stimulus in time as it crosses or
does not cross the threshold for subjective report. In the 1960s

(A) An early fMRI experiment contrasting the fMRI


activations evoked by brief presentations of words
that were either readable (left) or made invisible by
masking (right) (adapted from Dehaene et al., 2001).
Nonconscious word processing activated the left
occipito-temporal visual word form area, but
conscious perception was characterized by (a) an
intense amplification of activation in relevant
nonconscious processors, here the visual word
form area (left occipito-temporal cortex; see middle
graph); (b) an additional spread of activation to
a distributed, though restricted set of associative
cortices including inferior parietal, prefrontal, and
cingulate areas.
(B) fMRI study of threshold-level noises, approximately half of which were consciously detected
(Sadaghiani et al., 2009). Bilateral auditory areas
showed a nonconscious activation, which was
amplified and spread to distributed inferior parietal,
prefrontal, and cingulate areas (for similar results
with tactile stimuli, see Boly et al., 2007).
(C) fMRI study of inhibitory control by a visible or
invisible cue (van Gaal et al., 2011). Subjects were
presented with masked visual shapes, at the
threshold for conscious perception, some of which
occasionally required inhibiting a response (go/
no-go task). Small activations to the nonconscious
no-go signal were detected in the inferior frontal
and preSMA cortices, but inhibitory control by
a conscious no-go signal was associated with fMRI
signal amplification (see the difference between nogo and go signals in middle graphs), and massive
spread of the activation to additional and more
anterior areas including prefrontal, anterior cingulate, and inferior parietal cortices.

already, ERP studies showed that early


visual activation can be fully preserved
during masking (Schiller and Chorover,
1966). This early finding has been supported by animal electrophysiology
(Bridgeman, 1975, 1988; Kovacs et al.,
1995; Lamme et al., 2002; Rolls et al.,
1999) and by essentially all recent ERP
and MEG studies (Dehaene et al., 2001;
Del Cul et al., 2007; Fahrenfort et al.,
2007; Koivisto et al., 2006, 2009; Lamy
et al., 2009; Melloni et al., 2007; Railo
and Koivisto, 2009; van Aalderen-Smeets
et al., 2006). Evidence from the attentional
blink also confirms that the first 200 ms of
initial visual processing can be fully preserved on trials in which
subjects deny seeing a stimulus (Sergent et al., 2005; Vogel
et al., 1998) (see Figure 2).
In ERPs, the most consistent correlate of visibility appears to
be a late (300500 ms) and broadly distributed positive component called P3 or sometimes P3b (to distinguish it from the focal
anterior P3a, which is thought to reflect automatic attention
attraction and can occur nonconsciously [e.g., Muller-Gass
et al., 2007; Salisbury et al., 1992]). A similarly slow and late
waveform is seen in MEG (van Aalderen-Smeets et al., 2006).
The generators of the P3b ERP have been shown by intracranial

Neuron 70, April 28, 2011 2011 Elsevier Inc. 203

Neuron

Review

Figure 2. Electro- and Magneto-encephalography Measures of Conscious Access


(A) Time course of scalp event-related potentials evoked by an identical visual stimulus, presented during the attentional blink, as a function of whether it was
reported as seen or unseen (Sergent et al., 2005). Early events (P1 and N1) were strictly identical, but the N2 event was amplified and the P3 events (P3a and P3b)
were present essentially only during conscious perception.
(B) Manipulation of visibility by varying the temporal asynchrony between a visual stimulus and a subsequent mask (Del Cul et al., 2007). A nonlinearity, defining
a threshold value for conscious access, was seen in both subjective visibility reports and the P3b event amplitude. Source modeling related this P3b to a sudden
nonlinear ignition, about 300 ms after stimulus presentation, of distributed sources including inferior prefrontal cortex, with a simultaneous reactivation of early
visual areas. Note the two-stage pattern of fusiform activation, with an early linear activation followed by a late nonlinear ignition.
(C) Magneto-encephalography correlates of the attentional blink (Gross et al., 2004). On perceived trials, induced power and phase synchrony increased in the
low beta band (1318 Hz), in a broad network dominated by right inferior parietal and left prefrontal sites.

recordings and ERP-fMRI correlation to involve a highly distributed set of nearly simultaneous active areas including hippocampus and temporal, parietal, and frontal association cortices
(Halgren et al., 1998; Mantini et al., 2009). The P3b has been
reproducibly observed as strongly correlated with subjective
reports, both when varying stimulus parameters (e.g., Del Cul
et al., 2007) and when comparing identical trials with or without
conscious perception (e.g., Babiloni et al., 2006; Del Cul et al.,
2007; Fernandez-Duque et al., 2003; Koivisto et al., 2008;
Lamy et al., 2009; Niedeggen et al., 2001; Pins and Ffytche,
2003; Sergent et al., 2005) (however, this effect may disappear
when the subject already has a conscious working memory
representation of the target: Melloni et al., 2011). The effect is
not easily imputable to increased postperceptual processing or
other task confounds, as many studies equated attention and
response requirements on conscious and nonconscious trials
(e.g., Del Cul et al., 2007; Gaillard et al., 2009; Lamy et al.,
2009; Sergent et al., 2005). For instance, Lamy et al. (2009)

204 Neuron 70, April 28, 2011 2011 Elsevier Inc.

compared correct aware versus correct unaware trials in


a forced-choice localization task on a masked stimulus, thus
equating for stimuli and responses, and again observed a tight
correlation with the P3b component.
Human ERP and MEG recordings also revealed that conscious
perception is also accompanied, during a similar time window, by
increases in the power of high-frequency fluctuations, primarily in
the gamma band (>30 Hz), as well as their phase synchronization
across distant cortical sites (Doesburg et al., 2009; Melloni et al.,
2007; Rodriguez et al., 1999; Schurger et al., 2006; Wyart and
Tallon-Baudry, 2009). In lower frequencies belonging to the alpha
and low beta bands (1020 Hz), the data are more ambiguous, as
both power increases (Gross et al., 2004) and decreases (Gaillard
et al., 2009; Wyart and Tallon-Baudry, 2009) have been reported,
perhaps due to paradigm-dependent variability in the deployment of dorsal parietal attention networks associated with
decreases in alpha-band power (Sadaghiani et al., 2010). Even
when power decreases in these low frequencies, however, their

Neuron

Review
A Late event-related

B Late gamma-band

C Beta phase

D Long-distance

potentials

power

synchrony

causality

1.4

100

Freq. (Hz)

invisible

80

0.7

60
0

40
-0.7

20
-1.4

-200 0 200 400 600 ms

100

1.4

Freq. (Hz)

visible

80

0.7

60
0

40
-0.7

20
0.5

-1.4

-200 0 200 400 600 ms

Spatial distribution of late


high-gamma
high
gamma power on visible trials
als

Time-frequency map of phase


synchrony on visible trials

80

300

200

visible

100

Time course of occipital to


frontal causal gain (%)
0.016

100

400

Freq. (Hz)

Voltage Power

Time course of Voltage power


over frontal electrodes

0.01

100

300

invisible
500
700 ms

0.03
0.02

60

visible

0.003

0.01

40
-0.003

20
0

0.04

-0.01

-200 0 200 400 600 ms

0
-0.01
-500

invisible
0

500

ms

Figure 3. Intracranial Potentials during Conscious Access


Intracranial local-field potentials were recorded during stimulation with masked or unmasked words from a total of ten patients implanted with deep intracortical
electrodes (Gaillard et al., 2009). Four intracranial signatures of conscious access were identified.
(A) Although invisible words elicited event-related potentials, mostly early (<300 ms) and at posterior sites, only visible words elicited massive and durable
voltages in a late time window, particularly from the few available frontal electrodes.
(B) Gamma-band power increases were detectable for invisible words, but in a late time window (>300 ms) gamma power was massively amplified when the
words were visible, particularly in the high-gamma range (50100 Hz). Reduced power was seen in the alpha and lower beta bands.
(C) Phase synchrony increased for invisible words in a late time window (300500 ms) in the beta frequency range (1330 Hz).
(D) Causal relations across distant electrodes, assessed by Granger causality gain due to word presence, increased massively during the same time window. The
bottom row shows causal gain for a particular electrode pair as a function of time. Increases were bidirectional but dominant in the bottom-up direction (e.g.,
occipital-to-frontal), compatible with the idea of posterior information accessing more anterior sites. All time scales are relative to stimulus onset.

long-distance phase synchrony is consistently increased during


conscious perception (Gaillard et al., 2009; Gross et al., 2004;
see also Hipp et al., 2011). The globally distributed character of
these power and synchrony increases seems essential, because
recent results indicate that localized increases in these parameters can be evoked by nonconscious stimuli, particularly during
the first 200 ms of stimulus processing (Fisch et al., 2009; Gaillard
et al., 2009; Melloni et al., 2007). Thus, short-lived focal increases
in gamma-band power are not unique to conscious states but
track activation of both conscious and nonconscious local
cortical circuits (Ray and Maunsell, 2010). However, their significant enhancement on consciously perceived trials, turning into
an all-or-none pattern after 200 ms, appears as a potentially
more specific marker of conscious access (Fisch et al., 2009;
Gaillard et al., 2009).
The high spatial precision and signal-to-noise ratio afforded by
intracranial recording in epileptic patients provides essential

data on this point. Gaillard et al. (2009) contrasted the fate of


masked (subliminal) versus unmasked (conscious) words while
recording from a total of 176 local sites using intracortical depth
electrodes in ten epileptic patients. Four objective signatures of
conscious perception were identified (Figure 3): (1) late (>300 ms)
and distributed event-related potentials contacting sites in
prefrontal cortex; (2) large and late (>300 ms) increases in
induced power (indexing local synchrony) in high-gamma
frequencies (50100 Hz), accompanied by a decrease in lowerfrequency power (centered around 10 Hz); (3) increases in
long-distance cortico-cortical synchrony in the beta frequency
band 1330 Hz; (4) increases in causal relations among distant
cortical areas, bidirectionally but more strongly in the bottomup direction (as assessed by Granger causality, a statistical technique that measures whether the time course of signals at one
site can forecast the future evolution of signals at another distant
site). Gaillard et al. (2009) noted that all four signatures coincided

Neuron 70, April 28, 2011 2011 Elsevier Inc. 205

Neuron

Review
Figure 4. Human Single-Cell Recordings
during Conscious Access
Single cells were recorded from the human medial
temporal lobe and hippocampus during presentation of masked pictures, with a variable targetmask delay (Quiroga et al., 2008). The example at
left shows a single cell that fired specifically to
pictures of the World Trade Center, and did so only
on trials when the patient recognized the picture
(dark blue raster plots), not on trials when recognition failed (red raster plots). Graphs at right show
the average firing rate across all neurons. Although
a small transient firing could be seen on unrecognized trials, conscious perception was characterized by a massive and durable amplification of
activity (for complementary results using electrocorticography (ECoG) in human occipito-temporal
areas, see also Fisch et al., 2009).

in the same time window (300500 ms) and suggested that they
might constitute different measures of the same state of distributed ignition of a large cortical network including prefrontal
cortex. Indeed, seen stimuli had a global impact on late evoked
activity virtually anywhere in the cortex: 68.8% of electrode sites,
although selected for clinical purposes, were modulated by the
presence of conscious words (as opposed to 24.4% of sites
for nonconscious words).
Neuronal recordings. A pioneering research program was conducted by Logothetis and collaborators using monkeys trained
to report their perception during binocular rivalry (Leopold and
Logothetis, 1996; Sheinberg and Logothetis, 1997; Wilke et al.,
2006). By recording from V1, V2, V4, MT, MST, IT, and STS
neurons and presenting two rivaling images, only one of which
led to high neural firing, they identified a fraction of cells whose
firing rate increased when their preferred stimuli was perceived,
thus participating in a conscious neuronal assembly. The proportion of such cells increased from about 20% in V1/V2 to 40% in
V4, MT, or MST to as high as 90% in IT and STS. This finding
supports the hypothesis that subjective perception is associated
with distributed cell assemblies whose neurons are denser in
higher associative cortices than in primary and secondary visual
cortices. Surprisingly, fMRI signals correlated quite strongly with
conscious perception during rivalry in area V1 (Haynes and Rees,
2005; Polonsky et al., 2000) and even in the lateral geniculate
nucleus of the thalamus (Haynes et al., 2005a; Wunderlich
et al., 2005). The discrepancy between fMRI and single-cell
recordings was addressed in a recent electrophysiological study
(Maier et al., 2008; see also Wilke et al., 2006): within area V1 of
the same monkeys, fMRI signals and low-frequency (530 Hz)
local field potentials (LFPs) correlated with subjective visibility
while high-frequency (3090 Hz) LFPs and single-cell firing rate
did not. One interpretation of this finding is that V1 neurons
receive additional top-down synaptic signals during conscious
perception compared to nonconscious perception, although
these signals need not be translated into changes in average
firing rate (Maier et al., 2008).

206 Neuron 70, April 28, 2011 2011 Elsevier Inc.

The masking paradigm afforded


a more precise measurement of the
timing of conscious information progression in the visual system. In area V1, multiunit recordings during
both threshold judgments (Supe`r et al., 2001) and masking paradigms (Lamme et al., 2002) identified two successive response
periods. The first period was phasic, was time-locked to stimulus
onset, and reflected objective properties such as stimulus orientation, whether or not they were detectable by the animal. The
second period was associated with a late, slow, and long-lasting
amplification of firing rate, called figure-ground modulation
because it was specific to neurons whose receptive field fell on
the foreground figure part of the stimulus. Crucially, only this
second phase of late amplification correlated tightly with stimulus detectability in awake animals (Lamme et al., 2002; Supe`r
et al., 2001) and vanished under anesthesia (Lamme et al.,
1998). Thus, although different forms of masking can affect
both initial and late neural responses (Macknik and Haglund,
1999; Macknik and Livingstone, 1998), the work of Lamme and
colleagues suggests that it is the late sustained phase that is
most systematically correlated with conscious visibility. A similar
conclusion was reached from earlier recordings in inferotemporal cortex (Kovacs et al., 1995; Rolls et al., 1999) and
frontal eye fields (Thompson and Schall, 1999, 2000).
Only a single study to date has explored single-neuron
responses to seen or unseen stimuli in human cortex (Quiroga
et al., 2008). Pictures followed at a variable delay by a mask
were presented while recording from the antero-medial temporal
lobe in five patients with epilepsy. A very late response was seen,
peaking around 300 ms and extending further in time. This late
firing reflected tightly the persons subjective report, to such an
extent that individual trials reported as seen or unseen could
be categorically distinguished by the neurons firing train (see
Figure 4). Such a late categorical response is consistent with
the hypothesis that conscious access is all-or-none, leading
either to a high degree of reverberation in higher association
cortex (conscious trial) or to a vanishing response (Dehaene
et al., 2003b; Sergent et al., 2005; Sergent and Dehaene, 2004).
Single-cell electrophysiology has also contributed to a better
description of the postulated role of synchrony in conscious

Neuron

Review
perception (Rodriguez et al., 1999; Varela et al., 2001). Within
a single area such as V4, the degree to which single neurons
synchronize with the ongoing fluctuations in local-field potential
is a predictor of stimulus detection (Womelsdorf et al., 2006).
Across distant areas such as FEF and V4 (Gregoriou et al.,
2009) or PFC and LIP (Buschman and Miller, 2007), synchrony
is enhanced when the stimulus in the receptive field is attended
and is thus presumably accessed consciously. Consistent with
human MEG and intracranial studies (e.g., Gaillard et al., 2009;
Gross et al., 2004), synchronization involves both gamma and
beta bands, the latter being particularly enhanced during topdown attention (Buschman and Miller, 2007). During the late
phase of attention-driven activity, causal relations between
distant areas are durably enhanced in both directions, but
more strongly so in the bottom-up direction from V4 to FEF (Gregoriou et al., 2009), again similar to human findings (Gaillard
et al., 2009) and compatible with the idea that sensory information needs to be propagated anteriorily, particularly to PFC,
before becoming consciously reportable.
Experiments with Perceived and Unperceived Stimuli
outside the Visual Modality
Although vision remains the dominant paradigm, remarkably
similar signatures of conscious access have been obtained in
other sensory or motor modalities (see Figure 1).
In the tactile modality, threshold-level stimuli were studied
both in humans with fMRI and magneto-encephalography
(Boly et al., 2007; Jones et al., 2007) and in awake monkeys
with single-cell electrophysiology (de Lafuente and Romo,
2005, 2006). In the monkey, the early activity of neurons in the
primary somatosensory area S1 was identical on detected and
undetected trials, but within 180 ms the activation expanded
into parietal and medial frontal cortices (MFC) where it showed
a large difference predictive of behavioral reports (high activation
on detected trials and low activity on undetected trials, even for
constant stimuli). In humans, a similar two-phase pattern was
identified within area S1 (Jones et al., 2007). According to the
authors, modeling of these S1 potentials required the postulation
of a late top-down input from unknown distant areas to supragranular and granular layers, specific to detected stimuli. Thus,
as in the visual modality (Del Cul et al., 2007; Supe`r et al.,
2001), tactile cortices may be mobilized into a conscious
assembly only during a later phase of top-down amplification,
synchronous to the activation of higher association cortices.
In the auditory modality, similarly, stimuli that are not
consciously detected still trigger considerable sensory processing, including 40 Hz steady-state responses (Gutschalk et al.,
2008) and mismatch negativities (MMN), i.e., electrophysiological
responses that arise primarily from the temporal lobe in response
to rare, deviant, or otherwise unpredictable auditory stimuli (Allen
et al., 2000; Bekinschtein et al., 2009a; Diekhof et al., 2009; Naatanen, 1990). Once again, conscious and nonconscious stimuli
differ in a late (>200 ms) and global P3 wave arising from bilateral
prefronto-parietal generators, with joint enhancement of
temporal auditory cortices (Bekinschtein et al., 2009a; Diekhof
et al., 2009). These localizations are confirmed by an fMRI study
that contrasted detected versus undetected near-threshold noise
bursts (Sadaghiani et al., 2009) (Figure 1). Similarly, an fMRI study
of speech listening at different levels of sedation showed partially

preserved responses in temporal cortices but the total disappearance of activation in the left inferior frontal gyrus during deep
sedation (Davis et al., 2007). A study by Hasson et al. (2007)
further suggests that the content of what we consciously hear
does not depend on early modality-specific responses in auditory
cortex, but rather on late fronto-parietal cross-modal computations. Using the McGurk illusion (perception of a syllable ta
when simultaneously hearing pa and seeing a face saying
ka), they dissociated the objective auditory and visual stimuli
from the subjective percept. Using fMRI repetition suppression,
they then showed that early auditory cortices coded solely for
the objective auditory stimulus, while the perceived subjective
conscious content was reflected in the activation of the left posterior inferior frontal gyrus and anterior inferior parietal lobule. In this
instance, at least, PFC activation could not be attributed to
a generic process of attention, detection, or memory but demonstrably encoded the specific syllable perceived.
Turning to the action domain, several studies have demonstrated that the awareness of ones action, surprisingly, is not
associated with primary or premotor cortices but arises from
a higher-level representation of intentions and their expected
sensory consequences; this representation involves prefrontal
and parietal cortices, notably the angular gyrus (AG) (Desmurget
et al., 2009; Farrer et al., 2008). Using direct cortical stimulation,
Desmurget et al. (2009) observed a double dissociation: premotor stimulation often led to overt movements that the subject was
not aware of performing, while angular gyrus stimulation led to
a subjective perception of movement intention and performance
even in the absence of any detectable muscle activation. In
normal subjects, disrupted sensori-motor feedback has also
been used to define a minimal contrast between subliminal
versus conscious gestures. For instance, when a temporal delay
or a spatial bias was introduced in the visual feedback provided
to participants about their own hand movements, they continuously adjusted their behavior, but these motor adjustments
were only perceived consciously when the disruption exceeded
a certain threshold (Farrer et al., 2008; Slachevsky et al., 2001).
fMRI revealed that this nonlinearity related to a bilateral distributed network involving AG and PFC cortices (Farrer et al., 2008).
Perhaps the clearest evidence for a two-stage process in
action awareness comes from studies of error awareness (Nieuwenhuis et al., 2001). In an antisaccade paradigm, participants
were instructed to move their eyes in the direction opposite to
a visual target. This instruction generated frequent errors, where
the eyes first moved toward the stimulus and then away from it.
Many of these erroneous eye movements remained undetected.
Remarkably, immediately after such undetected errors, a strong
and early (80 ms) ERP component called the error-related
negativity arose from midline frontal cortices (anterior cingulate
or pre-SMA). Only when the error was consciously detected
was this early waveform amplified and followed by a massive
P3-like waveform, which fMRI associated with the expansion
of activation into a broader network including left inferior
frontal/anterior insula activity (Klein et al., 2007).
Convergence with Studies of Inattention and Dual Tasks
The experiments reviewed so far considered primarily subliminal
paradigms where access to conscious reportability was modulated by reducing the incoming sensory information. However,

Neuron 70, April 28, 2011 2011 Elsevier Inc. 207

Neuron

Review
Figure 5. Recruitment of Global FrontoParietal Networks in Effortful Serial Tasks
(A) Simulations of the original global neuronal
workspace proposal before, during, and after
learning of an effortful Stroop-like task (adapted
from Dehaene et al., 1998a). The figure shows the
activity of various processor and workspace units
as a function of time. Workspace units show
strong activation (a) during the search for a taskappropriate configuration of workspace units; (b)
during the effortful execution of a novel task (but
not after its routinization); and (c) after errors, or
whenever higher control is needed.
(BD) Example of corresponding global frontoparietal activations as seen with fMRI. (B) Strong
activation of a distributed network involving PFC
during effortful search for the solution of
a master-mind type problem, with a sudden
collapse as soon as a routine solution is found
(adapted from Landmann et al., 2007). (C) Activation of inferior PFC during dual-task performance which diminishes with training (adapted
from Dux et al., 2009). (D) Activation of a distribution parieto-prefrontal-cingulate network on error
and conflict trials (adapted from the meta-analysis
by Klein et al., 2007).

similar findings arise from preconscious paradigms where withdrawal of attentional selection is used to modulate conscious
access (Dehaene et al., 2006), resulting in either failed (attentional
blink, AB) or delayed (psychological refractory period or PRP)
conscious access. In such states, initial visual processing, indexed by P1 and N1 waves, can be largely or even entirely unaffected (Sergent et al., 2005; Sigman and Dehaene, 2008; Vogel
et al., 1998). However, only perceived stimuli exhibit an amplification of activation in task-related sensory areas (e.g., parahippocampal place area for pictures of places) as well as the unique
emergence of lateral and midline prefrontal and parietal areas
(see also Asplund et al., 2010; Marois et al., 2004; Slagter et al.,
2010; Williams et al., 2008). Temporally resolved fMRI studies
indicate that, during the dual-task bottleneck, PFC activity
evoked by the second task is delayed (Dux et al., 2006; Sigman
and Dehaene, 2008). With electrophysiology, the P3b waveform
again appears as a major correlate of conscious processing
that is both delayed during the PRP (Dellacqua et al., 2005;
Sigman and Dehaene, 2008) and absent during AB (Kranczioch
et al., 2007; Sergent et al., 2005). Seen versus blinked trials are
also distinguished by another marker, the synchronization of
distant frontoparietal areas in the beta band (Gross et al., 2004).
William James (1890) noted how conscious attention and
effort are required for the controlled execution of novel nonroutine sequential tasks but is no longer needed or even detrimental
once routine sets in. Thus, the comparison of effortful versus
automatic tasks provides another contrast that, although not
quite as minimal as the previous ones, should at least provide
signatures of conscious-level processing consistent with other
paradigms. Indeed, a broad network including inferior and
dorsolateral prefrontal, anterior cingulated, and lateral parietal

208 Neuron 70, April 28, 2011 2011 Elsevier Inc.

and intraparietal components is activated


whenever human subjects perform effortful single or dual tasks (Marois and Ivanoff, 2005), and its activation diminishes with training in parallel
to the reduction in behavioral cost (Dux et al., 2009). Strikingly,
it suddenly drops as soon as subjects move into a routine
mode of task execution (Landmann et al., 2007; Procyk et al.,
2000) (Figure 5). On the contrary, focal cortical regions associated with automatized processing of the relevant sensory or
motor attributes remain invariant or may even increase their activation in the course of routinization (e.g., Sigman et al., 2005).
Broad fronto-parietal networks also figure prominently among
the distributed networks of coactive areas that can be isolated
during spontaneous brain activity in the absence of an explicit
task goal (Beckmann et al., 2005; Fox et al., 2006; Greicius
et al., 2003; Mantini et al., 2007; Vincent et al., 2008). How this
activity relates to conscious processing remains debated, since
it can still be observed, to some extent, during sleep (He et al.,
2008), vegetative state (Boly et al., 2009), or sedation in both humans (Greicius et al., 2008) and monkeys (Vincent et al., 2007),
though interestingly with reduced functional connectivity
(Schrouff et al., 2011). To resolve this issue, a direct test consists
in identifying participants with a given spontaneous activity
pattern and asking them whether they were experiencing a particular conscious content (Christoff et al., 2009; Mason et al., 2007).
Such studies reveal a tight correlation between default-mode
network activity and self-reported mind-wandering into
episodic memory and self-oriented thought. Smallwood et al.
(2008) further demonstrated that, during such mind-wandering
periods, the P3 wave evoked by external events is reduced. Overall, these findings indicate that spontaneous activity, like external
goal-driven activity, invades large-scale fronto-parietal networks
and impose a strong limitation on the processing of external
events, with the same signature as the attentional blink.

Neuron

Review
Evaluative
Systems
(VALUE)

Attentional
Systems
(FOCUSING)

Long-Term
Memory
(PAST)

Global
Workspace

Perceptual
systems
(PRESENT)

Motor
systems
(FUTURE)

frontal
sensory

II
III

II
III

Figure 6. Historical Steps in the Development of Models of Conscious Processing


In the Norman and Shallice (1980) model (top left), conscious processing is involved in the supervisory attentional regulation, by prefrontal cortices, of lower-level
sensori-motor chains. According to Baars (1989), conscious access occurs once information gains access to a global workspace (bottom left), which broadcasts
it to many other processors. The global neuronal workspace (GNW) hypothesis (right) proposes that associative perceptual, motor, attention, memory, and value
areas interconnect to form a higher-level unified space where information is broadly shared and broadcasted back to lower-level processors. The GNW is
characterized by its massive connectivity, made possibly by thick layers II/III with large pyramidal cells sending long-distance cortico-cortical axons, particularly
dense in prefrontal cortex (Dehaene et al., 1998a).

In conclusion, human neuroimaging methods and electrophysiological recordings during conscious access, under
a broad variety of paradigms, consistently reveal a late amplification of relevant sensory activity, long-distance cortico-cortical
synchronization at beta and gamma frequencies, and ignition
of a large-scale prefronto-parietal network.
III. Theoretical Modeling of Conscious Access
The above experiments provide a convergent database of
observations. In the present section, we examine which theoretical principles may account for these findings. We briefly survey
the major theories of conscious processing, with the goal to try to
isolate a core set of principles that are common to most theories
and begin to make sense of existing observations. We then
describe in more detail a specific theory, the Global Neuronal
Workspace (GNW), whose simulations coarsely capture the contrasting physiological states underlying nonconscious versus
conscious processing.

Convergence toward a Set of Core Concepts for


Conscious Access
Although consciousness research includes wildly speculative
proposals (Eccles, 1994; Jaynes, 1976; Penrose, 1990), research
of the past decades has led to an increasing degree of convergence toward a set of concepts considered essential in most
theories (for review, see Seth, 2007). Four such concepts can
be isolated.
A supervision system. In the words of William James,
consciousness appears as an organ added for the sake of
steering a nervous system grown too complex to regulate itself
(James, 1890, chapter 5). Posner (Posner and Rothbart, 1998;
Posner and Snyder, 1975) and Shallice (Shallice, 1972, 1988;
Norman and Shallice, 1980) first proposed that information is
conscious when it is represented in an executive attention or
supervisory attentional system that controls the activities of
lower-level sensory-motor routines and is associated with
prefrontal cortex (Figure 6). In other words, a chain of sensory,

Neuron 70, April 28, 2011 2011 Elsevier Inc. 209

Neuron

Review
semantic, and motor processors can unfold without our awareness, as reviewed in the previous section, but conscious perception seems needed for the flexible control of their execution, such
as their onset, termination, inhibition, repetition, or serial chaining.
A serial processing system. Descartes (1648) first observed
that ideas impede each other. Broadbent (1958) theorized
conscious perception as involving access to a limited-capacity
channel where processing is serial, one object at a time. The
attentional blink and psychological refractory period effects
indeed confirm that conscious processing of a first stimulus
renders us temporarily unable to consciously perceive other
stimuli presently shortly thereafter. Several psychological
models now incorporate the idea that initial perceptual processing is parallel and nonconscious and that conscious access is
serial and occurs at the level of a later central bottleneck (Pashler, 1994) or second processing stage of working memory
consolidation (Chun and Potter, 1995).
A coherent assembly formed by re-entrant or top-down loops.
In the context of the maintenance of invariant representations of
the body/world through reafference (von Holst and Mittelstaedt,
1950), Edelman (1987) proposed re-entry as an essential component of the creation of a unified percept: the bidirectional
exchange of signals across parallel cortical maps coding for
different aspects of the same object. More recently, the dynamic
core hypothesis (Tononi and Edelman, 1998) proposes that information encoded by a group of neurons is conscious only if it
achieves not only differentiation (i.e., the isolation of one specific
content out of a vast repertoire of potential internal representations) but also integration (i.e., the formation of a single,
coherent, and unified representation, where the whole carries
more information than each part alone). A notable feature of
the dynamic core hypothesis is the proposal of a quantitative
mathematical measure of information integration called F, high
values of which are achieved only through a hierarchical recurrent connectivity and would be necessary and sufficient to
sustain conscious experience: consciousness is integrated
information (Tononi, 2008). This measure has been shown to
be operative for some conscious/nonconscious distinctions
such as anesthesia (e.g., Lee et al., 2009b; Schrouff et al.,
2011), but it is computationally complicated and, as a result,
has not yet been broadly applied to most of the minimal empirical
contrasts reviewed above.
In related proposals, Crick and Koch (1995, 2003, 2005) suggested that conscious access involves forming a stable global
neural coalition. They initially introduced reverberating gammaband oscillations around 40 Hz as a crucial component, then
proposed an essential role of connections to prefrontal cortex.
Lamme and colleagues (Lamme and Roelfsema, 2000; Supe`r
et al., 2001) produced data strongly suggesting that feedforward
or bottom-up processing alone is not sufficient for conscious
access and that top-down or feedback signals forming recurrent
loops are essential to conscious visual perception. Llinas and
colleagues (Llinas et al., 1998; Llinas and Pare, 1991) have also
argued that consciousness is fundamentally a thalamocortical
closed-loop property in which the ability of cells to be intrinsically
active plays a central role.
A global workspace for information sharing. The theater metaphor (Taine, 1870) compares consciousness to a narrow scene

210 Neuron 70, April 28, 2011 2011 Elsevier Inc.

that allows a single actor to diffuse his message. This view has
been criticized because, at face value, it implies a conscious
homunculus watching the scene, thus leading to infinite regress
(Dennett, 1991). However, capitalizing on the earlier concept of
a blackboard system in artificial intelligence (a common data
structure shared and updated by many specialized modules),
Baars (1989) proposed a homunculus-free psychological model
where the current conscious content is represented within
a distinct mental space called global workspace, with the
capacity to broadcast this information to a set of other processors (Figure 6). Anatomically, Baars speculated that the neural
bases of his global workspace might comprise the ascending
reticular formation of the brain stem and midbrain, the outer shell
of the thalamus and the set of neurons projecting upward
diffusely from the thalamus to the cerebral cortex.
We introduced the Global Neuronal Workspace (GNW) model
as an alternative cortical mechanism capable of integrating the
supervision, limited-capacity, and re-entry properties (Changeux
and Dehaene, 2008; Dehaene and Changeux, 2005; Dehaene
et al., 1998a, 2003b, 2006; Dehaene and Naccache, 2001).
Our proposal is that a subset of cortical pyramidal cells with
long-range excitatory axons, particularly dense in prefrontal,
cingulate, and parietal regions, together with the relevant thalamocortical loops, form a horizontal neuronal workspace interconnecting the multiple specialized, automatic, and nonconscious processors (Figure 6). A conscious content is assumed
to be encoded by the sustained activity of a fraction of GNW
neurons, the rest being inhibited. Through their numerous reciprocal connections, GNW neurons amplify and maintain a specific
neural representation. The long-distance axons of GNW neurons
then broadcast it to many other processors brain-wide. Global
broadcasting allows information to be more efficiently processed
(because it is no longer confined to a subset of nonconscious
circuits but can be flexibly shared by many cortical processors)
and to be verbally reported (because these processors include
those involved in formulating verbal messages). Nonconscious
stimuli can be quickly and efficiently processed along automatized or preinstructed processing routes before quickly decaying
within a few seconds. By contrast, conscious stimuli would be
distinguished by their lack of encapsulation in specialized
processes and their flexible circulation to various processes of
verbal report, evaluation, memory, planning, and intentional
action, many seconds after their disappearance (Baars, 1989;
Dehaene and Naccache, 2001). Dehaene and Naccache (2001)
postulate that this global availability of information (.) is what
we subjectively experience as a conscious state.
Explicit Simulations of Conscious Ignition
The GNW has been implemented as explicit computer simulations of neural networks (Dehaene and Changeux, 2005; Dehaene et al., 1998a, 2003b; see also Zylberberg et al., 2009).
These simulations incorporate spiking neurons and synapses
with detailed membrane, ion channel, and receptor properties,
organized into distinct cortical supragranular, granular, infragranular, and thalamic sectors with reasonable connectivity and
temporal delays. Although the full GNW architecture was not
simulated, four areas were selected and hierarchically interconnected (Figure 7). Bottom-up feed-forward connections linked
each area to the next, while long-distance top-down

Neuron

Review
A

Feed-forward propagation
(subliminal processing)

Reverberating global neuronal workspace


(conscious access)

Feedforward connections
(AMPA)

Feedback connections
(NMDA)

Area D

Area C
Area B1
Area A1
T1

Supra
granular

T2

layer IV
Infra
granular

Thalamocortical
column

Thalamus
neuromodulation

Simulated areas

Propagation with failure of ignition

Global workspace ignition

Time

connections projected to all preceding areas. Moreover, in


a simplifying assumption, bottom-up connections impinged on
glutamate AMPA receptors while the top-down ones, which
are slower, more numerous, and more diffuse, primarily involved
glutamate NMDA receptors (the plausibility of this hypothesis is
discussed further below). In higher areas, inputs competed with
each other through GABAergic inhibitory interneurons, and it
was assumed (though not explicitly simulated) that the winning
representation would be broadcasted by additional longdistance connections to yet other cortical regions.
Initial simulations explored the sequence of activity leading to
conscious access. When sensory stimulation was simulated as

Figure 7. Schematic Representation of the


Hypothesized Events Leading to Conscious
Access According to the GNW Model
(A) Schema illustrating the main postulated differences between subliminal and conscious processing (adapted from Dehaene et al., 2006).
During feed-forward propagation, sensory inputs
progress through a hierarchy of sensory areas in
a feedforward manner, successively contacting
diverse and nonnecessarily compatible representations corresponding to all probabilistic interpretations of the stimuli. Multiple signals converge to
support each others interpretation in higher-level
cortical areas. Higher areas feedback onto lowerlevel sensory representations, favoring a convergence toward a single coherent representation
compatible with current goals. Such a self-connected system exhibits a dynamical threshold: if
the incoming activity carries sufficient weight, it
leads to the ignition of a self-supporting, reverberating, temporary, metastable, and distributed
cell assembly that represents the current
conscious contents and broadcasts it to virtually
all distant sites.
(B) Architecture of an explicit neuronal simulation
model of a small part of the GNW architecture
(adapted from Dehaene and Changeux, 2005;
Dehaene et al., 2003b). The model contains
thalamic and cortical excitatory and inhibitory
neurons, organized in layers with realistic interconnections (inset). Stimuli T1 and T2 can be
presented at the lower level of a hierarchy of four
successive areas, linked by feedforward (AMPA)
nearest-neighbor connections and by global
feedback (NMDA connections).
(C) Simulation of two single trials in which a identical pulse of brief stimulation was applied to
sensory inputs for T1 (Dehaene and Changeux,
2005). Fluctuations in ongoing activity prevented
ignition in the left diagram, resulting in a purely
feedforward propagation dying out in higher-level
areas. In the right diagram, the same stimulus
crossed the threshold for ignition, resulting in selfamplification, a global state of activation, oscillation and synchrony, and a late long-lasting wave of
late activation reaching back to early sensory
areas.

a brief depolarizing current at the lowest


thalamic level, activation propagated according to two successive phases (see
Figure 7): (1) initially, a brief wave of excitation progressed into the simulated
hierarchy through fast AMPA-mediated
feedforward connections, with an amplitude and duration
directly related to the initial input; (2) in a second stage, mediated
by the slower NMDA-mediated feedback connections, the
advancing feed-forward wave amplified its own inputs in
a cascading manner, quickly leading the whole stimulus-relevant
network into a global self-sustained reverberating or ignited
state. This ignition was characterized by an increased power of
local cortico-thalamic oscillations in the gamma band and their
synchrony across areas (Dehaene et al., 2003b). This second
phase of the simulation reproduces most of the empirical signatures of conscious access: late, all-or-none, cortically distributed
potentials involving prefrontal cortex and other high-level

Neuron 70, April 28, 2011 2011 Elsevier Inc. 211

Neuron

Review
associative cortices, with simultaneous increases in highfrequency power and synchrony (e.g., de Lafuente and Romo,
2006; Del Cul et al., 2007; Gaillard et al., 2009).
In GNW simulations, ignition manifests itself, at the cortical
level, as a depolarization of layer II/III apical dendrites of pyramidal dendrites in a subset of activated GNW neurons defining
the conscious contents, the rest being inhibited. In a geometrically accurate model of the pyramidal cell, the summed postsynaptic potentials evoked by long-distance signaling among these
distributed sets of active cells would create slow intracellular
currents traveling from the apical dendrites toward the cells
soma, summing up on the cortical surface as negative slow
cortical potentials (SCPs) over regions coding for the conscious
stimulus (see He and Raichle, 2009). Simultaneously, many other
GNW neurons are strongly suppressed by lateral inhibition via
GABAergic interneurons and define what the current conscious
content is not. As already noted by Rockstroh et al. (1992, p.
175), assuming that many more neurons are inhibited than activated, The surface positivity corresponding to these inhibited
networks would then dominate over the relatively smaller spots
of negativity caused by the reverberating excitation. Thus, the
model can explain why, during conscious access, the resulting
event-related potential is dominated by a positive waveform,
the P3b. This view also predicts that scalp negativities should
appear specifically over areas dense in neurons coding for the
current conscious content. Indeed, in a spatial working memory
task, all stimuli evoke a broad P3b, but when subtracting ERPs
ipsilateral and controlateral to the side of the memorized items,
negative potentials appeared over parietal cortex contralateral
to the memorized locations (Vogel and Machizawa, 2004).
Further GNW simulations showed that ignition could fail to be
triggered under specific conditions, thus leading to simulated
nonconscious states. For very brief or low-amplitude stimuli,
a feedforward wave was seen in the initial thalamic and cortical
stages of the simulation, but it died out without triggering the
late global activation, because it was not able to gather sufficient
self-sustaining reverberant activation (Dehaene and Changeux,
2005). Even at higher stimulus amplitudes, the second global
phase could also be disrupted if another incoming stimulus
had been simultaneously accessed (Dehaene et al., 2003b).
Such a disruption occurs because during ignition, the GNW is
mobilized as a whole, some GNW neurons being active while
the rest is actively inhibited, thus preventing multiple simultaneous ignitions. A strict seriality of conscious access and processing is therefore predicted and has been simulated (Dehaene
and Changeux, 2005; Dehaene et al., 2003b; Zylberberg et al.,
2010). Overall, these simulations capture the two main types of
experimental conditions known to lead to nonconscious processing: subliminal states due to stimulus degradation (e.g.,
masking), and preconscious states due to distraction by a simultaneous task (e.g., attentional blink).
The transition to the ignited state can be described, in theoretical physics terms, as a stochastic phase transitiona sudden
change in neuronal dynamics whose occurrence depends in
part on stimulus characteristics and in part on spontaneous fluctuations in activity (Dehaene and Changeux, 2005; Dehaene
et al., 2003b). In GNW simulations, prestimulus fluctuations in
neural discharges only have a small effect on the early sensory

212 Neuron 70, April 28, 2011 2011 Elsevier Inc.

stage, which largely reflects objective stimulus amplitude and


duration, but they have a large influence on the second slower
stage, which is characterized by NMDA-based reverberating
integration and ultimately leads to a bimodal all-or-none distribution of activity, similar to empirical observations (Quiroga et al.,
2008; Sergent et al., 2005; Sergent and Dehaene, 2004). Due to
these fluctuations, across trials, the very same stimulus does or
does not lead to global ignition, depending in part on the precise
phase of the stimulus relative to ongoing spontaneous activity.
This notion that prestimulus baseline fluctuations partially predict
conscious perception is now backed up by considerable empirical data (e.g., Boly et al., 2007; Palva et al., 2005; Sadaghiani
et al., 2009; Supe`r et al., 2003; Wyart and Tallon-Baudry, 2009).
More generally, these simulations provide a partial neural implementation of the psychophysical framework according to which
conscious access corresponds to a decision based on the
accumulation of stimulus-based evidence, prior knowledge,
and biases (Dehaene, 2008; for specific implementations, see
Lau, 2008, and the mathematical appendix in Del Cul et al., 2009).
Modeling Spontaneous Activity and Serial Goal-Driven
Processing
An original feature of the GNW model, absent from many other
formal neural network models, is the occurence of highly structured spontaneous activity (Dehaene and Changeux, 2005).
Even in the absence of external inputs, the simulated GNW
neurons are assumed to fire spontaneously, in a top-down
manner, starting from the highest hierarchical levels of the simulation and propagating downward to form globally synchronized
ignited states. When the ascending vigilance signal is large,
several such spontaneous ignitions follow each other in a neverending stream and can block ignition by incoming external
stimuli (Dehaene and Changeux, 2005). These simulations
capture some of the empirical observations on inattentional
blindness (Mack and Rock, 1998) and mind wandering (Christoff
et al., 2009; Mason et al., 2007; Smallwood et al., 2008). More
complex network architectures have also been simulated in
which a goal state is set and continuously shapes the structured
patterns of activity that are spontaneously generated, until the
goal is ultimately attained (Dehaene and Changeux, 1997; Zylberberg et al., 2010). In these simulations, ignited states are
stable only for a transient time period and can be quickly destabilized by a negative reward signal that indicates deviation from
the current goal, in which case they are spontaneously and
randomly replaced by another discrete combination of workspace neurons. The dynamics of such networks is thus characterized by a constant flow of individual coherent episodes of variable duration, selected by reward signals in order to achieve
a defined goal state. Architectures based on these notions
have been applied to a variety of tasks (delayed response: Dehaene and Changeux, 1989; Wisconsin card sorting: Dehaene
and Changeux, 1991; Tower of London: Dehaene and
Changeux, 1997; Stroop: Dehaene et al., 1998a), although
a single architecture common to all tasks is not yet in sight (but
see Rougier et al., 2005). As illustrated in Figure 5, they provide
a preliminary account of why GNW networks are spontaneously
active, in a sustained manner, during effortful tasks that require
series of conscious operations, including search, dual-task,
and error processing.

Neuron

Review
In summary, we propose that a core set of theoretical concepts
lie at the confluence of the diverse theories that have been
proposed to account for conscious access: high-level supervision; serial processing; coherent stability through re-entrant
loops; and global information availability. Furthermore, once implemented in the specific neuronal architecture of the GNW
model, these concepts begin to provide a schematic account of
the neurophysiological signatures that, empirically, distinguish
conscious access from nonconscious processing. In particular,
simulations of the GNW architecture can explain the close similarity of the brain activations seen during (1) conscious access to
a single external stimulus; (2) effortful serial processing; and (3)
spontaneous fluctuations in the absence of any stimulus or task.
IV. Present Experimental and Theoretical Challenges
The existing empirical data on conscious access still present
many challenges for theorizing. Indeed, the above theoretical
synthesis may still be refuted if some of its key neural components were found to be implausible or altogether absent in
primate cerebral architecture, or if its predicted patterns of
activity (the late ignition) were found to be unnecessary, artifactual, noncoding, or noncausally related to conscious states.
We consider each of these potential challenges in turn.
Connectivity and Architecture of Long-Distance Cortical
Networks
Pyramidal neurons with long-distance axons. The main anatomical premise of the GNW model is that it consists of a distributed
set of cortical neurons characterized by their ability to receive
from and send back to homologous neurons in other cortical
areas horizontal projections through long-range excitatory axons
mostly originating from the pyramidal cells of layers II and III
(Dehaene et al., 1998a) and more densely distributed in
prefrontal and inferior parietal cortices. Do these units actually
exist? The special morphology of the pyramidal cells from
the cerebral cortex was already noted by Cajal (18991904),
who mentioned their long axons with multiple collaterals and
their very numerous and complex dendrites. Von Economo
(1929) further noted that these large pyramidal cells in layers III
and V are especially abundant in areas spread over the anterior
two-thirds of the frontal lobe, (.) the superior parietal lobule
and the cingulate cortex, among other cortical areas. Recent
investigations have confirmed that long-distance corticocortical and callosal fibers primarily (though not exclusively) arise
from layer II-III pyramids. Furthermore, quantitative analyses of
the dendritic field morphology of layer III pyramidal neurons revealed a continuous increase of complexity of the basal
dendrites from the occipital up to the prefrontal cortex within
a given species (DeFelipe and Farinas, 1992; Elston and Rosa,
1997, 1998) and from lower species (owl monkey, marmoset)
up to humans (Elston, 2003). Layer IV PFC pyramidal neurons
have as many as 16 times more spines in PFC than in V1 and,
as a result, the highly spinous cells in prefrontal areas may integrate many more inputs than cells in areas such as V1, TE, and
7a (Elston, 2000). These observations confirm that PFC cells
exhibit the morphological adaptations needed for massive
long-distance communication, information integration, and
broadcasting postulated in the GNW model and suggest that
this architecture is particularly developed in the human species.

Global brain-scale white matter networks involving PFC. The


GNW model further assumes that long-distance neurons form
brain-scale networks involving prefrontal cortex as a key node.
PFC indeed receives the most diverse set of corticocortical
inputs from areas involved in processing all sensory modalities
(Cavada et al., 2000; Fuster, 2008; Kringelbach and Rolls,
2004; Pandya and Yeterian, 1990; Petrides and Pandya, 2009).
In the monkey cerebral cortex, long-range connections link,
among others, the prefrontal cortex (area 46), the superior
temporal sulcus, parietal area 7a, and the hippocampus together
with the contralateral anterior and posterior cingulum, area 19,
and the parahippocampal gyrus (Goldman-Rakic, 1988). In addition, areas within PFC are multiply interconnected (Barbas and
Pandya, 1989; Preuss and Goldman-Rakic, 1991), and the
superficial layers in PFC are characterized by an abundance of
horizontal intrinsic axon projections that arise from supragranular pyramidal cells (Kritzer and Goldman-Rakic, 1995; Melchitzky
et al., 1998, 2001; Pucak et al., 1996), thus exhibiting the massive
and recurrent interconnectivity needed to sustain GNW ignition.
In humans, the course of cortical tracts can now be confirmed
by diffusion tensor imaging (DTI) and tractography algorithms
(Figure 8), yet with important limitations. Measurements typically
average over relatively large voxels (a few millimeters aside) that
contain a diversity of criss-crossing fibers. Even recent articles
claiming to study the entire connectome (e.g., Hagmann et al.,
2008) suffer for underestimation of the true long-distance
connectivity of areas 46, 6, FEF, and LIP, critical to GNW theory
and known from macaque invasive tracer studies and careful
human anatomical dissections dating from the end of the 19th
century (Dejerine, Meynert, Fleschig). In a still up-to-date
volume, Dejerine (1895) distinguished five main tracts of long
association fibers running deeply in the human white matter.
Consistent with the GNW hypothesis, four of them connect
prefrontal cortex with other cortical areas and are confirmed
by diffusion tensor tractography (Catani and Thiebaut de Schotten, 2008) and by correlation of cortical thickness measures
(Bassett et al., 2008; He et al., 2009). The networks thus identified converge well with those extracted by fMRI intercorrelation
patterns during the resting state or by phase synchrony in the
beta band during either working memory (Bassett et al., 2009)
or attentional blink (Gross et al., 2004).
The importance of long-distance cortical projection pathways
in conscious perception was recently tested in patients at the
very first clinical stage of multiple sclerosis (MS), a neurological
disease characterized by extensive white matter damage
leading to perturbed long-distance connectivity (He et al.,
2009; Reuter et al., 2007; Reuter et al., 2009). As predicted,
MS patients showed abnormal conscious perception of masked
stimuli: they needed a longer target-mask delay before
conscious access occurred. Furthermore, this behavioral
anomaly correlated with structural damage in the dorsolateral
prefrontal white matter and the right occipito-frontal fasciculus
(Figure 8). Importantly, subliminal priming was preserved.
While recent results thus support the existence of massive
long-distance cortical networks involving PFC and their role in
conscious perception, two points should be stressed. First, the
PFC is increasingly being decomposed into multiple specialized
and lateralized subnetworks (e.g., Koechlin et al., 2003; Voytek

Neuron 70, April 28, 2011 2011 Elsevier Inc. 213

Neuron

Review
A

Spatial neglect

Multiple sclerosis
Inferior longitudinal
fasciculus

30

70

90 ms

Dorsolateral white matter


and occipitofrontal fasciculus
magnetization transfer ratio

Neglect

50

30

50

70

90 ms

masking visibility threshold

Figure 8. Role of Long-Distance Connections in Conscious Access


(A) Diffusion-based tracking of human brain connectivity reveals long-distance fiber tracts, both callosal (left) and intrahemispheric (right), forming an anatomical
substrate for the proposed GNW (images courtesy of Michel Thiebaut de Schotten and Flavio DellAcqua).
(B and C) Pathologies of long-distance fiber tracts can be associated with deficits in conscious access. Spatial neglect patients (B) showing perturbed conscious
processing of left-sided stimuli exhibit impaired right-hemispheric communication between occipital and parietal regions and frontal cortex along the inferior
fronto-occipito fasciculus (IFOF), shown in yellow (image courtesy of Michel Thiebaut de Schotten; see Thiebaut de Schotten et al., 2005; Urbanski et al., 2008).
Multiple sclerosis patients (C) in the very first stages of the disease exhibit impairments in the threshold for conscious detection of a masked visual target,
correlating with impaired magnetization transfer, a measure of white matter integrity, in several long-distance fiber tracts (adapted from Reuter et al., 2009).

and Knight, 2010). These findings need not, however, be seen as


contradicting the GNW hypothesis that these subnetworks,
through their tight interconnections, interact so strongly as to
make any information coded in one area quickly available to all
others. Second, in addition to PFC, the nonspecific thalamic
nuclei, the basal ganglia, and some cortical nodes are likely to
contribute to global information broadcasting (Voytek and
Knight, 2010). The precuneus, in particular, may also operate
as a cortical hub with a massive degree of interconnectivity
(Hagmann et al., 2008; Iturria-Medina et al., 2008). This region,
plausibly homologous to the highly connected macaque posteromedial cortex (PMC) (Parvizi et al., 2006), is an aggregate of
convergence-divergence zones (Meyer and Damasio, 2009)

214 Neuron 70, April 28, 2011 2011 Elsevier Inc.

and is tightly connected to PFC area 46 and other workspace


regions (Goldman-Rakic, 1999). In humans, the PMC may play
a critical role in humans in self-referential processing (Cavanna
and Trimble, 2006; Damasio, 1999; Vogt and Laureys, 2005),
thus allowing any conscious content to be integrated into
a subjective first-person perspective.
NMDA receptors and GNW simulations. GNW simulations
assume that long-distance bottom-up connections primarily
impinge on fast glutamate AMPA receptors while top-down
ones primarily concern the slower glutamate NMDA receptor.
This assumption contributes importantly to the temporal
dynamics of the model, particularly the separation between
a fast phasic bottom-up phase and a late sustained integration

Neuron

Review
phase, mimicking experimental observations. It can be criticized
as both receptor types are known to be present in variable
proportions at glutamatergic synapses (for pioneering data on
human receptor distribution, see Amunts et al., 2010). However,
in agreement with the model, physiological recordings suggest
that NMDA antagonists do not interfere with early bottom-up
sensory activity, but only affect later integrative events such as
the mismatch negativity in auditory cortex (Javitt et al., 1996).
Thus, although GNW simulations adopted a highly simplified
anatomical assumption of radically distinct distributions of
NMDA and AMPA, which may have to be qualified in more realistic models, the notion that NMDA receptors contribute
primarily to late, slow, and top-down integrative processes is
plausible (for a related argument, see Wong and Wang, 2006).
Is Conscious Perception Slow and Late?
A strong statement of the proposed theoretical synthesis is that
early bottom-up sensory events, prior to global ignition (<200
300 ms), contribute solely to nonconscious percept construction
and do not systematically distinguish consciously seen from
unseen stimuli. In apparent contradiction with this view, certain
experiments, using both visual (Pins and Ffytche, 2003) or tactile
stimuli (Palva et al., 2005), have observed that the early incoming
wave of sensory-evoked activity (e.g., P1 component) is already
enhanced on conscious compared to nonconscious trials.
Lamme and collaborators (Fahrenfort et al., 2007) found amplification in visual cortex, just posterior to the P1 wave (110
140 ms). More frequently, at around 200300 ms, surrounding
the P2 ERP component, more negative voltages are reported
over posterior cortices on visible compared to invisible trials
(Del Cul et al., 2007; Fahrenfort et al., 2007; Koivisto et al.,
2008, 2009; Railo and Koivisto, 2009; Sergent et al., 2005). Koivisto and collaborators have called this event the visual awareness negativity (VAN).
Several arguments, however, mitigate the possibility that
these early or midlatency differences already reflect conscious
perception. First, they may not be necessary and sufficient, as
they are absent from several experiments (e.g., Lamy et al.,
2009; van Aalderen-Smeets et al., 2006) (although one cannot
exclude that they failed to be detected). Second, and most
crucially, their profile of variation with stimulus variables such
as target-mask delay does not always track the variations in
subjects conscious reports (Del Cul et al., 2007; van AalderenSmeets et al., 2006). Third, they typically consist only in small
modulations that ride on top of early sensory activations that
are still strongly present on nonconscious trials (Del Cul et al.,
2007; Fahrenfort et al., 2007; Sergent et al., 2005). Fourth, in
this respect they resemble the small electrophysiological modulations that have been found to partially predict later perception
even prior to the stimulus (e.g., Boly et al., 2007; Palva et al.,
2005; Sadaghiani et al., 2009; Supe`r et al., 2003; Wyart and Tallon-Baudry, 2009). The timing of these events makes it logically
impossible that they already participate in the neural mechanism
of conscious access. Similar, early differences in sensory activation between conscious and nonconscious trials may reflect fluctuations in prestimulus priors and in sensory evidence that
contribute to subsequent conscious access, rather than be
constitutive of a conscious state per se (Dehaene and Changeux,
2005; Wyart and Tallon-Baudry, 2009).

The evidence on this topic is still evolving, however, as a recent


study found strong correlation of visibility with the P3b component when participants had no expectation of the stimuli, but
a shift to the earlier P2 component when they already had
a working memory representation of the target (Melloni et al.,
2011). This study suggests that the timing of conscious access
may vary with the experimental paradigm and that a Bayesian
perspective, taking into account the subjects prior knowledge
at multiple hierarchical cortical levels (Del Cul et al., 2009; Kiebel
et al., 2008), may be an essential conceptual ingredient that still
needs to be integrated to the above synthesis.
Whether it takes 200 ms, 300 ms, or even more, the slow and
integrative nature of conscious perception is confirmed behaviorally by observations such as the rabbit illusion and its variants (Dennett, 1991; Geldard and Sherrick, 1972; Libet et al.,
1983), where the way in which a stimulus is ultimately perceived
is influenced by poststimulus events arising several hundreds of
milliseconds after the original stimulus. Psychophysical paradigms that rely on quickly alternating stimuli confirm that
conscious perception integrates over 100 ms or more, while
nonconscious perception is comparatively much faster (e.g.,
Forget et al., 2010; Vul and MacLeod, 2006).
Interestingly, recent research also suggests that spontaneous
brain activity, as assessed by resting-state EEG recordings, may
be similarly parsed into a stochastic series of slow microstates, stable for at least 100 ms, each exclusive of the other,
and separated by sharp transitions (Lehmann and Koenig,
1997; Van de Ville et al., 2010). These microstates have recently
been related to some of the fMRI resting-state networks (Britz
et al., 2010). Crucially, they are predictive of the thought contents
reported by participants when they are suddenly interrupted
(Lehmann et al., 1998, 2010). Thus, whether externally induced
or internally generated, the stream of consciousness may
consist in a series of slow, global, and transiently stable cortical
states (Changeux and Michel, 2004).
Can Nonconscious Stimuli Produce a Global Ignition?
Another pillar of the proposed theoretical synthesis is that global
ignition is unique to conscious states. This view would be challenged if some nonconscious stimuli were found to reproducibly
evoke intense PFC activations, P3b waves, or late and distributed patterns of brain-scale synchronization. Taking up this challenge, some studies have indeed reported small but significant
activations of prefrontal regions and a P3-like wave evoked by
infrequent nonconscious stimuli (Brazdil et al., 1998, 2001;
Muller-Gass et al., 2007; Salisbury et al., 1992). However, this
wave is usually a novelty P3a response, with a sharp midline
anterior positivity suggesting focal anterior midline generators,
rather than the global P3 or late positive complex response
evoked by novel stimuli. Similarly, van Gaal et al. (2011) used
fMRI to examine which areas contributed to subliminal versus
conscious processing of no-go signalsrare visual cues that
instructed subjects to refrain from responding on this particular
trial. Their initial observations suggested, provocatively, that
subliminal no-go signals evoked prefrontal potentials corresponding to nonconscious executive processing (van Gaal
et al., 2008). Subsequent fMRI, however, indicated that the
generators of the subliminal response inhibition effect were
restricted to a small set of specialized processors in midline

Neuron 70, April 28, 2011 2011 Elsevier Inc. 215

Neuron

Review
preSMA and the junction of the bilateral anterior insula with the
inferior frontal gyrus. Only conscious no-go signals triggered
a broad and more anterior activation expanding into anterior
cingulate, inferior, and middle frontal gyrus, dorsolateral
prefrontal cortex, and inferior parietal cortexa network fully
compatible with the GNW model (see Figure 1).
Identifying the limits of nonconscious processing remains an
active area of research, as new techniques for presentation of
nonconscious stimuli are constantly appearing (e.g., Arnold
et al., 2008; Wilke et al., 2003). A recent masking study observed
that subliminal task-switching cues evoked detectable activations in premotor, prefrontal, and temporal cortices (Lau and
Passingham, 2007), but with a much reduced amplitude
compared to conscious cues. Another more challenging study
(Diaz and McCarthy, 2007) reported a large network of cortical
perisylvian regions (inferior frontal, inferior temporal, and angular
gyrus) activated by subliminal words relative to subliminal pseudowords, and surprisingly more extended than in previous
reports (e.g., Dehaene et al., 2001). Attentional blink studies
also suggest that unseen words may cause surprisingly longlasting ERP components (N400) (see also Gaillard et al., 2007;
Vogel et al., 1998). A crucial question for future research is
whether these activations remain confined to specialized subcircuits, for instance in the left temporal lobe (Sergent et al., 2005),
or whether they constitute true instances of global cortical processing without consciousness.
Do Prefrontal and Parietal Networks Play a Causal Role
in Conscious Access?
Brain imaging is only correlational in nature, and leaves open the
possibility that distributed ignition involving PFC is a mere
epiphenomenon or a consequence of conscious access, rather
than being one of its necessary causes. Causality is a demanding
concept that can only be assessed by systematic lesion or interference methods, which are of very limited applicability in human
subjects. Nevertheless, one prediction of the GNW model is testable: lesioning or interfering with prefrontal or parietal cortex
activity, at sites quite distant from visual areas, should disrupt
conscious vision. This prediction was initially judged as so counterintuitive as to be immediately refuted by clinical observations,
because frontal lobe patients do not appear to be unconscious
(Pollen, 1999). However, recent evidence actually supports the
GNW account. In normal subjects, transcranial magnetic stimulation (TMS) over either parietal or prefrontal cortex can prevent
conscious perception and even trigger a sudden subjective
disappearance of visual stimulis during prolonged fixation (Kanai
et al., 2008), change blindness (Beck et al., 2006), binocularly
rivalry (Carmel et al., 2010), inattentional blindness (Babiloni
et al., 2007), and attentional blink paradigms (Kihara et al.,
2011). Over prefrontal cortex, bilateral theta-burst TMS leads
to a reduction of subjective visibility with preserved objective
sensori-motor performance (Rounis et al., 2010). We recently
made similar observations in patients with focal prefrontal
lesions (Del Cul et al., 2009): their masking threshold was significantly elevated, in tight correlation with the degree of expansion
of the lesions into left anterior prefrontal cortex, while subliminal
performance on not-seen trials did not differ from normal. In
more severe and diffuse cases, following traumatic brain injury,
bilateral lesions of fronto-parietal cortices or, characteristically,

216 Neuron 70, April 28, 2011 2011 Elsevier Inc.

of the underlying white matter, can cause coma or vegetative


state (Tshibanda et al., 2009). Frontal-lobe patients also suffer
from impaired conscious processing, in such syndromes as
hemineglect, abulia, akinetic mutism, anosognosia, or impaired
autonoetic memory, while they frequently exhibit preserved or
even heightened capacity for automatic action as indexed by
utilization and imitation behaviors (Husain and Kennard, 1996;
Lhermitte, 1983; Passingham, 1993). Indeed, spatial hemineglect, in which conscious access fails for stimuli contralateral
to the lesion, can arise from focal frontal lesions as well as
from impairments of the long-distance fiber tracts linking posterior visual areas with the frontal lobe (Bartolomeo et al., 2007; He
et al., 2007; Thiebaut de Schotten et al., 2005; Urbanski et al.,
2008) (Figure 8).
While suggestive, these observations do not quite suffice to
establish that a frontal contribution is causally necessary for
conscious perception. Arguably, the above effects may not
necessarily indicate a direct or central contribution of PFC to
conscious access, but rather could be mediated by another
brain structure under the influence of PFC or parietal networks,
such as the thalamic nuclei. Also, it is difficult to exclude a contribution of reduced top-down attention or enhanced distractibility
in frontal patients or TMS subjectsalthough some studies have
attempted to control for these factors by equalizing primary
task performance (Rounis et al., 2010) or by demonstrating
a preserved capacity for attentional modulation (Del Cul et al.,
2009). Ultimately, the crucial experiment would involve inducing
a change in the actual conscious content, rather than a mere
elevation of the reportability threshold, by stimulating PFC or
other components of the GNW networks. While we know of no
such experiment yet, microstimulation and optogenetic methods
now make it feasible, at least in nonhuman animals.
Does the Theory Lead to Clinical Applications?
A strong test for any theory of consciousness is whether it can be
clinically used. Conscious access is altered or reduced in three
clinicial situations: schizophrenia, anesthesia, and loss of
consciousness due to coma or vegetative state. Can the
proposed theoretical synthesis shed some light on these issues?
Schizophrenia. Friston and Frith (1995) first hypothesized that
schizophrenia results from a functional disconnection of longdistance prefrontal cortex projection affecting primarily the
strength of N-methyl-D-aspartate receptor (NMDAR)-mediated
synaptic transmission (Niswender and Conn, 2010; see also
Roopun et al., 2008; Stephan et al., 2009). Bullmore et al.
(1997) further suggested a disruption of anatomical connectivity
possibly associated with an aberrant synaptic elimination during
late adolescence and early adulthood (Changeux and Danchin,
1976; McGlashan and Hoffman, 2000), a possibility consistent
with the fact that many potential risk genes are involved in
neuronal and connectivity development (Karlsgodt et al., 2008).
The volume or density of white matter tracks is, indeed, reduced
in a number of regions, including the temporal and prefrontal
lobes, the anterior limb of the internal capsule, and the cingulum
bundle (Lynall et al., 2010; Oh et al., 2009). The cingulate fasciculus disconnection would, secondarily, impair the link to reward
and emotional systems (Holland and Gallagher, 2004), thus
possibly accounting for the known effect of dopaminergic neuroleptics.

Neuron

Review
A

Reduced metabolism during loss of consciousness


Slow-wave sleep
Anesthesia
Vegetative state

Reduced restingstate activity in


vegetative state

Reduced complexity of cortical signals during anesthesia

Figure 9. Cortical Measures of Loss of Consciousness in Sleep, Anesthesia, and Vegetative State
(A) Massive drops in cortical metabolism observed with PET rCBF measurements in slow-wave sleep (Maquet et al., 1997), anesthesia (Kaisti et al., 2002), and
vegetative state (Laureys et al., 2004).
(B) Reduced activity in a resting-state distributed cortical network in three vegetative state patients, as measured by independent component analysis of fMRI
data (adapted from Cauda et al., 2009).
(C) Sudden change in dimensional activation, a nonlinear dynamics measure of EEG complexity, at the precise point of loss of consciousness during anesthesia
(adapted from Velly et al., 2007). Signals were measured from the scalp as well as from the thalamus using depth electrodes (left). Only the scalp (cortical) EEG
showed a dramatic and discontinuous change accompanying loss of consciousness (right).

Schizophrenia thus provides another possible test of the


hypothesis that disruption of PFC long-distance connections
impairs conscious access. Indeed, there is direct evidence for
impaired neural signatures of conscious access, together with
normal subliminal processing, in schizophrenic patients (Dehaene et al., 2003a; Del Cul et al., 2006; Luck et al., 2006). As
in frontal patients, the threshold for conscious access to masked
visual stimuli is elevated in schizophrenia (Del Cul et al., 2006).
The P3b wave is typically delayed and reduced in amplitude, in
both chronic and first-episode schizophrenics (Demiralp et al.,
2002; van der Stelt et al., 2004) and their siblings (Groom et al.,
2008). Frontal slow waves associated with working memory
are similarly impaired (Kayser et al., 2006). Gamma- and betaband power and long-distance phase synchrony are drastically
reduced, even during simple perceptual tasks (Uhlhaas et al.,
2006; Uhlhaas and Singer, 2006). By applying graph-theoretical
tools to MEG recordings, Bassett et al. (2009) observed that activation in the beta and gamma bands failed to organize into longdistance parieto-frontal networks that were cost-efficient, i.e.,

had close to the minimal number of connections needed to


confer a high efficiency of information transmission. In summary,
the neuronal processes of conscious access appear systematically deteriorated in schizophrenia.
Anesthesia. A classical question concerns whether general
anesthetics alter consciousness by binding to molecular target
sites, principally ion channels and ligand-gated ion channels (Forman and Miller, 2011; Li et al., 2010; Nury et al., 2011) present all
over the cortex, in specific and nonspecific thalamic nuclei, or, as
suggested by intracerebral microinjections (Sukhotinsky et al.,
2007), localized to specific sets of brain stem neurons (for review,
see Alkire et al., 2008; Franks, 2008). Anesthetic-induced loss of
consciousness usually coincides with the disruption of activity in
extensive regions of cerebral cortex, particularly the precuneus,
posterior cingulate cortex, cuneus, localized regions of the lateral
frontal and parietal cortices, and occasionally the cerebellum
(Franks, 2008; Kaisti et al., 2002; Schrouff et al., 2011; Veselis
et al., 2004) (Figure 9). Consistent with these views, Velly et al.
(2007) found that during induction of anesthesia by sevofurane

Neuron 70, April 28, 2011 2011 Elsevier Inc. 217

Neuron

Review
and propofol in human patients with Parkinson disease, cortical
EEG complexity decreased dramatically at the precise time where
consciousness was lost, while for several minutes there was little
change in subcortical signals, and eventually a slow decline
(Figure 9). These data suggest that in humans, the early stage of
anesthesia correlates with cortical disruption, and that the effects
on the thalamus are indirectly driven by cortical feedback (Alkire
et al., 2008). Indeed, in the course of anesthesia induction, there
is a decrease in EEG coherence in the 20 to 80 Hz frequency range
between right and left frontal cortices and between frontal and
occipital territories (John and Prichep, 2005). Quantitative analysis
of EEG under propofol induction further indicates a reduction of
mean information integration, as measured by Tononis Phi
measure, around the g-band (40 Hz) and a breakdown of the
spatiotemporal organization of this particular band (Lee et al.,
2009b). In agreement with experiments carried out with rats
(Imas et al., 2005; Imas et al., 2006), quantitative EEG analysis in
humans under propofol anesthesia induction noted a decrease
of directed feedback connectivity with loss of consciousness
and a return with responsiveness to verbal command (Lee et al.,
2009a). Also, during anesthesia induced by the benzodiazepine
midazolam, an externally induced transcranial pulse evoked reliable initial activity monitored by ERPs in humans, but the subsequent late phase of propagation to distributed areas was abolished (Ferrarelli et al., 2010). These observations are consistent
with the postulated role of top-down frontal-posterior amplification in conscious access (see also Supe`r et al., 2001).
Coma and vegetative state. The clinical distinctions between
coma, vegetative state (Laureys, 2005), and minimal consciousness (Giacino, 2005) remain poorly defined, and even fully
conscious but paralyzed patients with locked-in syndrome can
remain undetected. It is therefore of interest to see whether
objective neural measures and GNW theory can help discriminate
them. In coma and vegetative state, as with general anesthesia,
global metabolic activity typically decreases to 50% of normal
levels (Laureys, 2005). This decrease is not homogeneous,
however, but particularly pronounced in GNW areas including
lateral and mesial prefrontal and inferior parietal cortices
(Figure 9). Spontaneous recovery from VS is accompanied by
a functional restoration of this broad frontoparietal network (Laureys et al., 1999) and some of its cortico-thalamo-cortical
connections (Laureys et al., 2000; see also Voss et al., 2006).
Anatomically, prediction of recovery from coma relies on the
comprehensive assessment of all structures involved in arousal
and awareness functions, namely, the ascending reticular activating system located in the postero-superior part of the brainstem and structures encompassing thalamus, basal forebrain,
and fronto-parietal association cortices (Tshibanda et al.,
2009). Lesion or inhibition of part of this system suffices to cause
immediate coma (e.g., Parvizi and Damasio, 2003). Studies on
traumatic coma patients with conventional MRI showed that
lesions of the pons, midbrain, and basal ganglia were predictive
of poor outcome especially when they were bilateral (Tshibanda
et al., 2009). In relation with the GNW model, it is noteworthy that
prediction of nonrecovery after 1 year could be calculated with
up to 86% sensitivity and 97% specificity when taking into
account both diffusion tensor and spectroscopic measures of
cortical white matter integrity (Tshibanda et al., 2009).

218 Neuron 70, April 28, 2011 2011 Elsevier Inc.

The objective neural measures of conscious processing


demonstrated earlier in this review should be applicable to the
difficult clinical problem of detecting consciousness in noncommunicating patients. Using fMRI, a few patients initially classified
as vegetative by clinical signs showed essentially normal activations of distributed long-distance cortical networks during
speech processing and mental imagery tasks (Owen et al.,
2006; Monti et al., 2010), and one patient proved able to voluntarily control them to provide yes/no answers to simple personal
questions, clearly indicating some degree of preserved
conscious processing (Monti et al., 2010). In an effort to isolate
a more theoretically validated scalp signature of conscious
sensory processing, Bekinschtein et al. (2009a) recorded ERPs
to local versus global violations of an auditory regularity. When
hearing a deviant tone after a sequence of repeated standard
tones (sequence XXXXY), a local mismatch response was elicited
nonconsciously even in coma and vegetative-state patients, as
previously demonstrated (e.g., Fischer et al., 2004). However,
when this sequence XXXXY was repeatedly presented, such
that the final tone change could be expected, the presentation
of a deviant monotonic sequence (XXXXX) engendered a P3b
wave in normal subjects that was absent in coma patients and
in most vegetative-state patients but could still be observed in
minimally conscious and locked-in patients. This paradigm,
founded upon previous identification of the P3b component as
a signature of conscious processing, is now undergoing validation as a means of identifying residual conscious processing in
patients (Faugeras et al., 2011).
V. Conclusion and Future Research Directions
The present review was deliberately limited to conscious access.
Several authors argue, however, for additional, higher-order
concepts of consciousness. For Damasio and Meyer (2009),
core consciousness of incoming sensory information requires
integrating it with a sense of self (the specific subjective point of
view of the perceiving organism) to form a representation of how
the organism is modified by the information; extended consciousness occurs when this representation is additionally related to the
memorized past and anticipated future (see also Edelman, 1989).
For Rosenthal (2004), a higher-order thought, coding for the very
fact that the organism is currently representing a piece of information, is needed for that information to be conscious. Indeed, metacognition, or the ability to reflect upon thoughts and draw judgements upon them, is often proposed as a crucial ingredient of
consciousness (Cleeremans et al., 2007; Lau, 2008) (although
see Kanai et al., 2010, for evidence that metacognitive judgements can occur without conscious perception). In humans, as
opposed to other animals, consciousness may also involve the
construction of a verbal narrative of the reasons for our behavior
(Gazzaniga et al., 1977). Although this narrative can be fictitious
(Wegner, 2003), it would be indispensable to interindividual
communication (Bahrami et al., 2010; Frith, 2007).
Metacognition and self-representation have only recently
begun to be studied behaviorally with paradigms simple enough
to extend to nonhuman species (Kiani and Shadlen, 2009; Terrace
and Son, 2009) and to be related to specific brain measurements,
notably anterior prefrontal cortex (Fleming et al., 2010). Thus, our
view is that these concepts, although essential, have not yet

Neuron

Review
received a sufficient empirical and neurophysiological definition to
figure in this review. Following Crick and Koch (1990), we focused
solely here on the simpler and well-studied question of what
neurophysiological mechanisms differentiate conscious access
to some information from nonconscious processing of the same
information. Additional work will be needed to explore, in the
future, these important aspects of higher-order consciousness.
In the present state of investigations, experimental measures
of conscious access identified in this review include: (1) sudden,
all-or-none ignition of prefronto-parietal networks; (2) concomitant all-or-none amplification of sensory activation; (3) a late
global P3b wave in event-related potentials; (4) late amplification
of broad-band power in the gamma range; (5) enhanced longdistance phase synchronization, particularly in the beta range;
and (6) enhanced causal relations between distant areas,
including a significant top-down component. Many of these
measures are also found during complex serial computations
and in spontaneous thought. There is evidence that they rely
on an anatomical network of long-distance connections that is
particularly developed in the human brain. Finally, pathologies
of these networks or their long-distance connections are associated with impairments of conscious access.
In the future, as argued by Haynes (2009), the mapping of
conscious experiences onto neural states will ultimately require
not only a neural distinction between seen and not-seen trials,
but also a proof that the proposed conscious neural state actually encodes all the details of the participants current subjective
experience. Criteria for a genuine one-to-one mapping should
include verifying that the proposed neural state has the same
perceptual stability (for instance over successive eye movements) and suffers from the same occasional illusions as the
subjects own report. Multivariate decoding techniques provide
pertinent tools to address this question and have already been
used to infer conscious mental images from early visual areas
(Haynes and Rees, 2005; Thirion et al., 2006) and from inferotemporal cortex (Schurger et al., 2010; Sterzer et al., 2008).
However, decoding the more intermingled neural patterns expected from PFC and other associative cortices is clearly a challenge for future research (though see Fuentemilla et al., 2010).
Another important question concerns the genetic mechanisms
that, in the coure of biological evolution, have led to the development of the GNW architecture, particularly the relative expansion
of PFC, higher associative cortices, and their underlying longdistance white matter tracts in the course of hominization (see
Avants et al., 2006; Schoenemann et al., 2005; Semendeferi
et al., 2002). Finally, now that measures of conscious processing
have been identified in human adults, it should become possible
to ask how they transpose to lower animal species (Changeux,
2006, 2010) and to human infants and fetuses (Dehaene-Lambertz et al., 2002; Gelskov and Kouider, 2010; Lagercrantz and
Changeux, 2009), in whom genuine but immature long-distance
networks have been described (Fair et al., 2009; Fransson et al.,
2007).
ACKNOWLEDGMENTS
We gratefully acknowledge extensive discussions with Lionel Naccache,
Sid Kouider, Jerome Sackur, Bechir Jarraya, and Pierre-Marie Lledo as

well as commens on previous drafts by Stuart Edelstein, Raphael Gaillard,


Biyu He, Henri Korn, and two expert referees. This work was supported
by Colle`ge de France, INSERM, CNRS, Human Frontiers Science
Program, European Research Council (S.D.), and Skaggs Research Foundation at UCSD School of Pharmacy (J.P.C.).
REFERENCES
Alkire, M.T., Hudetz, A.G., and Tononi, G. (2008). Consciousness and anesthesia. Science 322, 876880.
Allen, J., Kraus, N., and Bradlow, A. (2000). Neural representation of
consciously imperceptible speech sound differences. Percept. Psychophys.
62, 13831393.
Amunts, K., Lenzen, M., Friederici, A.D., Schleicher, A., Morosan, P., Palomero-Gallagher, N., and Zilles, K. (2010). Brocas region: Novel organizational
principles and multiple receptor mapping. PLoS Biol. 8, e1000489.
Arnold, D.H., Law, P., and Wallis, T.S. (2008). Binocular switch suppression: A
new method for persistently rendering the visible invisible. Vision Res. 48,
9941001.
Asplund, C.L., Todd, J.J., Snyder, A.P., and Marois, R. (2010). A central role for
the lateral prefrontal cortex in goal-directed and stimulus-driven attention. Nat.
Neurosci. 13, 507512.
Avants, B.B., Schoenemann, P.T., and Gee, J.C. (2006). Lagrangian frame diffeomorphic image registration: Morphometric comparison of human and
chimpanzee cortex. Med. Image Anal. 10, 397412.
Baars, B.J. (1989). A Cognitive Theory of Consciousness (Cambridge, Mass:
Cambridge University Press).
Babiloni, C., Vecchio, F., Miriello, M., Romani, G.L., and Rossini, P.M. (2006).
Visuo-spatial consciousness and parieto-occipital areas: A high-resolution
EEG study. Cereb. Cortex 16, 3746.
Babiloni, C., Vecchio, F., Rossi, S., De Capua, A., Bartalini, S., Ulivelli, M., and
Rossini, P.M. (2007). Human ventral parietal cortex plays a functional role on
visuospatial attention and primary consciousness. A repetitive transcranial
magnetic stimulation study. Cereb. Cortex 17, 14861492.
Bahrami, B., Olsen, K., Latham, P.E., Roepstorff, A., Rees, G., and Frith, C.D.
(2010). Optimally interacting minds. Science 329, 10811085.
Barbas, H., and Pandya, D.N. (1989). Architecture and intrinsic connections of
the prefrontal cortex in the rhesus monkey. J. Comp. Neurol. 286, 353375.
Bartolomeo, P., Thiebaut de Schotten, M., and Doricchi, F. (2007). Left unilateral neglect as a disconnection syndrome. Cereb. Cortex 17, 24792490.
Bassett, D.S., Bullmore, E., Verchinski, B.A., Mattay, V.S., Weinberger, D.R.,
and Meyer-Lindenberg, A. (2008). Hierarchical organization of human cortical
networks in health and schizophrenia. J. Neurosci. 28, 92399248.
Bassett, D.S., Bullmore, E.T., Meyer-Lindenberg, A., Apud, J.A., Weinberger,
D.R., and Coppola, R. (2009). Cognitive fitness of cost-efficient brain functional
networks. Proc. Natl. Acad. Sci. USA 106, 1174711752.
Beck, D.M., Muggleton, N., Walsh, V., and Lavie, N. (2006). Right parietal
cortex plays a critical role in change blindness. Cereb. Cortex 16, 712717.
Beckmann, C.F., DeLuca, M., Devlin, J.T., and Smith, S.M. (2005). Investigations into resting-state connectivity using independent component analysis.
Philos. Trans. R. Soc. Lond. B Biol. Sci. 360, 10011013.
Bekinschtein, T.A., Dehaene, S., Rohaut, B., Tadel, F., Cohen, L., and Naccache, L. (2009a). Neural signature of the conscious processing of auditory
regularities. Proc. Natl. Acad. Sci. USA 106, 16721677.
Bekinschtein, T.A., Shalom, D.E., Forcato, C., Herrera, M., Coleman, M.R.,
Manes, F.F., and Sigman, M. (2009b). Classical conditioning in the vegetative
and minimally conscious state. Nat. Neurosci. 12, 13431349.
Boly, M., Balteau, E., Schnakers, C., Degueldre, C., Moonen, G., Luxen, A.,
Phillips, C., Peigneux, P., Maquet, P., and Laureys, S. (2007). Baseline brain
activity fluctuations predict somatosensory perception in humans. Proc.
Natl. Acad. Sci. USA 104, 1218712192.

Neuron 70, April 28, 2011 2011 Elsevier Inc. 219

Neuron

Review
Boly, M., Tshibanda, L., Vanhaudenhuyse, A., Noirhomme, Q., Schnakers, C.,
Ledoux, D., Boveroux, P., Garweg, C., Lambermont, B., Phillips, C., et al.
(2009). Functional connectivity in the default network during resting state is
preserved in a vegetative but not in a brain dead patient. Hum. Brain Mapp.
30, 23932400.
Brazdil, M., Rektor, I., Dufek, M., Jurak, P., and Daniel, P. (1998). Effect of
subthreshold target stimuli on event-related potentials. Electroencephalogr.
Clin. Neurophysiol. 107, 6468.
Brazdil, M., Rektor, I., Daniel, P., Dufek, M., and Jurak, P. (2001). Intracerebral
event-related potentials to subthreshold target stimuli. Clin. Neurophysiol.
112, 650661.
Breitmeyer, B. (2006). Visual Masking: Time Slices through Conscious and
Unconscious Vision (New York: Oxford University Press).
Bressan, P., and Pizzighello, S. (2008). The attentional cost of inattentional
blindness. Cognition 106, 370383.
Bridgeman, B. (1975). Correlates of metacontrast in single cells of the cat
visual system. Vision Res. 15, 9199.
Bridgeman, B. (1988). Visual evoked potentials: Concomitants of metacontrast
in late components. Percept. Psychophys. 43, 401403.
Britz, J., Van De Ville, D., and Michel, C.M. (2010). BOLD correlates of EEG
topography reveal rapid resting-state network dynamics. Neuroimage 52,
11621170.
Broadbent, D.E. (1958). Perception and Communication (London: Pergamon).
Bullmore, E.T., Frangou, S., and Murray, R.M. (1997). The dysplastic net
hypothesis: An integration of developmental and dysconnectivity theories of
schizophrenia. Schizophr. Res. 28, 143156.
Buschman, T.J., and Miller, E.K. (2007). Top-down versus bottom-up control
of attention in the prefrontal and posterior parietal cortices. Science 315,
18601862.
Cajal, S.R. (18991904). Cajal on the Cerebral Cortex: An Annotated Translation of the Complete Writings, J. DeFelipe and E.G. Jones, trans. and eds.
(New York: Oxford University Press, 1988).

Christoff, K., Gordon, A.M., Smallwood, J., Smith, R., and Schooler, J.W.
(2009). Experience sampling during fMRI reveals default network and executive system contributions to mind wandering. Proc. Natl. Acad. Sci. USA
106, 87198724.
Chun, M.M., and Potter, M.C. (1995). A two-stage model for multiple target
detection in rapid serial visual presentation. J. Exp. Psychol. Hum. Percept.
Perform. 21, 109127.
Clark, R.E., Manns, J.R., and Squire, L.R. (2002). Classical conditioning,
awareness, and brain systems. Trends Cogn. Sci. (Regul. Ed.) 6, 524531.
Cleeremans, A., Timmermans, B., and Pasquali, A. (2007). Consciousness and
metarepresentation: A computational sketch. Neural Netw. 20, 10321039.
Corallo, G., Sackur, J., Dehaene, S., and Sigman, M. (2008). Limits on introspection: Distorted subjective time during the dual-task bottleneck. Psychol.
Sci. 19, 11101117.
Crick, F., and Koch, C. (1990). Some reflections on visual awareness. Cold
Spring Harb. Symp. Quant. Biol. 55, 953962.
Crick, F., and Koch, C. (1995). Are we aware of neural activity in primary visual
cortex? Nature 375, 121123.
Crick, F., and Koch, C. (2003). A framework for consciousness. Nat. Neurosci.
6, 119126.
Crick, F.C., and Koch, C. (2005). What is the function of the claustrum? Philos.
Trans. R. Soc. Lond. B Biol. Sci. 360, 12711279.
Damasio, A. (1999). The feeling of what happens (New York: Harcourt
Brace & Co.).
Damasio, A., and Meyer, D.E. (2009). Consciousness: An overview of the
phenomenon and of its possible neural basis. In The neurology of consciousness, S. Laureys and G. Tononi, eds. (Amsterdam: Elsevier), pp. 314.
Davis, M.H., Coleman, M.R., Absalom, A.R., Rodd, J.M., Johnsrude, I.S.,
Matta, B.F., Owen, A.M., and Menon, D.K. (2007). Dissociating speech
perception and comprehension at reduced levels of awareness. Proc. Natl.
Acad. Sci. USA 104, 1603216037.

Carmel, D., Walsh, V., Lavie, N., and Rees, G. (2010). Right parietal TMS
shortens dominance durations in binocular rivalry. Curr. Biol. 20, R799R800.

de Lafuente, V., and Romo, R. (2005). Neuronal correlates of subjective


sensory experience. Nat. Neurosci. 8, 16981703.

Catani, M., and Thiebaut de Schotten, M. (2008). A diffusion tensor imaging


tractography atlas for virtual in vivo dissections. Cortex 44, 11051132.

de Lafuente, V., and Romo, R. (2006). Neural correlate of subjective sensory


experience gradually builds up across cortical areas. Proc. Natl. Acad. Sci.
USA 103, 1426614271.

Cauda, F., Micon, B.M., Sacco, K., Duca, S., DAgata, F., Geminiani, G., and
Canavero, S. (2009). Disrupted intrinsic functional connectivity in the vegetative state. J. Neurol. Neurosurg. Psychiatry 80, 429431.
Cavada, C., Company, T., Tejedor, J., Cruz-Rizzolo, R.J., and Reinoso-Suarez,
F. (2000). The anatomical connections of the macaque monkey orbitofrontal
cortex. A review. Cereb. Cortex 10, 220242.
Cavanna, A.E., and Trimble, M.R. (2006). The precuneus: A review of its functional anatomy and behavioural correlates. Brain 129, 564583.
Changeux, J.P. (2006). The Ferrier Lecture 1998. The molecular biology of
consciousness investigated with genetically modified mice. Philos. Trans. R.
Soc. Lond. B Biol. Sci. 361, 22392259.
Changeux, J.P. (2010). Nicotine addiction and nicotinic receptors: Lessons
from genetically modified mice. Nat. Rev. Neurosci. 11, 389401.
Changeux, J.P., and Danchin, A. (1976). Selective stabilisation of developing
synapses as a mechanism for the specification of neuronal networks. Nature
264, 705712.
Changeux, J.P., and Dehaene, S. (2008). The neuronal workspace model:
Conscious processing and learning. In Learning Theory and Behavior.
Volume 1 of Learning and Memory: A Comprehensive Reference, J. Byrne
and R. Menzel, eds. (Oxford: Elsevier), pp. 729758.
Changeux, J.P., and Michel, C.M. (2004). Mechanisms of neural integration at
the brain-scale level. The neuronal workspace and microstate models. In
Microcircuits: The Interface between Neurons and Global Brain Function, S.
Grillner and A.M. Graybiel, eds. (Cambridge, MA: MIT Press), pp. 347370.

220 Neuron 70, April 28, 2011 2011 Elsevier Inc.

DeFelipe, J., and Farinas, I. (1992). The pyramidal neuron of the cerebral
cortex: Morphological and chemical characteristics of the synaptic inputs.
Prog. Neurobiol. 39, 563607.
Dehaene, S. (2008). Conscious and nonconscious processes: Distinct forms of
evidence accumulation? In Better Than Conscious? Decision Making, the
Human Mind, and Implications for Institutions. Strungmann Forum Report,
C. Engel and W. Singer, eds. (Cambridge: MIT Press).
Dehaene, S., and Changeux, J.P. (1989). A simple model of prefrontal cortex
function in delayed-response tasks. J. Cogn. Neurosci. 1, 244261.
Dehaene, S., and Changeux, J.P. (1991). The Wisconsin Card Sorting Test:
Theoretical analysis and modeling in a neuronal network. Cereb. Cortex 1,
6279.
Dehaene, S., and Changeux, J.P. (1997). A hierarchical neuronal network for
planning behavior. Proc. Natl. Acad. Sci. USA 94, 1329313298.
Dehaene, S., and Changeux, J.P. (2005). Ongoing spontaneous activity
controls access to consciousness: A neuronal model for inattentional blindness. PLoS Biol. 3, e141.
Dehaene, S., and Naccache, L. (2001). Towards a cognitive neuroscience of
consciousness: Basic evidence and a workspace framework. Cognition 79,
137.
Dehaene, S., Kerszberg, M., and Changeux, J.P. (1998a). A neuronal model of
a global workspace in effortful cognitive tasks. Proc. Natl. Acad. Sci. USA 95,
1452914534.

Neuron

Review
Dehaene, S., Naccache, L., Le ClecH, G., Koechlin, E., Mueller, M., DehaeneLambertz, G., van de Moortele, P.F., and Le Bihan, D. (1998b). Imaging unconscious semantic priming. Nature 395, 597600.
Dehaene, S., Naccache, L., Cohen, L., Bihan, D.L., Mangin, J.F., Poline, J.B.,
and Rivie`re, D. (2001). Cerebral mechanisms of word masking and unconscious repetition priming. Nat. Neurosci. 4, 752758.
Dehaene, S., Artiges, E., Naccache, L., Martelli, C., Viard, A., Schurhoff, F., Recasens, C., Martinot, M.L., Leboyer, M., and Martinot, J.L. (2003a). Conscious
and subliminal conflicts in normal subjects and patients with schizophrenia:
The role of the anterior cingulate. Proc. Natl. Acad. Sci. USA 100, 13722
13727.
Dehaene, S., Sergent, C., and Changeux, J.P. (2003b). A neuronal network
model linking subjective reports and objective physiological data during
conscious perception. Proc. Natl. Acad. Sci. USA 100, 85208525.
Dehaene, S., Changeux, J.P., Naccache, L., Sackur, J., and Sergent, C. (2006).
Conscious, preconscious, and subliminal processing: A testable taxonomy.
Trends Cogn. Sci. (Regul. Ed.) 10, 204211.
Dehaene-Lambertz, G., Dehaene, S., and Hertz-Pannier, L. (2002). Functional
neuroimaging of speech perception in infants. Science 298, 20132015.
Dejerine, J. (1895). Anatomie des Centres Nerveux, Volume 1 (Paris: Rueff
et Cie).
Del Cul, A., Dehaene, S., and Leboyer, M. (2006). Preserved subliminal processing and impaired conscious access in schizophrenia. Arch. Gen. Psychiatry 63, 13131323.
Del Cul, A., Baillet, S., and Dehaene, S. (2007). Brain dynamics underlying the
nonlinear threshold for access to consciousness. PLoS Biol. 5, e260.
Del Cul, A., Dehaene, S., Reyes, P., Bravo, E., and Slachevsky, A. (2009).
Causal role of prefrontal cortex in the threshold for access to consciousness.
Brain 132, 25312540.
Dellacqua, R., Jolicoeur, P., Vespignani, F., and Toffanin, P. (2005). Central
processing overlap modulates P3 latency. Exp. Brain Res. 165, 5468.
Demiralp, T., Ucok, A., Devrim, M., Isoglu-Alkac, U., Tecer, A., and Polich, J.
(2002). N2 and P3 components of event-related potential in first-episode
schizophrenic patients: Scalp topography, medication, and latency effects.
Psychiatry Res. 111, 167179.

Edelman, G.M. (1989). The Remembered Present (New York: Basic Books).
Elston, G.N. (2000). Pyramidal cells of the frontal lobe: All the more spinous to
think with. J. Neurosci. 20, RC95.
Elston, G.N. (2003). Cortex, cognition and the cell: New insights into the pyramidal neuron and prefrontal function. Cereb. Cortex 13, 11241138.
Elston, G.N., and Rosa, M.G. (1997). The occipitoparietal pathway of the
macaque monkey: Comparison of pyramidal cell morphology in layer III of
functionally related cortical visual areas. Cereb. Cortex 7, 432452.
Elston, G.N., and Rosa, M.G. (1998). Morphological variation of layer III pyramidal neurones in the occipitotemporal pathway of the macaque monkey
visual cortex. Cereb. Cortex 8, 278294.
Fahrenfort, J.J., Scholte, H.S., and Lamme, V.A. (2007). Masking disrupts
reentrant processing in human visual cortex. J. Cogn. Neurosci. 19, 1488
1497.
Fair, D.A., Cohen, A.L., Power, J.D., Dosenbach, N.U., Church, J.A., Miezin,
F.M., Schlaggar, B.L., and Petersen, S.E. (2009). Functional brain networks
develop from a local to distributed organization. PLoS Comput. Biol. 5,
e1000381.
Farrer, C., Frey, S.H., Van Horn, J.D., Tunik, E., Turk, D., Inati, S., and Grafton,
S.T. (2008). The angular gyrus computes action awareness representations.
Cereb. Cortex 18, 254261.
Faugeras, F., Rohaut, B., Weiss, N., Bekinschtein, T., Galanaud, D., Puybasset, L., Bolgert, F., Sergent, C., Cohen, L., Dehaene, S., and Naccache, L.
(2011). Probing consciousness in clinically defined vegetative patients with
event-related potentials. Neurology, in press.
Fernandez-Duque, D., Grossi, G., Thornton, I.M., and Neville, H.J. (2003).
Representation of change: Separate electrophysiological markers of attention,
awareness, and implicit processing. J. Cogn. Neurosci. 15, 491507.
Ferrarelli, F., Massimini, M., Sarasso, S., Casali, A., Riedner, B.A., Angelini, G.,
Tononi, G., and Pearce, R.A. (2010). Breakdown in cortical effective connectivity during midazolam-induced loss of consciousness. Proc. Natl. Acad.
Sci. USA 107, 26812686.
Fisch, L., Privman, E., Ramot, M., Harel, M., Nir, Y., Kipervasser, S., Andelman,
F., Neufeld, M.Y., Kramer, U., Fried, I., and Malach, R. (2009). Neural ignition:
Enhanced activation linked to perceptual awareness in human ventral stream
visual cortex. Neuron 64, 562574.

Dennett, D.C. (1991). Consciousness Explained (London: Penguin).


Descartes, R. (1648). Traite de lhomme. In Descartes: Oeuvres et Lettres
(Paris: Gallimard), 1937.

Fischer, C., Luaute, J., Adeleine, P., and Morlet, D. (2004). Predictive value of
sensory and cognitive evoked potentials for awakening from coma. Neurology
63, 669673.

Desmurget, M., Reilly, K.T., Richard, N., Szathmari, A., Mottolese, C., and
Sirigu, A. (2009). Movement intention after parietal cortex stimulation in humans. Science 324, 811813.

Fleming, S.M., Weil, R.S., Nagy, Z., Dolan, R.J., and Rees, G. (2010). Relating
introspective accuracy to individual differences in brain structure. Science
329, 15411543.

Diaz, M.T., and McCarthy, G. (2007). Unconscious word processing engages


a distributed network of brain regions. J. Cogn. Neurosci. 19, 17681775.

Forget, J., Buiatti, M., and Dehaene, S. (2010). Temporal integration in visual
word recognition. J. Cogn. Neurosci. 22, 10541068.

Diekhof, E.K., Biedermann, F., Ruebsamen, R., and Gruber, O. (2009). Topdown and bottom-up modulation of brain structures involved in auditory
discrimination. Brain Res. 1297, 118123.

Forman, S.A., and Miller, K.W. (2011). Anesthetic sites and allosteric mechanisms of action on Cys-loop ligand-gated ion channels. Can. J. Anaesth. 58,
191205.

Doesburg, S.M., Green, J.J., McDonald, J.J., and Ward, L.M. (2009). Rhythms
of consciousness: Binocular rivalry reveals large-scale oscillatory network
dynamics mediating visual perception. PLoS ONE 4, e6142.

Fox, M.D., Corbetta, M., Snyder, A.Z., Vincent, J.L., and Raichle, M.E. (2006).
Spontaneous neuronal activity distinguishes human dorsal and ventral attention systems. Proc. Natl. Acad. Sci. USA 103, 1004610051.

Dupoux, E., de Gardelle, V., and Kouider, S. (2008). Subliminal speech perception and auditory streaming. Cognition 109, 267273.

Franks, N.P. (2008). General anaesthesia: From molecular targets to neuronal


pathways of sleep and arousal. Nat. Rev. Neurosci. 9, 370386.

Dux, P.E., Ivanoff, J., Asplund, C.L., and Marois, R. (2006). Isolation of a central
bottleneck of information processing with time-resolved FMRI. Neuron 52,
11091120.

Fransson, P., Skiold, B., Horsch, S., Nordell, A., Blennow, M., Lagercrantz, H.,
and Aden, U. (2007). Resting-state networks in the infant brain. Proc. Natl.
Acad. Sci. USA 104, 1553115536.

Dux, P.E., Tombu, M.N., Harrison, S., Rogers, B.P., Tong, F., and Marois, R.
(2009). Training improves multitasking performance by increasing the speed
of information processing in human prefrontal cortex. Neuron 63, 127138.

Friston, K.J., and Frith, C.D. (1995). Schizophrenia: A disconnection


syndrome? Clin. Neurosci. 3, 8997.

Eccles, J.C. (1994). How the Self Controls Its Brain (New York: Springer
Verlag).
Edelman, G. (1987). Neural Darwinism (New York: Basic Books).

Frith, C. (2007). Making up the Mind. How the Brain Creates Our Mental World
(London: Blackwell).
Fuentemilla, L., Penny, W.D., Cashdollar, N., Bunzeck, N., and Duzel, E. (2010).
Theta-coupled periodic replay in working memory. Curr. Biol. 20, 606612.

Neuron 70, April 28, 2011 2011 Elsevier Inc. 221

Neuron

Review
Fuster, J.M. (2008). The Prefrontal Cortex, Fourth Edition (London: Academic
Press).

Haynes, J.D., and Rees, G. (2005). Predicting the stream of consciousness


from activity in human visual cortex. Curr. Biol. 15, 13011307.

Gaillard, R., Del Cul, A., Naccache, L., Vinckier, F., Cohen, L., and Dehaene, S.
(2006). Nonconscious semantic processing of emotional words modulates
conscious access. Proc. Natl. Acad. Sci. USA 103, 75247529.

Haynes, J.D., Deichmann, R., and Rees, G. (2005a). Eye-specific effects of


binocular rivalry in the human lateral geniculate nucleus. Nature 438, 496499.

Gaillard, R., Cohen, L., Adam, C., Clemenceau, S., Hasboun, D., Baulac, M.,
Willer, J.C., Dehaene, S., and Naccache, L. (2007). Subliminal words durably
affect neuronal activity. Neuroreport 18, 15271531.
Gaillard, R., Dehaene, S., Adam, C., Clemenceau, S., Hasboun, D., Baulac, M.,
Cohen, L., and Naccache, L. (2009). Converging intracranial markers of
conscious access. PLoS Biol. 7, e61.
Gazzaniga, M.S., LeDoux, J.E., and Wilson, D.H. (1977). Language, praxis, and
the right hemisphere: Clues to some mechanisms of consciousness.
Neurology 27, 11441147.
Geldard, F.A., and Sherrick, C.E. (1972). The cutaneous rabbit: A perceptual
illusion. Science 178, 178179.
Gelskov, S.V., and Kouider, S. (2010). Psychophysical thresholds of face visibility during infancy. Cognition 114, 285292.
Giacino, J.T. (2005). The minimally conscious state: Defining the borders of
consciousness. Prog. Brain Res. 150, 381395.
Goldman-Rakic, P.S. (1988). Topography of cognition: Parallel distributed
networks in primate association cortex. Annu. Rev. Neurosci. 11, 137156.
Goldman-Rakic, P.S. (1999). The psychic neuron of the cerebral cortex. Ann.
N Y Acad. Sci. 868, 1326.
Greenwald, A.G., Draine, S.C., and Abrams, R.L. (1996). Three cognitive
markers of unconscious semantic activation. Science 273, 16991702.
Gregoriou, G.G., Gotts, S.J., Zhou, H., and Desimone, R. (2009). Highfrequency, long-range coupling between prefrontal and visual cortex during
attention. Science 324, 12071210.
Greicius, M.D., Krasnow, B., Reiss, A.L., and Menon, V. (2003). Functional
connectivity in the resting brain: A network analysis of the default mode
hypothesis. Proc. Natl. Acad. Sci. USA 100, 253258.
Greicius, M.D., Kiviniemi, V., Tervonen, O., Vainionpaa, V., Alahuhta, S., Reiss,
A.L., and Menon, V. (2008). Persistent default-mode network connectivity
during light sedation. Hum. Brain Mapp. 29, 839847.

Haynes, J.D., Driver, J., and Rees, G. (2005b). Visibility reflects dynamic
changes of effective connectivity between V1 and fusiform cortex. Neuron
46, 811821.
He, B.J., and Raichle, M.E. (2009). The fMRI signal, slow cortical potential and
consciousness. Trends Cogn. Sci. (Regul. Ed.) 13, 302309.
He, B.J., Snyder, A.Z., Vincent, J.L., Epstein, A., Shulman, G.L., and Corbetta,
M. (2007). Breakdown of functional connectivity in frontoparietal networks
underlies behavioral deficits in spatial neglect. Neuron 53, 905918.
He, B.J., Snyder, A.Z., Zempel, J.M., Smyth, M.D., and Raichle, M.E. (2008).
Electrophysiological correlates of the brains intrinsic large-scale functional
architecture. Proc. Natl. Acad. Sci. USA 105, 1603916044.
He, Y., Dagher, A., Chen, Z., Charil, A., Zijdenbos, A., Worsley, K., and Evans,
A. (2009). Impaired small-world efficiency in structural cortical networks in
multiple sclerosis associated with white matter lesion load. Brain 132, 3366
3379.
Heinemann, A., Kunde, W., and Kiesel, A. (2009). Context-specific primecongruency effects: On the role of conscious stimulus representations for
cognitive control. Conscious. Cogn. 18, 966976.
Hipp, J.F., Engel, A.K., and Siegel, M. (2011). Oscillatory synchronization in
large-scale cortical networks predicts perception. Neuron 69, 387396.
Holland, P.C., and Gallagher, M. (2004). Amygdala-frontal interactions and
reward expectancy. Curr. Opin. Neurobiol. 14, 148155.
Huang, L. (2010). What is the unit of visual attention? Object for selection, but
Boolean map for access. J. Exp. Psychol. Gen. 139, 162179.
Husain, M., and Kennard, C. (1996). Visual neglect associated with frontal lobe
infarction. J. Neurol. 243, 652657.
Imas, O.A., Ropella, K.M., Ward, B.D., Wood, J.D., and Hudetz, A.G. (2005).
Volatile anesthetics enhance flash-induced gamma oscillations in rat visual
cortex. Anesthesiology 102, 937947.
Imas, O.A., Ropella, K.M., Wood, J.D., and Hudetz, A.G. (2006). Isoflurane
disrupts anterio-posterior phase synchronization of flash-induced field potentials in the rat. Neurosci. Lett. 402, 216221.

Grill-Spector, K., Kushnir, T., Hendler, T., and Malach, R. (2000). The dynamics
of object-selective activation correlate with recognition performance in
humans. Nat. Neurosci. 3, 837843.

Iturria-Medina, Y., Sotero, R.C., Canales-Rodrguez, E.J., Aleman-Gomez, Y.,


and Melie-Garca, L. (2008). Studying the human brain anatomical network via
diffusion-weighted MRI and Graph Theory. Neuroimage 40, 10641076.

Groom, M.J., Bates, A.T., Jackson, G.M., Calton, T.G., Liddle, P.F., and Hollis,
C. (2008). Event-related potentials in adolescents with schizophrenia and their
siblings: A comparison with attention-deficit/hyperactivity disorder. Biol.
Psychiatry 63, 784792.

James, W. (1890). The Principles of Psychology (New York: Holt).

Gross, J., Schmitz, F., Schnitzler, I., Kessler, K., Shapiro, K., Hommel, B., and
Schnitzler, A. (2004). Modulation of long-range neural synchrony reflects
temporal limitations of visual attention in humans. Proc. Natl. Acad. Sci. USA
101, 1305013055.
Gutschalk, A., Micheyl, C., and Oxenham, A.J. (2008). Neural correlates of
auditory perceptual awareness under informational masking. PLoS Biol.
6, e138.
Hagmann, P., Cammoun, L., Gigandet, X., Meuli, R., Honey, C.J., Wedeen,
V.J., and Sporns, O. (2008). Mapping the structural core of human cerebral
cortex. PLoS Biol. 6, e159.
Halgren, E., Marinkovic, K., and Chauvel, P. (1998). Generators of the late
cognitive potentials in auditory and visual oddball tasks. Electroencephalogr.
Clin. Neurophysiol. 106, 156164.

Javitt, D.C., Steinschneider, M., Schroeder, C.E., and Arezzo, J.C. (1996). Role
of cortical N-methyl-D-aspartate receptors in auditory sensory memory and
mismatch negativity generation: Implications for schizophrenia. Proc. Natl.
Acad. Sci. USA 93, 1196211967.
Jaynes, J. (1976). The Origin of Consciousness in the Breakdown of the Bicameral Mind (New York: Houghton Mifflin Company).
John, E.R., and Prichep, L.S. (2005). The anesthetic cascade: A theory of how
anesthesia suppresses consciousness. Anesthesiology 102, 447471.
Jolicoeur, P. (1999). Concurrent response-selection demands modulate the
attentional blink. J. Exp. Psychol. Hum. Percept. Perform. 25, 10971113.
Jones, S.R., Pritchett, D.L., Stufflebeam, S.M., Hamalainen, M., and Moore,
C.I. (2007). Neural correlates of tactile detection: A combined magnetoencephalography and biophysically based computational modeling study. J. Neurosci. 27, 1075110764.

Hasson, U., Skipper, J.I., Nusbaum, H.C., and Small, S.L. (2007). Abstract
coding of audiovisual speech: Beyond sensory representation. Neuron 56,
11161126.

Kaisti, K.K., Metsahonkala, L., Teras, M., Oikonen, V., Aalto, S., Jaaskelainen,
S., Hinkka, S., and Scheinin, H. (2002). Effects of surgical levels of propofol and
sevoflurane anesthesia on cerebral blood flow in healthy subjects studied with
positron emission tomography. Anesthesiology 96, 13581370.

Haynes, J.D. (2009). Decoding visual consciousness from human brain


signals. Trends Cogn. Sci. (Regul. Ed.) 13, 194202.

Kanai, R., Muggleton, N.G., and Walsh, V. (2008). TMS over the intraparietal
sulcus induces perceptual fading. J. Neurophysiol. 100, 33433350.

222 Neuron 70, April 28, 2011 2011 Elsevier Inc.

Neuron

Review
Kanai, R., Walsh, V., and Tseng, C.H. (2010). Subjective discriminability of
invisibility: A framework for distinguishing perceptual and attentional failures
of awareness. Conscious. Cogn. 19, 10451057.

Kringelbach, M.L., and Rolls, E.T. (2004). The functional neuroanatomy of the
human orbitofrontal cortex: Evidence from neuroimaging and neuropsychology. Prog. Neurobiol. 72, 341372.

Karlsgodt, K.H., Sun, D., Jimenez, A.M., Lutkenhoff, E.S., Willhite, R., van Erp,
T.G., and Cannon, T.D. (2008). Developmental disruptions in neural connectivity in the pathophysiology of schizophrenia. Dev. Psychopathol. 20, 1297
1327.

Kritzer, M.F., and Goldman-Rakic, P.S. (1995). Intrinsic circuit organization of


the major layers and sublayers of the dorsolateral prefrontal cortex in the rhesus monkey. J. Comp. Neurol. 359, 131143.

Kayser, J., Tenke, C.E., Gates, N.A., Kroppmann, C.J., Gil, R.B., and Bruder,
G.E. (2006). ERP/CSD indices of impaired verbal working memory subprocesses in schizophrenia. Psychophysiology 43, 237252.
Kentridge, R.W., Nijboer, T.C., and Heywood, C.A. (2008). Attended but
unseen: Visual attention is not sufficient for visual awareness. Neuropsychologia 46, 864869.
Kiani, R., and Shadlen, M.N. (2009). Representation of confidence associated
with a decision by neurons in the parietal cortex. Science 324, 759764.
Kiebel, S.J., Daunizeau, J., and Friston, K.J. (2008). A hierarchy of time-scales
and the brain. PLoS Comput. Biol. 4, e1000209.
Kiefer, M., and Brendel, D. (2006). Attentional modulation of unconscious
automatic processes: Evidence from event-related potentials in a masked
priming paradigm. J. Cogn. Neurosci. 18, 184198.

Kunde, W. (2003). Sequential modulations of stimulus-response correspondence effects depend on awareness of response conflict. Psychon. Bull.
Rev. 10, 198205.
Lagercrantz, H., and Changeux, J.P. (2009). The emergence of human
consciousness: From fetal to neonatal life. Pediatr. Res. 65, 255260.
Lamme, V.A., and Roelfsema, P.R. (2000). The distinct modes of vision offered
by feedforward and recurrent processing. Trends Neurosci. 23, 571579.
Lamme, V.A., Zipser, K., and Spekreijse, H. (1998). Figure-ground activity in
primary visual cortex is suppressed by anesthesia. Proc. Natl. Acad. Sci.
USA 95, 32633268.
Lamme, V.A., Zipser, K., and Spekreijse, H. (2002). Masking interrupts figureground signals in V1. J. Cogn. Neurosci. 14, 10441053.
Lamy, D., Salti, M., and Bar-Haim, Y. (2009). Neural correlates of subjective
awareness and unconscious processing: An ERP study. J. Cogn. Neurosci.
21, 14351446.

Kihara, K., Ikeda, T., Matsuyoshi, D., Hirose, N., Mima, T., Fukuyama, H., and
Osaka, N. (2011). Differential contributions of the intraparietal sulcus and the
inferior parietal lobe to attentional blink: Evidence from transcranial magnetic
stimulation. J. Cogn. Neurosci. 23, 247256.

Landmann, C., Dehaene, S., Pappata, S., Jobert, A., Bottlaender, M., Roumenov, D., and Le Bihan, D. (2007). Dynamics of prefrontal and cingulate activity
during a reward-based logical deduction task. Cereb. Cortex 17, 749759.

Kim, C.Y., and Blake, R. (2005). Psychophysical magic: Rendering the visible
invisible. Trends Cogn. Sci. (Regul. Ed.) 9, 381388.

Lau, H.C. (2008). A higher order Bayesian decision theory of consciousness.


Prog. Brain Res. 168, 3548.

Kinoshita, S., Forster, K.I., and Mozer, M.C. (2008). Unconscious cognition
isnt that smart: Modulation of masked repetition priming effect in the word
naming task. Cognition 107, 623649.

Lau, H.C., and Passingham, R.E. (2006). Relative blindsight in normal


observers and the neural correlate of visual consciousness. Proc. Natl.
Acad. Sci. USA 103, 1876318768.

Klein, T.A., Endrass, T., Kathmann, N., Neumann, J., von Cramon, D.Y., and
Ullsperger, M. (2007). Neural correlates of error awareness. Neuroimage 34,
17741781.
Koch, C., and Tsuchiya, N. (2007). Attention and consciousness: Two distinct
brain processes. Trends Cogn. Sci. (Regul. Ed.) 11, 1622.
Koechlin, E., Ody, C., and Kouneiher, F. (2003). The architecture of cognitive
control in the human prefrontal cortex. Science 302, 11811185.
Koivisto, M., Revonsuo, A., and Lehtonen, M. (2006). Independence of visual
awareness from the scope of attention: An electrophysiological study. Cereb.
Cortex 16, 415424.
Koivisto, M., Lahteenmaki, M., Srensen, T.A., Vangkilde, S., Overgaard, M.,
and Revonsuo, A. (2008). The earliest electrophysiological correlate of visual
awareness? Brain Cogn. 66, 91103.
Koivisto, M., Kainulainen, P., and Revonsuo, A. (2009). The relationship
between awareness and attention: Evidence from ERP responses. Neuropsychologia 47, 28912899.
Kouider, S., and Dehaene, S. (2007). Levels of processing during nonconscious perception: A critical review of visual masking. Philos. Trans. R.
Soc. Lond. B Biol. Sci. 362, 857875.
Kouider, S., Dehaene, S., Jobert, A., and Le Bihan, D. (2007). Cerebral bases of
subliminal and supraliminal priming during reading. Cereb. Cortex 17, 2019
2029.
Kouider, S., Eger, E., Dolan, R., and Henson, R.N. (2009). Activity in faceresponsive brain regions is modulated by invisible, attended faces: Evidence
from masked priming. Cereb. Cortex 19, 1323.

Lau, H.C., and Passingham, R.E. (2007). Unconscious activation of the cognitive control system in the human prefrontal cortex. J. Neurosci. 27, 58055811.
Laureys, S. (2005). The neural correlate of (un)awareness: Lessons from the
vegetative state. Trends Cogn. Sci. (Regul. Ed.) 9, 556559.
Laureys, S., Lemaire, C., Maquet, P., Phillips, C., and Franck, G. (1999). Cerebral metabolism during vegetative state and after recovery to consciousness.
J. Neurol. Neurosurg. Psychiatry 67, 121.
Laureys, S., Faymonville, M.E., Luxen, A., Lamy, M., Franck, G., and Maquet,
P. (2000). Restoration of thalamocortical connectivity after recovery from
persistent vegetative state. Lancet 355, 17901791.
Laureys, S., Owen, A.M., and Schiff, N.D. (2004). Brain function in coma, vegetative state, and related disorders. Lancet Neurol. 3, 537546.
Lee, U., Kim, S., Noh, G.J., Choi, B.M., Hwang, E., and Mashour, G.A. (2009a).
The directionality and functional organization of frontoparietal connectivity
during consciousness and anesthesia in humans. Conscious. Cogn. 18,
10691078.
Lee, U., Mashour, G.A., Kim, S., Noh, G.J., and Choi, B.M. (2009b). Propofol
induction reduces the capacity for neural information integration: Implications
for the mechanism of consciousness and general anesthesia. Conscious.
Cogn. 18, 5664.
Lehmann, D., and Koenig, T. (1997). Spatio-temporal dynamics of alpha brain
electric fields, and cognitive modes. Int. J. Psychophysiol. 26, 99112.
Lehmann, D., Strik, W.K., Henggeler, B., Koenig, T., and Koukkou, M. (1998).
Brain electric microstates and momentary conscious mind states as building
blocks of spontaneous thinking: I. Visual imagery and abstract thoughts. Int.
J. Psychophysiol. 29, 111.

Kovacs, G., Vogels, R., and Orban, G.A. (1995). Cortical correlate of pattern
backward masking. Proc. Natl. Acad. Sci. USA 92, 55875591.

Lehmann, D., Pascual-Marqui, R.D., Strik, W.K., and Koenig, T. (2010). Core
networks for visual-concrete and abstract thought content: A brain electric
microstate analysis. Neuroimage 49, 10731079.

Kranczioch, C., Debener, S., Maye, A., and Engel, A.K. (2007). Temporal
dynamics of access to consciousness in the attentional blink. Neuroimage
37, 947955.

Leopold, D.A., and Logothetis, N.K. (1996). Activity changes in early visual
cortex reflect monkeys percepts during binocular rivalry. Nature 379,
549553.

Neuron 70, April 28, 2011 2011 Elsevier Inc. 223

Neuron

Review
Leuthold, H., and Kopp, B. (1998). Mechanisms of priming by masked stimuli:
Inferences from event-related potentials. Psychol. Sci. 9, 263269.
Lhermitte, F. (1983). Utilization behaviour and its relation to lesions of the
frontal lobes. Brain 106, 237255.
Li, G.D., Chiara, D.C., Cohen, J.B., and Olsen, R.W. (2010). Numerous classes
of general anesthetics inhibit etomidate binding to gamma-aminobutyric acid
type A (GABAA) receptors. J. Biol. Chem. 285, 86158620.
Libet, B., Gleason, C.A., Wright, E.W., and Pearl, D.K. (1983). Time of
conscious intention to act in relation to onset of cerebral activity (readinesspotential). The unconscious initiation of a freely voluntary act. Brain 106,
623642.
Llinas, R.R., and Pare, D. (1991). Of dreaming and wakefulness. Neuroscience
44, 521535.
Llinas, R., Ribary, U., Contreras, D., and Pedroarena, C. (1998). The neuronal
basis for consciousness. Philos. Trans. R. Soc. Lond. B Biol. Sci. 353, 1841
1849.
Logan, G.D., and Crump, M.J. (2010). Cognitive illusions of authorship reveal
hierarchical error detection in skilled typists. Science 330, 683686.
Luck, S.J., Fuller, R.L., Braun, E.L., Robinson, B., Summerfelt, A., and Gold,
J.M. (2006). The speed of visual attention in schizophrenia: Electrophysiological and behavioral evidence. Schizophr. Res. 85, 174195.
Lynall, M.E., Bassett, D.S., Kerwin, R., McKenna, P.J., Kitzbichler, M., Muller,
U., and Bullmore, E. (2010). Functional connectivity and brain networks in
schizophrenia. J. Neurosci. 30, 94779487.
Mack, A., and Rock, I. (1998). Inattentional Blindness (Cambridge, Mass: MIT
Press).
Macknik, S.L., and Haglund, M.M. (1999). Optical images of visible and invisible percepts in the primary visual cortex of primates. Proc. Natl. Acad. Sci.
USA 96, 1520815210.
Macknik, S.L., and Livingstone, M.S. (1998). Neuronal correlates of visibility
and invisibility in the primate visual system. Nat. Neurosci. 1, 144149.
Maier, A., Wilke, M., Aura, C., Zhu, C., Ye, F.Q., and Leopold, D.A. (2008).
Divergence of fMRI and neural signals in V1 during perceptual suppression
in the awake monkey. Nat. Neurosci. 11, 11931200.
Mantini, D., Perrucci, M.G., Del Gratta, C., Romani, G.L., and Corbetta, M.
(2007). Electrophysiological signatures of resting state networks in the human
brain. Proc. Natl. Acad. Sci. USA 104, 1317013175.
Mantini, D., Corbetta, M., Perrucci, M.G., Romani, G.L., and Del Gratta, C.
(2009). Large-scale brain networks account for sustained and transient activity
during target detection. Neuroimage 44, 265274.
Maquet, P., Degueldre, C., Delfiore, G., Aerts, J., Peters, J.M., Luxen, A., and
Franck, G. (1997). Functional neuroanatomy of human slow wave sleep. J.
Neurosci. 17, 28072812.
Marois, R., and Ivanoff, J. (2005). Capacity limits of information processing in
the brain. Trends Cogn. Sci. (Regul. Ed.) 9, 296305.
Marois, R., Yi, D.J., and Chun, M.M. (2004). The neural fate of consciously
perceived and missed events in the attentional blink. Neuron 41, 465472.
Marti, S., Sackur, J., Sigman, M., and Dehaene, S. (2010). Mapping introspections blind spot: Reconstruction of dual-task phenomenology using quantified
introspection. Cognition 115, 303313.
Mason, M.F., Norton, M.I., Van Horn, J.D., Wegner, D.M., Grafton, S.T., and
Macrae, C.N. (2007). Wandering minds: The default network and stimulusindependent thought. Science 315, 393395.
Mattler, U. (2005). Inhibition and decay of motor and nonmotor priming.
Percept. Psychophys. 67, 285300.

Melchitzky, D.S., Sesack, S.R., Pucak, M.L., and Lewis, D.A. (1998). Synaptic
targets of pyramidal neurons providing intrinsic horizontal connections in
monkey prefrontal cortex. J. Comp. Neurol. 390, 211224.
Melchitzky, D.S., Gonzalez-Burgos, G., Barrionuevo, G., and Lewis, D.A.
(2001). Synaptic targets of the intrinsic axon collaterals of supragranular pyramidal neurons in monkey prefrontal cortex. J. Comp. Neurol. 430, 209221.
Melloni, L., Molina, C., Pena, M., Torres, D., Singer, W., and Rodriguez, E.
(2007). Synchronization of neural activity across cortical areas correlates
with conscious perception. J. Neurosci. 27, 28582865.
Melloni, L., Schwiedrzik, C.M., Muller, N., Rodriguez, E., and Singer, W. (2011).
Expectations change the signatures and timing of electrophysiological correlates of perceptual awareness. J. Neurosci. 31, 13861396.
Merikle, P.M., and Joordens, S. (1997). Parallels between perception without
attention and perception without awareness. Conscious. Cogn. 6, 219236.
Meyer, K., and Damasio, A. (2009). Convergence and divergence in a neural
architecture for recognition and memory. Trends Neurosci. 32, 376382.
Monti, M.M., Vanhaudenhuyse, A., Coleman, M.R., Boly, M., Pickard, J.D.,
Tshibanda, L., Owen, A.M., and Laureys, S. (2010). Willful modulation of brain
activity in disorders of consciousness. N. Engl. J. Med. 362, 579589.
Muller-Gass, A., Macdonald, M., Schroger, E., Sculthorpe, L., and Campbell,
K. (2007). Evidence for the auditory P3a reflecting an automatic process: Elicitation during highly-focused continuous visual attention. Brain Res. 1170,
7178.
Naatanen, R. (1990). The role of attention in auditory information processing as
revealed by event-related potentials and other brain measures of cognitive
function. Behav. Brain Sci. 13, 201288.
Naccache, L., and Dehaene, S. (2001). Unconscious semantic priming extends
to novel unseen stimuli. Cognition 80, 215229.
Naccache, L., Blandin, E., and Dehaene, S. (2002). Unconscious masked
priming depends on temporal attention. Psychol. Sci. 13, 416424.
Niedeggen, M., Wichmann, P., and Stoerig, P. (2001). Change blindness and
time to consciousness. Eur. J. Neurosci. 14, 17191726.
Nieuwenhuis, S., Ridderinkhof, K.R., Blom, J., Band, G.P., and Kok, A. (2001).
Error-related brain potentials are differentially related to awareness of
response errors: Evidence from an antisaccade task. Psychophysiology 38,
752760.
Nieuwenstein, M., Van der Burg, E., Theeuwes, J., Wyble, B., and Potter, M.
(2009). Temporal constraints on conscious vision: On the ubiquitous nature
of the attentional blink. J. Vis. 9, 18.118.14.
Niswender, C.M., and Conn, P.J. (2010). Metabotropic glutamate receptors:
Physiology, pharmacology, and disease. Annu. Rev. Pharmacol. Toxicol. 50,
295322.
Norman, D.A., and Shallice, T. (1980). Attention to action: Willed and automatic
control of behavior. In Consciousness and Self-Regulation, R.J. Davidson,
G.E. Schwartz, and D. Shapiro, eds. (New York: Plenum Press), pp. 118.
Nury, H., Van Renterghem, C., Weng, Y., Tran, A., Baaden, M., Dufresne, V.,
Changeux, J.P., Sonner, J.M., Delarue, M., and Corringer, P.J. (2011). X-ray
structures of general anaesthetics bound to a pentameric ligand-gated ion
channel. Nature 469, 428431.
Oh, J.S., Kubicki, M., Rosenberger, G., Bouix, S., Levitt, J.J., McCarley, R.W.,
Westin, C.F., and Shenton, M.E. (2009). Thalamo-frontal white matter alterations in chronic schizophrenia: A quantitative diffusion tractography study.
Hum. Brain Mapp. 30, 38123825.
Owen, A.M., Coleman, M.R., Boly, M., Davis, M.H., Laureys, S., and Pickard,
J.D. (2006). Detecting awareness in the vegetative state. Science 313, 1402.

McCormick, P.A. (1997). Orienting attention without awareness. J. Exp.


Psychol. Hum. Percept. Perform. 23, 168180.

Palva, S., Linkenkaer-Hansen, K., Naatanen, R., and Palva, J.M. (2005). Early
neural correlates of conscious somatosensory perception. J. Neurosci. 25,
52485258.

McGlashan, T.H., and Hoffman, R.E. (2000). Schizophrenia as a disorder of


developmentally reduced synaptic connectivity. Arch. Gen. Psychiatry 57,
637648.

Pandya, D.N., and Yeterian, E.H. (1990). Prefrontal cortex in relation to other
cortical areas in rhesus monkey: Architecture and connections. Prog. Brain
Res. 85, 6394.

224 Neuron 70, April 28, 2011 2011 Elsevier Inc.

Neuron

Review
Parvizi, J., and Damasio, A.R. (2003). Neuroanatomical correlates of brainstem
coma. Brain 126, 15241536.
Parvizi, J., Van Hoesen, G.W., Buckwalter, J., and Damasio, A. (2006). Neural
connections of the posteromedial cortex in the macaque. Proc. Natl. Acad.
Sci. USA 103, 15631568.
Pashler, H. (1994). Dual-task interference in simple tasks: Data and theory.
Psychol. Bull. 116, 220244.
Passingham, R. (1993). The Frontal Lobes and Voluntary Action, Volume 21
(New York: Oxford University Press).
Penrose, R. (1990). The Emperors New Mind. Concerning Computers, Minds,
and the Laws of Physics (London: Vintage books).

Robitaille, N., and Jolicoeur, P. (2006). Fundamental properties of the N2pc as


an index of spatial attention: Effects of masking. Can. J. Exp. Psychol. 60, 101
111.
Rockstroh, B., Muller, M., Cohen, R., and Elbert, T. (1992). Probing the functional brain state during P300 evocation. J. Psychophysiol. 6, 175184.
Rodriguez, E., George, N., Lachaux, J.P., Martinerie, J., Renault, B., and Varela, F.J. (1999). Perceptions shadow: Long-distance synchronization of
human brain activity. Nature 397, 430433.
Rolls, E.T., Tovee, M.J., and Panzeri, S. (1999). The neurophysiology of backward visual masking: Information analysis. J. Cogn. Neurosci. 11, 300311.

Persaud, N., McLeod, P., and Cowey, A. (2007). Post-decision wagering


objectively measures awareness. Nat. Neurosci. 10, 257261.

Roopun, A.K., Cunningham, M.O., Racca, C., Alter, K., Traub, R.D., and Whittington, M.A. (2008). Region-specific changes in gamma and beta2 rhythms in
NMDA receptor dysfunction models of schizophrenia. Schizophr. Bull. 34,
962973.

Pessiglione, M., Schmidt, L., Draganski, B., Kalisch, R., Lau, H., Dolan, R.J.,
and Frith, C.D. (2007). How the brain translates money into force: A neuroimaging study of subliminal motivation. Science 316, 904906.

Rosenthal, D.M. (2004). Varieties of higher-order theory. In Higher-Order Theories of Consciousness, R.J. Gennaro, ed. (Philadelphia: John Benjamins),
pp. 1944.

Petrides, M., and Pandya, D.N. (2009). Distinct parietal and temporal pathways
to the homologues of Brocas area in the monkey. PLoS Biol. 7, e1000170.

Rougier, N.P., Noelle, D.C., Braver, T.S., Cohen, J.D., and OReilly, R.C.
(2005). Prefrontal cortex and flexible cognitive control: Rules without symbols.
Proc. Natl. Acad. Sci. USA 102, 73387343.

Pins, D., and Ffytche, D. (2003). The neural correlates of conscious vision.
Cereb. Cortex 13, 461474.
Pollen, D.A. (1999). On the neural correlates of visual perception. Cereb.
Cortex 9, 419.
Polonsky, A., Blake, R., Braun, J., and Heeger, D.J. (2000). Neuronal activity in
human primary visual cortex correlates with perception during binocular
rivalry. Nat. Neurosci. 3, 11531159.
Posner, M.I., and Dehaene, S. (1994). Attentional networks. Trends Neurosci.
17, 7579.
Posner, M.I., and Rothbart, M.K. (1998). Attention, self-regulation and
consciousness. Philos. Trans. R. Soc. Lond. B Biol. Sci. 353, 19151927.
Posner, N.I., and Snyder, C.R.R. (1975). Attention and cognitive control. In
Information Processing and Cognition: The Loyola Symposium, R.L. Solso,
ed. (Hillsdale: L. Erlbaum), pp. 205223.

Rounis, E., Maniscalco, B., Rothwell, J.C., Passingham, R., and Lau, H. (2010).
Theta-burst transcranial magnetic stimulation to the prefrontal cortex impairs
metacognitive visual awareness. Cognitive Neuroscience 1, 165175.
Sackur, J., and Dehaene, S. (2009). The cognitive architecture for chaining of
two mental operations. Cognition 111, 187211.
Sadaghiani, S., Hesselmann, G., and Kleinschmidt, A. (2009). Distributed and
antagonistic contributions of ongoing activity fluctuations to auditory stimulus
detection. J. Neurosci. 29, 1341013417.
Sadaghiani, S., Scheeringa, R., Lehongre, K., Morillon, B., Giraud, A.L., and
Kleinschmidt, A. (2010). Intrinsic connectivity networks, alpha oscillations,
and tonic alertness: A simultaneous electroencephalography/functional
magnetic resonance imaging study. J. Neurosci. 30, 1024310250.
Salisbury, D., Squires, N.K., Ibel, S., and Maloney, T. (1992). Auditory eventrelated potentials during stage 2 NREM sleep in humans. J. Sleep Res. 1,
251257.

Preuss, T.M., and Goldman-Rakic, P.S. (1991). Ipsilateral cortical connections


of granular frontal cortex in the strepsirhine primate Galago, with comparative
comments on anthropoid primates. J. Comp. Neurol. 310, 507549.

Schiller, P.H., and Chorover, S.L. (1966). Metacontrast: Its relation to evoked
potentials. Science 153, 13981400.

Procyk, E., Tanaka, Y.L., and Joseph, J.P. (2000). Anterior cingulate activity
during routine and non-routine sequential behaviors in macaques. Nat. Neurosci.
3, 502508.

Schoenemann, P.T., Sheehan, M.J., and Glotzer, L.D. (2005). Prefrontal white
matter volume is disproportionately larger in humans than in other primates.
Nat. Neurosci. 8, 242252.

Pucak, M.L., Levitt, J.B., Lund, J.S., and Lewis, D.A. (1996). Patterns of intrinsic
and associational circuitry in monkey prefrontal cortex. J. Comp. Neurol. 376,
614630.

Schrouff, J., Perlbarg, V., Boly, M., Marrelec, G., Boveroux, P., Vanhaudenhuyse, A., Bruno, M.A., Laureys, S., Phillips, C., Pelegrini-Issac, M., Maquet,
P., and Benali, H. (2011). Brain functional integration decreases during
propofol-induced loss of consciousness. NeuroImage, in press. 10.1016/j.
neuroimage.2011.04.020.

Quiroga, R.Q., Mukamel, R., Isham, E.A., Malach, R., and Fried, I. (2008).
Human single-neuron responses at the threshold of conscious recognition.
Proc. Natl. Acad. Sci. USA 105, 35993604.
Railo, H., and Koivisto, M. (2009). The electrophysiological correlates of stimulus visibility and metacontrast masking. Conscious. Cogn. 18, 794803.
Ray, S., and Maunsell, J.H. (2010). Differences in gamma frequencies across
visual cortex restrict their possible use in computation. Neuron 67, 885896.
Raymond, J.E., Shapiro, K.L., and Arnell, K.M. (1992). Temporary suppression
of visual processing in an RSVP task: An attentional blink? J. Exp. Psychol.
Hum. Percept. Perform. 18, 849860.
Reuter, F., Del Cul, A., Audoin, B., Malikova, I., Naccache, L., Ranjeva, J.P.,
Lyon-Caen, O., Ali Cherif, A., Cohen, L., Dehaene, S., and Pelletier, J.
(2007). Intact subliminal processing and delayed conscious access in multiple
sclerosis. Neuropsychologia 45, 26832691.
Reuter, F., Del Cul, A., Malikova, I., Naccache, L., Confort-Gouny, S., Cohen,
L., Cherif, A.A., Cozzone, P.J., Pelletier, J., Ranjeva, J.P., et al. (2009). White
matter damage impairs access to consciousness in multiple sclerosis. Neuroimage 44, 590599.

Schurger, A., and Sher, S. (2008). Awareness, loss aversion, and post-decision
wagering. Trends Cogn. Sci. (Regul. Ed.) 12, 209210, author reply 210.
Schurger, A., Cowey, A., and Tallon-Baudry, C. (2006). Induced gamma-band
oscillations correlate with awareness in hemianopic patient GY. Neuropsychologia 44, 17961803.
Schurger, A., Pereira, F., Treisman, A., and Cohen, J.D. (2010). Reproducibility
distinguishes conscious from nonconscious neural representations. Science
327, 9799.
Semendeferi, K., Lu, A., Schenker, N., and Damasio, H. (2002). Humans and
great apes share a large frontal cortex. Nat. Neurosci. 5, 272276.
Sergent, C., and Dehaene, S. (2004). Is consciousness a gradual phenomenon? Evidence for an all-or-none bifurcation during the attentional blink.
Psychol. Sci. 15, 720728.
Sergent, C., Baillet, S., and Dehaene, S. (2005). Timing of the brain events
underlying access to consciousness during the attentional blink. Nat. Neurosci.
8, 13911400.

Neuron 70, April 28, 2011 2011 Elsevier Inc. 225

Neuron

Review
Seth, A.K. (2007). Models of consciousness. Scholarpedia 2, 1328.
Shallice, T. (1972). Dual functions of consciousness. Psychol. Rev. 79,
383393.

confined to areas in the occipital cortex beyond human V1/V2. Proc. Natl.
Acad. Sci. USA 102, 1717817183.

Shallice, T. (1988). From Neuropsychology to Mental Structure (Cambridge,


UK: Cambridge University Press).

Tshibanda, L., Vanhaudenhuyse, A., Galanaud, D., Boly, M., Laureys, S., and
Puybasset, L. (2009). Magnetic resonance spectroscopy and diffusion tensor
imaging in coma survivors: Promises and pitfalls. Prog. Brain Res. 177,
215229.

Sheinberg, D.L., and Logothetis, N.K. (1997). The role of temporal cortical
areas in perceptual organization. Proc. Natl. Acad. Sci. USA 94, 34083413.

Tsuchiya, N., and Koch, C. (2005). Continuous flash suppression reduces


negative afterimages. Nat. Neurosci. 8, 10961101.

Sigman, M., and Dehaene, S. (2008). Brain mechanisms of serial and parallel
processing during dual-task performance. J. Neurosci. 28, 75857598.

Uhlhaas, P.J., and Singer, W. (2006). Neural synchrony in brain disorders:


Relevance for cognitive dysfunctions and pathophysiology. Neuron 52,
155168.

Sigman, M., Pan, H., Yang, Y., Stern, E., Silbersweig, D., and Gilbert, C.D.
(2005). Top-down reorganization of activity in the visual pathway after learning
a shape identification task. Neuron 46, 823835.
Simons, D.J., and Ambinder, M.S. (2005). Change blindness: Theory and
consequences. Curr. Dir. Psychol. Sci. 14, 4448.
Slachevsky, A., Pillon, B., Fourneret, P., Pradat-Diehl, P., Jeannerod, M., and
Dubois, B. (2001). Preserved adjustment but impaired awareness in a sensorymotor conflict following prefrontal lesions. J. Cogn. Neurosci. 13, 332340.
Slagter, H.A., Johnstone, T., Beets, I.A., and Davidson, R.J. (2010). Neural
competition for conscious representation across time: An fMRI study. PLoS
ONE 5, e10556.
Smallwood, J., Beach, E., Schooler, J.W., and Handy, T.C. (2008). Going
AWOL in the brain: Mind wandering reduces cortical analysis of external
events. J. Cogn. Neurosci. 20, 458469.
Stephan, K.E., Friston, K.J., and Frith, C.D. (2009). Dysconnection in schizophrenia: From abnormal synaptic plasticity to failures of self-monitoring.
Schizophr. Bull. 35, 509527.
Sterzer, P., Haynes, J.D., and Rees, G. (2008). Fine-scale activity patterns in
high-level visual areas encode the category of invisible objects. J. Vis. 8,
10.110.12.
Sukhotinsky, I., Zalkind, V., Lu, J., Hopkins, D.A., Saper, C.B., and Devor, M.
(2007). Neural pathways associated with loss of consciousness caused by
intracerebral microinjection of GABA A-active anesthetics. Eur. J. Neurosci.
25, 14171436.
Supe`r, H., Spekreijse, H., and Lamme, V.A. (2001). Two distinct modes of
sensory processing observed in monkey primary visual cortex (V1). Nat.
Neurosci. 4, 304310.
Supe`r, H., van der Togt, C., Spekreijse, H., and Lamme, V.A. (2003). Internal
state of monkey primary visual cortex (V1) predicts figure-ground perception.
J. Neurosci. 23, 34073414.
Taine, H. (1870). De lintelligence (Paris: Hachette).
Terrace, H.S., and Son, L.K. (2009). Comparative metacognition. Curr. Opin.
Neurobiol. 19, 6774.
Thiebaut de Schotten, M., Urbanski, M., Duffau, H., Volle, E., Levy, R., Dubois,
B., and Bartolomeo, P. (2005). Direct evidence for a parietal-frontal pathway
subserving spatial awareness in humans. Science 309, 22262228.
Thirion, B., Duchesnay, E., Hubbard, E., Dubois, J., Poline, J.B., Lebihan, D.,
and Dehaene, S. (2006). Inverse retinotopy: Inferring the visual content of
images from brain activation patterns. Neuroimage 33, 11041116.

Uhlhaas, P.J., Linden, D.E., Singer, W., Haenschel, C., Lindner, M., Maurer, K.,
and Rodriguez, E. (2006). Dysfunctional long-range coordination of neural
activity during Gestalt perception in schizophrenia. J. Neurosci. 26, 8168
8175.
Urbanski, M., Thiebaut de Schotten, M., Rodrigo, S., Catani, M., Oppenheim,
C., Touze, E., Chokron, S., Meder, J.F., Levy, R., Dubois, B., and Bartolomeo,
P. (2008). Brain networks of spatial awareness: Evidence from diffusion tensor
imaging tractography. J. Neurol. Neurosurg. Psychiatry 79, 598601.
van Aalderen-Smeets, S.I., Oostenveld, R., and Schwarzbach, J. (2006). Investigating neurophysiological correlates of metacontrast masking with magnetoencephalography. Adv. Cogn. Psychol. 2, 2135.
Van de Ville, D., Britz, J., and Michel, C.M. (2010). EEG microstate sequences
in healthy humans at rest reveal scale-free dynamics. Proc. Natl. Acad. Sci.
USA 107, 1817918184.
Van den Bussche, E., Segers, G., and Reynvoet, B. (2008). Conscious and
unconscious proportion effects in masked priming. Conscious. Cogn. 17,
13451358.
Van den Bussche, E., Notebaert, K., and Reynvoet, B. (2009a). Masked primes
can be genuinely semantically processed: A picture prime study. Exp. Psychol.
56, 295300.
Van den Bussche, E., Van den Noortgate, W., and Reynvoet, B. (2009b). Mechanisms of masked priming: A meta-analysis. Psychol. Bull. 135, 452477.
van der Stelt, O., Frye, J., Lieberman, J.A., and Belger, A. (2004). Impaired P3
generation reflects high-level and progressive neurocognitive dysfunction in
schizophrenia. Arch. Gen. Psychiatry 61, 237248.
van Gaal, S., Ridderinkhof, K.R., Fahrenfort, J.J., Scholte, H.S., and Lamme,
V.A. (2008). Frontal cortex mediates unconsciously triggered inhibitory control.
J. Neurosci. 28, 80538062.
van Gaal, S., Lamme, V.A., and Ridderinkhof, K.R. (2010). Unconsciously triggered conflict adaptation. PLoS ONE 5, e11508.
van Gaal, S., Lamme, V.A., Fahrenfort, J.J., and Ridderinkhof, K.R. (2011).
Dissociable brain mechanisms underlying the conscious and unconscious
control of behavior. J. Cogn. Neurosci. 23, 91105.
Varela, F., Lachaux, J.P., Rodriguez, E., and Martinerie, J. (2001). The brainweb: Phase synchronization and large-scale integration. Nat. Rev. Neurosci.
2, 229239.

Thompson, K.G., and Schall, J.D. (1999). The detection of visual signals by
macaque frontal eye field during masking. Nat. Neurosci. 2, 283288.

Velly, L.J., Rey, M.F., Bruder, N.J., Gouvitsos, F.A., Witjas, T., Regis, J.M.,
Peragut, J.C., and Gouin, F.M. (2007). Differential dynamic of action on cortical
and subcortical structures of anesthetic agents during induction of anesthesia.
Anesthesiology 107, 202212.

Thompson, K.G., and Schall, J.D. (2000). Antecedents and correlates of visual
detection and awareness in macaque prefrontal cortex. Vision Res. 40, 1523
1538.

Veselis, R.A., Feshchenko, V.A., Reinsel, R.A., Dnistrian, A.M., Beattie, B., and
Akhurst, T.J. (2004). Thiopental and propofol affect different regions of the
brain at similar pharmacologic effects. Anesth. Analg. 99, 399408.

Tononi, G. (2008). Consciousness as integrated information: A provisional


manifesto. Biol. Bull. 215, 216242.

Vincent, J.L., Patel, G.H., Fox, M.D., Snyder, A.Z., Baker, J.T., Van Essen,
D.C., Zempel, J.M., Snyder, L.H., Corbetta, M., and Raichle, M.E. (2007).
Intrinsic functional architecture in the anaesthetized monkey brain. Nature
447, 8386.

Tononi, G., and Edelman, G.M. (1998). Consciousness and complexity.


Science 282, 18461851.
Tse, P.U., Martinez-Conde, S., Schlegel, A.A., and Macknik, S.L. (2005). Visibility, visual awareness, and visual masking of simple unattended targets are

226 Neuron 70, April 28, 2011 2011 Elsevier Inc.

Vincent, J.L., Kahn, I., Snyder, A.Z., Raichle, M.E., and Buckner, R.L. (2008).
Evidence for a frontoparietal control system revealed by intrinsic functional
connectivity. J. Neurophysiol. 100, 33283342.

Neuron

Review
Vogel, E.K., and Machizawa, M.G. (2004). Neural activity predicts individual
differences in visual working memory capacity. Nature 428, 748751.
Vogel, E.K., Luck, S.J., and Shapiro, K.L. (1998). Electrophysiological evidence
for a postperceptual locus of suppression during the attentional blink. J. Exp.
Psychol. Hum. Percept. Perform. 24, 16561674.
Vogt, B.A., and Laureys, S. (2005). Posterior cingulate, precuneal and retrosplenial cortices: Cytology and components of the neural network correlates
of consciousness. Prog. Brain Res. 150, 205217.
Von Economo, C. (1929). The Cytoarchitectonics of the Human Cerebral
Cortex (London: Oxford University Press).
von Holst, E., and Mittelstaedt, H. (1950). Das Reafferenzprinzip. Naturwissenschaften 37, 464476.
Voss, H.U., Uluc, A.M., Dyke, J.P., Watts, R., Kobylarz, E.J., McCandliss, B.D.,
Heier, L.A., Beattie, B.J., Hamacher, K.A., Vallabhajosula, S., et al. (2006).
Possible axonal regrowth in late recovery from the minimally conscious state.
J. Clin. Invest. 116, 20052011.
Voytek, B., and Knight, R.T. (2010). Prefrontal cortex and basal ganglia contributions to visual working memory. Proc. Natl. Acad. Sci. USA 107, 18167
18172.
Vul, E., and MacLeod, D.I. (2006). Contingent aftereffects distinguish
conscious and preconscious color processing. Nat. Neurosci. 9, 873874.
Wegner, D.M. (2003). The Illusion of Conscious Will (Cambridge: MIT Press).
Welford, A.T. (1952). The psychological refractory period and the timing of
high speed performanceA review and a theory. Br. J. Psychol. 43, 219.

Williams, M.A., Visser, T.A., Cunnington, R., and Mattingley, J.B. (2008). Attenuation of neural responses in primary visual cortex during the attentional blink.
J. Neurosci. 28, 98909894.
Womelsdorf, T., Fries, P., Mitra, P.P., and Desimone, R. (2006). Gamma-band
synchronization in visual cortex predicts speed of change detection. Nature
439, 733736.
Wong, K.F.E. (2002). The Relationship Between Attentional Blink and Psychological Refractory Period. J. Exp. Psychol. Hum. Percept. Perform. 28, 5471.
Wong, K.F., and Wang, X.J. (2006). A recurrent network mechanism of time
integration in perceptual decisions. J. Neurosci. 26, 13141328.
Woodman, G.F., and Luck, S.J. (2003). Dissociations among attention,
perception, and awareness during object-substitution masking. Psychol.
Sci. 14, 605611.
Wunderlich, K., Schneider, K.A., and Kastner, S. (2005). Neural correlates of
binocular rivalry in the human lateral geniculate nucleus. Nat. Neurosci. 8,
15951602.
Wyart, V., and Tallon-Baudry, C. (2008). Neural dissociation between visual
awareness and spatial attention. J. Neurosci. 28, 26672679.
Wyart, V., and Tallon-Baudry, C. (2009). How ongoing fluctuations in human
visual cortex predict perceptual awareness: Baseline shift versus decision
bias. J. Neurosci. 29, 87158725.

Wilke, M., Logothetis, N.K., and Leopold, D.A. (2003). Generalized flash
suppression of salient visual targets. Neuron 39, 10431052.

Zylberberg, A., Dehaene, S., Mindlin, G.B., and Sigman, M. (2009). Neurophysiological bases of exponential sensory decay and top-down memory retrieval:
A model. Front Comput Neurosci 3, 4.

Wilke, M., Logothetis, N.K., and Leopold, D.A. (2006). Local field potential
reflects perceptual suppression in monkey visual cortex. Proc. Natl. Acad.
Sci. USA 103, 1750717512.

Zylberberg, A., Fernandez Slezak, D., Roelfsema, P.R., Dehaene, S., and
Sigman, M. (2010). The brains router: A cortical network model of serial processing in the primate brain. PLoS Comput. Biol. 6, e1000765.

Neuron 70, April 28, 2011 2011 Elsevier Inc. 227

Consciousness Might Emerge from a Data Broadcast - Scientif...

http://www.scienticamerican.com/article/consciousness-migh...

ADVERTISEMENT

Permanent Address: http://www.scientificamerican.com/article/consciousness-might-emerge-from-a-data-broadcast/


Mind & Brain Scientific American Mind Volume 25, Issue 3 Consciousness Redux

Consciousness Might Emerge from a Data Broadcast


What is consciousness? A neuroscientist's new book argues that it arises when information is broadcast throughout the brain
By Christof Koch | May 1, 2014 |

Quantum physicist Wolfgang Pauli expressed disdain for sloppy, nonsensical


theories by denigrating them as not even wrong, meaning they were just empty
conjectures that could be quickly dismissed. Unfortunately, many remarkably
popular theories of consciousness are of this ilkthe idea, for instance, that our
experiences can somehow be explained by the quantum theory that Pauli himself
helped to formulate in the early 20th century. An even more far-fetched idea holds
that consciousness emerged only a few thousand years ago, when humans realized
that the voices in their head came not from the gods but from their own internal
spoken narratives.
Not every theory of consciousness, however, can be dismissed as just so much
intellectual flapdoodle. During the past several decades, two distinct frameworks for
explaining what consciousness is and how the brain produces it have emerged, each
compelling in its own way. Each framework seeks to explain a vast storehouse of
observations from both neurological patients and sophisticated laboratory
experiments.
One of thesethe Integrated Information Theorydevised by psychiatrist and
neuroscientist Giulio Tononi, which I have described before in these pages [see
Ubiquitous Minds; Scientific American Mind, January/February 2014], uses a
mathematical expression to represent conscious experience and then derives
predictions about which circuits in the brain are essential to produce these

Cognitive scientists Stanislas Dehaene and Bernard


Baars have suggested that memories, sensory
perceptions, judgments and other inputs are stored
in a type of short-term memory called the global
workspace. This buffer gives rise to consciousness
when the collected information is broadcast
throughout the brain to stimulate cognitive
processes that then engage the motor system,
spurring the body to action.
Credit: Raymond Biesinger

ADVERTISEMENT

experiences. [Full disclosure: I have worked with Tononi on this theory.] In contrast,
the Global Workspace Model of consciousness moves in the opposite direction. Its starting point is behavioral experiments that
manipulate conscious experience of people in a very controlled setting. It then seeks to identify the areas of the brain that underlie these
experiences.
Stanislas Dehaene, the French cognitive neuroscientist at the Collge de France in Paris who has devoted much of his career to studying
the psychology of consciousness, has just published a compelling book on his investigations into how the Global Workspace Model
maps onto the brain.
The model derives from the realization that whenever we become conscious of somethingwhether a familiar face in a crowd or the
voice of a strangerwe can retain what we perceive in our mind for a brief period. This perception can remain in this short-term
memory storage, a kind of mental scratch pad, even after the face has disappeared or the voice has died away. Cognitive scientist
Bernard Baars of the Neurosciences Institute in La Jolla, Calif., who came up with the Global Workspace Model, took his central insight
from the early days of artificial intelligence, in which specialized programs accessed a shared repository of information, the blackboard.
According to Baars, it is the act of broadcasting data from the blackboard throughout a computational system, whether cybernetic or
biological, that makes it conscious. Consciousness is just brain-wide sharing of information that is in the memory buffer of the
blackboard.

1 of 4

29/07/2015 11:05 am

Consciousness Might Emerge from a Data Broadcast - Scientif...

http://www.scienticamerican.com/article/consciousness-migh...

This neural buffer does more than process recent sensory inputs. It can also call up a memory from long ago and move it into the buffer.
Once information is loaded into this workspace, a host of powerful cognitive processes can make use of it. The data can be sent off to a
particular brain area that processes languagea language modulewhere this knowledge can be readied for sharing with other people
by formulating a spoken explanation: Guess who I just saw over there. It can also be forwarded to a planning module to be reasoned
about, and it can be stored in long-term memory. The act of transmitting these data from the brain's memory buffers to its various
functional modules is what gives rise to consciousness.
Unfortunately, this workspace has extremely limited capacity. At any one time, we can be conscious of only one or a few items or events,
although we can quickly shift things into and out of consciousness. New information competes with the old and may ultimately
overwrite it. This limitation probably is an unavoidable design characteristic of any information-processing system that is overwhelmed
by inflowing data streams and has to concentrate its most precious resources on dealing with a couple of critical items as fast as
possible.
The brain compensates for the dearth of neural bandwidth by calling on a host of unconscious processes that either totally bypass this
central scratch pad or interact with it below the level of awareness. The vast subliminal onslaught of data thereby turns sounds into
meaningful words and photons into objects and identifiable people. These processes evaluate and weigh evidence, pass judgment and
synchronize the movements initiated by the musculoskeletal system so that an organism can survive in a constantly and rapidly
changing world. They are sophisticated and act quickly but do not share information with one another, nor do they transfer it into the
common workspace. As with an intelligence agency, information is shared only on a need-to-know basis.
Yet these myriad agents of the unconscious shape our daily routines. Because we have, by definition, no access to these subliminal
events, we consistently underestimate their importance. Yet occasionally they manifest themselves quite dramatically. Japanese novelist
Haruki Murakami put it well in a striking interview: We have rooms in ourselves. Most of them we have not visited yet. Forgotten
rooms. From time to time we can find the passage. We find strange things ... old phonographs, pictures, books ... they belong to us, but
it is the first time we have found them.
Dehaene probes these unconscious lairs using a technique called masking. A picture, say, of a face or a word is briefly flashed onto a
monitor, preceded and followed by images of a bunch of randomly drawn lines or a cloud of X's. These masks prevent the displayed
face or word from becoming consciousa subject reports seeing only a mask. Combining versions of this technique with recordings
from electrodes implanted deep into the brain of patients monitored for epilepsy seizures, Dehaene and his colleagues demonstrated
that the unconscious can process the meaning of word combinationsthe brain responds differently to happy war than to happy
loveimplying that it has noticed the incongruence of having a word with a positive emotional meaning followed by a word with a
negative one.
Dehaene and the distinguished molecular biologist Jean-Pierre Changeux have gone beyond this rather abstract model and are
searching for the specific brain areas and populations of neurons that correspond to the global workspace. Their ongoing research using
functional brain imaging and electroencephalographic electrodes placed on the skull has uncovered distinct neural signatures in these
regions that appear to represent the theorized mental buffer.
SEE ALSO:

Energy & Sustainability: 5 Steps to Feed the World and Sustain the Planet | Evolution: Clues to How Homo sapiens Conquered the Earth
Emerge from Digs in South Africa [Slide Show] | Health: The Conflicted History of Alcohol in Western Civilization | Space: Europa's
"Brown Gunk" Suggests a Briny Sea | Technology: Timeline: The Amazing Multimillion-Year History of Processed Food | More
Science: The Flavor Connection

In one classic experiment, Dehaene and his colleagues had volunteers lie inside a magnetic resonance imaging scanner while they
watched a stream of words on a computer screen, each one displayed for 29 milliseconds. Some of the words were masked, which
triggered only a slight brain response. But when the words were legible, an avalanche of neural activity occurred.
The activated regions make up a dense tapestry of interlocking brain cellsspecifically pyramidal neuronsthat tie together the
prefrontal cortex, the inferior parietal lobe, the middle and anterior temporal lobes and other brain regions. Axons, the wirelike
extensions from a neuron's cell body, fan out from the brain's fissured surface, the cerebral cortex, to bind together vast reaches of
neural topography. This network is where Dehaene and his colleagues have started to look both for the brain's scratch pad and for how
signals streaming through this web of connections are communicated to the rest of the brain.
Whenever a stimulus is consciously perceived, its neuronal footprinta particular type of brain activityshows up in many parts of the
cerebral cortex. Take, for instance, the intense electrical activity triggered by an image that passes into the primary visual cortex at the
back of the head and from there to many cortical regions. As it reaches anterior regions of cortex, the signals increase in amplitude,
prompting Dehaene to call it a neuronal avalanche.

2 of 4

29/07/2015 11:05 am

Consciousness Might Emerge from a Data Broadcast - Scientif...

http://www.scienticamerican.com/article/consciousness-migh...

The intense neuronal firing can be caught in the act with EEG electrodes by measuring the P300 wave, a brain wave that, in
experiments, starts about 300 milliseconds after an image is projected onto a computer screen. As Dehaene's experiments demonstrate,
becoming conscious of a sight or sound by having it broadcast throughout the brain from areas postulated to make up the global
workspace often goes hand in hand with the presence of a P300 wave in the prefrontal cortex, a brain area associated with higher
mental processes. Conversely, without the signature P300 wave, electrical activity dies out, and the image displayed is not consciously
perceived. The information fails to enter the global workspace and so remains subliminal.
First Glimmers
Dehaene and his colleagues used this electrophysiological marker of conscious perception to map when consciousness first arises in
five- to 15-month-old infants [see The Conscious Infant; Scientific American Mind, September/October 2013] and to devise a clever
test for consciousness in severely brain-injured patients with whom no reliable communication using speech, eyes or gestures is
possible. The tests depend on the ability of a conscious individual to detect a novel stimulusimagine reading a book when your cell
phone abruptly rings. This unexpected event can trigger a massive P300 wave that is readily noticeable. Yet when you do not pick up the
phone and it rings again and again, you come to expect it, and the P300 becomes fainter until it cannot be detected.
In the laboratory, the researchers play a sequence of five simple tones: beep beep beep beep boop. The last odd-man-out tone generates
a strong P300. When the entire sequence of five tones is repeated three times, the brain adapts to the deviant sound, and the
consciousness marker disappears.
Then, along comes a beep beep beep beep beep sequence. As an attentive subject becomes conscious of the lack of a deviating sound in
the fourth sequence, her brain responds with a P300 to the final beep because it was conditioned to expect a boop.
Preliminary trials using this test with brain-injured patients are intriguing. Patients in whom behavioral evidence indicates a minimal
level of consciousness show this pattern of P300 activity on their EEGs, whereas those in a coma, thought to be without any sensation
whatsoever, do not. Ongoing experiments seek to exploit the same odd-man-out paradigm in monkeys and in mice.
Proposing that what we consciously experience can be defined as the brain's ability to distribute information from the global workspace
to the rest of the brain brings up several questions. Why and how, for instance, does broadcasting information from the global
workspace give rise to consciousness? What message is being broadcast? Blood-borne hormones and chemicals that regulate neural
activity also relay information throughout the body and brain. Yet we are not aware of them. Why not? And can data transmitted over
the Internet or information coursing through the nervous system of a roundworm represent conscious activity? For now the Global
Workspace Model avoids such thorny questions.
When the molecular-biologist-turned-neuroscientist Francis Crick and I started our joint work in the late 1980s on trying to understand
the brain activity underlying vision and other mental processes, scant experimental work was dedicated to empirical studies of the
hallmarks of consciousness.
As the work by Dehaene, Changeux and their colleagues makes abundantly clear, this sorry situation has changed radically. Their
research program is beginning to untangle how the firing of networks of brain cells translates into this most mysterious of all
phenomena.

ABOUT THE AUTHOR(S)


Christof Koch is chief scientific officer at the Allen Institute for Brain Science in Seattle. He serves on Scientific American Mind's board
of advisers.
This article was originally published with the title "Keep it in Mind."

Buy this digital issue or subscribe to access other articles from the May 2014 publication.
Already have an account? Sign In

3 of 4

Digital Issue
$5.99

Digital Subscription
$19.99

Add To Cart

Subscribe

29/07/2015 11:05 am

Consciousness Might Emerge from a Data Broadcast - Scientif...

http://www.scienticamerican.com/article/consciousness-migh...

Recommended For You


1.

The Problem with Female Superheroes a month ago scienticamerican.com ScienticAmerican.com Mind & Brain
2.

Your Facial Bone Structure Has a Big Inuence on How People See You a month ago scienticamerican.com ScienticAmerican.com
Everyday Science

What Kind of Introvert Are You? 9 months ago blogs.scienticamerican.com ScienticAmerican.com Jennifer Odessa Grimes

2015 Scientific American, a Division of Nature America, Inc.


All Rights Reserved.
YYEESS!! Send me a free issue of Scientific
American with no obligation to continue
the subscription. If I like it, I will be billed
for the one-year subscription.

Subscribe Now

4 of 4

29/07/2015 11:05 am

REVIEWS
Consciousness and Anesthesia
Michael T. Alkire,1 Anthony G. Hudetz,2 Giulio Tononi3*
When we are anesthetized, we expect consciousness to vanish. But does it always? Although
anesthesia undoubtedly induces unresponsiveness and amnesia, the extent to which it causes
unconsciousness is harder to establish. For instance, certain anesthetics act on areas of the brains
cortex near the midline and abolish behavioral responsiveness, but not necessarily consciousness.
Unconsciousness is likely to ensue when a complex of brain regions in the posterior parietal area is
inactivated. Consciousness vanishes when anesthetics produce functional disconnection in this
posterior complex, interrupting cortical communication and causing a loss of integration; or when
they lead to bistable, stereotypic responses, causing a loss of information capacity. Thus, anesthetics
seem to cause unconsciousness when they block the brains ability to integrate information.
ow consciousness arises in the brain remains unknown. Yet, for nearly two centuries our ignorance has not hampered
the use of general anesthesia for routinely extinguishing consciousness during surgery. Unfortunately, once in every 1000 to 2000 operations
a patient may temporarily regain consciousness or
even remain conscious during surgery (1). Such
intraoperative awareness arises in part because our
ability to evaluate levels of consciousness remains
limited. Nevertheless, progress is being made in
identifying general principles that underlie how
anesthetics bring about unconsciousness (26) and
how, occasionally, they may fail to do so.

Cellular Actions of Anesthetics


The cellular and molecular pharmacology of
anesthetics has been reviewed extensively (68).
General anesthetics fall into two main classes:
intravenous agents used to induce anesthesia,
generally administered together with sedatives or
narcotics; and volatile agents, generally used for
anesthesia maintenance (Table 1). Anesthetics
are thought to work by interacting with ion channels that regulate synaptic transmission and membrane potentials in key regions of the brain and
spinal cord. These ion-channel targets are differentially sensitive to various anesthetic agents
(Table 1).
Anesthetics hyperpolarize neurons by increasing inhibition or decreasing excitation (9) and alter
neuronal activity: The sustained firing typical of
the aroused brain changes to a bistable burst-pause
pattern (10) that is also observed in nonrapideye-movement (NREM) sleep. At intermediate
anesthetic concentrations, neurons begin oscillating, roughly once a second, between a depolarized up-state and a hyperpolarized down-state (11).
The up-state is similar to the sustained depolar-

ization of wakefulness. The down-state shows


complete cessation of synaptic activity for a tenth
of a second or more, after which neurons revert to
another up-state. As anesthetic doses increase, the
up-state turns to a short burst and the down-state
becomes progressively longer. These changes in
neuronal firing patterns are reflected in the electroencephalogram (EEG) (electrical recording
from the scalp) as a transition from the lowvoltage, high-frequency pattern of wakefulness
(known as activated EEG), to the slow-wave
EEG of deep NREM sleep, and finally to an
EEG burst-suppression pattern (12).
The Anesthetized Patient: Unconscious
or Unresponsive?
Clinically, at low-sedative doses anesthetics cause
a state similar to drunkenness, with analgesia, amnesia, distorted time perception, depersonalization,
and increased sleepiness. At slightly higher doses,
a patient fails to move in response to a command
and is considered unconscious. This behavioral
definition of unconsciousness, which was introduced with anesthesia over 160 years ago, while

convenient, has drawbacks. For instance, unresponsiveness can occur without unconsciousness.
When we dream, we have vivid conscious experiences, but are unresponsive because inhibition
by the brainstem induces muscle paralysis (13).
Similarly, paralyzing agents used to prevent unwanted movements during anesthesia do not remove consciousness (14).
Certain anesthetics may impair a persons
willfulness to respond by affecting brain regions
where executive decisions are made. This is not
an issue for anesthetics that globally deactivate
the brain, but it may be problematic for dissociative anesthetics like ketamine. Low doses of
ketamine cause depersonalization, out-of-body experiences, forgetfulness, and loss of motivation to
follow commands (15). At higher doses, ketamine
causes a characteristic state in which the eyes are
open and the face takes on a disconnected blank
stare. Neuroimaging data show a complex pattern
of regional metabolic changes (16), including a
deactivation of executive circuits in anterior
cingulate cortex and basal ganglia (Fig. 1) (17).
A similar open-eyed unresponsiveness is seen in
akinetic mutism after bilateral lesions around the
anterior cingulate cortex (18). In at least some
of these cases, patients understand questions, but
may fail to respond. Indeed, a woman with large
frontal lesions who was clinically unresponsive
was asked to imagine playing tennis or to navigate her room, and she showed cortical activation
patterns indistinguishable from those of healthy
subjects (19). Thus, clinical unresponsiveness is not
necessarily synonymous with unconsciousness.
At doses near the unconsciousness threshold,
some anesthetics block working memory (20).
Thus, patients may fail to respond because they
immediately forget what to do. At much lower
doses, anesthetics cause profound amnesia. Studies
with the isolated forearm technique, in which a
tourniquet is applied to the arm before paralysis is

Potassium channels
Nicotinic Muscarinic
Two Inwardly Voltage
Serotonin AMPA Kainate
GABAA NMDA pore rectifying gated Glycine Ach
Ach

Intravenous
anesthestics
Barbiturates
Propofol
Etomidate
Ketamine
Inhalational
anesthestics
Nitrous oxide
Isoflurane
Sevoflurane

Department of Anesthesiology and the Center for the Neurobiology of Learning and Memory, University of California,
Irvine, CA 92868, USA. 2Department of Anesthesiology, Medical College of Wisconsin, Milwaukee, WI 53226, USA.
3
Department of Psychiatry, University of Wisconsin, Madison,
WI 53719, USA.
*To whom correspondence should be addressed. E-mail:
gtononi@wisc.edu

876

Desflurane
Major potentiation

Minor potentiation

Major inhibition

Minor inhibition

Biphasic

No effect

Table 1. Ionic mechanisms and targets of current clinical anesthetics (6, 8). Abbreviations: Ach,
acetylcholine; AMPA, a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid; GABAA, g-aminobutyric
acid, type A; NMDA, N-methyl-D-aspartate.

7 NOVEMBER 2008

VOL 322

SCIENCE

www.sciencemag.org

REVIEWS
induced (to allow the hand to
Anterior
move while the rest of the body
cingulate
is paralyzed), show that patients
cortex
under general anesthesia can
sometimes carry on a conversation using hand signals, but postoperatively deny ever being awake
(21). Thus, retrospective oblivion
is no proof of unconsciousness.
Nevertheless, at some level of
anesthesia between behavioral unresponsiveness and the induction
of a flat EEG [indicating the cessation of the brains electrical activity, one of the criteria for brain
Fig. 1. Brain
death (22)], consciousness must
vanish. Therefore, the use of brain-function monitors could improve consciousness assessment
during anesthesia (23). For instance, bispectral
index monitors record the EEG signal over the
forehead and reduce the complex signal into a
single number that tracks a patients depth of
anesthesia over time (12). Such devices help guide
anesthetic delivery and may reduce cases of intraoperative awareness (24), but they remain limited at directly indicating the presence or absence
of consciousness, especially around the transition
point. The isolated forearm technique has shown
that individual patients can be aware and responsive during surgery even though their bispectral
index value suggests they are not (25). Either the
EEG is not sensitive enough to the neural processes underlying consciousness, or we still do
not yet fully understand what to look for.
The ThalamusSwitch or Readout?
The most consistent regional effect produced by
anesthetics at (or near) loss of consciousness is a
reduction of thalamic metabolism and blood flow
(Fig. 1), suggesting that the thalamus may serve as
a consciousness switch (2). Indeed, switchlike effects
have been found with a number of thalamic manipulations. For example, g-aminobutyric acid (GABA)
agonists (mimicking anesthetic action) injected into
the intralaminar nuclei cause rats to rapidly fall
asleep, with a corresponding slowing of the EEG (26).
Conversely, rats under anesthetic concentrations of
sevoflurane can be awakened by a minute injection
of nicotine into the intralaminar thalamus (27). In
humans, midline thalamic damage can result in a
vegetative state (18). Conversely, recovery from the
vegetative state is heralded by the restoration of functional connectivity between thalamus and cingulate
cortex (28). Also, deep brain electrical stimulation of
the central thalamus improved behavioral responsiveness in a patient who was minimally conscious (29).
Nevertheless, thalamic activity does not decrease with all anesthetics. Ketamine increases
global metabolism, especially in the thalamus
(16). Other anesthetics can substantially reduce
thalamic activity at doses that cause sedation, not
unconsciousness. For instance, sevoflurane sedation causes a 23% reduction of relative thalamic
metabolism when subjects are still awake and
responsive (30). Indeed, anesthetic effects on the

anesthetics? Evoked responses


in primary sensory corticesthe
first relay for incoming stimuli
Parietal
lobe
are often unchanged during anFrontal
lobe
esthesia, deep sleep, and in vegetative patients. Also, activity in
primary sensory areas often does
Thalamus
not correlate with perceptual experience (43). Frontal cortex too
may not be essential for anesthetBasal
ic unconsciousness, because difTemporoganglia
ferent anesthetics have variable
parietooccipital
effects on this area. For instance,
junction
at equivalent hypnotic doses, both
propofol and thiopental deactivate
areas associated with anesthetic effects [see text and (2)].
posterior brain areas, but only prothalamus may be largely indirect (6, 31, 32). Spon- pofol deactivates frontal cortex (44). Furthermore,
taneous thalamic firing during anesthesia is largely large lesions of the frontal cortex do not by themdriven by feedback from cortical neurons (33), selves produce unconsciousness (45).
Anesthetic-induced unconsciousness is usually
especially anesthetic-sensitive layer V cells (34).
Many of these cells also project onto brainstem associated with deactivation of mesial parietal corarousal centers, so cortical deactivation can re- tex, posterior cingulate cortex, and precuneus (Fig. 1)
duce both thalamic activity and arousal (35). Also, (46). These same areas are deactivated in vegetative
the metabolic and electrophysiological effects of patients but are the first to reactivate in those who
anesthetics on the thalamus in animals are abol- recover (28). Moreover, neural activity in these
ished by removal of the cortex (33, 34, 36). By areas is altered during seizures associated with an
contrast, after thalamic ablation, the cortex still impairment of consciousness (47) and in sleep
produces an activated EEG (37), suggesting that (48). These mesial cortical areas are strategicalthe thalamus is not the sole mediator of cortical ly located at the main hub of the brains connecarousal, nor perhaps is it the most direct one. In tional core (49). They are also part of a default
patients with implanted brain electrodes undergoing network that is especially active at rest and may
a second surgery to place a deep brain stimulator, be involved in global monitoring of the internal
the cortical EEG changed dramatically the instant environment and several functions related to the
the patients lost consciousness (38). However, there self (50). Nevertheless, mesial cortical areas are
was little change in thalamic EEG activity until deactivated in REM sleep (48), when subjects
10 min later. Conversely, in epileptic patients, dur- experience vivid dreams. Intriguingly, at intermeing REM sleep (usually associated with dreaming) diate doses, certain anesthetics, such as nitrous
the cortical EEG was activated as if patients were oxide, produce a fairly selective deactivation of
awake, but the thalamic EEG showed slow wave posterior mesial cortex (51), yet when these areas
activity, as if patients were asleep (39). Thus, the start to turn off, subjects report dreamlike feelings
effects of anesthetics on the thalamus may with depersonalization and out-of-body experirepresent a readout of global cortical activity rather ences, rather than unconsciousness.
In addition to mesial cortical areas, many
than a consciousness switch, and thalamic activity
anesthetics also deactivate or disconnect a latmay not be a sufficient basis for consciousness.
Nonetheless, it is premature to write off the eral temporo-parieto-occipital complex of multhalamus altogether. Perhaps efficient communi- timodal associative areas centered on the inferior
cation among cortical areas requires a thalamic parietal cortex (Fig. 1). In this case, lesion and anrelay (40), in which case thalamic lesions would esthesia data are mutually supportive: Patients with
lead to a functional disconnection despite an acti- bilateral lesions at the temporo-parieto-occipital
vated cortex. A functional thalamic disconnection junction show no sign of perceptual experience,
during anesthesia has been found with neuroim- despite a flurry of undirected motor activity, a
aging (41). Subthreshold depolarization to many condition called hyperkinetic mutism (18). Thus,
cortical areas may be provided by calbindin-positive a complex of posterior brain areas comprising
matrix cells, which are especially concentrated within the lateral temporo-parieto-occipital junction and
some intralaminar thalamic nuclei and project dif- perhaps a mesial cortical core are most likely
fusely to superficial layers of cortex (42). Cells in the final common target for anesthetic-induced
intralaminar nuclei can fire at high frequencies, unconsciousness.
thus providing a coherent oscillatory bias that may
facilitate long-range cortico-cortical interactions. Disruption of Cortical Integration
Therefore, whereas cortical arousal may occur Loss of consciousness may not necessarily require
that neurons in these posterior brain areas be inwithout the thalamus, consciousness may not.
activated. Instead, it may be sufficient that dyCortical Effects of Anesthetics
namic aspects of neural activity change, especially
Are some cortical areas more important than if these affect the brains ability to integrate inothers for the induction of unconsciousness by formation (Fig. 2) (3, 5).
Posterior
cingulate
cortex

www.sciencemag.org

Mesial parietal
cortex, precuneus

SCIENCE

VOL 322

7 NOVEMBER 2008

877

REVIEWS
(56). Also, anesthesia suppresses the late component (>100 ms) of visual responses, possibly by
inhibiting feedback connections (57), but not the
early feedforward components. Moreover, anesthesia abolishes contextual and attentional modulation of firing, presumably mediated by feedback
connections (58). The corticothalamic system may
be especially vulnerable to anesthetics due to its
small-world organization. Small-world networks
have mostly local connectivity with comparatively
few long-range connections. Augmented with hubs,
such networks maximize interactions while minimizing wiring. By the same token, anesthetics need
only disrupt a few long-range connections to produce a set of disconnected components. Indeed,
computer simulations demonstrate a rapid state
transition at a critical anesthetic dose (59), consistent with a breakdown in network integration.

Consider first large-scale integration, loosely


defined as the ability of different cortical regions
to interact effectively (52). When consciousness
fades during anesthesia, there is a drop in EEG
coherence in the g-frequency range (20 to 80 Hz)
between right and left frontal cortices and between frontal and occipital regions (4). Anesthetics
also suppress fronto-occipital g coherence in animals, both under visual stimulation and at rest (53).
The effect is gradual and much stronger for longrange than for local coherence (53). Anesthetics
may disrupt cortical integration (5) by acting on
structures that facilitate long-range cortico-cortical
interactions, such as the posterior cortical connectional hub (49), certain thalamic nuclei (42), or
possibly the claustrum (54). Anesthetics may also
disrupt synchronization among distant areas by
slowing neural responses (55).
The loss of feedback interactions in the cortex
may be especially critical. When rats become unresponsive under anesthesia, information transfer
first decreases in the feedback direction (Fig. 2B)

Disruption of Cortical Information Capacity


Consider next how anesthetics affect information,
defined loosely as the number of discriminable

Loss of cortical integration

Integrated information

activity patterns. When the repertoire of discriminable firing patterns available to the corticothalamic
system shrinks, neural activity becomes less informative, even though it may be globally integrated
(52). As described above, at high enough doses
several anesthetics produce a burst-suppression
pattern in which a near-flat EEG is interrupted
every few seconds by brief, quasi-periodic bursts
of global activationa stereotypic, global on-off
pattern. Such stereotypic burst-suppression can also
be elicited by visual, auditory, and mechanical
stimuli (Fig. 3B) (60, 61). Thus, during deep anesthetic unconsciousness, the corticothalamic system can still be activein fact, hyperexcitable
and can produce global responses. However, the
repertoire of responses has shrunk to a stereotypic burst-suppression pattern, with a corresponding
loss of information, essentially creating a system having only two possible states (on or off).
Generalized convulsive seizures provide another
example in which consciousness can be lost even
though neural activity remains high and highly

Human

Premotor cortex
Awake
12 V

TMS

50 ms

Rat

A/mm2

0.01

Anesthetized

Awake

1.1% Isoflurane
Fr

Fr

15 ms

Oc
Par

250 ms

0.002

Occipital (Oc)

12 V

Parietal (Par)

50 ms

0.20

TMS

0.15
0.01

0.10

A/mm2

Gamma power ( V2)

150 ms

0.25

0.05
0.00
-200

-100

100

200

300 -200

-100

100

200

300

Time (ms)

15 ms

Fig. 2. Unconsciousness is associated with a loss of cortical integration.


(A) The corticothalamic system is represented metaphorically as a large
die having many faces, each corresponding to a different brain firing
pattern. During conscious waking, the die rolls on a particular face, ruling out all the others and thus generating integrated information. If
integration is lost (as in anesthesia or sleep), the die disintegrates into
many two-faced dice, each generating 1 bit of information. (B) Anesthesia
reduces cortical integration in the rat. (Top) During waking, transfer entropy,
a measure of directional interactions among brain areas, is balanced in the
feedforward (green) and feedback (red) directions. During anesthesia,
feedback transfer entropy (red) is reduced, implying a decrease in front-toback interactions. (Bottom) Responses to a flashing light delivered at 0.2 Hz
(arrow) from a representative rat when awake and under 1.1% isoflurane

878

100 ms

Asleep

Par

Frontal (Fr)

50 ms

Oc

7 NOVEMBER 2008

VOL 322

50 ms

100 ms

150 ms

250 ms

0.002

anesthesia (56). When the rat is awake, each flash evokes a sustained gfrequency (20 to 60 Hz) response in visual occipital cortex (blue) and a later
response in parietal association cortex (red). During anesthesia, the occipital
response is preserved, although it is shorter (blue), and the parietal response
is attenuated, indicating that anesthesia reduces cortical interactions and thus
reduces integration. (C) Sleeping reduces cortical integration in humans. EEG
voltages and current densities are shown from a representative subject in
which the premotor cortex was stimulated with transcranial magnetic stimulation (TMS) (black arrow). During waking (top), stimulation evokes EEG responses first near the stimulation site (black circle; the white cross is the site of
maximum evoked current) and then in sequence at other cortical locations.
During deep sleep (bottom), the stimulus-evoked response remains local,
indicating a loss of cortical integration.
SCIENCE

www.sciencemag.org

REVIEWS
(TMS) applied to premotor cortex and other cortical areas induces a sustained response (300 ms)
involving the sequential activation of specific brain
areas, the identity of which depends upon the precise site of stimulation (63, 64). During early
NREM sleep, possibly due to the induction of a
local down-state, TMS pulses produce instead a
short (<150 ms) local response (64), suggesting
a loss of integration. Intriguingly, TMS pulses to
mesial parietal regions, overlying the main hub in
the cortical connectional core (49), trigger a stereotypic, high-amplitude slow wave closely resembling spontaneous slow waves (63). This stereotypic
response, presumably due to the simultaneous activation of the cortical connectional core and to the
induction of a global down-state, reflects a limited
repertoire of activity patterns and thus a loss of
information.

synchronized: A large portion of the corticothalamic complex is engaged in strong, hypersynchronous activity, but this activity is stereotypic
(60, 61).
A Bit Like Sleep
Sleep is the only time when healthy humans regularly lose consciousness. Subjects awakened during slow wave sleep early in the night may report
short, thoughtlike fragments of experience, or often
nothing at all (13). Although anesthesia is not the
same as natural sleep, brain-arousal systems are
similarly deactivated (6, 62). Also, as under anesthesia, during slow wave sleep, cortical and
thalamic neurons become bistable and undergo
slow oscillations (1 Hz or less) between up- and
down-states. Like animal studies during anesthesia (Fig. 2B and 3B), human studies during slow
wave sleep suggest that the bistability of cortical
neurons has consequences for the brains capacity
to integrate information (Figs. 2C and 3C). During wakefulness, transcranial magnetic stimulation

Consciousness and Integrated Information


The evidence from anesthesia and sleep states
(Fig. 2 and 3) converges to suggest that loss of

C Human

Loss of
information capacity

Integrated information

consciousness is associated with a breakdown


of cortical connectivity and thus of integration, or
with a collapse of the repertoire of cortical activity patterns and thus of information (Figs. 2
and 3). Why should this be the case? A recent
theory suggests a principled reason: Information
and integration may be the very essence of consciousness (52). Classically, information is the reduction of uncertainty among alternatives: When
a coin falls on one of its two sides, it provides
1 bit of information, whereas a die falling on one
of six faces provides ~2.6 bits. But then having
any conscious experience, even one of pure darkness, must be extraordinarily informative, because
we could have had countless other experiences
instead (think of all the frames of every possible
movie). Having any experience is like throwing a
die with a trillion faces and identifying which
number came up (Figs. 2A and 3A). On the other
hand, every experience is an integrated whole that
cannot be subdivided into independent components. For example, with an intact brain you cannot
Mesial parietal cortex

Awake

20 V

50 ms

TMS

B Rat

A/mm2

0.01

Anesthetized

Awake

1.8% Isoflurane
5 V

Fr

0.002

Fr

Par

15 ms

Par

Oc

50 ms

100 ms

150 ms

250 ms

Asleep
20 V

Oc
25s

Occipital (Oc)

50 ms

TMS

Parietal (Par)

0.15
0.01

0.10

A/mm2

Gamma power ( V2)

Frontal (Fr)
0.20

0.05
0.00
-200

200

400

-200

200

400
0.002

Time (ms)

15 ms

Fig. 3. Unconsciousness is associated with a loss of information capacity.


(A) As in Fig 2, the corticothalamic system is represented metaphorically
as a large die having many faces, each corresponding to a different brain
firing pattern. During conscious waking, the die rolls on a particular face,
ruling out all the others and thus generating integrated information. If
information is lost (as in anesthesia or sleeping), the die is flattened so
that it has only two faces (firing patterns). Due to the loss of repertoire, it
generates only 1 bit of information. (B) Anesthesia reduces information
capacity in rat cortex. (Top) Field potentials recorded before and during
light flashes (marks below each trace). During waking (left), flash-evoked
field potentials (blue) (light flashes indicated by marks below each trace)
are small and variable, being masked by spontaneous neuronal activity.
www.sciencemag.org

SCIENCE

50 ms

100 ms

150 ms

250 ms

During deep anesthesia (right), bursts of activity occur spontaneously and


after each light flash. (Bottom) During anesthesia, the g-burst response is
uniform across all three brain regions. Thus, responses are stereotypic
and lack regional specificity, indicating a loss of information capacity. (C)
Sleeping reduces cortical information carrying capacity in humans. (Top)
During waking, stimulation over the mesial parietal cortex produces a
specific, sequential pattern of activation. (Bottom) During sleep, stimulation produces a global, stereotypic response that spreads from the stimulation site to most of the cortex, indicating a loss of information capacity.
Black traces represent averaged voltage potentials recorded at all electrodes
and superimposed, while estimated current density is displayed in absolute
scale (63, 64).
VOL 322

7 NOVEMBER 2008

879

REVIEWS
experience the left half of the visual field independently of the right half, or visual shapes independently of their color. In other words, the die
of experience is a single onethrowing multiple
dice and combining the numbers will not do.
Less metaphorically, the theory claims that
the level of consciousness of a physical system is
related to the repertoire of different states (information) that can be discriminated by the system
as a whole (integration). A measure of integrated
information, called phi (F), can be used to quantify the information generated when a system enters one particular state of its repertoire, above
and beyond the information generated independently by its parts (52, 65). In practice, F can
only be measured rigorously for small, simulated
systems. However, empirical measures could be
devised to evaluate integrated information on the
basis of EEG data, resting functional connectivity,
or TMS-evoked responses. This approach could
allow the development of consciousness monitors
that evaluate both loss of integration, as revealed
by reduced functional or effective connectivity, and
loss of information, as evidenced by stereotypic
responses.
This theory has some interesting implications
for anesthesia. For example, it explains why a corticothalamic complex is essential for consciousness and is thus the proper target for anesthesia:
By conjoining functional specialization (each cortical area and neuronal group within each area is
exquisitely specialized) with functional integration
(thanks to extensive corticocortical and corticothalamocortical connectivity), a corticothalamic
complex is well suited to behave as a single dynamic entity endowed with a large number of discriminable states. By contrast, parts of the brain
made up of small, quasi-independent modules,
such as the cerebellum, and parallel loops through
the basal ganglia, are not sufficiently integrated,
which is perhaps why they can be lesioned without
loss of consciousness (18, 52). The theory suggests
that one should not interpret individual motor responses, or localized activations, as signs of consciousness, and conversely, should not interpret
the absence of motor responses as a sure sign of
unconsciousness. Finally, from this theoretical
perspective, consciousness is not an all-or-none
property, but it is graded: Specifically, it increases
in proportion to a systems repertoire of discriminable states. The shrinking or dimming of the
field of consciousness during sedation is consistent with this idea. On the other hand, the
abrupt loss of consciousness at a critical concentration of anesthetics suggests that the integrated
repertoire of neural states underlying consciousness may collapse nonlinearly.
Conclusions
Despite different mechanisms and sites of action,
most anesthetic agents appear to cause unconsciousness by targeting, directly or indirectly, a
posterior lateral corticothalamic complex centered
around the inferior parietal lobe, and perhaps a
medial cortical core. Whether the medial or lateral

880

component is more important, and whether anterior


cortical regions are critical primarily for executive functions and perhaps self-reflection, remain
questions for future work. Second, anesthetics can
cause unconsciousness not just by deactivating
this posterior corticothalamic complex, but also
by producing a functional disconnection between
subregions of this complex. Third, although assessing loss of consciousness with verbal commands
may usually be adequate, it may occasionally be
misleading. Finally, one theoretical framework
that seems to fit well with current empirical data
suggests that consciousness requires an integrated
system with a large repertoire of discriminable
states. According to this framework, anesthetics
would produce unconsciousness either by preventing integration (blocking the interactions among
specialized brain regions) or by reducing information (shrinking the number of activity patterns
available to cortical networks). Other frameworks
for consciousness, emphasizing access to a global
workspace (66, 67), or the formation of large coalitions of neurons (43), are also consistent with
many of the findings described here, especially
those concerning the role of cortical integration.
Together, these ideas should help in developing
agents with more specific actions, in better monitoring their effects on consciousness, and in using
anesthesia as a tool for characterizing the neural
substrates of consciousness.
References and Notes
1. P. S. Sebel et al., Anesth. Analg. 99, 833 (2004).
2. M. T. Alkire, J. Miller, Prog. Brain Res. 150, 229
(2005).
3. A. G. Hudetz, Sem. Anesth. Perioper. Med. Pain 25, 196
(2006).
4. E. R. John, L. S. Prichep, Anesthesiology 102, 447
(2005).
5. G. A. Mashour, Anesthesiology 100, 428 (2004).
6. N. P. Franks, Nat. Rev. Neurosci. 9, 370 (2008).
7. J. A. Campagna, K. W. Miller, S. A. Forman, N. Engl. J. Med.
348, 2110 (2003).
8. U. Rudolph, B. Antkowiak, Nat. Rev. Neurosci. 5, 709
(2004).
9. C. R. Ries, E. Puil, J. Neurophysiol. 81, 1795 (1999).
10. R. R. Llinas, M. Steriade, J. Neurophysiol. 95, 3297
(2006).
11. M. Steriade, I. Timofeev, F. Grenier, J. Neurophysiol. 85,
1969 (2001).
12. L. Voss, J. Sleigh, Best Pract. Res. Clin. Anaesthesiol. 21,
313 (2007).
13. J. A. Hobson, E. F. Pace-Schott, R. Stickgold, Behav. Brain
Sci. 23, 793 (2000).
14. G. P. Topulos, R. W. Lansing, R. B. Banzett, J. Clin. Anesth.
5, 369 (1993).
15. M. R. Tucker, J. R. Hann, C. L. Phillips, J. Oral Maxillofac.
Surg. 42, 668 (1984).
16. J. W. Langsjo et al., Anesthesiology 103, 258 (2005).
17. M. T. Alkire, R. L. Gruver, J. W. Lngsj, K. Kaisti,
H. Scheinin, Anesthesiology 104, A1219 (2007).
18. J. B. Posner, C. B. Saper, N. D. Schiff, F. Plum, Plum and
Posner's Diagnosis of Stupor and Coma (Oxford Univ.
Press, Oxford, New York, ed. 4, 2007), Contemporary
Neurology Ser. 71.
19. A. M. Owen et al., Science 313, 1402 (2006).
20. R. A. Veselis, R. A. Reinsel, V. A. Feshchenko, A. M. Dnistrian,
Anesthesiology 97, 329 (2002).
21. I. F. Russell, M. Wang, Br. J. Anaesth. 78, 3 (1997).
22. E. F. Wijdicks, N. Engl. J. Med. 344, 1215 (2001).
23. American Society of Anesthesiologists, Anesthesiology
104, 847 (2006).

7 NOVEMBER 2008

VOL 322

SCIENCE

24. P. S. Myles, K. Leslie, J. McNeil, A. Forbes, M. T. Chan,


Lancet 363, 1757 (2004).
25. G. Schneider, A. W. Gelb, B. Schmeller, R. Tschakert,
E. Kochs, Br. J. Anaesth. 91, 329 (2003).
26. J. W. Miller, J. A. Ferrendelli, Neuropharmacology 29,
649 (1990).
27. M. T. Alkire, J. R. McReynolds, E. L. Hahn, A. N. Trivedi,
Anesthesiology 107, 264 (2007).
28. S. Laureys, M. Boly, P. Maquet, J. Clin. Invest. 116, 1823
(2006).
29. N. D. Schiff et al., Nature 448, 600 (2007).
30. M. T. Alkire et al., Proc. Natl. Acad. Sci. U.S.A. 105, 1722
(2008).
31. M. T. Alkire, R. J. Haier, J. H. Fallon, Conscious. Cogn. 9,
370 (2000).
32. N. D. Schiff, F. Plum, J. Clin. Neurophysiol. 17, 438
(2000).
33. C. Vahle-Hinz, O. Detsch, M. Siemers, E. Kochs,
Exp. Brain Res. 176, 159 (2007).
34. A. Angel, Br. J. Anaesth. 71, 148 (1993).
35. J. D. French, R. Hernandez-Peon, R. B. Livingston,
J. Neurophysiol. 18, 74 (1955).
36. K. Nakakimura, T. Sakabe, N. Funatsu, T. Maekawa,
H. Takeshita, Anesthesiology 68, 777 (1988).
37. J. Villablanca, M. E. Salinas-Zeballos, Arch. Ital. Biol.
110, 383 (1972).
38. L. J. Velly et al., Anesthesiology 107, 202 (2007).
39. M. Magnin, H. Bastuji, L. Garcia-Larrea, F. Mauguiere,
Cereb. Cortex 14, 858 (2004).
40. R. W. Guillery, S. M. Sherman, Neuron 33, 163
(2002).
41. N. S. White, M. T. Alkire, Neuroimage 19, 402 (2003).
42. E. G. Jones, Philos. Trans. R. Soc. London B Biol. Sci.
357, 1659 (2002).
43. F. Crick, C. Koch, Nat. Neurosci. 6, 119 (2003).
44. R. A. Veselis et al., Anesth. Analg. 99, 399 (2004).
45. H. J. Markowitsch, J. Kessler, Exp. Brain Res. 133, 94
(2000).
46. K. K. Kaisti et al., Anesthesiology 96, 1358 (2002).
47. H. Blumenfeld, Prog. Brain Res. 150, 271 (2005).
48. P. Maquet, J. Sleep Res. 9, 207 (2000).
49. P. Hagmann et al., PLoS Biol. 6, e159 (2008).
50. K. Vogeley et al., J. Cogn. Neurosci. 16, 817 (2004).
51. F. E. Gyulai, L. L. Firestone, M. A. Mintun, P. M. Winter,
Anesth. Analg. 83, 291 (1996).
52. G. Tononi, BMC Neurosci. 5, 42 (2004).
53. O. A. Imas, K. M. Ropella, J. D. Wood, A. G. Hudetz,
Neurosci. Lett. 402, 216 (2006).
54. F. C. Crick, C. Koch, Philos. Trans. R. Soc. London B Biol. Sci.
360, 1271 (2005).
55. R. Munglani, J. Andrade, D. J. Sapsford, A. Baddeley,
J. G. Jones, Br. J. Anaesth. 71, 633 (1993).
56. O. A. Imas, K. M. Ropella, B. D. Ward, J. D. Wood,
A. G. Hudetz, Neurosci. Lett. 387, 145 (2005).
57. A. G. Hudetz, Int. Anesthesiol. Clin. 46, 25 (2008).
58. H. Super, H. Spekreijse, V. A. Lamme, Nat. Neurosci.
4, 304 (2001).
59. M. L. Steyn-Ross, D. A. Steyn-Ross, J. W. Sleigh,
L. C. Wilcocks, Phys. Rev. E Stat. Nonlin. Soft Matter Phys.
64, 011917 (2001).
60. A. G. Hudetz, O. A. Imas, Anesthesiology 107, 983
(2007).
61. D. Kroeger, F. Amzica, J. Neurosci. 27, 10597
(2007).
62. R. Lydic, H. A. Baghdoyan, Anesthesiology 103, 1268
(2005).
63. M. Massimini et al., Proc. Natl. Acad. Sci. U.S.A. 104,
8496 (2007).
64. M. Massimini et al., Science 309, 2228 (2005).
65. D. Balduzzi, G. Tononi, PLoS Comput. Biol. 4, e1000091
(2008).
66. B. J. Baars, Prog. Brain Res. 150, 45 (2005).
67. S. Dehaene, C. Sergent, J. P. Changeux, Proc. Natl. Acad.
Sci. U.S.A. 100, 8520 (2003).
68. Supported by the NIH Directors Pioneer Award
and the James S. McDonnell Foundation (G.T.). G.T.
has a patent pending on the use of TMS-EEG in
anesthesia.
10.1126/science.1149213

www.sciencemag.org

Intrinsic Brain Activity in Altered States


of Consciousness
How Conscious Is the Default Mode of Brain Function?
M. BOLY,a,b C. PHILLIPS,a L. TSHIBANDA,c A. VANHAUDENHUYSE,a M. SCHABUS,a,d
T.T. DANG-VU,a,b G. MOONEN,b R. HUSTINX,e P. MAQUET,a,b AND S. LAUREYSa,b
a

Coma Science Group, Cyclotron Research Center, University of Li`ege, Li`ege, Belgium

Neurology Department, CHU Hospital, Li`ege, Belgium

Radiology Department, CHU Hospital, Li`ege, Belgium

University of Salzburg, Department of Physiological Psychology, Salzburg, Austria


e

Nuclear Medicine Department, CHU Hospital, Li`ege, Belgium

Spontaneous brain activity has recently received increasing interest in the neuroimaging community. However, the value of resting-state studies to a better understanding of brainbehavior
relationships has been challenged. That altered states of consciousness are a privileged way to
study the relationships between spontaneous brain activity and behavior is proposed, and common resting-state brain activity features observed in various states of altered consciousness are
reviewed. Early positron emission tomography studies showed that states of extremely low or high
brain activity are often associated with unconsciousness. However, this relationship is not absolute,
and the precise link between global brain metabolism and awareness remains yet difficult to assert.
In contrast, voxel-based analyses identified a systematic impairment of associative frontoparieto
cingulate areas in altered states of consciousness, such as sleep, anesthesia, coma, vegetative
state, epileptic loss of consciousness, and somnambulism. In parallel, recent functional magnetic
resonance imaging studies have identified structured patterns of slow neuronal oscillations in the
resting human brain. Similar coherent blood oxygen leveldependent (BOLD) systemwide patterns
can also be found, in particular in the default-mode network, in several states of unconsciousness,
such as coma, anesthesia, and slow-wave sleep. The latter results suggest that slow coherent spontaneous BOLD fluctuations cannot be exclusively a reflection of conscious mental activity, but may
reflect default brain connectivity shaping brain areas of most likely interactions in a way that
transcends levels of consciousness, and whose functional significance remains largely in the dark.
Key words: functional neuroimaging; resting state; disorders of consciousness; vegetative state

Introduction
In recent years, there has been a growing interest
from the neuroscientific community concerning spontaneous brain activity and its relation to cognition and
behavior. The concept of a default mode of brain
function arose from the need to explain consistent
brain-activity decreases in a set of areas during cognitive processing as compared to a passive resting baseline.1 These areas, encompassing the posterior cingulate cortex/precuneus, the medial prefrontal cortex,
and bilateral temporoparietal junctions, began to be

Address for correspondence: Melanie Boly, Cyclotron Research Center,


B30, Allee du 6 aout, Sart Tilman, 4000 Li`ege, Belgium.
mboly@student.ulg.ac.be

known as the default network. Furthermore, Raichle


et al.2 showed that most brain areas at rest manifest a
high level of default functional activity. This work
has called attention to the importance of intrinsic functional activity in assessing brain behavior relationships,
and has now been extended in several functional magnetic resonance imaging (fMRI) studies.
An ongoing controversy concerns the value and interpretability of resting-state studies and their contribution to a better understanding of brainbehavior
relationships.3,4 It has been suggested that intrinsic
brain activity would have a limited role for behavioral outcomes. In this view, observations made under resting conditions have no privileged status as a
fundamental metric of brain functioning, and the link
between the processing taking place at rest and its
physiology would be one without direct relevance to

C 2008 New York Academy of Sciences.


Ann. N.Y. Acad. Sci. 1129: 119129 (2008). 
doi: 10.1196/annals.1417.015
119

120

neuroscience. In contrast, the aims of cognitive neuroscience would be best served by the study of specific
task manipulations, rather than of rest.3
In response to this criticism, Raichle and Snyder5
argued that there is likely much more to brain function than that revealed by experiments manipulating
momentary demands of the environment. In their
view, a first argument in this direction is the cost of intrinsic brain activity, which far exceeds that of evoked
activity.6 Indeed, relative to the high rate of ongoing
or basal brain metabolism,6 the amount dedicated
to task-evoked regional imaging signals is remarkably
small (estimated to be less than 5%). The brain continuously expends a considerable amount of energy, even
in the absence of a particular task (i.e., when a subject is
awake and at rest). A significant fraction of the energy
consumed by the brain (quite possibly the majority) has
been shown to be a result of functionally significant
spontaneous neuronal activity.7 From this cost-based
analysis of brain functional activity, it seems reasonable
to conclude that intrinsic activity may be as significant,
if not more so, than evoked activity in terms of overall brain function.6 Another argument for the interest
of studying spontaneous brain activity is the striking
degree of functional organization exhibited by this intrinsic activity.5 The first clue of this organization is the
consistent activity decreases in default network during
cognitive tasks.1 Even more striking data recently arose
from fMRI blood oxygen leveldependent (BOLD)
connectivity studies in awake resting subjects, which
will be discussed later in this chapter. Maps of spontaneous network correlations have also been proposed
to provide tools for functional localization, for the understanding of clinical conditions such as Alzheimers
disease8,9 and autism,10 or for the study of comparative anatomy between primate species.11 However, the functional significance of the observed patterns of intrinsic brain activity remains actually poorly
understood.
We here propose that disorders of consciousness
are a privileged way to investigate the links between spontaneous brain activity and behavior. These
states are indeed mainly characterized by the alteration of intrinsic brain activity, which induces dramatic changes in the contents of awareness and
responses to environmental stimuli and demands.
We will illustrate our view, reviewing common features of spontaneous brain-activity patterns in altered states of consciousness, as shown by metabolic
positron emission tomography (PET) data as well
as recent BOLD fMRI studies. We will also discuss
methodological issues of resting-state neuroimaging
experiments.

Annals of the New York Academy of Sciences

Consciousness as a Multidimensional
Concept
Consciousness has two major components: awareness (i.e., the content of consciousness) and arousal (i.e.,
the level of consciousness).12 Arousal and awareness
are usually positively correlated: when your arousal
decreases, so does your awareness [rapid eye movement (REM) sleep being a notable exception]. Awareness can also be divided into two components: selfawareness and external awareness. Self- and external
awareness usually behave in an anti-correlated manner. When you are engaged in self-related processes,
you are less receptive to environmental demands, and
vice versa.13,14 A number of studies have compared
brain activation in circumstances that do or do not give
rise to consciousness in either of its two main senses of
awareness and arousal. Very few groups, however, have
studied situations in which wakefulness and arousal are
dissociated.
The vegetative state (VS) is a classic example of a
dissociated state of unconsciousness. VS patients are
fully aroused, but are unaware of themselves and their
environment. They can show automatic reactions like
moving their eyes, head, and limbs in a meaningless manner, and may even grimace, cry, or smile (albeit never contingently upon specific external stimuli).
Some patients might evolve toward full recovery or remain in the minimally conscious state,15 where some
nonreflexive or nonmeaningful behaviors are shown,
but patients are still unable to communicate. In addition to their clinical and ethical importance, the study
of vegetative and minimally conscious states offers a
still widely unexploited means of studying human consciousness.12 In contrast to other unconscious states,
such as general anesthesia and deep sleep, where impairment in arousal cannot be disentangled from impairment in awareness, these states represent a unique
lesional approach enabling us to identify the neural
correlates of (un)awareness.

PET Studies of Brain Metabolism in


Altered States of Consciousness
PET studies modulating arousal, and hence awareness, by means of anaesthetic drugs such as halothane16
or propofol17 have shown a drop in global brain
metabolism to around half of normal values. Similar
global decreases in metabolic activity are observed in
deep slow-wave sleep,18 although in rapid eye movement (REM) sleep brain metabolism returns to normal
waking values. On average, grey-matter metabolism

Boly et al.: Intrinsic Brain Activity in Altered States of Consciousness

FIGURE 1. Relationships between global brain


metabolism and awareness. The link between global brain
energy consumption and awareness is complex. Evidence
exists that both states of extremely low and extremely high
global brain metabolism are associated with small amounts
of awareness. An intermediate level of brain metabolism,
corresponding to a proper balance between inhibitory and
excitatory neural activity, seems to be necessary to allow
the genesis of awareness.

is 5070% of the normal range in comatose patients


of traumatic or hypoxic origin.19 In VS, that is, in
arousal without awareness, global brain metabolic
activity also decreases to about 50% of normal levels.20,21 As vegetative patients are fully aroused, global
brain metabolism seems to correlate with awareness
rather than with arousal in altered consciousness states.
However, contradictory data exist concerning the
positive correlation between global brain metabolism
and levels of consciousness. First, not all anesthetics suppress global cerebral metabolism. Some studies have also reported that ketamine, a so-called dissociative anesthetic agent, increases global cerebral
metabolism and fast EEG rhythms at doses associated
with a loss of consciousness.22,23 In the same line, in
comatose patients with traumatic diffuse axonal injury,
hyperglycolysis, leading to increased brain metabolism,
has sometimes been reported.19 Another counterexample is the loss of consciousness induced by generalized epilepsia24 and some absence seizures,25 where
global brain metabolism is diffusely increased. Finally,
a similar return from abnormally high global brain
metabolism to a normal balance of activation and inhibition could be evoked in rare cases of zolpidemevoked (a GABAergic agent) clinical improvement in
VS patients.26
Looking at these data as a whole (summarized in
FIG. 1), the complex relationships between global brain
metabolism and awareness could be linked to Tononis
theory of information integration.27,28 This theory
claims that consciousness is reflected in a systems capacity to integrate information, and proposes a way
the value to measure such a capacity.28 Computer
simulation shows that is very low in the case of an hy-

121

perpolarized state of the system, is maximized in conditions of intermediate neural activity, and decreases
in states where the neural activity is extremely high
and near-synchronous.27 It has been proposed that a
proper balance between excitatory and inhibitory activity would be necessary to allow neurons to respond
appropriately to correlational changes in their input,
and to establish the functional connectivity as required
for a particular cognitive task or behavior.29 In the
same line, Raichle and Gusnard suggested that a large
part of the brains default activity could be devoted to
ongoing synaptic processes associated with the maintenance of this balance.30 In this view, global brain
metabolism would have a potentially decisive role by
allowing the presence of conscious perception or behavior.
The equivocal link between global brain activity and
consciousness is, however, further challenged by the
fact that in some patients who subsequently recovered
from a VS to normal consciousness, global metabolic
rates for glucose metabolism did not show substantial
changes.31 Moreover, some awake healthy volunteers
have global brain metabolism values comparable to
those observed in some patients in a VS.12 Inversely,
some well-documented vegetative patients have shown
close to normal global cortical metabolism.32 These
data led us to focus rather on regional metabolism
in our quest for a better understanding of the links
between consciousness and resting-brain activity.
Voxel-based statistical analyses have sought to identify regions showing metabolic dysfunction in VS patients as compared with the conscious resting state
in healthy controls. These studies have identified a
systematic metabolic dysfunction, not in one brain
region but in a wide frontoparietal network encompassing the polymodal associative cortices in
VS: lateral and medial frontal regions bilaterally, parietotemporal and posterior parietal areas bilaterally, posterior cingulated, and precuneal
cortices,12,21 known to be the most active by
default in resting nonstimulated conditions.33 In contrast, arousal structures (encompassing the pedunculopontine reticular formation, the hypothalamus, and
the basal forebrain) are relatively preserved in these
patients.19 The same frontoparietal functional impairment is found in various other states of unconsciousness, that is, in sleep,34 coma,35 general
anesthesia,36 generalized seizures,37 or in other dissociated unconscious states like absence seizures,38,39
complex partial seizures,40 or somnambulism.41
FIGURE 2 illustrates the involvement of the frontoparietal cortical network in awareness, while arousal rather
relies on subcortical structures.

122

Annals of the New York Academy of Sciences

FIGURE 2. (Left ) Consciousness has two main components: arousal, or the level of consciousness, and awareness,
corresponding to the contents of consciousness per se. Arousal and awareness are usually positively correlated. However,
they involve different brain structures. Arousal involves the activity of subcortical structures encompassing brain-stem reticular
formation, hypothalamus, and basal forebrain. Awareness is related to the activity of a widespread set of frontoparietal
associative areas, both on the convexity and on the midline. (Right ) Awareness can in turn be divided into two main
components: self and external awareness. In healthy volunteers, self- and external awareness are usually negatively
correlated. Similarly, the frontoparietal awareness network can in turn be divided into two sub-systems, involved in selfand external awareness. Self-awareness networks encompass the posterior cingulate/precuneal cortices, medial frontal
cortex, and bilateral temporoparietal junctions. The external awareness network encompasses lateral frontal and parietal
cortices. In healthy volunteers, self- and external awareness networks usually show an anticorrelated pattern of activity.

These findings emphasize the importance of frontoparietal association areas in consciousness, and are
in line with the global workspace theory as introduced
by Baars.42,43 This theory views the brain as a massive parallel set of specialized processors. Consciousness might be a gateway to brain integration, enabling
access between otherwise separate neuronal functions.
In such a system, coordination and control may take
place by way of a central information exchange, allowing some processorssuch as sensory systems in
the brainto distribute information to the system as
a whole. According to Baars,35 frontoparietal association areas would be an ideal candidate for being the
global workspace processor.
As reported below, awareness can in turn be divided
in two main components: self- and external awareness. In the same line, FIGURE 2 illustrates that frontoparietal network can be subdivided in areas involved
in external awareness, and in self-awareness. External
awareness network activity is crucial for conscious external stimuli perception, as documented in healthy
awake volunteers.13,44 Self-awareness network encompasses the so-called default network and has been
involved in various aspects of self-related processes.45
In awake healthy volunteers, self- and external awareness networks usually show an anticorrelated pattern
of activity. Anticorrelations between self- and external
awareness networks have indeed been observed during cognitive tasks,46 sensory perception,13 as well as
in studies of resting-state brain activity.45 In contrast, in
most states of altered consciousness, both the activity
of subnetworks is similarly impaired.

In addition to activity in frontoparietal network,


awareness seems also to relate to the functional
connectivity within this network, and with the thalami. Functional disconnections in long-range cortico
cortical (between laterofrontal and midline-posterior
areas) and corticothalamocortical (between nonspecific thalamic nuclei and lateral and medial frontal
cortices) pathways have been identified in the vegetative state.21,47 In the same line, disruptions of
thalamocortical48 and corticocortical49 connectivity have been reported during other unconscious states
like sleep or anesthesia. Moreover, recovery from VS
is accompanied by a functional restoration of the frontoparietal network31 and some of its corticothalamo
cortical connections.47
These results are in line with Dehaene and
Changeuxs recent computational model of the relationships between spontaneous brain activity and
external stimuli awareness.50 This model emphasizes
the importance of both thalamocortical and cortico
cortical cerebral connections to create patterns of spontaneous brain activity hypothesized to allow conscious
perception.

Methodological Considerations in the


Study of Spontaneous Brain Activity
Using Functional Magnetic Rersonance
Up to now, a large majority of functional neuroimaging focused on brain activity elicited by external stimuli or evoked responses. Likewise, current

Boly et al.: Intrinsic Brain Activity in Altered States of Consciousness

mathematical tools have been mainly devised for the


analysis of data acquired during external stimulation
and not for spontaneous activity. Since the late 1990s,
brain activity fluctuations in the default resting state
have received increasing interest. Analytical tools have
therefore been developed for the processing of spontaneous fMRI (and electroencephalographic) data.
Spontaneous brain activity is by definition not triggered by external stimuli. We therefore have no control
and, importantly, no a priori knowledge about when a
spontaneous event occurs. The problem of analyzing
spontaneous data is thus twofold, as we must simultaneously determine the where and when of brain
activity. Knowing one or the other, that is, location or
timing, allows a more complete characterization of the
spontaneous activity. Most methods rely on this idea,
that is, make assumptions about the timing or location
of some activity pattern to determine the location or
timing, respectively, of the activated brain system.
The BOLD signal recorded with fMRI allows the
mapping of brain activity with good spatial resolution
(submillimetric) but poor timing (around 1 s, at best),
due to the physiological origin of the hemodynamic
signal. Unfortunately, the signal recorded also contains
some noise on top of the brain signal. This noise component has typically two origins: the scanner itself (such
as scanner instability)51 and nonneural physiological
fluctuations due to, for example, cardiac or respiratory
artefacts.52,53 Before investigating spontaneous brain
activity, it is necessary to correct the fMRI data for
these artefacts. One option to account for part of the
noise is to use a high sampling rate. With a very short
repetition time, the higher-frequency spurious (mainly
cardiac and respiratory) signal will not be aliased and
can be directly filtered out.5456 There are technical
limitations, though, as a compromise must be found
between speed of acquisition, field of view, and spatial
resolution. High-pass filtering is the method of choice
to remove the slowly varying scanner drift signal but,
of course, as for any filtering method, this removes possible information carrying signals. Alternatively, linear
regression can be employed to remove nonneural signal from the data.52,57 For example, if physiological
parameters are measured along side the fMRI acquisition, any BOLD signal correlated with (functions of)
these measurements can be regressed out. Other regressors generated directly from the fMRI data are
also possible, such as the mean BOLD signal over all
voxels (usually called global), or the BOLD signal
from areas where there should be little or no neural
activity (e.g., the ventricles or white matter). As hinted
by its name, linear regression assumes a linear relationship between the noise regressor and the noise

123

part of the signal in all the voxels. If, by chance, some


neural signal was also correlated with the noise regressor, that part of the signal would be lost for further
analysis.
Finally, independent component analysis (ICA)
a relatively new approachis capable of directly separating the signals of interest (due to brain activity) from
the noise.58,59 This approach is discussed later on in
more detail. The goal of the described procedures is
to ensure that the further analyzed signal is neurobiologically meaningful and corresponds to the spontaneous brain activity of interest. We now focus on ways
to identify and characterize patterns of spontaneous
activity.
The most straightforward approach is correlation
or functional connectivity analyses. After choosing
a seed region, that is, a region of interest, the time
course of the BOLD signal is extracted (averaged over
the region or the first principal component) and a correlation coefficient is calculated for all the other voxels, providing a correlation map. This method has the
advantage of being simple, sensitive, and easily interpretable, but is limited to one seed region at a time.45,60
Results rely heavily on the a priori choice of seed region
and provide no information about the causality of the
observed correlated activation. Indeed the activity in
two disconnected areas can be correlated because they
are driven by a third independent area. To study the
interaction between two seed regions, physiophysiological interaction models are useful.61 These models provide evidence for the interaction between distributed brain systems: voxels whose correlation with
one seed region is modulated by the other seed region
are highlighted.
A more mathematically sophisticated approach to
analyze spontaneous fMRI data is the previously mentioned ICA. ICA is a data-driven blind source separation algorithm that tries to decompose the entire
data set into components, spatial and temporal, that
are statistically independent.6264 The way statistical
independence is defined and reached leads to different
flavors of ICA decompositions. The main advantage
of ICA is the direct extractions of spatial maps, with
their associated time course: the sources of interest,
that is, spontaneous brain activity, can be automatically separated from the noise components. However,
there remain two major difficulties with ICA. First,
the number of components to be extracted has to
be defined a priori, and results are highly dependent
on that chosen number. Second, components are not
ranked during the decomposition. It is the investigators duty to manually and subjectively decide, based
on his or her experience, knowledge, or priors, which

124

Annals of the New York Academy of Sciences

components correspond to noise or neural systems. Solutions to these practical problems have been proposed,
for example, the probabilistic ICA.59
So far, we have only considered fMRI recordings,
but the electroencephalogram (EEG) is more and more
routinely recorded alongside fMRI to study spontaneous brain activity.65 Importantly, EEG data provide
access to very useful information regarding the timing of spontaneous brain activity. Features can be detected in the EEG signal and used to build an activation regressor for fMRI. For example, during sleep
studies, typical waves (e.g., slow waves or spindles; or
epileptic spikes) are easily detected on the EEG trace
and a spontaneous hemodynamic event is associated
with each occurrence of such wave. The analysis of the
fMRI data can then proceed as usual in stimulus induced tasks.6668 EEG data can also be processed to
yield a continuous regressor associated with spontaneous brain dynamics. For example, correlation between the BOLD signal of each voxel and the EEG
power in a specific frequency band, convoluted with
the standard hemodynamic response function, provides a correlation map, similar to what is done with the
seed-region activity. Typically, the spectrogram, that is,
the power spectrum evolving over time, of some or all
EEG channels is calculated and resampled at the fMRI
acquisition frequency. Then the time course of power
within frequency bands of interest is used to build a
correlations map.69,70

Functional Magnetic Resonance


Imaging Resting-State Studies in
Awake Healthy Subjects
Even in the absence of sensory inputs, structured
patterns of ongoing spontaneous activity can be observed in cortical and thalamic neurons.7173 In parallel, recent fMRI studies have identified spontaneous
fluctuations in neural activity in the resting human
brain. These slow BOLD fluctuations (in the range
of 0.1 Hz) are not random but coherent within specific neuroanatomical systems. Biswal and colleagues60
were the first to describe correlations between the activity of bilateral somatomotor cortices in the awake
resting human brain. The finding that spontaneous
fluctuations in the fMRI BOLD signal at rest in one
area of the cerebral cortex exhibited system-relevant
correlations with signal fluctuations in other areas
has then been replicated several times for motor cortices54,7476 and extended to other neuroanatomical
systems, including visual,54,77 auditory,77 default-mode
network,45,7880 memory,57,81 language,77,82 and atten-

tion systems.80,83 Similar results were derived from


other methods like hierarchical clustering84,85 and
ICA.56,58,63,64,86 The joint finding of these studies is
that regions similarly modulated by tasks or stimuli
tend to exhibit correlated spontaneous fluctuations
even in the absence of these tasks or stimuli.7 Restingstate fMRI patterns have also been shown to be spatially very consistent across subjects.87
In parallel, other resting-state fMRI studies showed
anticorrelated patterns of spontaneous fluctuations, in
regions with apparent opposing functionality. In particular, two independent studies45,79 recently showed
that even in the absence of any task or behavior, in
the so-called conscious resting state of the human
brain, two networks very similar to self- and external
awareness networks show a pattern of anticorrelated
activity (illustrated in FIG. 3). It has been suggested
that these anticorrelations could be a reflection of periodical shifts from introspective or self-oriented processes into a state-of-mind of extrospectively oriented
attention, and an engagement of networks that support sensorimotor planning.79 Another recent fMRI
study88 investigated the positive and negative correlations of three regions of interest (ROIs) located in the
auditory, visual, and somatosensory systems by using
resting-state fMRI. They found that all three sensory
systems exhibited significant negative correlation with
the default network (self-awareness or intrinsic system). This study extends former findings by indicating
that multiple subsystems rather than a single subsystem of the extrinsic system are inherently negatively
correlated with the self-awareness network. These negative correlations may explain the phenomenon that
externally and internally oriented processes can always disturb or even interrupt each other. These data
resemble our recent findings of a competitive effect
between self-awareness network activity and conscious
somatosensory stimuli perception.13
To date, only a few studies combined EEG and
fMRI data to better characterize spontaneous brain
activity fluctuations in the awake resting state. A simultaneous EEG/fMRI study showed a strong negative correlation of parietal and frontal cortical activity
with spontaneous fluctuations in EEG alpha power
(812 Hz).69 Beta activity was shown to be positively
correlated with activity in retrosplenial, temporoparietal, and dorsomedial prefrontal cortices, in the default
network.80 These data were interpreted as alpha oscillations signaling a neural baseline with inattention,
whereas beta rhythms index spontaneous cognitive operations during conscious rest. Conversely, a recent
EEG/fMRI study showed that each fMRI resting-state

Boly et al.: Intrinsic Brain Activity in Altered States of Consciousness

125

FIGURE 3. Spontaneous anticorrelations between self- and external awareness networks in the conscious resting state, as observed in an individual volunteer. (Left ) Areas correlated (above) and anticorrelated (below ) with the blood oxygen leveldependent (BOLD) time course of a seed voxel located in
the posterior cingulate/precuneus. (Right ) Plot of the BOLD time courses of posterior cingulate/precuneus
(PCC, red/gray line) and of middle frontal gyrus (MFG, blue/dark line) in the same volunteer. As previously reported, anticorrelations between these area time courses occur in slow frequencies with a period
below 0.1 Hz. (In color in Annals online.)

network as identified by ICA was associated not to


only one electrophysiological frequency, but to a coalescence of several brain rhythms in the delta, theta,
alpha, beta, and gamma ranges.89 Furthermore, each
functional network was shown to be characterized by
a specific electrophysiological signature that involved
the combination of these different brain rhythms. This
neurophysiological signature was suggested to constitute a baseline for evaluating changes in oscillatory
signals during active behavior.

Functional Magnetic Resonance


Imaging Resting-state Studies in
Altered States of Consciousness
It has been suggested that coherent spontaneous
BOLD fluctuations observed in the resting state reflect
unconstrained but consciously directed mental activity.3 One could argue that intrinsic activity simply represents unconstrained, spontaneous cognition, mindwandering, or stimulus-independent thoughts.5,90 Alternatively, coherent BOLD fluctuations may persist
in the absence of normal perception and behavior, reflecting a more fundamental or intrinsic property of
functional brain organization.91 Importantly, the former view predicts that coherent BOLD fluctuations
should be absent in coma, sleep, or deep anesthesia,
in which conscious mental activity is thought to be
absent.

Peltier et al.92 assessed the effect of sevoflurane anesthesia on the temporal BOLD correlations in activity
in the motor cortices of healthy humans. Across all
volunteers, they found that the number of significant
voxels in the functional connectivity maps was reduced
by 78% for light anesthesia and by 98% for deep anesthesia, compared with the awake state. Additionally,
significant correlations in the connectivity maps were
bilateral in the awake state, but unilateral in the light
anesthesia state. Interestingly, this loss of interhemispheric connectivity was also found in an independent
resting-state fMRI study on a minimally conscious patient compared to healthy volunteers.85
In contrast to these findings, recent data from several independent BOLD fMRIs suggest that lowfrequency systemwide BOLD coherent spontaneous
activity can be preserved in various states of unconsciousness. First, low-frequency BOLD fluctuations
have recently been investigated using ICA during light
sleep in humans.93,94 In this work (collapsing nonREM sleep stages 1 and 2), significant increases in the
fluctuation level of the BOLD signal were observed
in several cortical areas, among which visual cortex
was the most significant.93 Furthermore, correlations
among brain regions involved in the default network
(encompassing posterior cingulate/precuneus, medial
frontal cortex, and bilateral temporoparietal junctions) persisted during light non-REM sleep.94 Vincent
et al.91 demonstrated in deeply isoflurane-anesthetised

126

Annals of the New York Academy of Sciences

FIGURE 4. Preserved coherent blood oxygen leveldependent (BOLD) oscillations in the default
network persist in three documented states of unawareness. Brain areas showing correlations with a seed
voxel in the posterior cingulate cortex, after correction for spurious variance as described in Reference
91. From the left to the right, results of 12 volunteers random-effect analysis, from an individual sleeping
volunteer scanned during sleep stage 2 (from Ref. 68), from a patient in coma due to a nontraumatic
origin, and anaesthetized monkey data (reproduced by permission from Vincent et al.91 ). Sleep and
coma patients were masked inclusively with healthy volunteers results to check for spatial consistency of
the resting-state connectivity patterns.

monkeys preserved and coherent resting-state spontaneous fluctuations within three well-known neuroanatomical systems (oculomotor, somatomotor, and
visual) and within a network very close to the human
default system (see FIG. 4), a set of brain regions
thought by some to support uniquely human capabilities. These results demonstrate that cortical systems
previously associated with performance in sensory, motor, and/or cognitive tasks are manifest in the correlation structure of spontaneous BOLD fluctuations
observed in the absence of normal perception or behavior. Finally, using a method similar to that used in
Vincent et al.,91 we could identify persisting coherent
BOLD oscillations within the default-mode network
in coma, and during stage 2 slow-wave sleep (FIG. 4,
unpublished results).
All these results indicate that coherent systemwide
fluctuations probably reflect an aspect of brain
functional organization that transcends levels of consciousness.91 Thus, coherent spontaneous BOLD
fluctuations cannot be exclusively a reflection of
conscious mental activity,3 but may reflect a more
fundamental or intrinsic property of functional brain
organization. They should be considered as certainly
necessary, but not sufficient to support consciousness.
One could argue that the temporal dynamics of our
ongoing stream of consciousness (classically considered around 500 ms95 ) is much faster than the slow
fMRI BOLD oscillations occurring at around the 10-s
time period (0.1 Hz) observed here.
The physiological origin and functional significance
of low-frequency spontaneous brain activity fluctu-

ations remain to be assessed. Even if at least part


of these systemwide default interactions correspond
to unconscious processes, these fluctuations are likely
to shape brain responses to environmental demands
and to ongoingly modulate perception and behavior. Systemwide correlations in the absence of consciousness could also be seen as reflecting preserved
anatomical connections dissociated from higher cognitive functions. According to the hypothesis of a
tight correlate between low-frequency BOLD fluctuations and neuroanatomical connectivity,7 resting-state
fMRI data would also be likely to bring prognostic
information in acute brain-damaged patients. Further
studies correlating diffusion tensor imaging measures
to slow BOLD correlations are ongoing to test this
hypothesis.

Conclusion
Even if states of extremely low or high brain activity
are often associated with unconsciousness, the precise
link between global brain metabolism and awareness
remains difficult to assert. On the contrary, regional
brain activity in a widespread frontoparietal associative
network has been shown to be systematically altered in
all documented states of unconsciousness. In line with
studies in awake volunteers, these data emphasize the
potential role of frontoparietal association cortices in
the genesis of awareness.
Recent functional MRI studies have identified
coherent low-frequency fluctuations among welldocumented neuroanatomical networks. We, however,

Boly et al.: Intrinsic Brain Activity in Altered States of Consciousness

showed that these correlations can be similarly found


in three documented states of unawareness, namely,
sleep, coma, and deep anesthesia. We conclude that
the presence of slow BOLD fluctuations is unlikely to
merely reflect ongoing changes in the contents of consciousness and may be related to a more basic principle
of brain function.

9.

10.

11.
Acknowledgments

This work was supported by grants from the Belgian Fonds National de la Recherche Scientifique
(FNRS) and from the Centre Hospitalier Universitaire Sart Tilman, the University of Li`ege, the Mind
Science Foundation, the European Comission, the
French Speaking Community Concerted Research Action, and the Fondation Medicale Reine Elisabeth.
M.B. and T.D.V., C.P., S.L., and P.M. are, respectively, Research Fellows, Research Associate, Senior
Research Associate, and Research Director at FNRS.
M.S. was supported by an Austrian Science Fund
Erwin-Schrodinger Fellowship J2470-B02 (to M.S.).
We thank Dimitri Haye, Jacques Trantsieaux, and
Charlemagne Noukoua for their assistance in acquiring the fMRI data.

12.

13.

14.
15.
16.

17.

18.
Competing Interest

The authors declare no competing interest.

References
1. SHULMAN, G.L. et al. 1997. Common blood flow changes
across visual tasks: II. Decreases in cerebral cortex. J.
Cogn. Neurosci. 9: 648663.
2. RAICHLE, M.E. et al. 2001. A default mode of brain function.
Proc. Natl. Acad. Sci. USA 98: 676682.
3. MORCOM, A.M. & P.C. FLETCHER. 2007. Does the brain
have a baseline? Why we should be resisting a rest. NeuroImage 37: 10731082.
4. MORCOM, A.M. & P.C. FLETCHER. 2007. Cognitive neuroscience: the case for design rather than default. NeuroImage 37: 10971099.
5. RAICHLE, M.E. & A.Z. SNYDER. 2007. A default mode of
brain function: a brief history of an evolving idea. NeuroImage 37: 10831090.
6. RAICHLE, M.E. & M.A. MINTUN. 2006. Brain work and
brain imaging. Annu. Rev. Neurosci. 29: 449476.
7. FOX, M.D. & M.E. RAICHLE. 2007. Spontaneous fluctuations in brain activity observed with functional magnetic
resonance imaging. Nat. Rev. Neurosci. 8: 700711.
8. GREICIUS, M.D. et al. 2004. Default-mode network activity
distinguishes Alzheimers disease from healthy aging: evi-

19.

20.

21.

22.

23.

24.

25.

26.

27.
28.

127

dence from functional MRI. Proc. Natl. Acad. Sci. USA


101: 46374642.
HE, Y. et al. 2007. Regional coherence changes in the early
stages of Alzheimers disease: a combined structural and
resting-state functional MRI study. NeuroImage 35: 488
500.
CHERKASSKY, V.L. et al. 2006. Functional connectivity in a
baseline resting-state network in autism. Neuroreport 17:
16871690.
BUCKNER, R.L. & J.L. VINCENT. 2007. Unrest at rest: Default
activity and spontaneous network correlations. NeuroImage 37: 10911096.
LAUREYS, S. 2005. The neural correlate of (un)awareness:
lessons from the vegetative state. Trends Cogn. Sci. 9:
556559.
BOLY, M. et al. 2007. Baseline brain activity fluctuations
predict somatosensory perception in humans. Proc. Natl.
Acad. Sci. USA 104: 1218712192.
DUVAL, S. & R. WICKLUND. 1972. A Theory of Objective
Self-awareness. Academy Press. New York.
GIACINO, J.T. et al. 2002. The minimally conscious state:
definition and diagnostic criteria. Neurology 58: 349353.
ALKIRE, M.T. et al. 1999. Functional brain imaging during
anesthesia in humans: effects of halothane on global and
regional cerebral glucose metabolism. Anesthesiology 90:
701709.
ALKIRE, M.T. et al. 1995. Cerebral metabolism during propofol anesthesia in humans studied with positron emission
tomography. Anesthesiology 82: 393403.
MAQUET, P. et al. 1997. Functional neuroanatomy of human
slow wave sleep. J. Neurosci. 17: 28072812.
LAUREYS, S., A.M. OWEN & N.D. SCHIFF. 2004. Brain function in coma, vegetative state, and related disorders. Lancet
Neurol. 3: 537546.
SCHIFF, N.D. et al. 2002. Residual cerebral activity and behavioural fragments can remain in the persistently vegetative brain. Brain 125: 12101234.
LAUREYS, S. et al. 1999. Impaired effective cortical connectivity in vegetative state: preliminary investigation using
PET. NeuroImage 9: 377382.
ITOH, T. et al. 2005. Effects of anesthesia upon 18F-FDG
uptake in rhesus monkey brains. Ann. Nucl. Med. 19:
373377.
MAKSIMOW, A. et al. 2006. Increase in high frequency EEG
activity explains the poor performance of EEG spectral
entropy monitor during S-ketamine anesthesia. Clin. Neurophysiol 117: 16601668.
BLUMENFELD, H. 2005. Consciousness and epilepsy: Why
are patients with absence seizures absent? Prog. Brain Res.
150: 271286.
ENGEL, J., JR. et al. 1985. Local cerebral metabolic rate for
glucose during petit mal absences. Ann. Neurol. 17: 121
128.
CLAUSS, R. & W. NEL. 2006. Drug induced arousal from
the permanent vegetative state. NeuroRehabilitation 21:
2328.
TONONI, G. 2005. Consciousness, information integration,
and the brain. Prog. Brain Res. 150: 109126.
TONONI, G. 2004. An information integration theory of consciousness. BMC Neurosci. 5: 42.

128
29. SALINAS, E. & T.J. SEJNOWSKI. 2001. Correlated neuronal
activity and the flow of neural information. Nat. Rev. Neurosci. 2: 539550.
30. RAICHLE, M.E. & D.A. GUSNARD. 2002. Appraising the
brains energy budget. Proc. Natl. Acad. Sci. USA 99:
1023710239.
31. LAUREYS, S. et al. 1999. Cerebral metabolism during vegetative state and after recovery to consciousness. J. Neurol.
Neurosurg. Psychiatry 67: 121.
32. SCHIFF, N.D. et al. 2002. Residual cerebral activity and behavioural fragments can remain in the persistently vegetative brain. Brain 125: 12101234.
33. GUSNARD, D.A. & M.E. RAICHLE. 2001. Searching for a
baseline: functional imaging and the resting human brain.
Nat. Rev. Neurosci. 2: 685694.
34. MAQUET, P. 2000. Functional neuroimaging of normal human sleep by positron emission tomography. J. Sleep Res.
9: 207231.
35. BAARS, B.J., T.Z. RAMSOY & S. LAUREYS. 2003. Brain, conscious experience and the observing self. Trends Neurosci.
26: 671675.
36. KAISTI, K.K. et al. 2002. Effects of surgical levels of propofol and sevoflurane anesthesia on cerebral blood flow in
healthy subjects studied with positron emission tomography. Anesthesiology 96: 13581370.
37. BLUMENFELD, H. et al. 2003. Selective frontal, parietal, and
temporal networks in generalized seizures. NeuroImage
19: 15561566.
38. SALEK-HADDADI, A. et al. 2003. Functional magnetic resonance imaging of human absence seizures. Ann. Neurol.
53: 663667.
39. LAUFS, H. et al. 2006. Linking generalized spike-andwave discharges and resting state brain activity by using
EEG/fMRI in a patient with absence seizures. Epilepsia
47: 444448.
40. BLUMENFELD, H. et al. 2004. Positive and negative network
correlations in temporal lobe epilepsy. Cereb. Cortex. 14:
892902.
41. BASSETTI, C. et al. 2000. SPECT during sleepwalking. Lancet
356: 484485.
42. BAARS, B.J. 1988. A Cognitive Theory of Consciousness.
Cambridge University Press. Cambridge, UK.
43. BAARS, B.J. 2002. The conscious access hypothesis: origins
and recent evidence. Trends Cogn. Sci. 6: 4752.
44. DEHAENE, S. et al. 2001. Cerebral mechanisms of word masking and unconscious repetition priming. Nat. Neurosci. 4:
752758.
45. FOX, M.D. et al. 2005. The human brain is intrinsically organized into dynamic, anticorrelated functional networks.
Proc. Natl. Acad. Sci. USA 102: 96739678.
46. GUSNARD, D.A., M.E. RAICHLE & M.E. RAICHLE. 2001.
Searching for a baseline: functional imaging and the resting human brain. Nat. Rev. Neurosci. 2: 685694.
47. LAUREYS, S. et al. 2000. Restoration of thalamocortical connectivity after recovery from persistent vegetative state.
Lancet 355: 17901791.
48. WHITE, N.S. & M.T. ALKIRE. 2003. Impaired thalamocortical connectivity in humans during general-anestheticinduced unconsciousness. NeuroImage 19: 402
411.

Annals of the New York Academy of Sciences


49. MASSIMINI, M. et al. 2005. Breakdown of cortical effective
connectivity during sleep. Science 309: 22282232.
50. DEHAENE, S. & J.P. CHANGEUX. 2005. Ongoing spontaneous
activity controls access to consciousness: a neuronal model
for inattentional blindness. PLoS Biol. 3: e141.
51. ZARAHN, E., G.K. AGUIRRE & M. DESPOSITO. 1997. Empirical analyses of BOLD fMRI statistics. I. Spatially unsmoothed data collected under null-hypothesis conditions.
NeuroImage 5: 179197.
52. BIRN, R.M. et al. 2006. Separating respiratory-variationrelated fluctuations from neuronal-activity-related fluctuations in fMRI. NeuroImage 31: 15361548.
53. LUND, T.E. et al. 2006. Non-white noise in fMRI: Does modelling have an impact? NeuroImage 29: 5466.
54. LOWE, M.J., B.J. MOCK & J.A. SORENSON. 1998. Functional
connectivity in single and multislice echoplanar imaging
using resting-state fluctuations. NeuroImage 7: 119132.
55. KIVINIEMI, V., J. RUOHONEN & O. TERVONEN. 2005. Separation of physiological very low frequency fluctuation from
aliasing by switched sampling interval fMRI scans. Magn.
Reson. Imaging 23: 4146.
56. DE LUCA, M. et al. 2006. fMRI resting state networks define
distinct modes of long-distance interactions in the human
brain. NeuroImage 29: 13591367.
57. ROMBOUTS, S.A. et al. 2003. Identifying confounds to increase specificity during a no task condition. Evidence
for hippocampal connectivity using fMRI. NeuroImage
20: 12361245.
58. KIVINIEMI, V. et al. 2003. Independent component analysis
of nondeterministic fMRI signal sources. NeuroImage 19:
253260.
59. BECKMANN, C.F. & S.M. SMITH. 2004. Probabilistic independent component analysis for functional magnetic resonance imaging. IEEE Trans. Med. Imaging 23: 137152.
60. BISWAL, B. et al. 1995. Functional connectivity in the motor
cortex of resting human brain using echo-planar MRI.
Magn. Reson. Med. 34: 537541.
61. FRISTON, K.J. et al. 1997. Psychophysiological and modulatory interactions in neuroimaging. NeuroImage 6: 218
229.
62. MCKEOWN, M.J. et al. 1998. Analysis of fMRI data by blind
separation into independent spatial components. Hum.
Brain Mapp. 6: 160188.
63. VAN DE VEN, V.G. et al. 2004. Functional connectivity as
revealed by spatial independent component analysis of
fMRI measurements during rest. Hum. Brain Mapp. 22:
165178.
64. BECKMANN, C.F. et al. 2005. Investigations into resting-state
connectivity using independent component analysis. Philos. Trans. R. Soc. Lond. B Biol. Sci. 360: 10011013.
65. SALEK-HADDADI, A. et al. 2003. Studying spontaneous EEG
activity with fMRI. Brain Res. Brain Res. Rev. 43: 110
133.
66. LEMIEUX, L. et al. 2001. Event-related fMRI with simultaneous and continuous EEG: description of the method and
initial case report. NeuroImage 14: 780787.
67. BAGSHAW, A.P. et al. 2004. EEG-fMRI of focal epileptic spikes: analysis with multiple haemodynamic functions and comparison with gadolinium-enhanced MR angiograms. Hum. Brain Mapp. 22: 179192.

Boly et al.: Intrinsic Brain Activity in Altered States of Consciousness


68. SCHABUS, M. et al. 2007. Hemodynamic cerebral correlates
of sleep spindles during human non-rapid eye movement
sleep. Proc. Natl. Acad. Sci. USA 104: 1316413169.
69. LAUFS, H. et al. 2003. EEG-correlated fMRI of human alpha
activity. NeuroImage 19: 14631476.
70. CZISCH, M. et al. 2004. Functional MRI during sleep: BOLD
signal decreases and their electrophysiological correlates.
Eur. J. Neurosci. 20: 566574.
71. STERIADE, M., D.A. MCCORMICK & T.J. SEJNOWSKI. 1993.
Thalamocortical oscillations in the sleeping and aroused
brain. Science 262: 679685.
72. TSODYKS, M. et al. 1999. Linking spontaneous activity of
single cortical neurons and the underlying functional architecture. Science 286: 19431946.
73. KENET, T. et al. 2003. Spontaneously emerging cortical representations of visual attributes. Nature 425: 954956.
74. XIONG, J. et al. 1999. Interregional connectivity to primary
motor cortex revealed using MRI resting state images.
Hum. Brain Mapp. 8: 151156.
75. FOX, M.D. et al. 2006. Coherent spontaneous activity accounts for trial-to-trial variability in human evoked brain
responses. Nat. Neurosci. 9: 2325.
76. DE LUCA, M. et al. 2005. Blood oxygenation level dependent
contrast resting state networks are relevant to functional
activity in the neocortical sensorimotor system. Exp. Brain
Res. 167: 587594.
77. CORDES, D. et al. 2000. Mapping functionally related regions
of brain with functional connectivity MR imaging. AJNR
Am. J. Neuroradiol. 21: 16361644.
78. GREICIUS, M.D. et al. 2003. Functional connectivity in the
resting brain: a network analysis of the default mode hypothesis. Proc. Natl. Acad. Sci. USA 100: 253258.
79. FRANSSON, P. 2005. Spontaneous low-frequency BOLD signal fluctuations: an fMRI investigation of the resting-state
default mode of brain function hypothesis. Hum. Brain
Mapp. 26: 1529.
80. LAUFS, H. et al. 2003. Electroencephalographic signatures
of attentional and cognitive default modes in spontaneous
brain activity fluctuations at rest. Proc. Natl. Acad. Sci.
USA 100: 1105311058.
81. VINCENT, J.L. et al. 2006. Coherent spontaneous activity identifies a hippocampal-parietal memory network. J.
Neurophysiol. 96: 35173531.

129

82. HAMPSON, M. et al. 2002. Detection of functional connectivity using temporal correlations in MR images. Hum.
Brain Mapp. 15: 247262.
83. FOX, M.D. et al. 2006. Spontaneous neuronal activity distinguishes human dorsal and ventral attention systems. Proc.
Natl. Acad. Sci. USA 103: 1004610051.
84. CORDES, D. et al. 2002. Hierarchical clustering to measure
connectivity in fMRI resting-state data. Magn. Reson.
Imaging 20: 305317.
85. SALVADOR, R. et al. 2005. Neurophysiological architecture
of functional magnetic resonance images of human brain.
Cereb. Cortex. 15: 13321342.
86. BARTELS, A. & S. ZEKI. 2005. The chronoarchitecture of the
cerebral cortex. Philos. Trans. R Soc. Lond. B Biol. Sci.
360: 733750.
87. DAMOISEAUX, J.S. et al. 2006. Consistent resting-state networks across healthy subjects. Proc. Natl. Acad. Sci. USA
103: 1384813853.
88. TIAN, L. et al. 2007. The relationship within and between
the extrinsic and intrinsic systems indicated by resting state
correlational patterns of sensory cortices. NeuroImage 36:
684690.
89. MANTINI, D. et al. 2007. Electrophysiological signatures of
resting state networks in the human brain. Proc. Natl.
Acad. Sci. USA 104: 1317013175.
90. MASON, M.F. et al. 2007. Wandering minds: the default
network and stimulus-independent thought. Science 315:
393395.
91. VINCENT, J.L. et al. 2007. Intrinsic functional architecture
in the anaesthetized monkey brain. Nature 447: 83
86.
92. PELTIER, S.J. et al. 2005. Functional connectivity changes
with concentration of sevoflurane anesthesia. Neuroreport
16: 285288.
93. FUKUNAGA, M. et al. 2006. Large-amplitude, spatially correlated fluctuations in BOLD fMRI signals during extended
rest and early sleep stages. Magn. Reson. Imaging 24:
979992.
94. HOROVITZ, S.G. et al. 2007. Low frequency BOLD fluctuations during resting wakefulness and light sleep: a simultaneous EEG-fMRI study. Hum. Brain Mapp. (In press.)
95. LIBET, B. 2006. Reflections on the interaction of the mind
and brain. Prog. Neurobiol. 78: 322326.

REVIEWS

The neuroscience of mindfulness


meditation
YiYuan Tang1,2*, Britta K.Hlzel3,4* and Michael I.Posner2

Abstract | Research over the past two decades broadly supports the claim that mindfulness
meditation practiced widely for the reduction of stress and promotion of health
exertsbeneficial effects on physical and mental health, and cognitive performance. Recent
neuroimaging studies have begun to uncover the brain areas and networks that mediate
these positive effects. However, the underlying neural mechanisms remain unclear, and it is
apparent that more methodologically rigorous studies are required if we are to gain a full
understanding of the neuronal and molecular bases of the changes in the brain that
accompany mindfulness meditation.
Longitudinal studies
Study designs that compare
data from one or more groups
at several time points and that
ideally include a (preferably
active) control condition and
random assignment to
conditions.

Cross-sectional studies
Study designs that compare
data from an experimental
group with those from a control
group at one point in time.

1
Department of Psychological
Sciences, Texas Tech
University, Lubbock, Texas
79409, USA.
2
Department of Psychology,
University of Oregon, Eugene,
Oregon 97403, USA.
3
Department of
Neuroradiology, Technical
University of Munich, 81675
Munich, Germany.
4
Massachusetts General
Hospital, Charlestown,
Massachusetts 02129, USA.
*These authors contributed
equally to this work.
Correspondence to Y.Y.T.
email: yiyuan.tang@ttu.edu
doi:10.1038/nrn3916
Published online
18 March 2015

Meditation can be defined as a form of mental training


that aims to improve an individuals core psychological
capacities, such as attentional and emotional self-regulation. Meditation encompasses a family of complex
practices that include mindfulness meditation, mantra
meditation, yoga, tai chi and chi gong 1. Of these practices, mindfulness meditation often described as nonjudgemental attention to present-moment experiences
(BOX1) has received most attention in neuroscience
research over the past two decades28.
Although meditation research is in its infancy, a
number of studies have investigated changes in brain
activation (at rest and during specific tasks) that are
associated with the practice of, or that follow, training
in mindfulness meditation. These studies have reported
changes in multiple aspects of mental function in beginner and advanced meditators, healthy individuals and
patient populations914.
In this Review, we consider the current state of
research on mindfulness meditation. We discuss the
methodological challenges that the field faces and point
to several shortcomings in existing studies. Taking into
account some important theoretical considerations, we
then discuss behavioural and neuroscientific findings in
light of what we think are the core components of meditation practice: attention control, emotion regulation
and self-awareness (BOX1). Within this framework, we
describe research that has revealed changes in behaviour,
brain activity and brain structure following mindfulness
meditation training. We discuss what has been learned
so far from this research and suggest new research
strategies for the field. We focus here on mindfulness
meditation practices and have excluded studies on other

types of meditation. However, it is important to note that


other styles of meditation may operate via distinct neural
mechanisms15,16.

Challenges in meditation research


Findings on the effects of meditation on the brain are
often reported enthusiastically by the media and used by
clinicians and educators to inform their work. However,
most of the findings have not yet been replicated. Many
researchers are enthusiastic meditators themselves.
Although their insider perspective may be valuable for a
deep understanding of meditation, these researchers must
ensure that they take a critical view of study outcomes. In
fact, for meditation studies there is a relatively strong bias
towards the publication of positive or significant results,
as was shown in a meta-analysis17.
The methodological quality of many meditation
research studies is still relatively low. Few are actively
controlled longitudinal studies, and sample sizes are small.
As is typical for a young research field, many experiments
are not yet based on elaborated theories, and conclusions
are often drawn from post-hoc interpretations. These
conclusions therefore remain tentative, and studies must
be carefully replicated. Meditation research also faces
several specific methodological challenges.
Cross-sectional versus longitudinal studies. Early meditation studies were mostly cross-sectional studies: that is,
they compared data from a group of meditators with
data from a control group at one point in time. These
studies investigated practitioners with hundreds or
thousands of hours of meditation experience (such as
Buddhist monks) and compared them with control

NATURE REVIEWS | NEUROSCIENCE

VOLUME 16 | APRIL 2015 | 213


2015 Macmillan Publishers Limited. All rights reserved

REVIEWS
Box 1 | Mindfulness meditation
Different styles and forms of meditation are found in almost all cultures and religions. Mindfulness meditation originally
stems from Buddhist meditation traditions3. Since the 1990s, mindfulness meditation has been applied to multiple mental
and physical health conditions, and has received much attention in psychological research2,47. In current clinical and
research contexts, mindfulness meditation is typically described as non-judgemental attention to experiences in the
present moment38. This definition encompasses the Buddhist concepts of mindfulness and equanimity158 and describes
practices that require both the regulation of attention (in order to maintain the focus on immediate experiences, such as
thoughts, emotions, body posture and sensations) and the ability to approach ones experiences with openness and
acceptance210,53,158161. Mindfulness meditation can be subdivided into methods involving focused attention and those
involving open monitoring of present-moment experience9.
The mindfulness practices that have been the subject of neuroscientific research comprise a broad range of methods
and techniques, including Buddhist meditation traditions, such as Vipassana meditation, Dzogchen and Zen, as well as
mindfulness-based approaches such as integrative bodymind training (IBMT), mindfulness-based stress reduction
(MBSR) and clinical interventions based on MBSR46. Both
MBSR and IBMT have adopted mindfulness practices from a
Mindfulness meditation
the Buddhist traditions and aim to develop
moment-to-moment, non-judgemental awareness through
various techniques8-11,53,73. IBMT has been categorized in
Attention
Emotion
Selfthe literature as open-monitoring mindfulness
control
regulation
awareness
9,10,161
meditation
, whereas MBSR includes both focused
attention and open-monitoring practices8.
It has been suggested that mindfulness meditation
Self-regulation
includes at least three components that interact closely to
constitute a process of enhanced self-regulation:
enhanced attention control, improved emotion regulation b
and altered self-awareness (diminished self-referential
Early stage
Middle stage
Advanced stage
processing and enhanced body awareness)10 (see the
figure, part a). Mindfulness meditation can be roughly
Eort to
Eortful
Eortless
divided into three different stages of practice early,
reduce mind
doing
being
middle (intermediate) and advanced that involve
wandering
different amounts of effort11 (see the figure, part b).

Correlational studies
Studies that assess the
co-variation between two
variables: for example,
co-variation of functional or
structural properties of the
brain and a behavioural
variable, such as reported
stress.

groups of non-meditators matched on various dimensions9,18. The rationale was that any effects of meditation
would be most easily detectable in highly experienced
practitioners.
A number of cross-sectional studies revealed differences in brain structure and function associated with
meditation (see below). Although these differences may
constitute training-induced effects, a cross-sectional
study design precludes causal attribution: it is possible
that there are pre-existing differences in the brains of
meditators, which might be linked to their interest in
meditation, personality or temperament 2,19. Although
correlational studies have attempted to discover whether
more meditation experience is related to larger changes
in brain structure or function, such correlations still
cannot prove that meditation practice has caused the
changes because it is possible that individuals with these
particular brain characteristics may be drawn to longer
meditation practice.
More recent research has used longitudinal designs,
which compare data from one or more groups at several time points and ideally include a (preferably active)
control condition and random assignment to conditions1114,2025. In meditation research, longitudinal studies are still relatively rare. Among those studies, some
have investigated the effects of mindfulness training
over just a few days, whereas others have investigated
programmes of 1 to 3months. Some of these studies
have revealed changes in behaviour, brain structure and
function1114,2025. A lack of similar changes in the control

group suggests that meditation has caused the observed


changes, especially when other potentially confounding
variables are controlled for properly2022.
Novice meditators versus expert meditators. Although
most cross-sectional studies included long-term meditators9,17, longitudinal studies are often conducted in
beginners or naive subjects. Thus, differences in the
results of cross-sectional and longitudinal analyses
might be attributed to the different brain regions used
during learning of meditation versus those used during
the continued practice of an acquired skill. It would be
interesting to follow subjects over a long-term period
of practice to determine whether changes induced by
meditation training persist in the absence of continued
practice. However, such long-term longitudinal studies
would be compromised by feasibility constraints, and
it is likely that future longitudinal studies will remain
restricted to relatively short training periods2.
Control groups and interventions. It is important to control for variables that may be confounded with meditation
training, such as changes in lifestyle and diet that might
accompany the meditation practice or the expectancy
and intention that meditation beginners bring to their
practice. Researchers must carefully determine which
variables are integral aspects of the meditation training
and which can be controlled for. Some earlier studies only
controlled for the length of time that the individual has
practised meditation and the effects of repeated testing,

214 | APRIL 2015 | VOLUME 16

www.nature.com/reviews/neuro
2015 Macmillan Publishers Limited. All rights reserved

REVIEWS
Blood-oxygen-leveldependent contrasts
(BOLD contrasts). Signals that
can be extracted with
functional MRI and that reflect
the change in the amount of
deoxyhaemoglobin that is
induced by changes in the
activity of neurons and their
synapses in a region of the
brain. The signals thus reflect
the activity in a local brain
region.

Arterial spin labelling


(ASL). An MRI technique that is
capable of measuring cerebral
blood flow invivo. It provides
cerebral perfusion maps
without requiring the
administration of a contrast
agent or the use of ionizing
radiation because it uses
magnetically labelled
endogenous blood water as a
freely diffusible tracer.

Brain state
The reliable patterns of brain
activity that involve the
activation and/or connectivity
of multiple large-scale brain
networks.

Fractional anisotropy
A parameter in diffusion tensor
imaging, which images brain
structures by measuring the
diffusion properties of water
molecules. It provides
information about the
microstructural integrity of
white matter.

but more recent studies have developed and included


active interventions in control groups such as stress
management education26, relaxation training 14,23,27 or
health enhancement programmes2022 that can control
for variables such as social interaction with the group and
teachers, amount of home exercise, physical exercise and
psychoeducation. These studies are therefore better able
to extract and delineate the meditation-specific effects.
For example, one study investigating short-term meditation training used a sham meditation condition in which
participants thought they were meditating, but did not
receive proper meditation instructions, which allowed
the researchers to control for factors such as expectancy, body posture and attention from the teacher 28.
Mechanistic studies ideally need to use interventions that
are as effective as mindfulness meditation in producing
the beneficial effects on target variables but that allow
for assessment of the unique mechanism underlying the
mindfulness practice23,29.
Control conditions in functional imaging. Although all
functional neuroimaging studies must use appropriate
comparison conditions, this challenge is particularly
important when imaging meditative states (BOX2). The
comparison condition should be one in which a state
of mindfulness meditation is not present. Many studies
use resting comparison conditions, but a problem with
this is that experienced practitioners are likely to enter
into a state of meditation when at rest. However, other
active tasks introduce additional brain activity that renders the comparison difficult to interpret. Using imaging
protocols that do not rely on blood-oxygen-level-dependent
contrasts (BOLD contrasts), such as arterial spin labelling,
might be a possible solution for this problem30.

Changes in brain structure


In the past decade, 21 studies have investigated alterations in brain morphometry related to mindfulness
meditation17. These studies varied in regard to the
exact mindfulness meditation tradition under investigation3134,3652, and multiple measurements have been

Box 2 | Imaging the meditative state


A brain state can be defined as a reliable pattern of activity and/or connectivity in
multiple large-scale brain networks11,73. Meditation training involves obtaining a
meditative state, and measurements of behaviour and/or brain activity can be made
while participants are thought to be in such a state15,76,162,163. These studies can elucidate
how the state influences the brain and behaviour2,10,14,73. To identify brain regions
activated during the state of meditation (compared to a baseline state) across multiple
studies in experienced healthy meditators, an activation likelihood estimate
meta-analysis of 10 studies with 91 subjects published before January 2011 was
performed108. This revealed three areas in which there were clusters of activity: the
caudate, which is thought (together with the putamen) to have a role in attentional
disengagement from irrelevant information, allowing a meditative state to be achieved
and maintained; the entorhinal cortex (parahippocampus), which is thought to control
the mental stream of thoughts and possibly stop mind wandering; and the medial
prefrontal cortex, which is thought to support the enhanced self-awareness during
meditation108 (also see REF.162). It was suggested that these regions of activity might
represent a core cortical network for the meditative state, independent of the
meditation technique. It is important to note, however, that this meta-analysis included
mostly papers from traditions other than mindfulness.

used to investigate effects on both grey and white matter.


Studies have captured cortical thickness32,51, grey-matter
volume and/or density 33,40, fractional anisotropy and axial
and radial diffusivity38,39. These studies have also used different research designs. Most have made cross-sectional
comparisons between experienced meditators and controls3234; however, a few recent studies have investigated
longitudinal changes in novice practitioners3840. Some
further studies have investigated correlations between
brain changes and other variables related to mindfulness
practice, such as stress reduction40, emotion regulation39
or increased well-being 47. Most studies include small
sample sizes of between 10 and 34 subjects per group31-42.
Because the studies vary in regard to study design,
measurement and type of mindfulness meditation, it is
not surprising that the locations of reported effects are
diverse and cover multiple regions in the brain3134,3652.
Effects reported by individual studies have been found
in multiple brain regions, including the cerebral cortex,
subcortical grey and white matter, brain stem and cerebellum, suggesting that the effects of meditation might
involve large-scale brain networks. This is not surprising
because mindfulness practice involves multiple aspects
of mental function that use multiple complex interactive
networks in the brain. TABLE1 summarizes the main findings of structural neuroimaging studies on mindfulness
meditation (grey and white matter).
An activation likelihood estimation meta-analysis, which
also included studies from traditions other than mindfulness meditation, was conducted to investigate which
regions were consistently altered in meditators across
studies17. The findings demonstrated a global medium
effect size, and eight brain regions were found to be
consistently altered in meditators: the frontopolar
cortex, which the authors suggest might be related to
enhanced meta-awareness following meditation practice; the sensory cortices and insula, areas that have
been related to body awareness; the hippocampus, a
region that has been related to memory processes; the
anterior cingulate cortex (ACC), mid-cingulate cortex
and orbitofrontal cortex, areas known to be related to
self and emotion regulation; and the superior longitudinal fasciculus and corpus callosum, areas involved in
intra- and inter-hemispherical communication17.
Thus, some initial attempts have been undertaken to
investigate the brain regions that are structurally altered
by the practice of meditation. However, our knowledge
of what these changes actually mean will remain trivial
until we gain a better understanding of how such structural changes are related to the reported improvements
in affective, cognitive and social function. Very few
studies have begun to relate findings in the brain to selfreported variables and behavioural measures34,39,47,48,51.
Future studies therefore need to replicate the reported
findings and begin to unravel how changes in the neural
structure relate to changes in well-being and behaviour.
Growing evidence also demonstrates changes in the
functional properties of the brain following meditation.
Below, we summarize such findings in the context of
the framework of core mechanisms of mindfulness
meditation (BOX1; FIG.1).

NATURE REVIEWS | NEUROSCIENCE

VOLUME 16 | APRIL 2015 | 215


2015 Macmillan Publishers Limited. All rights reserved

REVIEWS
Table 1 | Structural changes in the brain associated with mindfulness meditation
Meditation
tradition*

Control

Sample size of
meditation (M) and
control (C) groups

Type of measurement

Key areas affected

Refs

Cross-sectional studies (non-clinical samples)


Insight

Non-meditators

M: 20, C: 15

Cortical thickness

Right anterior insula and right middle and


superior frontal sulci

32

Zen

Non-meditators

M: 13, C: 13

Grey-matter volume

Less age-related decline at left putamen

34

Insight

Non-meditators

M: 20, C: 20

Grey-matter density

Right anterior insula, left inferior temporal gyrus


and right hippocampus

31

Tibetan
Dzogchen

Non-meditators

M: 10, C: 10

Grey-matter density

Medulla oblongata, left superior and inferior


frontal gyrus, anterior lobe of the cerebellum
(bilateral) and left fusiform gyrus

33

Zen

Non-meditators

M: 17, C: 18

Cortical thickness

Right dorsal anterior cingulate cortex and


secondary somatosensory cortices (bilateral)

51

MBSR

Non-meditators

M: 20, C: 16

Grey-matter volume

Left caudate nucleus

52

Zen

Non-meditators

M: 10, C: 10

DTI: mean diffusivity and


fractional anisotropy

Lower mean diffusivity in left posterior parietal


white matter and lower fractional anisotropy in
left primary sensorimotor cortex grey matter

37

Longitudinal studies (non-clinical samples)


IBMT
(4weeks)

Active control:
relaxation training

M: 22, C: 23

DTI: FA and grey-matter


volume

FA increased for left anterior corona radiata,


superior corona radiata (bilateral), left superior
longitudinal fasciculus, genu and body of corpus
callosum. No effect on grey-matter volume

38

MBSR

Individuals on a
waiting list

M: 16, C: 17

Grey-matter density

Left hippocampus, left posterior cingulate gyrus,


cerebellum and left middle temporal gyrus

40

IBMT
(2weeks)

Active control:
relaxation training

M: 34, C: 34

DTI: FA, radial diffusivity


and axial diffusivity

Decrease of axial diffusivity in corpus callosum,


corona radiata, superior longitudinal fasciculus,
posterior thalamic radiation and sagittal striatum

39

Longitudinal studies (clinical samples)


MBI

Usual care
(patients with
Parkinson disease)

M: 14, C: 13

Grey-matter density

Caudate (bilateral), left inferior temporal lobe,


hippocampus (bilateral), left occipital cuneus
and other small clusters; anterior cerebellum
increased in usual care group

42

MBSR

Waiting list
(patients with
mild cognitive
impairment)

M: 8, C: 5

Hippocampal volume
(region of interest analysis)

Trend towards less hippocampal atrophy

41

DTI, diffusion tensor imaging; FA, fractional anisotropy; IBMT, integrative bodymind training; MBI, mindfulness-based intervention; MBSR, mindfulness-based
stress reduction. *Studies that include meditators from traditions other than mindfulness or studies only investigating correlations with other variables are not
listed. Meditators show increased values, unless otherwise noted.

Axial and radial diffusivity


Derived from the eigenvalues
of the diffusion tensor, their
underlying biophysical
properties are associated with
axonal density and
myelination, respectively.

Activation likelihood
estimation meta-analysis
A technique for
coordinate-based
meta-analysis of neuroimaging
data. It determines the
convergence of foci reported
from different experiments,
weighted by the number of
participants in each study.

Mindfulness and attention


Many meditation traditions emphasize the necessity to
cultivate attention regulation early in the practice9,53.
A sufficient degree of attentional control is required to
stay engaged in meditation, and meditators often report
improved attention control as an effect of repeated practice10,14. Multiple studies have experimentally investigated
such effects54.
Components of attention. Attention is often subdivided
into three different components: alerting (readiness in
preparation for an impending stimulus, which includes
tonic effects that result from spending time on a task
(vigilance) and phasic effects that are due to brain
changes induced by warning signals or targets); orienting
(the selection of specific information from multiple sensory stimuli); and conflict monitoring (monitoring and
resolution of conflict between computations in different

neural areas, also referred to as executive attention)55,56.


Other distinctions between types of attention refer to
combinations of these three components. For example,
sustained attention refers to the sense of vigilance during
long continued tasks and may involve both tonic alerting
and orienting, whereas selective attention may involve
either orienting (when a stimulus is present) or executive
function (when stored information is involved).
Performance in these three basic domains can be
measured with the attention network test (ANT)57. This
test uses as a target an arrow pointing left or right. The
target is surrounded by flankers, and subtracting reaction times to congruent stimuli (that is, those on the side
of the screen indicated by the arrow) from reaction times
to incongruent stimuli produces a measure of the time to
resolve conflict. The inclusion of cues that indicate when
or where the target will occur allows the measurement
of alerting and orienting. These measures are used to

216 | APRIL 2015 | VOLUME 16

www.nature.com/reviews/neuro
2015 Macmillan Publishers Limited. All rights reserved

REVIEWS
Posterior cingulate
cortex/precuneus

Anterior cingulate
cortex
Multiple
prefrontal
regions
Medial
prefrontal
cortex
Striatum
Amygdala

Insula

Medial view

Lateral view

Figure 1 | Brain regions involved in the components of mindfulnessmeditation.


Schematic view of some of the brain regions involved in attention control (the anterior
cingulate cortex and the striatum), emotion regulation (multiple prefrontal regions, limbic
regions and the striatum) and self-awareness (the insula, medial prefrontal cortex and
posterior cingulate cortex and precuneus).

quantify efficiency in each of the three networks that


support the individual components of attention. Alerting
involves the brains noradrenaline system, which originates in the locus coeruleus. Orienting involves frontal
and parietal areas, including the frontal eye fields and
inferior and superior parietal lobe. The executive network involved in conflict resolution involves the ACC,
anterior insula and basal ganglia58,59.
Effects of mindfulness meditation on attention. The
ANT and other experimental paradigms have been
used to investigate the effects of meditation on attentional performance60. Improved conflict monitoring was
reported in several studies14,6164. For example, a longitudinal study showed that only 5days (20 min per day) of
integrative bodymind training (IBMT) led to improved
conflict monitoring 14. In addition, cross-sectional studies of 3months of mindfulness meditation showed a
reduced attentional blink (a lapse in attention following
a stimulus within a rapid stream of presented stimuli
that has been related to executive function65,66) following training 64 (also see REF.67). Better performance in
conflict monitoring has also been demonstrated in experienced meditators in cross-sectional studies68. However,
although altered attention is a common finding in these
well-designed meditation studies, some studies investigating mindfulness-based stress reduction (MBSR) have
not observed effects on conflict monitoring 69,70.
Most of the studies on the effects of short-term
(1week) mindfulness meditation on alerting have
not found significant effects, but studies investigating
long-term meditators (ranging from months to years)
did detect changes in alerting 27,7072. Enhanced orienting
has been reported in some cross-sectional studies using
longer periods of training. For example, 3months of
Shamatha mindfulness training improved tonic alertness (the ability to remain alert over time) and allowed
for improved orienting towards a visual target in comparison to controls71. However, 8weeks of MBSR did not
improve measures of sustained attention in a continuous

performance task that measured aspects of tonic alertness,


but did show some improvement in orienting 22. We do
not know whether the differences in the findings of these
studies are due to the type of training, type of control or
other subtle factors.
A systematic review that compiled the findings of
these studies (as well as the effects on other measures of
cognition) concluded that early phases of mindfulness
meditation might be associated with improvements in
conflict monitoring and orienting, whereas later phases
might be mainly associated with improved alerting 60. It is
currently still unclear how different meditation practices
differentially affect the specific attentional components2,9,53.
In addition, the length of practice needs to be defined more
consistently in future research.
Neural mechanisms of enhanced attention control.
Several functional and structural MRI studies on mindfulness training have investigated neuroplasticity in brain
regions supporting attention regulation. The brain region
to which the effects of mindfulness training on attention is most consistently linked is the ACC11,23,38,39,7376.
The ACC enables executive attention and control7779 by
detecting the presence of conflicts emerging from incompatible streams of information processing. The ACC and
the fronto-insular cortex form part of a network that
facilitates cognitive processing through long-range connections to other brain areas11,80. Cross-sectional studies
have reported enhanced activation of regions of the ACC
in experienced meditators compared to controls during
focused attention meditation76 or when mindfully anticipating delivery of a painful stimulus81. Greater activation
of the ventral and/or rostral ACC during the resting state
following 5days of IBMT was also found in an actively
controlled, randomized, longitudinal study 23. Although
ACC activation may be enhanced in earlier stages of
mindfulness meditation, it might decrease with higher
levels of expertise, as demonstrated in a cross-sectional
study 18. Structural MRI data suggest that mindfulness
meditation might be associated with greater cortical
thickness51 and might lead to enhanced white-matter
integrity in the ACC38,39.
Other attention-related brain regions in which functional changes have been observed following mindfulness meditation include the dorsolateral prefrontal cortex
(PFC), where responses were enhanced during executive
processing 82, as revealed by a randomized longitudinal study, and parietal attention regions, which showed
greater activation following an MBSR course in people
with social anxiety, as demonstrated by an uncontrolled
longitudinal study 83. Furthermore, a diminished agerelated decline of grey-matter volume in the putamen as
well as diminished age-related decline in sustained attention performance were found in a cross-sectional study of
Zen meditation practitioners34.
Although there is evidence that brain regions relevant for the regulation of attention show functional and
structural changes following mindfulness meditation
practice, it has not yet been determined whether these
changes are actually related to the improved attentional
performance. Longitudinal studies that use measures

NATURE REVIEWS | NEUROSCIENCE

VOLUME 16 | APRIL 2015 | 217


2015 Macmillan Publishers Limited. All rights reserved

REVIEWS
of attentional performance along with functional MRI
(fMRI) are needed. If supported by more rigorous
future research, the evidence of improved attention regulation and strengthened brain activity in the regions
underlying attentional control following mindfulness
meditation might be promising for the treatment of
psychiatric disorders in which there are deficiencies in
these functions74,84,85.

Mindfulness and emotion regulation


Enhanced emotion regulation has been suggested to
underlie many of the beneficial effects of mindfulness
meditation. Emotion regulation refers to strategies that
can influence which emotions arise and when, how long
they occur, and how these emotions are experienced and
expressed. A range of implicit and explicit emotion regulation processes has been proposed86, and mindfulnessbased emotion regulation may involve a mix of these
processes, including attentional deployment (attending
to mental processes, including emotions), cognitive
change (altering typical patterns of appraisal regarding
ones emotions) and response modulation (decreasing
tonic levels of suppression).
Effects of mindfulness meditation on emotion regulation.
Improvements in emotion regulation associated with
mindfulness meditation have been investigated through
various approaches, including experimental studies, selfreporting studies, measurement of peripheral physiology
and neuroimaging 10. These studies have reported various
positive effects of mindfulness meditation on emotional
processing, such as a reduction in emotional interference
by unpleasant stimuli87, decreased physiological reactivity and facilitated return to emotional baseline after
response to a stressor film88, and decreased self-reported
difficulties in emotion regulation89. Consequently, lowered intensity and frequency of negative affect 90,91 and
improved positive mood states14,91,92 are reported to be
associated with mindfulness meditation.
Neural mechanisms of improved emotion regulation.
Neuroimaging studies that have probed the enhanced
emotion regulation associated with mindfulness meditation in an attempt to identify the underlying brain activation patterns typically present study participants with
emotional pictures82,9397, words and/or statements29,98
and instruct them to encounter these with a state of
mindfulness or a simple baselinestate.
The hypothesis that drives many of these studies is
that mindful emotion regulation works by strengthening prefrontal cognitive control mechanisms and thus
downregulates activity in regions relevant to affect processing, such as the amygdala. Present-moment awareness
and non-judgemental acceptance through mindfulness
meditation8,10 are thought to be crucial in promoting
cognitive control because they increase sensitivity to
affective cues that help to signal the need for control99.
Studies have therefore investigated whether mindfulness
training exerts its effects through enhanced top-down
control or facilitated bottom-up processing 100. The findings (outlined below) suggest that the level of expertise

is important, with beginners showing a different pattern from expert meditators. However, although several
studies have pointed to the involvement of fronto-limbic
regions, very few studies have begun to relate changes
in these regions to changes in measures of behaviour or
well-being 10.
A frequently reported finding is that mindfulness
practice leads to (or is associated with) a diminished activation of the amygdala in response to emotional stimuli
during mindful states83,94,95 as well as in a resting state93,
suggesting a decrease in emotional arousal. However,
although such results have been reported for meditation
beginners, they have less consistently been detected in
experienced meditators95 (but see REF.18).
Prefrontal activations are often enhanced as an effect
of mindfulness meditation in novice meditators (but see
REF.29): for example, greater dorsolateral PFC responses
were found during executive processing within an emotional Stroop task in healthy individuals after 6weeks
of mindfulness training 82. Enhanced dorsomedial and
dorsolateral PFC activation was also detected when participants expected to see negative images while engaging
in a mindful state94. Moreover, after an MBSR course, an
enhanced activation in the ventrolateral PFC in people
suffering from anxiety was found when they labelled the
affect of emotional images97. By contrast, experienced
meditators have been found to show diminished activation in medial PFC regions95. This finding could be
interpreted as indicating reduced control (disengagement
of elaboration and appraisal) and greater acceptance of
affectivestates.
Neuroimaging studies of ameliorated pain processing through mindfulness meditation have also pointed
to expertise-related differences in the extent of cognitive control over sensory experience. Meditation beginners showed increased activity in areas involved in the
cognitive regulation of nociceptive processing (the ACC
and anterior insula) and areas involved in reframing the
evaluation of stimuli (the orbitofrontal cortex), along with
reduced activation in the primary somatosensory cortex
in a 4-day longitudinal study with no control group30,
whereas meditation experts were characterized by
decreased activation in dorsolateral and ventrolateral PFC
regions and enhancements in primary pain processing
regions (the insula, somatosensory cortex and thalamus)
compared with controls in two cross-sectional studies35,81.
These findings are in line with the assumption that
the process of mindfulness meditation is characterized
as an active cognitive regulation in meditation beginners, who need to overcome habitual ways of internally
reacting to ones emotions and might therefore show
greater prefrontal activation. Expert meditators might
not use this prefrontal control. Rather, they might have
automated an accepting stance towards their experience
and thus no longer engage in top-down control efforts
but instead show enhanced bottom-up processing 100.
In the early stages of meditation training, achieving
the meditation state seems to involve the use of attentional control and mental effort; thus, areas of the lateral prefrontal and parietal cortex are more active than
before training 11,16,100,101. This may reflect the higher level

218 | APRIL 2015 | VOLUME 16

www.nature.com/reviews/neuro
2015 Macmillan Publishers Limited. All rights reserved

REVIEWS
of effort often found when participants struggle to obtain
the meditation state in the early stages11,73,98,102. However,
in the advanced stages, prefrontalparietal activity is
often reduced or eliminated, but ACC, striatum and
insula activity remains9,10,53,73,76,101103. Whether effort has
a key role in PFC and ACC activation during or following
meditation needs further investigation.
Analysis of functional connectivity between regions
of the fronto-limbic network could help to further elucidate the regulatory function of executive control regions.
Only a few studies have included such analyses. One
cross-sectional study on pain processing in meditators
demonstrated decreased connectivity of executive and
pain-related brain regions35, and one study of mindfulness-naive smokers demonstrated reduced connectivity
between craving-related brain regions during a mindfulness condition compared to passive viewing of smokingrelated images during cigarette craving 96, suggesting
a functional decoupling of involved regions. Another
longitudinal, randomized study reported that people
suffering from anxiety showed a change from a negative
correlation between the activity of frontal regions and
that of the amygdala before intervention (that is, negative
connectivity) to a positive correlation between the activity
of these regions (positive connectivity) after a mindfulness intervention97. Because such a negative correlation
will occur when prefrontal regions downregulate limbic
activation104,105, it was speculated that the positive coupling
between the activity of the two regions after mindfulness
intervention might indicate that meditation involves
monitoring of arousal rather than a downregulation or
suppression of emotional responses, and that it might
be a unique signature of mindful emotion regulation.

Box 3 | Mindfulness meditation as exposure therapy


Exposure therapy aims for patients to extinguish a fear response and instead to
acquire a sense of safety in the presence of a formerly feared stimulus by exposing
them to that stimulus and preventing the usual response164. Mindfulness meditation
resembles an exposure situation because practitioners turn towards their emotional
experience, bring acceptance to bodily and affective responses, and refrain from
engaging in internal reactivity towards it. Research on fear conditioning has helped to
identify a network of brain regions that are crucial for the extinction of conditioned
fear responses and the retention of extinction165. This network includes the
ventromedial prefrontal cortex (vmPFC), which is important for a successful recall of
the extinction; the hippocampus166, which is related to signalling the extinguished
context (contextual safety); and the amygdala, which has a crucial role during the
acquisition and expression of conditioned fear167 and is thought to be downregulated
by the vmPFC and the hippocampus105,168. Activation in the vmPFC (subgenual anterior
cingulate cortex) is primarily linked to the expression of fear learning during a delayed
test of extinction and is critical for the retention of extinction169.
There is emerging evidence from MRI studies that the aforementioned brain regions
show structural and functional changes following mindfulness meditation training
(see main text). This overlap of involved brain regions, as well as the conceptual
similarity between mindfulness and an exposure situation, suggest that mindfulness
training might enhance the ability to extinguish conditioned fear by structurally and
functionally affecting the brain network that supports safety signalling. The capacity
for successful extinction memories reliably differentiates healthy from pathological
conditions170,171, and is crucial in order to overcome maladaptive states. It helps
individuals to learn to have no fear response to neutral stimuli when there is no
adaptive function for a fear response. Instead, individuals can experience a sense of
safety and can flexibly elicit other emotions and behaviours.

Importantly, this study also investigated the correlation


between neural and self-reported findings and demonstrated that the changes in PFCamygdala connectivity
were correlated with anxiety symptom improvement.
Further studies are needed to elucidate the complex
interplay between regions of the fronto-limbic network
in mindfulness meditation.
Although the proposed similarities between mindfulness and the reappraisal strategy of emotion regulation have been much debated, there is some evidence
that mindfulness also shares similarities with extinction
processes (BOX3).
Brain regions involved in motivation and reward processing also show functional alterations that are related to
mindfulness training, such as stronger activity of the putamen and caudate during a resting state following mindfulness training23 and lower activation in the caudate nucleus
during reward anticipation in experienced meditators106.
These studies might indicate altered self-regulation in
the motivational realm, with possibly reduced susceptibility to incentives and enhanced reward-related activity
duringrest.
Brain regions involved in the regulation of emotions
have also shown structural changes following mindfulness meditation31,32,3841,48,51. Although these findings provide some initial evidence that these brain regions are
related to mindfulness practice, the question of whether
they are involved in mediating the beneficial effects of
mindfulness meditation remains largely unanswered.

Mindfulness and self-awareness


According to Buddhist philosophy, the identification with
a static concept of self causes psychological distress. Disidentification from such a static self-concept results in
the freedom to experience a more genuine way of being.
Through enhanced meta-awareness (making awareness
itself an object of attention), mindfulness meditation is
thought to facilitate a detachment from identification
with the self as a static entity 3,10,107 and a tendency to identify with the phenomenon of experiencing itself is said
to emerge15,108112. Currently, empirical research into this
area is only just emerging111,113, and the few interpretations
of connections between neuroimaging findings and selfreported data which we will summarize briefly below
are suggestive atbest.
Self-referential processing. Altered self-representation has
been investigated with questionnaire studies. Early studies reported mindfulness training to be associated with
a more positive self-representation, higher self-esteem,
higher acceptance of oneself 114 and styles of self-concept
that are typically associated with less-severe pathological
symptoms115. Meditators have also been shown to score
higher than non-meditators on a scale that measures
non-attachment 116: a construct that is based on insight
into the constructed and impermanent nature of mental
representations. Although such concepts are not easy to
capture in experimental and neuroscientific studies, findings from a few recent studies seem to suggest that brain
structures supporting self-referential processing might be
affected by mindfulness meditation98,117,118.

NATURE REVIEWS | NEUROSCIENCE

VOLUME 16 | APRIL 2015 | 219


2015 Macmillan Publishers Limited. All rights reserved

REVIEWS
Although there is much debate about its exact function,
a widespread view holds that the default mode network
(DMN)119,120 is involved in self-referential processing. This
network includes midline structures of the brain, such as
areas of the medial PFC, posterior cingulate cortex (PCC),
anterior precuneus and inferior parietal lobule121,122. These
structures show high activity during rest, mind wandering
and conditions of stimulus-independent thought 121 and
have been suggested to support diverse mechanisms by
which an individual can project themselves into another
perspective123. fMRI studies have investigated activity
in the DMN in association with mindfulness practice.
Regions of the DMN (the medial PFC and PCC) showed
relatively little activity in meditators compared to controls across different types of meditation, which has
been interpreted as indicating diminished self-referential
processing 117. Functional connectivity analysis revealed
stronger coupling in experienced meditators between the
PCC, dorsal ACC and dorsolateral PFC, both at baseline
and during meditation, which was interpreted as indicating increased cognitive control over the function of
the DMN117. Increased functional connectivity was also
found between DMN regions and the ventromedial PFC
in participants with more compared to less meditation
experience118. It has been speculated that this increased
connectivity with ventromedial PFC regions supports
greater access of the default circuitry to information about
internal states because this region is highly interconnected
with limbic regions118.
Awareness of present-moment experiences. Evaluative
self-referential processing is assumed to decrease as an
effect of mindfulness meditation, whereas awareness of
present-moment experiences is thought to be enhanced.
Mindfulness practitioners often report that the practice of
attending to present-moment body sensations results in
an enhanced awareness of bodily states and greater perceptual clarity of subtle interoception. Empirical findings
to support this claim are mixed. Although studies that
assessed performance on a heartbeat detection task a
standard measure of interoceptive awareness found
no evidence that meditators had superior performance to
non-meditators124,125, other studies found that meditators
showed greater coherence between objective physiological
data and their subjective experience in regard to an emotional experience126 and the sensitivity of body regions127.
Multiple studies have shown the insula to be implicated in mindfulness meditation: it shows stronger activation during compassion meditation128 and following
mindfulness training 23,52,98, and has greater cortical thickness in experienced meditators32. Given its known role in
awareness129, it is conceivable that enhanced insula activity in meditators might represent the amplified awareness
of present-moment experience.
Similarly, a study reported an uncoupling of the right
insula and medial PFC and increased connectivity of the
right insula with dorsolateral PFC regions in individuals
after mindfulness training 98. The authors interpret their
findings as a shift in self-referential processing towards
a more self-detached and objective analysis of interoceptive and exteroceptive sensory events, rather than their

affective or subjective self-referential value. Furthermore,


a preliminary analysis from a study of a state of non-dual
awareness (a state of awareness in which perceived dualities, such as the distinction between subject and object, are
absent) showed a decreased functional connectivity of the
central precuneus with the dorsolateral PFC. The author
speculates that this finding might be indicative of a state in
which awareness is itself the subject of awareness111.
Together, the findings from these studies have been
taken to suggest that mindfulness meditation might alter
the self-referential mode so that a previous narrative,
evaluative form of self-referential processing is replaced
by greater awareness98,111. We suggest that this shift in
self-awareness is one of the major active mechanisms
of the beneficial effects of mindfulness meditation.
However, because these interpretations are built on a
still-fragmentary understanding of the function of the
involved brain regions, future research will need to test
and elaborate these assumptions.
Across the functional and structural MRI studies that
have been published to date, especially those based on
the longitudinal, randomized, controlled studies with
active control groups and meta-analyses, the ACC, PFC,
PCC, insula, striatum (caudate and putamen) and amygdala seem to show consistent changes associated with
mindfulness meditation911,13,17,23,34,73,108,130 (FIG.1; TABLE2).
We consider these areas to be the core regions involved
in self-regulation of attention, emotion and awareness
following mindfulness training. However, we acknowledge that many other brain areas are also involved in
mindfulness practice and warrant further investigation
using rigorous randomized and controlled designs.

Future questions
Mechanisms of mindfulness-induced changes. A number of studies seem to suggest that mindfulness meditation induces changes in brain structure and function,
raising the question of which underlying mechanisms
support these processes. It is possible that engaging the
brain in mindfulness affects brain structure by inducing
dendritic branching, synaptogenesis, myelinogenesis or
even adult neurogenesis. Alternatively, it is possible that
mindfulness positively affects autonomic regulation and
immune activity, which may result in neuronal preservation, restoration and/or inhibition of apoptosis14,23,131.
It is well known that mindfulness-based techniques are
highly effective in stress reduction, and it is possible
that such stress reduction may mediate changes in brain
function14,48,132137 (BOX4). A combination of all of these
mechanisms may evenoccur.
It is also important to realize that the direction of
the observed effects of mindfulness meditation has
not been consistent across all studies. Although larger
values in meditators compared to controls are predominantly reported, a cross-sectional study also revealed
smaller fractional anisotropy and cortical thickness
values in meditators in some brain regions, including
the medial PFC, postcentral and inferior parietal cortices, PCC and medial occipital cortex 138. Along these
lines, mindfulness-induced increases are predominantly
observed in longitudinal studies. However, it was also

220 | APRIL 2015 | VOLUME 16

www.nature.com/reviews/neuro
2015 Macmillan Publishers Limited. All rights reserved

REVIEWS
Table 2 | Evidence for changes in the core brain regions after mindfulness meditation
Brain region

Study design

Findings*

ACC
(self-regulation of
attention and emotion)

Cross-sectional, Vipassana meditators (N = 15)


versus controls (N = 15)

Enhanced ACC activation during breath awareness (focused


attention) meditation

76

Longitudinal, IBMT versus active control


(relaxation training) (N = 23 each group)

Enhanced ACC activity in resting state

23

Greater dorsolateral PFC activation during emotional Stroop


executive processing

82

Longitudinal, patients with generalized anxiety


disorder, MBSR (N = 15) versus active control
(N = 11)

Enhanced activation of ventrolateral PFC, enhanced


connectivity of several PFC regions with amygdala

97

Longitudinal, uncontrolled, before and after


mindfulness training (N = 15)

Anxiety relief following mindfulness training was related to


ventromedial PFC and ACC activation (along with insula)

157

Cross-sectional, expert meditators (N = 12)


versus controls (N = 13)

PCC deactivation during different types of meditation, increased


coupling with ACC and dorsolateral PFC

117

Cross-sectional, expert meditators (N = 14)


divided into high and low practice groups

Reduced connectivity between left PCC and medial PFC and


ACC at rest in high practice group

118

Longitudinal, IBMT, active control (relaxation


training) (N = 23 each group)

Enhanced right PCC activity at resting state

23

Cross-sectional, MBSR (N = 20) and waiting list


control (N = 16)

Greater anterior insula activation and altered coupling between


dorsomedial PFC and posterior insula during interoceptive
attention to respiratory sensations

52

Cross-sectional, expert Tibetan Buddhist


meditators (N = 15) and novices (N = 15)

Enhanced insula activation when presented with emotional


sounds during compassion meditation

Longitudinal, IBMT, active control (relaxation


training) (N = 23 each group)

Enhanced left insula activity at resting state

23

Striatum
(regulation of attention
and emotion)

Longitudinal, IBMT, active control (relaxation


training) (N = 23 each group)

Enhanced caudate and putamen activity at resting state

23

Cross-sectional, expert meditators (N = 34) and


controls (N = 44)

Lower activation in the caudate nucleus during reward


anticipation

Amygdala
(emotional processing)

Longitudinal, mindful attention training (N = 12), Decreased activation in right amygdala in response to emotional
compassion training (N = 12) and active control pictures in a non-meditative state
(N = 12)

93

Longitudinal, uncontrolled, patients with social


anxiety disorder before and after MBSR (N = 14)

Diminished right dorsal amygdala activity during reacting to


negative self-belief statements

83

Cross-sectional, beginner (N = 10) and expert


Zen meditators (N = 12)

Downregulation of the left amygdala when viewing emotional


pictures in a mindful state in beginner but not expert meditators

95

PFC
Longitudinal, mindfulness training (N = 30)
(attention and emotion) versus active control (N = 31)

PCC
(self-awareness)

Insula
(awareness and
emotional processing)

Refs

128

106

Exemplary studies for each region support its involvement in mindfulness (the list is not comprehensive). Future research will need to test the hypothesized
functions by relating behavioural and neuroimaging findings. ACC, anterior cingulate cortex; IBMT, integrative bodymind training; MBSR, mindfulness-based
stress reduction; PCC, posterior cingulate cortex; PFC, prefrontal cortex. *Meditators show increased values, unless otherwise noted.

reported, for example, that as a consequence of meditation, larger decreases in perceived stress were associated with larger decreases in grey-matter density in the
amygdala48. Thus, the underlying mechanisms seem to
be more complex than currently assumed, and further
research is necessary.
Although neuroimaging has advanced our understanding of the individual brain regions involved in
mindfulness meditation, most evidence supports the
idea that the brain processes information through the
dynamic interactions of distributed areas operating in
large-scale networks139,140. Because the complex mental
state of mindfulness is probably supported by alterations
in large-scale brain networks, future work should consider the inclusion of complex network analyses, rather
than restricting analyses to comparisons of the strength
of activations in single brain areas. Recent studies have
explored functional network architecture during the
resting state using these new tools141,142.

Decoding mental states. Mindfulness meditation


approaches can be divided into those involving focused
attention and those involving open monitoring. Even
within the same meditation style, practitioners can be
at different stages of mindfulness practice2. Investigating
the distinction between these different stages in terms of
brain function will require new advanced tools and methods. For instance, simultaneous multi-level recording
using fMRI and electrophysiology could provide
information on how the brain and body interact to support the meditation practice143. Electroencephalography
feedback has been used to aid training and study meditation by providing the practitioners with information on
the brain waves they are producing. Similarly, real-time
fMRI has been used to provide subjects with feedback
of the brain activity they are producing and allows the
experimenter to examine pain, cognitive control, emotion regulation and learning of meditation. This dynamic
recording and feedback technique may help to train the

NATURE REVIEWS | NEUROSCIENCE

VOLUME 16 | APRIL 2015 | 221


2015 Macmillan Publishers Limited. All rights reserved

REVIEWS
Box 4 | Mindfulness meditation and stress
Stress reduction might be a potential mediator of the effects of mindfulness practice on
neural function. Mindfulness meditation has been shown to reduce stress14,132137; this is
most consistently documented in self-reported data132,133. A review of
mindfulness-based stress reduction (MBSR) studies showed a non-specific effect on
stress reduction, which is similar to that of standard relaxation training134. However,
findings in studies that have examined biomarkers of stress, such as cortisol levels, are
less consistent: changes in cortisol levels have been found in association with
mindfulness training in some studies14,136 but not in others132,135.
The brain is a target for stress and stress-related hormones. It undergoes functional
and structural remodelling in response to stress in a manner that is adaptive under
normal circumstances but can lead to damage when stress is excessive172. Evidence
suggests that vulnerability to stress-induced brain plasticity is prominent in the
prefrontal cortex (PFC), hippocampus, amygdala and other areas associated with
fear-related memories and self-regulatory behaviours172,173. The interactions between
these brain regions determine whether life experiences lead to successful adaptation
or maladaptation and impaired mental and physical health173. A study has shown that
chronic stress induces less flexibility in attention shifting in the rodent and human
adult174. This was paralleled by a reduction in apical dendritic arborization in rodent
medial PFC (specifically, in the anterior cingulate cortex) and fewer feedforward PFC
connections in humans under stress, effects that recovered when the stressor was
removed174. This suggests that the effects of chronic psychosocial stress on PFC
function and connectivity are plastic and can change quickly as a function of mental
state174. Studies have also shown that moderate to severe stress seems to increase the
volume of the amygdala but reduce the volume of the PFC and hippocampus175.
Mindfulness training, however, has been shown to enhance grey-matter density in the
hippocampus40. Furthermore, after mindfulness training, reductions in perceived stress
correlate with reductions in amygdala grey-matter density48. These findings suggest
that mindfulness meditation might be a potential intervention and prevention
strategy176. Thus, it is possible that mindfulness meditation reduces stress by improving
self-regulation, which enhances neuroplasticity and leads to health benefits. It should be
noted that mindfulness meditation might also directly modulate stress processing via a
bottom-up pathway, through which it alters the sympatheticadrenalmedullary and
hypothalamicpituitaryadrenal axes by increasing activity in the parasympathetic
nervous system; thus, mindfulness meditation could prevent sympathetic nervous
system fight-or-flight stress responses177,178. Indeed, some research has suggested that
mindfulness leads to increased activity of the parasympathetic nervous system23,179.
Brain-derived neurotrophic factor (BDNF) has been linked to numerous aspects of
plasticity in the brain. Stress-induced remodelling of the PFC, hippocampus and
amygdala coincides with changes in the levels of BDNF, supporting its role as a trophic
factor modulating neuronal survival and regulating synaptic plasticity131. However,
glucocorticoids and other molecules have been shown to act in conjunction with BDNF
to facilitate both morphological and molecular changes. Because some forms of
mindfulness meditation training have been found to reduce stress-induced cortisol
secretion, this could potentially have neuroprotective effects by increasing levels of
BDNF, and future research should explore this possible causal relationship136,149,180.

Multivariate pattern
analysis
A method of analysing
functional MRI data that is
capable of detecting and
characterizing information
represented in patterns of
activity distributed within and
across multiple regions of the
brain. Unlike univariate
approaches, which only
identify magnitudes of activity
in localized parts of the brain,
this approach can monitor
multiple areas at once.

subjects effectively and allow their mental states at different stages of mindfulness training to be decoded from
their brain activity 144146, possibly by applying techniques
such as multivariate pattern analysis147.
Interpretations of study outcomes remain tentative until they are clearly linked to subjective reports or
behavioural findings. Future studies should therefore
increasingly draw connections between behavioural
outcomes and neuroimaging data using the advanced
multi-level analyses mentionedabove.
Investigating individual differences. People respond to
mindfulness meditation differently. These differences
may derive from temperamental, personality or genetic
differences. Studies in other fields have suggested that
genetic polymorphisms may interact with experience to

influence the success of training 148. Because mindfulness


meditation affects the activation and connectivity of the
ACC, PFC and other brain regions involved in cognitive
control and emotion regulation, it might be helpful to
examine these polymorphisms to determine their possible influence on the success of meditation practice2,59,149.
Moreover, individual differences in personality, lifestyle,
life events and trainertrainee dynamics are likely to have
substantial influence on training effects, although little is
known about these influences. Mood and personality have
been used to predict individual variation in the improvement of creative performance following mindfulness
meditation150. Capturing temperament and personality
differences may serve to predict success in mindfulness
training 150,151 because different temperament and personality traits are reported to be associated with different
electroencephalography patterns and heart-rate variability
in Zen meditators152.
Clinical application. Self-regulation deficits are associated with diverse behavioural problems and mental disorders, such as increased risk of school failure, attention
deficit disorder, anxiety, depression and drug abuse78,153.
Convergent findings indicate that mindfulness meditation could ameliorate negative outcomes resulting from
deficits in self-regulation and could consequently help
patient populations suffering from diseases and behavioural abnormalities. Several clinical trials have explored
the effects of mindfulness meditation on disorders such as
depression154, generalized anxiety26, addictions155, attention
deficit disorders156 and others42, and have begun to establish the efficiency of mindfulness practice for these conditions. Only a few recent studies, however, have investigated
the neuroplastic changes underlying these beneficial
effects of mindfulness in clinical populations29,41,42,74,97,142,157.
Although these studies are promising, future work needs
to replicate and expand the emerging findings to optimally
tailor interventions for clinical application.

Conclusions
Interest in the psychological and neuroscientific investigation of mindfulness meditation has increased markedly
over the past two decades. As is relatively common in a
new field of research, studies suffer from low methodological quality and present with speculative post-hoc interpretations. Knowledge of the mechanisms that underlie
the effects of meditation is therefore still in its infancy.
However, there is emerging evidence that mindfulness
meditation might cause neuroplastic changes in the structure and function of brain regions involved in regulation
of attention, emotion and self-awareness. Further research
needs to use longitudinal, randomized and actively controlled research designs and larger sample sizes to advance
the understanding of the mechanisms of mindfulness
meditation in regard to the interactions of complex brain
networks, and needs to connect neuroscientific findings
with behavioural data. If supported by rigorous research
studies, the practice of mindfulness meditation might be
promising for the treatment of clinical disorders and might
facilitate the cultivation of a healthy mind and increased
well-being.

222 | APRIL 2015 | VOLUME 16

www.nature.com/reviews/neuro
2015 Macmillan Publishers Limited. All rights reserved

REVIEWS
1.

2.

3.
4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

Ospina,M.B. etal. Meditation practices for health:


state of the research. Evid. Rep. Technol. Assess. (Full
Rep.) 155, 1263 (2007).
Tang,Y.-Y. & Posner,M.I. Theory and method in
mindfulness neuroscience. Soc. Cogn. Affect. Neurosci.
8, 118120 (2013).
Hart,W. The Art of Living: Vipassana Meditation
(Harper and Row, 1987).
Ivanovski,B. & Malhi,G.S. The psychological and
neurophysiological concomitants of mindfulness forms
of meditation. Acta Neuropsychiatr. 19, 7691 (2007).
Chiesa,A. & Malinowski,P. Mindfulness-based
approaches: are they all the same? J.Clin. Psychol.
67, 404424 (2011).
Baer,R.A. Mindfulness training as a clinical
intervention: a conceptual and empirical review. Clin.
Psychol. Sci. Practice 10, 125143 (2003).
Grossman,P. Defining mindfulness by how poorly I
think I pay attention during everyday awareness and
other intractable problems for psychologys (re)
invention of mindfulness: comment on Brown etal.
Psychol. Assess. 23, 10341040 (2011).
Kabat-Zinn,J. Full Catastrophe Living: Using the
Wisdom of Your Body and Mind to Face Stress, Pain,
and Illness (Delta Trade Paperbacks, 1990).
Lutz,A., Slagter,H.A., Dunne,J.D. & Davidson,R.J.
Attention regulation and monitoring in meditation.
Trends Cogn. Sci. 12, 163169 (2008).
Hlzel,B.K. etal. How does mindfulness meditation
work? Proposing mechanisms of action from a
conceptual and neural perspective. Perspect. Psychol.
Sci. 6, 537559 (2011).
A review of the mechanisms of meditation.
Tang,Y.Y., Rothbart,M.K. & Posner,M.I. Neural
correlates of establishing, maintaining and switching
brain states. Trends Cogn. Sci. 16, 330337 (2012).
A review of the mechanisms of brain states
associated with mental training.
Zeidan,F., Johnson,S.K., Diamond,B.J., David,Z. &
Goolkasian,P. Mindfulness meditation improves
cognition: evidence of brief mental training. Conscious.
Cogn. 19, 597605 (2010).
Ding,X. etal. Short-term meditation modulates brain
activity of insight evoked with solution cue. Soc. Cogn.
Affect. Neurosci. 10, 4349 (2014).
Tang,Y.Y. etal. Short-term meditation training
improves attention and self-regulation. Proc. Natl
Acad. Sci. USA 104, 1715217156 (2007).
The first longitudinal, randomized study to
document that brief training improves executive
attention, mood and immune function, and reduces
levels of stress hormones.
Manna,A. etal. Neural correlates of focused attention
and cognitive monitoring in meditation. Brain Res.
Bull. 82, 4656 (2010).
Tomasino,B., Fregona,S., Skrap,M. & Fabbro,F.
Meditation-related activations are modulated by the
practices needed to obtain it and by the expertise: an
ALE meta-analysis study. Front. Hum. Neurosci. 6,
346 (2012).
Fox,K.C. etal. Is meditation associated with altered
brain structure? A systematic review and meta-analysis of
morphometric neuroimaging in meditation practitioners.
Neurosci. Biobehav. Rev. 43, 4873 (2014).
A review of structural alterations in the brain
associated with meditation.
Brefczynski-Lewis,J.A., Lutz,A., Schaefer,H.S.,
Levinson,D.B. & Davidson,R.J. Neural correlates of
attentional expertise in long-term meditation
practitioners. Proc. Natl Acad. Sci. USA 104,
1148311488 (2007).
One of the first cross-sectional studies to document
the neural correlates of focused meditation.
Davidson,R.J. Empirical explorations of mindfulness:
conceptual and methodological conundrums. Emotion
10, 811 (2010).
MacCoon,D.G. etal. The validation of an active
control intervention for Mindfulness Based Stress
Reduction (MBSR). Behav. Res. Ther. 50, 312
(2012).
One of the first studies to validate the active
control conditions in mindfulness training.
Rosenkranz,M.A. etal. A comparison of mindfulnessbased stress reduction and an active control in
modulation of neurogenic inflammation. Brain Behav.
Immun. 27, 174184 (2013).
MacCoon,D.G., MacLean,K.A., Davidson,R.J.,
Saron,C.D. & Lutz,A. No sustained attention
differences in a longitudinal randomized trial
comparing mindfulness based stress reduction versus
active control. PLoS ONE 9, e97551 (2014).

23. Tang,Y.Y. etal. Central and autonomic nervous


system interaction is altered by short-term meditation.
Proc. Natl Acad. Sci. USA 106, 88658870 (2009).
24. Erisman,S.M. & Roemer,L. The effects of
experimentally induced mindfulness on emotional
responding to film clips. Emotion 10, 7282 (2010).
25. Leiberg,S., Klimecki,O. & Singer,T. Short-term
compassion training increases prosocial behaviour in a
newly developed prosocial game. PLoS ONE 6,
e17798 (2011).
26. Hoge,E.A. etal. Randomized controlled trial of
mindfulness meditation for generalized anxiety
disorder: effects on anxiety and stress reactivity.
J.Clin. Psychiatry 74, 786792 (2013).
27. Tang,Y.Y., Yang,L., Leve,L.D. & Harold,G.T.
Improving executive function and its neurobiological
mechanisms through a mindfulness-based intervention:
advances within the field of developmental
neuroscience. Child Dev. Perspect. 6, 361366
(2012).
28. Zeidan,F., Johnson,S.K., Gordon,N.S. &
Goolkasian,P. Effects of brief and sham mindfulness
meditation on mood and cardiovascular variables.
J.Altern. Complement. Med. 16, 867873 (2010).
29. Goldin,P., Ziv,M., Jazaieri,H., Hahn,K. & Gross,J.J.
MBSR versus aerobic exercise in social anxiety: fMRI
of emotion regulation of negative self-beliefs. Soc.
Cogn. Affect. Neurosci. 8, 6572 (2013).
One of the first randomized mindfulness studies to
document the neural mechanisms in social anxiety.
30. Zeidan,F. etal. Brain mechanisms supporting the
modulation of pain by mindfulness meditation.
J.Neurosci. 31, 55405548 (2011).
31. Hlzel,B.K. etal. Investigation of mindfulness
meditation practitioners with voxel-based
morphometry. Soc. Cogn. Affect. Neurosci. 3, 5561
(2008).
32. Lazar,S.W. etal. Meditation experience is associated
with increased cortical thickness. Neuroreport 16,
18931897 (2005).
The first cross-sectional study to document that
meditation is associated with structural changes in
the brain.
33. Vestergaard-Poulsen,P. etal. Long-term meditation is
associated with increased grey matter density in the
brain stem. Neuroreport 20, 170174 (2009).
34. Pagnoni,G. & Cekic,M. Age effects on grey matter
volume and attentional performance in Zen
meditation. Neurobiol. Aging 28, 16231627 (2007).
35. Grant,J.A., Courtemanche,J. & Rainville,P. A nonelaborative mental stance and decoupling of executive
and pain-related cortices predicts low pain sensitivity
in Zen meditators. Pain 152, 150156 (2010).
36. Grant,J.A. etal. Cortical thickness, mental absorption
and meditative practice: possible implications for
disorders of attention. Biol. Psychol. 92, 275281
(2013).
37. Fayed,N. etal. Brain changes in long-term zen
meditators using proton magnetic resonance
spectroscopy and diffusion tensor imaging: a
controlled study. PLoS ONE 8, e58476 (2013).
38. Tang,Y.Y. etal. Short-term meditation induces white
matter changes in the anterior cingulate. Proc. Natl
Acad. Sci. USA 107, 1564915652 (2010).
The first longitudinal study to document that brief
mindfulness training induces white-matter changes.
39. Tang,Y.Y., Lu,Q., Fan,M., Yang,Y. & Posner,M.I.
Mechanisms of white matter changes induced by
meditation. Proc. Natl Acad. Sci. USA 109,
1057010574 (2012).
40. Hlzel,B.K. etal. Mindfulness practice leads to
increases in regional brain grey matter density.
Psychiatry Res. 191, 3643 (2011).
41. Wells,R.E. etal. Meditations impact on default mode
network and hippocampus in mild cognitive impairment:
a pilot study. Neurosci. Lett. 556, 1519 (2013).
42. Pickut,B.A. etal. Mindfulness based intervention in
Parkinsons disease leads to structural brain changes
on MRI: a randomized controlled longitudinal trial.
Clin. Neurol. Neurosurg. 115, 24192425 (2013).
43. Luders,E., Toga,A.W., Lepore,N. & Gaser,C. The
underlying anatomical correlates of long-term
meditation: larger hippocampal and frontal volumes of
grey matter. Neuroimage 45, 672678 (2009).
44. Luders,E., Clark,K., Narr,K.L. & Toga,A.W.
Enhanced brain connectivity in long-term meditation
practitioners. Neuroimage 57, 13081316 (2011).
45. Luders,E. etal. Bridging the hemispheres in
meditation: thicker callosal regions and enhanced
fractional anisotropy (FA) in long-term practitioners.
Neuroimage 61, 181187 (2012).

NATURE REVIEWS | NEUROSCIENCE

46. Luders,E. etal. Global and regional alterations of


hippocampal anatomy in long-term meditation
practitioners. Hum. Brain Mapp. 34, 33693375
(2013).
47. Singleton,O. etal. Change in brainstem grey matter
concentration following a mindfulness-based
intervention is correlated with improvement in
psychological well-being. Front. Hum. Neurosci. 8, 33
(2014).
48. Hlzel,B.K. etal. Stress reduction correlates with
structural changes in the amygdala. Soc. Cogn. Affect.
Neurosci. 5, 1117 (2010).
49. Luders,E., Kurth,F., Toga,A.W., Narr,K.L. &
Gaser,C. Meditation effects within the hippocampal
complex revealed by voxel-based morphometry and
cytoarchitectonic probabilistic mapping. Front.
Psychol. 4, 398 (2013).
50. Luders,E. etal. The unique brain anatomy of
meditation practitioners: alterations in cortical
gyrification. Front. Hum. Neurosci. 6, 34 (2012).
51. Grant,J.A., Courtemanche,J., Duerden,E.G.,
Duncan,G.H. & Rainville,P. Cortical thickness and
pain sensitivity in zen meditators. Emotion 10, 4353
(2010).
52. Farb,N.A., Segal,Z.V. & Anderson,A.K. Mindfulness
meditation training alters cortical representations of
interoceptive attention. Soc. Cogn. Affect. Neurosci. 8,
1526 (2013).
53. Tang,Y.-Y. & Posner,M.I. Attention training and
attention state training. Trends Cogn. Sci. 13,
222227 (2009).
54. Tang,Y.Y. & Posner,M.I. in Handbook of Mindfulness:
Theory, Research, and Practice Ch. 5 (eds Brown, K.
W., Creswell, J. D. & Ryan, R. M.) 8189 (Guildford
Press, 2014).
55. Posner,M.I. & Petersen,S.E. The attention system of
the human brain. Annu. Rev. Neurosci. 13, 2542
(1990).
56. Petersen,S.E. & Posner,M.I. The attention system of
the human brain: 20years after. Annu. Rev. Neurosci.
35, 7389 (2012).
57. Fan,J., McCandliss,B.D., Sommer,T., Raz,A. &
Posner,M.I. Testing the efficiency and independence
of attentional networks. J.Cogn. Neurosci. 14,
340347 (2002).
58. Raz,A. & Buhle,J. Typologies of attentional networks.
Nature Rev. Neurosci. 7, 367379 (2006).
59. Posner,M.I. & Rothbart,M.K. Research on attention
networks as a model for the integration of
psychological science. Annu. Rev. Psychol. 58, 123
(2007).
60. Chiesa,A., Calati,R. & Serretti,A. Does mindfulness
training improve cognitive abilities? A systematic
review of neuropsychological findings. Clin. Psychol.
Rev. 31, 449464 (2011).
61. Chan,D. & Woollacott,M. Effects of level of
meditation experience on attentional focus: is the
efficiency of executive or orientation networks
improved? J.Altern. Complement. Med. 13, 651657
(2007).
62. Moore,A. & Malinowski,P. Meditation, mindfulness
and cognitive flexibility. Conscious. Cogn. 18,
176186 (2009).
63. Wenk-Sormaz,H. Meditation can reduce habitual
responding. Altern. Ther. Health Med. 11, 4258
(2005).
64. Slagter,H.A. etal. Mental training affects distribution
of limited brain resources. PLoS Biol. 5, e138 (2007).
65. Pashler,H. Overlapping mental operations in serial
performance with preview. Q.J.Exp. Psychol. A. 47,
161191 (discussion 193199, 201205) (1994).
66. Posner,M. I. Measuring alertness. Ann. NY Acad. Sci.
1129, 193199 (2008).
67. Van Leeuwen,S., Willer,N.G. & Melloni,L. Age effects
on attentional blink performance in meditation.
Conscious. Cogn. 18, 593599 (2009).
68. Van den Hurk,P.A., Giommi,F., Gielen,S.C.,
Speckens,A.E.M. & Barendregt,H.P. Greater
efficiency in attentional processing related to
mindfulness meditation. Q.J.Exp. Psychol. (Hove) 63,
11681180 (2010).
69. Anderson,N.D., Lau,M.A., Segal,Z.V. &
Bishop,S.R. Mindfulness-based stress reduction and
attentional control. Clin. Psychol. Psychother. 14,
449463 (2007).
70. Jha,A.P., Krompinger,J. & Baime,M.J. Mindfulness
training modifies subsystems of attention. Cogn.
Affect. Behav. Neurosci. 7, 109119 (2007).
71. MacLean,K.A. etal. Intensive meditation training
improves perceptual discrimination and sustained
attention. Psychol Sci, 21, 829839 (2010).

VOLUME 16 | APRIL 2015 | 223


2015 Macmillan Publishers Limited. All rights reserved

REVIEWS
72. Pagnoni,G. & Cekic,M. Age effects on grey matter
volume and attentional performance in Zen
meditation. Neurobiol. Aging 28, 16231627 (2007).
73. Tang,Y.Y. & Posner,M.I. Training brain networks and
states. Trends Cogn. Sci. 18, 345350 (2014).
74. Tang,Y.Y., Tang,R. & Posner,M.I. Brief meditation
training induces smoking reduction. Proc. Natl Acad.
Sci. USA 110, 1397113975 (2013).
75. Cahn,B.R. & Polich,J. Meditation states and traits:
EEG, ERP, and neuroimaging studies. Psychol. Bull.
132, 180211 (2006).
76. Hlzel,B.K. etal. Differential engagement of anterior
cingulate and adjacent medial frontal cortex in adept
meditators and non-meditators. Neurosci. Lett. 421,
1621 (2007).
77. Van Veen,V. & Carter,C.S. The anterior cingulate as a
conflict monitor: fMRI and ERP studies. Physiol.
Behav. 77, 477482 (2002).
78. Posner,M.I., Sheese,B., Rothbart,M. & Tang,Y.Y.
The anterior cingulate gyrus and the mechanism of
self-regulation. Cogn. Affect. Behav. Neurosci. 7,
391395 (2007).
79. Tang,Y.Y. & Tang,R. Ventral-subgenual anterior
cingulate cortex and self-transcendence. Front.
Psychol. 4, 1000(2014).
80. Sridharan,D., Levitin,D.J. & Menon,V. A critical role
for the right fronto-insular cortex in switching between
central-executive and default-mode networks. Proc.
Natl Acad. Sci. USA 105, 1256912574 (2008).
81. Gard,T. etal. Pain attenuation through mindfulness is
associated with decreased cognitive control and
increased sensory processing in the brain. Cereb.
Cortex 22, 26922702 (2012).
82. Allen,M. etal. Cognitive-affective neural plasticity
following active-controlled mindfulness intervention.
J.Neurosci. 32, 1560115610 (2012).
One of the first studies to document the effects of
mindfulness using active controls.
83. Goldin,P.R. & Gross,J.J. Effects of mindfulnessbased stress reduction (MBSR) on emotion regulation
in social anxiety disorder. Emotion 10, 8391
(2010).
84. Deckersbach,T., Hlzel,B.K., Eisner,L.R.,
Lazar,S.W. & Nierenberg,A.A. MindfulnessBased
Cognitive Therapy for Bipolar Disorder (Guildford
Press, 2014).
85. Passarotti,A.M., Sweeney,J.A. & Pavuluri,M.N.
Emotion processing influences working memory
circuits in pediatric bipolar disorder and attentiondeficit/hyperactivity disorder. J.Am. Acad. Child
Adolesc. Psychiatry 49, 10641080 (2010).
86. Gross,J.J. in Handbook of Emotion Regulation 2nd
edn (ed. Gross, J.J.) 320 (Guildford Press, 2014).
87. Ortner,C.N.M., Kilner,S.J. & Zelazo,P.D.
Mindfulness meditation and reduced emotional
interference on a cognitive task. Motiv. Emot. 31,
271283 (2007).
88. Goleman,D.J. & Schwartz,G.E. Meditation as an
intervention in stress reactivity. J.Consult. Clin.
Psychol. 44, 456466 (1976).
89. Robins,C.J., Keng,S.-L., Ekblad,A.G. &
Brantley,J.G. Effects of mindfulness-based stress
reduction on emotional experience and expression: a
randomized controlled trial. J.Clin. Psychol. 68,
117131 (2012).
90. Chambers,R., Lo,B.C.Y. & Allen,N.B. The impact of
intensive mindfulness training on attentional control,
cognitive style, and affect. Cogn. Ther. Res. 32,
303322 (2008).
91. Ding,X., Tang,Y.Y., Tang,R. & Posner,M.I. Improving
creativity performance by short-term meditation.
Behav. Brain Funct. 10, 9 (2014).
92. Jain,S. etal. A randomized controlled trial of
mindfulness meditation versus relaxation training:
effects on distress, positive states of mind,
rumination, and distraction. Ann. Behav. Med. 33,
1121 (2007).
93. Desbordes,G. etal. Effects of mindful-attention and
compassion meditation training on amygdala response
to emotional stimuli in an ordinary, non-meditative
state. Front. Hum. Neurosci. 6, 292 (2012).
94. Lutz,J. etal. Mindfulness and emotion regulation
an fMRI study. Soc. Cogn. Affect. Neurosci. 9,
776785 (2014).
95. Taylor,V.A. etal. Impact of mindfulness on the neural
responses to emotional pictures in experienced and
beginner meditators. Neuroimage 57, 15241533
(2011).
96. Westbrook,C. etal. Mindful attention reduces neural
and self-reported cue-induced craving in smokers. Soc.
Cogn. Affect. Neurosci. 8, 7384 (2013).

97. Hlzel,B.K. etal. Neural mechanisms of symptom


improvements in generalized anxiety disorder
following mindfulness training. Neuroimage Clin. 2,
448458 (2013).
One of the first longitudinal, randomized
mindfulness studies to document the neural
mechanisms in generalized anxiety disorder.
98. Farb,N.A.S. etal. Attending to the present:
mindfulness meditation reveals distinct neural modes
of self-reference. Soc. Cogn. Affect. Neurosci. 2,
313322 (2007).
99. Teper,R., Segal,Z.V. & Inzlicht,M. Inside the mindful
mind: how mindfulness enhances emotion regulation
through improvements in executive control. Curr. Dir.
Psychol. 22, 449454 (2013).
100. Chiesa,A., Serretti,A. & Jakobsen,J.C. Mindfulness:
top-down or bottom-up emotion regulation strategy?
Clin. Psychol. Rev. 33, 8296 (2013).
101. Malinowski,P. Neural mechanisms of attentional
control in mindfulness meditation. Front. Hum.
Neurosci. 7, 8 (2013).
102. Jensen,C.G. etal. Mindfulness training affects
attention or is it attentional effort? J.Exp. Psychol.
Gen. 141, 106123 (2012).
103. Posner,M.I., Rothbart,M.K., Reuda,M.R. &
Tang,Y.Y. in Effortless Attention: A New Perspective in
the Cognitive Science of Attention and Action (ed.
Bruya,B.) 410424 (MIT Press, 2010).
104. Banks,S.J., Eddy,K.T., Angstadt,M., Nathan,P.J. &
Phan,K.L. Amygdala-frontal connectivity during
emotion-regulation. Soc. Cogn. Affect Neurosci. 2,
303312 (2007).
105. Etkin,A., Egner,T., Peraza,D.M., Kandel,E.R. &
Hirsch,J. Resolving emotional conflict: a role for the
rostral anterior cingulate cortex in modulating activity
in the amygdala. Neuron 51, 871882 (2006).
106. Kirk,U., Brown,K.W. & Downar,J. Adaptive neural
reward processing during anticipation and receipt of
monetary rewards in mindfulness meditators. Soc.
Cogn. Affect. Neurosci. http://dx.doi.org/ 10.1093/
scan/nsu112 (2014).
107. Olendzki,A. Unlimiting Mind: The Radically
Experiential Psychology of Buddhism (Wisdom
Publications, 2010).
108. Sperduti,M., Martinelli,P. & Piolino,P. A
neurocognitive model of meditation based on
activation likelihood estimation (ALE) meta-analysis.
Conscious. Cogn. 21, 269276 (2012).
109. Fresco,D.M. etal. Initial psychometric properties of
the experiences questionnaire: validation of a selfreport measure of decentering. Behav. Ther. 38,
234246 (2007).
110. Shapiro,S.L., Carlson,L.E., Astin,J.A. & Freedman,B.
Mechanisms of mindfulness. J.Clin. Psychol. 62,
373386 (2006).
111. Josipovic,Z. Neural correlates of nondual awareness
in meditation. Ann. NY Acad. Sci. 1307, 918 (2014).
112. Kerr,C.E., Josyula,K. & Littenberg,R. Developing an
observing attitude: an analysis of meditation diaries in
an MBSR clinical trial. Clin. Psychol. Psychother. 18,
8093 (2011).
113. Dor-Ziderman,Y., Berkovich-Ohana,A., Glicksohn,J.
& Goldstein,A. Mindfulness-induced selflessness: a
MEG neurophenomenological study. Front. Hum.
Neurosci. 7, 582 (2013).
114. Emavardhana,T. & Tori,C.D. Changes in self-concept,
ego defense mechanisms, and religiosity following
seven-day Vipassana meditation retreats. J.Sci. Stud.
Relig. 36, 194206 (1997).
115. Haimerl,C.J. & Valentine,E.R. The effect of
contemplative practice on intrapersonal,
interpersonal, and transpersonal dimensions of the
self-concept. J.Transpers. Psychol. 33, 3752 (2001).
116. Sahdra,B.K., Shaver,P.R. & Brown,K.W. A scale to
measure nonattachment: a Buddhist complement to
Western research on attachment and adaptive
functioning. J.Pers. Assess. 92, 116127 (2010).
117. Brewer,J.A. etal. Meditation experience is associated
with differences in default mode network activity and
connectivity. Proc. Natl Acad. Sci. USA 108,
2025420259 (2011).
One of the first studies to document the alteration
of the DMN by meditation.
118. Hasenkamp,W. & Barsalou,L.W. Effects of
meditation experience on functional connectivity of
distributed brain networks. Front. Hum. Neurosci. 6,
38 (2012).
119. Buckner,R.L., Andrews-Hanna,J.R. & Schacter,D.L.
The brains default network: anatomy, function, and
relevance to disease. Ann. NY Acad. Sci. 1124, 138
(2008).

224 | APRIL 2015 | VOLUME 16

120. Raichle,M.E. etal. A default mode of brain function.


Proc. Natl Acad. Sci. USA 98, 676682 (2001).
121. Northoff,G. etal. Self-referential processing in our
brain: a meta-analysis of imaging studies on the self.
Neuroimage 31, 440457 (2006).
122. Sajonz,B. etal. Delineating self-referential processing
from episodic memory retrieval: common and
dissociable networks. Neuroimage 50, 16061617
(2010).
123. Buckner,R.L. & Carroll,D.C. Self-projection and the
brain. Trends Cogn. Sci. 11, 4957 (2007).
124. Khalsa,S.S. etal. Interoceptive awareness in
experienced meditators. Psychophysiology 45,
671677 (2008).
125. Nielsen,L. & Kaszniak,A.W. Awareness of subtle
emotional feelings: a comparison of long-term
meditators and nonmeditators. Emotion 6, 392405
(2006).
126. Sze,J.A., Gyurak,A., Yuan,J.W. & Levenson,R.W.
Coherence between emotional experience and
physiology: does body awareness training have an
impact? Emotion 10, 803814 (2010).
127. Fox,K. C. R. etal. Meditation experience predicts
introspective accuracy. PLoS ONE 7, e45370 (2012).
128. Lutz,A., Brefczynski-Lewis,J., Johnstone,T. &
Davidson,R.J. Regulation of the neural circuitry of
emotion by compassion meditation: effects of
meditative expertise. PLoS ONE 3, e1897 (2008).
129. Craig,A.D. How do you feel now? The anterior
insula and human awareness. Nature Rev. Neurosci.
10, 5970 (2009).
130. Monti,D. A. etal. Changes in cerebral blood flow and
anxiety associated with an 8-week mindfulness
programme in women with breast cancer. Stress
Health 28, 397407 (2012).
131. Grey,J.D., Milner,T.A. & McEwen,B.S. Dynamic
plasticity: the role of glucocorticoids, brain-derived
neurotrophic factor and other trophic factors.
Neuroscience 239, 214227 (2013).
132. Creswell,J.D., Pacilio,L.E., Lindsay,E.K. &
Brown,K.W. Brief mindfulness meditation training
alters psychological and neuroendocrine responses to
social evaluative stress. Psychoneuroendocrinology
44, 112 (2014).
133. Tang,Y.Y., Tang,R., Jiang,C. & Posner,M.I. Shortterm meditation intervention improves self-regulation
and academic performance. J.Child Adolesc. Behav. 2,
4 (2014).
134. Chiesa,A., Serretti,A. Mindfulness-based stress
reduction for stress management in healthy people: a
review and meta-analysis. J.Altern. Complement.
Med.15, 593600 (2009).
135. Jacobs,T. L. etal. Self-reported mindfulness and
cortisol during a Shamatha meditation retreat. Health
Psychol. 32, 11041109 (2013).
136. Fan,Y., Tang,Y.Y. & Posner,M.I. Cortisol level
modulated by integrative meditation in a dosedependent fashion. Stress Health 30, 6570 (2013).
137. Fan,Y., Tang,Y.Y., Ma,Y. & Posner,M.I. Mucosal
immunity modulated by integrative meditation in a
dose-dependent fashion. J.Altern. Complement. Med.
16, 151155 (2010).
138. Kang,D.H. etal. The effect of meditation on brain
structure: cortical thickness mapping and diffusion tensor
imaging. Soc. Cogn. Affect. Neurosci. 8, 2733 (2013).
139. Bressler,S.L. & Menon,V. Large-scale brain networks
in cognition: emerging methods and principles. Trends
Cogn. Sci. 14, 277290 (2010).
140. Menon,V. Large-scale brain networks and
psychopathology: a unifying triple network model.
Trends Cogn. Sci. 15, 483506 (2011).
141. Xue,S., Tang,Y.Y. & Posner,M.I. Short-term
meditation increases network efficiency of the anterior
cingulate cortex. Neuroreport 22, 570574 (2011).
142. Gard,T. etal. Fluid intelligence and brain functional
organization in aging yoga and meditation
practitioners. Front. Aging Neurosci. 6, 76 (2014).
143. Lane,R.D. & Wager,T.D. The new field of brain-body
medicine: what have we learned and where are we
headed? Neuroimage 47, 11351140 (2009).
144. Garrison,K.M. etal. Real-time fMRI links subjective
experience with brain activity during focused
attention. Neuroimage 81, 110118 (2013).
145. LaConte,S.M. Decoding fMRI brain states in realtime. Neuroimage 56, 440454 (2011).
146. Zotev,V. etal. Self-regulation of amygdala activation
using real-time fMRI neurofeedback. PLoS ONE 6,
e24522 (2011).
147. Haynes,J.D. & Rees,G. Decoding mental states from
brain activity in humans. Nature Rev. Neurosci. 7,
523534 (2006).

www.nature.com/reviews/neuro
2015 Macmillan Publishers Limited. All rights reserved

REVIEWS
148. van I.Jendoorn,M.H. etal. Gene-by-environment
experiments: a new approach to finding the missing
heritability. Nature Rev. Genet. 12, 881 (2011).
149. Jung,Y.H. etal. Influence of brain-derived
neurotrophic factor and catechol O-methyl transferase
polymorphisms on effects of meditation on plasma
catecholamines and stress. Stress 15, 97104 (2012).
150. Ding,X., Tang,Y.Y., Deng,Y., Tang,R. & Posner,M.I.
Mood and personality predict improvement in
creativity due to meditation training. Learn. Individ.
Differ. 37, 217221 (2014).
151. Rothbart,M.K. Becoming Who We Are (Guilford
Press, 2011).
152. Takahashi,T. etal. Changes in EEG and autonomic
nervous activity during meditation and their
association with personality traits. Int.
J.Psychophysiol. 55, 199207 (2005).
153. Moffitt,T.E. etal. A gradient of childhood self-control
predicts health, wealth, and public safety. Proc. Natl
Acad. Sci. USA 108, 26932698 (2011).
154. Hofmann,S.G., Sawyer,A.T., Witt,A. A. & Oh,D. The
effect of mindfulness-based therapy on anxiety and
depression: a meta-analytic review. J.Consult. Clin.
Psychol. 78, 169183 (2010).
A review of the effect of mindfulness-based therapy
on anxiety and mood symptoms.
155. Bowen,S. etal. Relative efficacy of mindfulness-based
relapse prevention, standard relapse prevention, and
treatment as usual for substance use disorders: a
randomized clinical trial. JAMA Psychiatry 71,
547556 (2014).
One of the first longitudinal studies to document the
effects of mindfulness on drug use and heavy drinking.
156. Schoenberg,P.L.A. etal. Effects of mindfulness-based
cognitive therapy on neurophysiological correlates of
performance monitoring in adult attention-deficit/
hyperactivity disorder. Clin. Neurophysiol. 125,
14071416 (2014).
157. Zeidan,F., Martucci,K.T., Kraft,R.A., McHaffie,J.G.
& Coghill,R.C. Neural correlates of mindfulness
meditation-related anxiety relief. Soc. Cogn. Affect.
Neurosci. 9, 751759 (2014).
158. Desbordes,G. etal. Moving beyond mindfulness:
defining equanimity as an outcome measure in
meditation and contemplative research. Mindfulness
http://dx.doi.org/10.1007/s12671-013-0269-8 (2014).

159. Smith,J. C. Alterations in brain and immune function


produced by mindfulness meditation: three caveats.
Psychosom. Med. 66, 148152 (2004).
160. Davidson,R.J., Kabat-Zinn,J. Response to
Smith,J.C. Psychosom. Med. 66,
148152 (2004).
161. Lippelt,D.P., Hommel,B. & Colzato,L.S. Focused
attention, open monitoring and loving kindness
meditation: effects on attention, conflict monitoring,
and creativity a review. Front. Psychol. 5, 1083
(2014).
162. Hasenkamp,W., Wilson-Mendenhall,C.D., Duncan,E.
& Barsalou,L.W. Mind wandering and attention
during focused meditation: a fine-grained temporal
analysis of fluctuating cognitive states. Neuroimage
59, 750760 (2012).
One of the first studies to document brain activity
during different phases of focused-attention
meditation.
163. Pagnoni,G. Dynamical properties of BOLD activity
from the ventral posteromedial cortex associated with
meditation and attentional skills. J.Neurosci. 32,
52425249 (2012).
164. st,L.G. in Phobias: A Handbook of Theory,
Research, and Treatment (ed. Davey,G.C.L.)
227247 (John Wiley, 1997).
165. Milad,M.R. & Quirk,G.J. Fear extinction as a model
for translational neuroscience: ten years of progress.
Annu. Rev. Psychol. 63, 129151 (2012).
166. Milad,M.R. etal. Recall of fear extinction in humans
activates the ventromedial prefrontal cortex and
hippocampus in concert. Biol. Psychiatry. 62,
446454 (2007).
167. LeDoux,J.E. Emotion circuits in the brain. Annu. Rev.
Neurosci. 23, 155184 (2000).
168. Davidson,R.J., Jackson,D.C. & Kalin,N.H. Emotion,
plasticity, context, and regulation: perspectives from
affective neuroscience. Psychol. Bull. 126, 890909
(2000).
169. Phelps,E.A., Delgado,M.R., Nearing,K.I. &
LeDoux,J.E. Extinction learning in humans: role of
the amygdala and vmPFC. Neuron 43, 897905
(2004).
170. Holt,D.J. etal. Extinction memory is impaired in
schizophrenia. Biol. Psychiatry 65, 455463
(2009).

NATURE REVIEWS | NEUROSCIENCE

171. Milad,M.R. etal. Presence and acquired origin of


reduced recall for fear extinction in PTSD: results of a
twin study. J.Psychiatr. Res. 42, 515520 (2008).
172. McEwen,B.S. & Morrison,J.H. The brain on stress:
vulnerability and plasticity of the prefrontal cortex
over the life course. Neuron 79, 1629 (2013).
173. McEwen,B.S. & Gianaros,P.J. Stress- and allostasisinduced brain plasticity. Annu. Rev. Med. 62,
431445 (2011).
174. Liston,C., McEwen,B.S. & Casey,B.J. Psychossocial
stress sreversibly disrupts prefrontal processing and
attentional control. Proc. Natl Acad. Sci. USA 106,
912917 (2009).
175. Davidson,R.J. & McEwen,B.S. Social influences on
neuroplasticity: stress and interventions to promote
well-being. Nature Neurosci.15, 689695 (2012).
176. McEwen,B. S. The brain on stress: toward an
integrative approach to brain, body and behaviour.
Perspect. Psychol. Sci. 8, 673675 (2013).
177. Thayer,J.F. & Lane,R.D. A model of neurovisceral
integration in emotion regulation and dysregulation.
J.Affective Disord. 61, 201216 (2000).
178. Creswell,J.D. in Handbook of Mindfulness: Theory,
Research, and Practice Ch. 23 (eds Brown, K. W.,
Creswell, J. D. & Ryan, R. M.) (Guildford Press,
2014).
179. Ditto,B., Eclache,M. & Goldman,N. Short-term
autonomic and cardiovascular effects of mindfulness
body scan meditation. Ann. Behav. Med. 32,
227234 (2006).
180. Xiong,G.L. & Doraiswamy,P.M. Does meditation
enhance cognition and brain plasticity? Ann. NY Acad.
Sci. 1172, 6369 (2009).

Acknowledgements
This work was supported by the US Office of Naval Research.
We thank E.Luders for her contributions to an earlier version
of this manuscript. We benefited from discussions with
R.Davidson and A.Chiesa. We thank four anonymous reviewers for their constructive comments and R.Tang for manuscript preparation.

Competing interests statement


The authors declare no competing financial interests.

VOLUME 16 | APRIL 2015 | 225


2015 Macmillan Publishers Limited. All rights reserved

CORRESPONDENCE

L I N K T O O R I G I N A L A RT I C L E

Neural mechanisms of mindfulness


meditation: bridging clinical and
neuroscience investigations
Anne Maj van der Velden and Andreas Roepstorff
Tang, Hlzel and Posner have written an
excellent Review of the literature on the
neural correlates and mechanisms of mindfulness meditation (The neuroscience of
mindfulness meditation. Nat. Rev. Neurosci.
16, 213225 (2015))1, and they provide much
needed methodological recommendations
for future research. However, on the basis of
the overview presented, it has become apparent that there is a gap between the clinical
and neuroscientific branches of research on
mindfulness meditation.
Although most clinical evidence has
been gathered from two interventions
that incorporate mindfulness meditation (mindfulness-based stress reduction
(MBSR) and mindfulness-based cognitive
therapy (MBCT)2), research on the neural
correlates of MBSR and MBCT and the prediction of clinical outcomes is still extremely
limited. Take, for instance, MBCT, which
was developed as a preventive treatment
for recurrent depression. Evidence for the
efficacy of MBCT has increased rapidly over
the past decade, and MBCT has been subject to rigorous clinical investigation in large
randomized controlled trials35. As a result,
MBCT is now recommended in several

national clinical guidelines, such as those


of the UK National Institute for Health and
Clinical Excellence6, as a preventive treatment for recurrent depression. However,
how MBCT treatment leads to reduced risk
of depressive relapse or recurrence is still yet
to be fully elucidated, and the underlying
neural mechanisms are particularly poorly
understood. In fact, in the rapidly growing literature on the neural mechanisms
of mindfulness meditation, only two studies have specifically looked at the neural
mechanisms of MBCT in the treatment of
recurrent major depressive disorder 7.
We find it remarkable that few neuroscientific investigations of mindfulness meditation have focused on the interventions
that have shown the most clinical promise.
By contrast, many of the mindfulness interventions cited in the Review by Tang, Hlzel
and Posner 1 have been subject to only
limited research on clinical and well-being
outcomes, and few have been investigated in
large rigorous clinical trials. As previously
called for, tighter integration and collaboration between the clinical and neuroscientific
branches of research are needed8. Clinicians
and neuroscientists need to work closer

NATURE REVIEWS | NEUROSCIENCE

together to understand and improve the


clinical potential of mindfulness-based
treatments.
Anne Maj van der Velden and Andreas Roepstorff are at
the Interacting Minds Centre, Aarhus University,
8000 Aarhus, Denmark.
Anne Maj van der Velden is also at the Danish Center for
Mindfulness, Aarhus University Hospital,
8000 Aarhus, Denmark.
Correspondence to A.M.V.D.V.
e-mail: wrp704@alumni.ku.dk
doi:10.1038/nrn3916-c1
Published online 17 June 2015

1.

2.

3.

4.

5.

6.

7.

8.

Tang,Y.Y., Hlzel,B.K. & Posner,M.I. The


neuroscience of mindfulness meditation. Nat. Rev.
Neurosci. 16, 213225 (2015).
Khoury,B. etal. Mindfulnessbased therapy:
acomprehensive metaanalysis. Clin. Psychol. Rev. 33,
763771 (2013).
Kuyken,W. etal. Effectiveness and costeffectiveness
of mindfulnessbased cognitive therapy compared
with maintenance antidepressant treatment in the
prevention of depressive relapse/recurrence:
arandomised controlled trial. Lancet http://dx.doi.
org/10.1016/S01406736(14)622224 (2015).
Piet,J. & Hougaard,E. The effect of mindfulness
based cognitive therapy for prevention of relapse in
recurrent major depressive disorder: a systematic
review and metaanalysis. Clin. Psychol. Rev. 31,
10321040 (2011).
Williams,J.M. etal. Mindfulnessbased cognitive
therapy for preventing relapse in recurrent depression:
a randomized dismantling trial. J.Consult. Clin.
Psychol. 82, 275286 (2014).
National Collaborating Centre for Mental Health.
Depression in adults: the treatment and management
of depression in adults. Ch. 1.9. NICE [online], http://
www.nice.org.uk/guidance/cg90/chapter/1
recommendations#continuationandrelapse
prevention (2009).
van der Velden,A.M. etal. A systematic review of
mechanisms in mindfulnessbased cognitive therapy in
the treatment of recurrent depressive disorder. Clin.
Psychol. Rev. 37, 2639 (2015).
Holmes,E.M., Craske,M.G. & Graybiel,A.M. A call for
mentalhealth science. Nature 511, 287289 (2014).

Competing interests statement


The authors declare no competing interests.

www.nature.com/reviews/neuro
2015 Macmillan Publishers Limited. All rights reserved

PeRSPecTiveS
OpiniOn

The neurobiology of psychedelic


drugs: implications for the treatment
of mood disorders
Franz X. Vollenweider and Michael Kometer

Abstract | After a pause of nearly 40 years in research into the effects of psychedelic
drugs, recent advances in our understanding of the neurobiology of psychedelics,
such as lysergic acid diethylamide (LSD), psilocybin and ketamine have led to
renewed interest in the clinical potential of psychedelics in the treatment of various
psychiatric disorders. Recent behavioural and neuroimaging data show that
psychedelics modulate neural circuits that have been implicated in mood and
affective disorders, and can reduce the clinical symptoms of these disorders. These
findings raise the possibility that research into psychedelics might identify novel
therapeutic mechanisms and approaches that are based on glutamate-driven
neuroplasticity.
Psychedelic drugs have long held a special
fascination for mankind because they produce an altered state of consciousness that is
characterized by distortions of perception,
hallucinations or visions, ecstasy, dissolution of self boundaries and the experience
of union with the world. As plant-derived
materials, they have been used traditionally
by many indigenous cultures in medical
and religious practices for centuries, if
not millennia1.
However, research into psychedelics
did not begin until the 1950s after the
breakthrough discovery of the classical
hallucinogen lysergic acid diethylamide
(LSD) by Albert Hofmann2 (TIMELINE). The
classical hallucinogens include indoleamines, such as psilocybin and LSD, and
phenethylamines, such as mescaline and
2,5-dimethoxy-4-iodo-amphetamine
(DOI). Research into psychedelics was
advanced in the mid 1960s by the finding
that dissociative anaesthetics such as ketamine and phencyclidine (PCP) also produce psychedelic-like effects3 (BOX 1). Given
their overlapping psychological effects,
both classes of drugs are included here
as psychedelics.

Depending on the individual taking the


drug, their expectations, the setting in which
the drug is taken and the drug dose, psychedelics produce a wide range of experiential
states, from feelings of boundlessness, unity
and bliss on the one hand, to the anxietyinducing experiences of loss of ego-control
and panic on the other hand47. Researchers
from different theoretical disciplines and
experimental perspectives have emphasized
different experiential states. One emphasis
has been placed on the LSD-induced perceptual distortions including illusions and
hallucinations, thought disorder and
experiences of split ego7,8 that are also
seen in naturally occurring psychoses911.
This perspective has prompted the use of
psychedelics as research tools for unravelling
the neuronal basis of psychotic disorders,
such as schizophrenia spectrum disorder.
The most recent work has provided compelling evidence that classical hallucinogens primarily act as agonists of serotonin
(5-hydroxytryptamine) 2A (5-HT2A)
receptors12 and mimic mainly the socalled positive symptoms (hallucinations
and thought disorder) of schizophrenia10.
Dissociative anaesthetics mimic the positive

642 | SEPTEMbER 2010 | VOLUME 11

and the negative symptoms (social withdrawal and apathy) of schizophrenia


through antagonism at NMDA (N-methyl-daspartate) glutamate receptors13,14.
Emphasis has also been placed on the
early observation that LSD can enhance
self-awareness and facilitate the recollection
of, and release from, emotionally loaded
memories15,16. This perspective appealed
to psychiatrists as a unique property that
could facilitate the psychodynamic process
during psychotherapy. In fact, by 1965 there
were more than 1,000 published clinical
studies that reported promising therapeutic
effects in over 40,000 subjects17. LSD,
psilocybin and, sporadically, ketamine have
been reported to have therapeutic effects in
patients with anxiety and obsessive
compulsive disorders (OCD), depression,
sexual dysfunction and alcohol addiction,
and to relieve pain and anxiety in
patients with terminal cancer 1823 (BOX 2).
Unfortunately, throughout the 1960s and
1970s LSD and related drugs became
increasingly associated with cultural rebellion; they were widely popularized as drugs
of abuse and were depicted in the media as
highly dangerous. Consequently, by about
1970, LSD and related drugs were placed
in Schedule I in many western countries.
Accordingly, research on the effects of
classical psychedelics in humans was
severely restricted, funding became
difficult and interests in the therapeutic
use of these drugs faded, leaving many
avenues of inquiry unexplored and
many questions unanswered.
With the development of sophisticated
neuroimaging and brain-mapping techniques and with the increasing understanding of the molecular mechanisms of action
of psychedelics in animals, renewed interest
in basic and clinical research with psychedelics in humans has steadily increased since
the 1990s. In this Perspective, we review
early and current findings of the therapeutic
effects of psychedelics and their mechanisms
of action in relation to modern concepts of
the neurobiology of psychiatric disorders.
We then evaluate the extent to which
psychedelics may be useful in therapy
aside from their established application as
models of psychosis3,11.

www.nature.com/reviews/neuro
2010 Macmillan Publishers Limited. All rights reserved

PersPectives
Current therapeutic studies
Several preclinical studies in the 1990s
revealed an important role for the NMDA
glutamate receptor in the mechanism of
action of antidepressants. These findings
consequently gave rise to the hypothesis that
the NMDA-antagonist ketamine might have
potential as an antidepressant24. This hypothesis was validated in an initial double-blind
placebo-controlled clinical study in seven
medication-free patients with major depression. Specifically, a significant reduction in
depression scores on the Hamilton depression
rating scale (HDRS) was observed 3 hours
after a single infusion of ketamine (0.5 mg
per kg), and this effect was sustained for at
least 72 hours25. Several studies have since
replicated this rapid antidepressant effect of
ketamine using larger sample sizes and treatment-resistant patients with depression2630.
Given that 71% of the patients met response
criteria (defined as a 50% reduction in HDRS
scores from baseline) within 24 hours26, this
rapid effect has a high therapeutic value. In
particular, patients with depression who are
suicidal might benefit from such a rapid and
marked effect as their acute mortality risk is
not considerably diminished with conventional antidepressants owing to their long
delay in onset of action (usually 23 weeks).
Indeed, suicidal ideations were reduced
24 hours after a single ketamine infusion28.
However, despite these impressive and
rapid effects, all but 2 of the patients relapsed
within 2 weeks after a single dose of ketamine26. Previous relapse prevention strategies,
such as the administration of either five
additional ketamine infusions29 or riluzole
(Rilutek; Sanofi-aventis) on a daily basis30,
yielded success only in some patients and

other strategies should be tested in further


studies. Moreover, the use of biomarkers
that are rooted in psychopathology, neuropsychology and/or genetics might help to
predict whether ketamine therapy will be
appropriate for a given patient with
depression31. In line with this idea, decreased
activation of the anterior cingulate cortex
(ACC) during a working memory task32 and
increased activation of the ACC during an
emotional facial processing task33, as well
as a positive family history of alcohol
abuse27, were associated with a stronger
antidepressant response to ketamine.
Ketamine therapy could be extended to
other disorders in which NMDA receptors
are implicated in the pathophysiology for
example, bipolar disorder 34 and addiction35. The use of ketamine for the treatment
of bipolar disorder is currently being tested
(Clinicaltrials.gov: NCT00947791). Its potential as a treatment for addiction is supported by
results from a double-blind, randomized clinical trial in which 90 heroin addicts received
either existentially oriented psychotherapy in
combination with a high dose (2.0 mg per kg)
or a low dose of ketamine (0.2 mg per kg).
Follow-up studies in the first 2 years revealed
a higher rate of abstinence, greater and
longer-lasting reductions in craving, and a
positive change in nonverbal, unconscious
emotional attitude in subjects who had been
treated with a high dose, compared with a low
dose, of ketamine36.
In contrast to the rapidly increasing
number of clinical studies with ketamine,
studies with classic hallucinogens are
emerging slowly. This slow progress may
be due to the fact that classic hallucinogens
are placed in Schedule 1 and therefore have

higher regulatory hurdles to overcome and


may have negative connotations as a drug
of abuse.
A recent study by Moreno and
colleagues37 evaluated case reports and
findings from studies performed in the
1960s that indicated that psilocybin and LSD
are effective in the treatment of OCD22,3840.
They subsequently carried out a study showing that psilocybin given on four different
occasions at escalating doses (ranging from
sub-hallucinogenic to hallucinogenic doses)
markedly decreased OCD symptoms
(by 23100%) on the Yalebrown obsessive
compulsive scale in patients with OCD who
were previously treatment resistant 37. The
reduction in symptoms occurred rapidly, at
about 2 h after the peak psychedelic effects,
and endured up to the 24-h post-treatment
rating 37. This symptom relief was not related
to the dose of the psychedelic drug or to the
intensity of the psychedelic experience, and
extended beyond the observed acute
psychological effect of 46 h, raising
intriguing questions regarding the mechanisms that underlie this protracted effect 37.
Further research on how this initial relief of
symptoms in response to psilocybin and
the subsequent return of symptoms is
linked to functional changes in the brain
could contribute not only to a mechanistic
explanation of the potentially beneficial
effects of psychedelics but also to the
development of novel treatments for OCD.
The chronicity and disease burden of
OCD, the suboptimal nature of available
treatments and the observation that
psilocybin was well tolerated in OCD
patients are clear indications that further
studies into the duration, efficacy and

Timeline | A brief history of psychedelic drugs


isolation and
identification
of mescaline
by A. Heffter

1897

Synthesis
of PcP

1919

Synthesis of
mescaline
by e. Spth

1926

Discovery of
psychoactive
effects of LSD
by A. Hofmann

1938

Synthesis
of LSD by
A. Hofmann

1943

First LSD study


in people with
depression by
c. Savage

1947

First LSD
study in
humans by
W. Stoll

1952

isolation and
synthesis of
psilocin and
psilocybin by
A. Hofmann

1953

First clinic using


LSD in psycholytic
therapy by
R. Sandison

1958

LSD appears on
the streets

1962

Synthesis
of ketamine

1963

Sandoz recalls
samples of
LSD and
ceases
supplying it

1965

introduction
of the term
dissociative
anaesthetic
by e. Domino

1966

Demonstration
of antagonistic
action of PcP at
NMDA receptors
by N. Anis

1970

LSD, psilocin
and mescaline
are placed in
Schedule i in
the US

1983

1988

First neuroimaging
study on psilocybin
and ketamine

1990

Demonstration of
agonistic action of
LSD at 5-HT2A
receptors; first
neuroimaging
study on mescaline

1999

Ketamine is
placed in
schedule iii
in the US

LSD, lysergic acid diethylamide; NMDA, N-methyl-d-aspartate; PcP, phencyclidine. Discoveries relating to classical hallucinogens and to dissociative anaesthetics are
shown by black and red boxes, respectively.

NATURE REVIEWS | NeuroscieNce

VOLUME 11 | SEPTEMbER 2010 | 643


2010 Macmillan Publishers Limited. All rights reserved

PersPectives
Box 1 | Assessing altered states of consciousness
Elementary
visual
alterations
Audiovisual
synesthaesia

Disembodiment
Impaired control
and cognition

Vivid imagery

Elementary
visual
alterations

Disembodiment

Audiovisual
synesthaesia

Impaired control
and cognition
Vivid imagery

10 20 30 40 50 60 70 Anxiety

Changed meaning
of percepts

20

30 40

50

60

Anxiety

Changed meaning
of percepts
Blissful state

Blissful state
Insightfulness

Insightfulness
Religious
experience

Experience
of unity

Psilocybin 115125 g per kg (n = 72)


Psilocybin 215270 g per kg (n = 214)
Psilocybin 315 g per kg (n = 41)

Experience
of unity

Ketamine 6 g per kg per min (n = 42)


Ketamine 12 g per kg per min (n = 92)

Quantifying altered states of consciousness was problematic in the early years


of hallucinogen research. Today, however, there are validated instruments
for assessing various aspects of consciousness. According to Dittrich133,
hallucinogen-induced altered states of consciousness can be reliably measured
by the five-dimensional altered states of consciousness (5DASC)
rating scale. This scale comprises five primary dimensions and their respective
subdimensions (see the figure). The primary dimensions are oceanic
boundlessness (shown by orange boxes), referring to positively experienced
loss of ego boundaries that are associated with changes in the sense of
time and emotions ranging from heightened mood to sublime happiness
and feelings of unity with the environment; anxious ego-disintegration
(shown by purple boxes), including thought disorder and loss of self-control;
visionary restructuralization (shown by blue boxes), referring to perceptual
alterations (such as visual illusions and hallucinations), and altered meaning of
percepts; acoustic alterations (not shown), including hypersensitivity to sound
and auditory hallucinations; and altered vigilance (not shown).

mechanisms of action of psilocybin or of


related compounds in the treatment of OCD
are warranted.
Encouraged by early findings (BOX 2),
several clinical centres have begun to investigate the potential beneficial effects of psilocybin (ClinicalTrials.gov: NCT00302744,
NCT00957359 and NCT00465595) and LSD
(ClinicalTrials.gov: NCT00920387) in the
treatment of anxiety and depression in
patients with terminal cancer, using state
of the art, double-blind, placebo-controlled
designs. One of these studies has recently
been completed and revealed that moderate doses of psilocybin improved mood and
reduced anxiety and that this relief variably
lasted between 2 weeks and 6 months in
patients with advanced cancer (C.S. Grob,
personal communication). Finally, another
recent study reported that psilocybin and LSD
aborted attacks, terminated the cluster period

Religious
experience

In general, the intensity of these psychedelic-induced alterations of


consciousness and perception is dose-dependent, so that hallucinations
that involve disorientation in person, place and time rarely, if ever, occur
with low to medium doses46. However, at larger doses and depending
on the individual, his or her expectations and the setting the same
hallucinogen might produce a pleasurable loss of ego boundaries combined
with feelings of oneness or might lead to a more psychotic ego dissolution
that involves fear and paranoid ideation 4,132,134. Such experiential
phenomena are otherwise rarely reported except in dreams, contemplative
or religious exaltation and acute psychoses11,135. The figure shows that the
classical hallucinogen psilocybin (0.0150.027 g per kg, by mouth) (see
the figure, left) and the dissociative s-ketamine (612 g per kg per min,
intravenously) (see the figure, right) produce a set of overlapping
psychological experiences, measured by the 5DASC rating scale and
respective subscales. The scales indicate the percentage scored of the
maximum score.

or extended the remission period in people


suffering from cluster headaches41. Taken
together, these findings support early observations in the 1960s that classical hallucinogens have antinociceptive potential and may
not only reduce symptoms but also induce
long-lasting adaptive processes.
neurobiology of psychedelic drugs
The enormous progress that has been made
in our understanding of the mechanisms of
action of psychedelics12,4245 and the neurobiology of affective disorders34,46,47 has enabled
us to postulate new hypotheses regarding the
therapeutic mechanisms of psychedelics and
their clinical applications. Here we focus on
the glutamatergic and serotonergic mechanisms of action of psychedelics with regard
to their most promising indications that
is, their use in the treatment of depression
and anxiety.

644 | SEPTEMbER 2010 | VOLUME 11

Classical hallucinogens. The classical hallucinogens are comprised of three main chemical classes: the plant-derived tryptamines
(for example, psilocybin) and phenethylamines (for example, mescaline), and the
semisynthetic ergolines (for example, LSD)48.
Although all classical hallucinogens display
high affinity for 5-HT2 receptors, they also
interact to some degree with 5-HT1, 5-HT4,
5-HT5, 5-HT6 and 5-HT7 receptors12. In contrast to the tryptamines, the ergolines also
show high intrinsic activity at dopamine D2
receptors and at -adrenergic receptors49.
Converging evidence from pharmacological50, electrophysiological51,52 and behavioural studies in animals53,54 suggests that
classical hallucinogens produce their effects
in animals and possibly in humans primarily
through agonistic actions at cortical 5-HT2A
receptors (FIG. 1a). Consistent with this view,
selectively restoring 5-HT2A receptors in
www.nature.com/reviews/neuro

2010 Macmillan Publishers Limited. All rights reserved

PersPectives
cortical pyramidal neurons is sufficient to
rescue hallucinogen-induced head shaking
in transgenic mice that lack 5-HT2A receptors53,55. Importantly, administration of the
5-HT2A receptor antagonist ketanserin abolishes virtually all of the psilocybin-induced
subjective effects in humans56. Recent studies have demonstrated that hallucinogenic
and non-hallucinogenic 5-HT2A agonists
differentially regulate intracellular signalling
pathways in cortical pyramidal neurons and
that this results in a differential expression
of downstream signalling proteins, such as
early growth response protein 1 (EGR1),
EGR2 and -arrestin 255,57. This suggests that
further elucidation of hallucinogen-specific
signalling pathways may aid the development of functionally selective ligands with
specific therapeutic properties for example, ligands that have antidepressant effects
but no hallucinogenic effects.
Several studies have demonstrated that
activation of 5-HT2A receptors by classical
hallucinogens or by serotonin leads to a
robust, glutamate-dependent increase in the
activity of pyramidal neurons, preferentially

those in layer V of the prefrontal cortex


(PFC)51,52,58,59 (FIG. 1a). This increase in
glutamatergic synaptic activity was initially
thought to result from stimulation of presynaptic 5-HT2A receptors located on glutamatergic thalamocortical afferents to the
PFC60,61. However, more recent studies suggest that stimulation of postsynaptic 5-HT2A
receptors55,58,59 on a subpopulation of pyramidal cells in the deep layers of the PFC59 leads
to an increase in glutamatergic recurrent
network activity 59,62. The increase in glutamatergic synaptic activity can be abolished
not only by specific 5-HT2A antagonists but
also by AMPA (-amino-3-hydroxyl-5methyl-4-isoxazole-propionic acid) receptor antagonists63, by agonists51 and positive
allosteric modulators of metabotropic
glutamate receptor 2 (mGluR2)64, and by
selective antagonists of the NR2b subunit
of NMDA receptors65. Taken together, these
findings indicate that classical hallucinogens
are potent modulators of prefrontal network
activity that involves a complex interaction
between the serotonin and glutamate
systems in prefrontal circuits.

Box 2 | Early therapeutic findings with psychedelics


By 1953, two forms of lysergic acid diethylamide (LSD) therapy based on different theoretical
frameworks were emerging. These have been named psychedelic (mind-manifesting)136 and
psycholytic (psyche-loosening)15 therapies. In psychedelic therapy, which was practised mostly in
North America, a large dose of LSD (200800 g) was applied in a single session. This was thought
to induce an overwhelming and supposedly conversion-like peak experience that would bring the
subject to a new level of awareness and self-knowledge. It was thought that that this would
facilitate self-actualization and lead to permanent changes that would be beneficial to the
subject128,129. Furthermore, it was claimed that intensive psychotherapeutic preparation of the
patient before the drug session and a follow-up integration of the peak experience in further
drug-free sessions were crucial for an optimal outcome130. Promising therapeutic effects of this
therapy were found in people with terminal cancer20,137, in severe alcoholics138,139, in people who
were addicted to narcotics140 and in patients with neurosis141. For example, a series of studies
showed that LSD could reduce depression and decrease apprehension towards death and,
surprisingly, that LSD had transient analgesic effects that were superior to those of
dihydromorphinone (also known as hydromorphone and Palladone SR (Napp)) and meperidine
(also known as pethidine)20. These effects were confirmed in later studies and the clinical efficacy
was linked with the intensity of the psychedelic experience129,141,142.
Psycholytic therapy was introduced by Ronald Sandison and applied in Europe at 18 treatment
centres143. In psycholytic therapy, low to moderate doses of LSD (50100 g), psilocybin (1015 mg)
or, sporadically, ketamine were used repeatedly as an adjunct in psychoanalytically oriented
psychotherapy to accelerate the therapeutic process by facilitating regression and the
recollection and release of emotionally loaded repressed memories, and by increasing the
transference reaction15,22,144147. A review of 42 studies reported impressive improvement rates in
(mostly treatment-resistant) patients with anxiety disorders (improvement in 70% of patients),
depression (in 62% of patients), personality disorders (in 5361% of patients), sexual dysfunction
(in 50% of patients) and obsessivecompulsive disorders (in 42% of patients)148.
Unfortunately, the majority of these studies had serious methodological flaws by contemporary
standards. In particular, with the absence of adequate control groups and follow-up measurements
and with vague criteria for therapeutic outcome, the studies did not clearly establish whether it
was the drug or the therapeutic engagement that produced the reported beneficial effect. It was
also difficult to draw firm conclusions regarding potential long-term efficacy. Nevertheless, the
studies provide a conceptual framework for the application of psychedelics, with the data
suggesting that the most promising indication for psychedelic use might be found in the treatment
of depression and anxiety disorders.

NATURE REVIEWS | NeuroscieNce

Activation of 5-HT2A and 5-HT1A receptors in the medial PFC (mPFC) also has
downstream effects on serotonergic and
dopaminergic activity through descending projections to the dorsal raphe and the
ventral tegmental area (VTA). For example,
activation of 5-HT2A receptors in the mPFC
increases the firing rate of 5-HT neurons in
the dorsal raphe and of dopamine neurons
in the VTA, resulting in an increased release
of 5-HT in the mPFC58,66 and of dopamine in
mesocortical areas67 in animals. In a study
in humans, the hallucinogenic 5-HT2A agonist psilocybin increased striatal dopamine
concentrations, and this increase correlated
with euphoria and depersonalization
phenomena68. blocking dopamine D2
receptors by haloperidol, however, reduced
these effects by only about 30%. This
suggests that the dopaminergic system contributes only moderately to the broad spectrum of psilocybin-induced psychological
alterations56.
Interestingly, 5-HT2A receptor activation
not only seems to underlie the preponderance of the acute psychedelic effects of hallucinogens but may also lead to neuroplastic
adaptations in an extended prefrontallimbic
network. For example, in rats a single dose
of the hallucinogen DOI transiently
increased the dendritic spine size in cortical neurons69 and repeated doses of LSD
downregulated cortical 5-HT2A but not
5-HT1A receptors; effects that were the most
pronounced in the frontomedial cortex and
ACC70,71. It is possible that such adaptations
and specifically a downregulation of prefrontal
5-HT2A receptors might underlie some of
the therapeutic effects of hallucinogens in the
treatment of depression, anxiety and chronic
pain. In favour of this hypothesis, 5-HT2A
receptor density was found to be increased
in the PFC in post-mortem samples72 and
in vivo73,74 in patients with major depression,
and to be reduced after chronic treatment with
various antidepressants the reduction coinciding with the onset of clinical efficacy7577. In
addition, chronic, antisense-mediated downregulation of 5-HT2A receptors in rats78 and in
5-HT2A knockout mice79 reduced anxiety-like
behaviour, and selective restoration of 5-HT2A
receptors in the PFC normalized anxiety-like
behaviour in these 5-HT2A knockout mice.
These findings suggest that prefrontal 5-HT2A
receptors might modulate the activity of subcortical structures, such as the amygdala79.
Anxiety and depression are interrelated with
stress80, which also affects the serotonin system81. Stress elevates corticotropin-releasing
factor (CRF)82, and administration of CRF
into the mPFC of mice enhanced anxiety-like
VOLUME 11 | SEPTEMbER 2010 | 645

2010 Macmillan Publishers Limited. All rights reserved

PersPectives
a
Cortical layer V

Deep cortical layers

Glutamate
release

NMDAR

Brainstem

5-HT neuron
5-HT2A

+
AMPAR
Psilocin/
LSD/DMT

5-HT2A

BDNF
Psilocin/
LSD/DMT

b
Cortex

Subcortical areas
Glutamate
release

Ketamine

NMDAR

AMPAR
Interneuron

+
BDNF
NMDAR

GABA

Ketamine

Figure 1 | Activation of the prefrontal network and glutamate release by psychedelics. a | The
figure shows a model in which hallucinogens, such as psilocin, lysergic acid diethylamide (LSD) and
dimethyltryptamine (DMT), increase extracellular glutamate levels in the prefrontal cortex through
stimulation of postsynaptic serotonin (5-hydroxytryptamine) 2A (5-HT2A) receptors that are located
on large glutamatergic pyramidal cells in deep cortical layers (v and vi) projecting to layer v pyramidal
neurons. This glutamate release leads to an activation of AMPA (-amino-3-hydroxy-5-methyl-4isoxazole propionic acid) and NMDA (N-methyl-d-aspartate) receptors on cortical pyramidal neurons. in
addition, hallucinogens directly activate 5-HT2A receptors located on cortical pyramidal neurons. This
activation is thought to ultimately lead to increased expression of brain-derived neurotrophic factor
(BDNF). b | The figure shows a model in which dissociative NMDA antagonists, such as ketamine, block
inhibitory GABA (-aminobutyric acid)-ergic interneurons in cortical and subcortical brain areas, leading to enhanced firing of glutamatergic projection neurons and increased extracellular glutamate
levels in the prefrontal cortex. As ketamine also blocks NMDA receptors on cortical pyramidal neurons,
the increased glutamate release in the cortex is thought to stimulate cortical AMPA more than NMDA
receptors. The increased AMPA-receptor-mediated throughput relative to NMDA-receptor-mediated
throughput is thought ultimately to lead to increased expression of BDNF.

behaviour in response to DOI through


sensitization of 5-HT2 receptor signalling in
the PFC83. In humans, fronto-limbic 5-HT2A
receptor density is correlated not only with
anxiety but also with an individuals difficulties in coping with stress84. Indeed, recent
studies showed that prefrontal 5-HT2A receptors located on descending projections that
control serotonergic activity in the dorsal

raphe are involved in stress responses67,85.


Together, these findings suggest that downregulation of prefrontal 5-HT2A receptors by
classical hallucinogens might underlie some
of the effects of hallucinogens on depression
and anxiety.
Finally, with regard to the finding that
LSD reduces anxiety and pain in cancer
patients20, it is of note that prefrontal 5-HT2A

646 | SEPTEMbER 2010 | VOLUME 11

density correlated with responses to tonic


pain but not with responses to short phasic pain stimuli. This suggests a role of the
5-HT2A receptors in the cognitive evaluation
of pain experiences86 and points to additional therapeutic potential for hallucinogens
in individuals with chronic pain.
Dissociative anaesthetics. At sub-anaesthetic
doses, dissociative anaesthetics, such as
ketamine, primarily block the NMDA receptor at the PCP binding site in the receptors
ionotropic channel14 (FIG. 1b). The psychoactive potency of the s-ketamine enantiomer is
three to four times higher than that of the
r-ketamine enantiomer. This is paralleled by
their relative affinities at the NMDA receptor
complex 87. Systemic administration of
non-competitive NMDA antagonists, such
as ketamine, PCP and MK-801 (also
known as dizocilpine), in rats markedly increases glutamate release in the
mPFC88,89 concomitant with an increase in
the firing rate of pyramidal neurons in this
area90. These effects are probably due to a
blockade of NMDA receptors on GAbA
(-aminobutyric acid)-ergic interneurons45,91
in cortical and/or subcortical structures and
to the subsequent reduction of inhibitory
control over prefrontal glutamatergic neurons92. The increased extracellular glutamate
levels in the mPFC seem to contribute to the
psychotropic effects of ketamine and PCP,
as AMPA receptor antagonists88 or agonists
of mGluR2 and mGluR3 (REF. 93) abolished
various behavioural effects of NMDA
antagonists in rats. Likewise, the behavioural
effects of selective NR2b antagonists such
as CP-101,606 (also known as Traxoprodil),
which produces dose-dependent psychotropic effects similar to those of ketamine in
humans94 can be blocked by administration of AMPA receptor antagonists95. Finally,
lamotrigine, which reduces presynaptic
glutamate release, attenuated the subjective
effects of s-ketamine in humans96.
In addition to having these glutamatergic
effects, non-competitive NMDA receptor
antagonists increase extracellular prefrontal
and mesolimbic dopamine89,93 and prefrontal serotonin89 levels in rats, presumably by stimulating corticofugal glutamate
release in the VTA97 and the dorsal raphe89,
respectively. Studies into the contribution of
this dopaminergic and serotonergic activation to the behavioural effects of NMDA
antagonists are scant and the results are
somewhat controversial. Specifically, in
two studies in humans, ketamine-induced
striatal dopamine release correlated with
the extent of ketamine-induced psychotic

www.nature.com/reviews/neuro
2010 Macmillan Publishers Limited. All rights reserved

PersPectives
symptoms98,99, but in another study systemic
administration of the dopamine D2 receptor antagonist haloperidol did not attenuate
ketamine-induced psychotic symptoms
in healthy volunteers100. Although 5-HT2A
receptor antagonists reverse the disruptive
effects of NMDA antagonists on sensorimotor gating 101 and on object recognition102 in
animals, no comparable studies of the role
of serotonin in the mechanism of action of
NMDA antagonists have been conducted
in humans.
The enhanced glutamate release that
results from NMDA receptor blockade
by ketamine leads to an increased activation of AMPA receptors relative to NMDA
receptors95. The antidepressant-like effects
of ketamine and the selective NR2b antagonist CP-101,606 in animals can be blocked
by administration of the AMPA receptor
antagonist 2,3-dihydroxy-6-nitro-7-sulphamoyl-benzo[f]quinoxaline-2,3-dione
(NbQX)95, suggesting that enhanced AMPA
activation in cortical circuits is crucial for
the therapeutic effect of NMDA receptor
antagonists34,95.

A common mechanism? There is accumulating evidence that, despite their different primary modes of action, classical hallucinogens
and dissociative anaesthetics both modulate
glutamatergic neurotransmission in the prefrontallimbic circuitry that is implicated in
the pathophysiology of mood disorders. This
modulation is evidenced by the observation
in rats that hallucinogens103,104 and dissociative anaesthetics88,89 have a similar effect in
enhancing extracellular glutamate release
in the PFC, leading to increased activation
of pyramidal cells63,65,105,106. Furthermore,
and congruent with these findings, human
neuroimaging studies have shown that both
psilocybin and ketamine markedly activate
prefrontal cortical areas, including the ACC
and insula and, to a lesser extent, temporal and
parieto-occipital regions107111 (FIG. 2).
According to current models of emotion
regulation the PFC, including the ACC, exerts
cognitive, top-down control over emotion
and stress responses through its connections to the amygdala and dorsal raphe47,85.
Reduced prefrontal glutamate levels that are
associated with attenuated PFC activation

b
s-Ketamine

Psilocybin

Figure 2 | Brain activity patterns in psychedelic-induced states of consciousness. a | Brain


imaging studies using 18fluorodeoxyglucose [18FDG] positron emission tomography (PeT) revealed that
moderate doses of s-ketamine (top) and psilocybin (bottom) in healthy volunteers increased neuronal
activity. This is shown by changes in the cerebral metabolic rate for glucose (cMRglu) in the prefrontal
cortex and associated limbic regions and in subcortical structures, including the thalamus107,109. This
similar prefrontallimbic activation pattern supports the view that both classes of drugs have converging effects on a final pathway or neurotransmitter system. b | Recent [18FDG] PeT brain imaging studies
have demonstrated that the degree to which each of the psychedelic-induced key dimensions of
altered states of consciousness (BOX 2) is manifested and correlated with functional alterations in
cortical and limbic regions and subcortical structures, including the basal ganglia and thalamus. For
example, the intensity of experience of the key dimension oceanic boundlessness correlated with
the s-ketamine- and psilocybin-induced activation (red) of a prefrontalparietal network and the
deactivation (blue) of a striatolimbic amygdalocentric network149.

NATURE REVIEWS | NeuroscieNce

in response to emotional stimuli34,112,113 have


been reported in patients with depression.
Further, depressed individuals46 and subjects
with high trait anxiety 114 show reduced PFC
activity when executive control is engaged,
and might suffer from decreased topdown inhibition of amygdala activity 115,116.
Conversely, chronic treatment with selective
serotonin reuptake inhibitors (SSRIs) increases
the functional connectivity between the amygdala and the PFC117, and attenuates the
amygdala response to the presentation of
images showing sad faces in patients with
depression118,119. This suggests that the normalization of this dysregulated network might be
important in the recovery from depression46.
Given that both psilocybin and ketamine
increase extracellular glutamate levels in
the prefrontallimbic circuitry in rats and
that the antidepressant effects of both drugs
outlast their acute psychotropic effects in
depressed patients, we propose that a
normalization of this network through
a glutamate-dependent neuroplastic adaptation is the common therapeutic mechanism
of these drugs. Specifically, we posit that
psychedelics enhance neuroplasticity by
increasing AMPA-type glutamate receptor
trafficking and by raising the level of brainderived neurotropic factor (bDNF). Deficits
in these neuroplastic mechanisms have been
implicated in the pathophysiology of depression34,120. Normalization of these neuroplastic
deficits might contribute not only to the
relatively sustained antidepressant effects of
ketamine121,122 but also to those of psilocybin.
In line with this view, both classes of drugs
have been demonstrated to stimulate AMPA
receptors by increasing extracellular glutamate levels6,95 and to increase bDNF levels in
prefrontal and limbic brain areas in rats123125.
A recent study in patients with depression,
however, failed to demonstrate an increase
in bDNF plasma levels in the first 4 h after
ketamine infusion122. Whether ketamine
treatment leads to an increase in bDNF levels
at a later time and whether such an increase
is associated with sustained antidepressant
effects warrants further investigation.
Conclusions and future directions
The clinical findings and current understanding of the mechanisms of action of
classical hallucinogens and dissociative
anaesthetics converge on the idea that
psychedelics might be useful in the treatment of major depression, anxiety disorders
and OCD. These are serious, debilitating,
life-shortening illnesses, and as the currently available treatments have high failure
rates, psychedelics might offer alternative

VOLUME 11 | SEPTEMbER 2010 | 647


2010 Macmillan Publishers Limited. All rights reserved

PersPectives
treatment strategies that could improve the
well-being of patients and the associated
economic burden on patients and society.
Accumulating evidence shows a crucial
role for the glutamate system in the regulation of neuronal plasticity, and indicates that
abnormalities in neuroplasticity contribute
to the pathophysiology of mood disorders.
Thus, drugs that target neuronal plasticity
may offer a novel approach to their treatment. This Perspective proposes that classical
Glossary
Cluster period
A period of time during which cluster headache attacks
occur regularly.

Enantiomers
Two stereoisomeric molecules that are mirror images of
each other and are not superimposable.

Existentially oriented psychotherapy


A form of therapy that emphasizes the development of a
sense of self-direction through choice and of awareness in
resolving existential conflicts (such as the inevitability of
death, isolation and meaninglessness).

Neurosis
A former term for a category of mental disorders
characterized by anxiety and a sense of distress. This
category includes disorders now classified as mood
disorders, anxiety disorders, dissociative disorders,
sexual disorders and somatoform disorders.

Psychoanalytically oriented psychotherapy


A therapy based on Freudian psychoanalysis in which
unconscious conflicts that are thought to cause the
patients symptoms are brought into consciousness to
create insight for the resolution of the problems.

Regression
In Freudian psychoanalytic theory this term describes a
psychological strategy to cope with reality by means of
a temporary reversion of the ego to an earlier stage of
development.

Riluzole
A drug used to treat amyotrophic lateral sclerosis and that
has NMDA (N-methyl-D-aspartate) receptor blocking
properties similar to those of ketamine.

Schedule 1
A legislative category containing controlled drugs that have
a high potential for abuse, a lack of accepted safety and no
currently accepted medical use in treatments.

Selective serotonin reuptake inhibitors


A class of compounds typically used as antidepressants.

Self-actualization
The motivation to realize all of ones potential.

Structureactivity relationship
(Often abbreviated to SAR.) This is the relationship between
the chemical structure of a molecule and its biological activity.

Transference
A phenomenon in psychoanalysis characterized by
unconscious redirection of feelings or desires from one
person to another.

psychedelics, such as psilocybin, and dissociative anaesthetics, such as ketamine,


alter glutamatergic neurotransmission in
prefrontallimbic circuitries, and that this
leads to neuroplastic adaptations, presumably
through enhancement of AMPA receptor
function. These adaptations may explain
some of the shared and relatively sustained
antidepressant effects that are observed in
clinical studies with ketamine and psilocybin.
To further validate this glutamate-induced
neuroplasticity hypothesis the relationship
between measures of glutamatergic activity
and clinical outcome needs to be established.
Moreover, the finding that classical hallucinogens (unlike dissociative anaesthetics) also
modulate 5-HT2A receptor signalling suggests
that they may improve subtypes of anxiety
and stress-related disorders. Studies that use
biomarkers for genotypes or that use expression levels of 5-HT2A receptors in parallel with
clinical end points would be essential not only
for clarifying the role of 5-HT2A receptors in
the therapeutic mechanism of classical hallucinogens but also for the development of
personalized medicines in the treatment
of anxiety and stress-related disorders.
In addition, to optimize the clinical
benefits of psychedelics and to reduce their
unwanted side effects, a deeper understanding of various factors is necessary. These
include structureactivity relationships, dose
response relationships and the influence of
psychotherapeutic approaches on the effects
of psychedelics. In this context, it is interesting to note that there was no indication of
prolonged psychosis, persisting perception
disorder or subsequent drug abuse after psilocybin126 or ketamine127 administration in a
large sample of psychotherapeutically wellprepared healthy subjects in a supportive
research setting. Similar observations were
reported in small samples of patients with
depression29 and OCD37. Nonetheless, it is
often claimed that the dissociative effects of,
for example, ketamine may limit clinical use,
despite its reported efficacy 24,94. In this sense,
understanding the molecular mechanism
of action could inform the development of
novel ligands for 5-HT2A or NMDA receptors
that display antidepressant properties but
have fewer dissociative effects than psilocybin and ketamine. Further evaluations of the
doseresponse relationship may be another
approach to minimize unwanted side effects.
For example, low to moderate oral doses of
psilocybin (<0.215 mg per kg) were found
to only rarely produce anxious dissociative
symptoms in controlled settings126 (BOX 1)
but to reduce anxiety, depression and OCD
symptoms in patients22,37. Similarly, a low

648 | SEPTEMbER 2010 | VOLUME 11

dose of the NR2b antagonist CP-101,606 (in


combination with an SSRI) had transient
antidepressant effects in a small sample of
patients with depression and only rarely
induced dissociative symptoms94.
To take the opposite perspective, it is
noteworthy that initial clinical applications of
psychedelics in psychedelic and psycholytic
therapy were based on the premise that the
drug-induced psychological experience had
an essential, facilitatory effect on the psychotherapeutic process that is, it was a form
of pharmacology-assisted psychotherapy.
Indeed, it has been shown that the transcendent peak (mystical-type) experience, which
has a key role in the therapeutic outcome
in psychedelic therapy 128130 and was rated
as among the most personally meaningful
experiences131,132, occurs in most cases only
in supportive settings and after high-dose
administration of psychedelics. One might
interpret this concept as an early example of
the neuroplasticity hypothesis in which the
drug-induced experience and its integration
in the psychotherapeutic process is the crucial mechanism that enables neuroplasticity
and behavioural changes. by contrast, current pharmacological strategies often assume
that medication alone produces neuroplastic
adaptations. However, drugs that increase
neuroplasticity, such as psychedelics, might
be particularly clinically efficient in combination with psychotherapeutic interventions121. In support of this notion, cognitive
behavioural therapy was shown to normalize
prefrontallimbic functioning in depressed
patients46, and could therefore enhance the
proposed neuroplastic effects of psychedelics
in prefrontallimbic structures as discussed
here. Thus, further blind, controlled studies
are obviously now needed to test these
alternative and opposing hypotheses.
The potential of drugs to target glutamatergic neurotransmission in prefrontal
limbic circuitries and to facilitate neuroplastic adaptations may translate into promising
new treatment approaches for affective disorders. The novel hypotheses presented here
now need to be investigated using wellcontrolled clinical studies, keeping in mind
the controversial history of this class of drugs.
Franz X. Vollenweider and Michael Kometer are at the
Neuropsychopharmacology and Brain Imaging
Research Unit, University Hospital of Psychiatry,
Zurich, Switzerland.
Franz X. Vollenweider is also at the School of Medicine,
University of Zurich, Switzerland.
Correspondence to F.X.V.
e-mail: vollen@bli.uzh.ch
doi:10.1038/nrn2884
Published online 18 August 2010

www.nature.com/reviews/neuro
2010 Macmillan Publishers Limited. All rights reserved

PersPectives
1.

2.

3.

4.

5.

6.

7.
8.

9.
10.

11.

12.
13.

14.

15.

16.
17.

18.

19.
20.

21.

22.

23.

24.

25.

26.

27.

Hofmann, A. & Schultes, R. E. Plants of the Gods


(McGraw-Hill Book Company, Maidenhead, UK,
1979).
Hofmann, A. in Chemical Constitution and
Pharmacodynamic Actions (ed. Burger, A.) 169235
(M.Dekker, New York, 1968).
Domino, E. F., Kamenka, J. M. & Gneste, P. The joint
FrenchUS seminar on phencyclidine and related
arylcyclohexylamines. Trends Pharmacol. Sci. 9,
363367 (1983).
Hasler, F., Grimberg, U., Benz, M. A., Huber, T. &
Vollenweider, F. X. Acute psychological and
physiological effects of psilocybin in healthy
humans: a double-blind, placebo-controlled doseeffect study. Psychopharmacology 172, 145156
(2004).
Dittrich, A. in 50 Years of LSD. Current Status and
Perspectives of Hallucinogens (eds Pletscher, A. &
Ladewig, D.) 101118 (Parthenon, New York, 1994).
Fischer, R., Marks, P. A., Hill, R. M. & Rockey, M. A.
Personality structure as the main determinant of drug
induced (model) psychoses. Nature 218, 296298
(1968).
Leuner, H. Die Experimentelle Psychose (Springer,
Berlin Gttingen Heidelberg, 1962).
Hoch, P. H., Cattell, J. P. & Pennes, H. H. Effects of
mescaline and lysergic acid (d-LSD-25). Am.
J. Psychiatry 108, 579584 (1952).
Chapman, J. The early symptoms of schizophrenia.
Br. J. Psychiatry 112, 225251 (1966).
Gouzoulis-Mayfrank, E. et al. Hallucinogenic drug
induced states resemble acute endogenous psychoses:
results of an empirical study. Eur. Psychiatry 13,
399406 (1998).
Geyer, M. A. & Vollenweider, F. X. Serotonin research:
contributions to understanding psychoses. Trends
Pharmacol. Sci. 29, 445453 (2008).
Nichols, D. E. Hallucinogens. Pharmacol. Ther. 101,
131181 (2004).
Krystal, J. H. et al. Subanesthetic effects of the
noncompetitive NMDA antagonist, ketamine, in
humans. Arch. Gen. Psychiatry 51, 199214
(1994).
Anis, N. A., Berry, S. C., Burton, N. R. & Lodge, D.
The dissociative anesthetics, ketamine and
phencyclidine selective reduce excitation of central
mammalian neurons by N-methyl-D-aspartate.
Br. J. Pharmacol. 79, 565575 (1983).
Sandison, R. A. Psychological aspects of the LSD
treatment of neuroses. J. Ment Sci. 100, 508515
(1954).
Schmiege, G. R. Jr. LSD as a therapeutic tool. J. Med.
Soc. N.J. 60, 203207 (1963).
Malleson, N. Acute adverse reactions to LSD in clinical
and experimental use in the United Kingdom.
Br. J. Psychiatry 118, 229230 (1971).
Hoffer, A. in The Uses and Implications of
Hallucinogenic Drugs (eds Aaronson, B. &
Osmond, H.) 357366 (Hogarth Press, London,
1970).
Abramson, H. The use of LSD in Psychotherapy and
Alcoholism (Bobbs-Merrill, New York, 1967).
Kast, E. in LSD: The Consciousness Expanding Drug
(ed. Solomon, D.) 241256 (G.P. Putman, New York,
1964).
Pahnke, W. N., Kurland, A. A., Goodman, L. E. &
Richards, W. A. LSD-assisted psychotherapy with
terminal cancer patients. Curr. Psychiatr. Ther. 9,
144152 (1969).
Leuner, H. in 50 Years of LSD: Current Status and
Perspectives of Hallucinogen Research (eds Pletscher,
A. & Ladewig, D.) 175189 (Parthenon, New York,
1994).
Kurland, A. A., Unger, S., Shaffer, J. W. & Savage, C.
Psychedelic therapy utilizing LSD in the treatment of
the alcoholic patient: a preliminary report. Am.
J. Psychiatry 123, 12021209 (1967).
Skolnick, P., Popik, P. & Trullas, R. Glutamate-based
antidepressants: 20 years on. Trends Pharmacol. Sci.
30, 563569 (2009).
Berman, R. M. et al. Antidepressant effects of
ketamine in depressed patients. Biol. Psychiatry 47,
351354 (2000).
Zarate, C. A. Jr et al. A randomized trial of an
N-methyl-D-aspartate antagonist in treatmentresistant major depression. Arch. Gen. Psychiatry 63,
856864 (2006).
Phelps, L. E. et al. Family history of alcohol
dependence and initial antidepressant response to an
N-methyl-D-aspartate antagonist. Biol. Psychiatry 65,
181184 (2009).

28. Price, R. B., Nock, M. K., Charney, D. S. & Mathew, S. J.


Effects of intravenous ketamine on explicit and implicit
measures of suicidality in treatment-resistant
depression. Biol. Psychiatry 66, 522526 (2009).
29. Aan het Rot, M. et al. Safety and efficacy of repeateddose intravenous ketamine for treatment-resistant
depression. Biol. Psychiatry 67, 139145 (2010).
30. Mathew, S. J. et al. Riluzole for relapse prevention
following intravenous ketamine in treatment-resistant
depression: a pilot randomized, placebo-controlled
continuation trial. Int. J. Neuropsychopharmacol. 13,
7182 (2010).
31. Holsboer, F. How can we realize the promise of
personalized antidepressant medicines? Nature Rev.
Neurosci. 9, 638646 (2008).
32. Salvadore, G. et al. Anterior cingulate
desynchronization and functional connectivity with the
amygdala during a working memory task predict rapid
antidepressant response to ketamine.
Neuropsychopharmacology 35, 14151422
(2010).
33. Salvadore, G. et al. Increased anterior cingulate
cortical activity in response to fearful faces: a
neurophysiological biomarker that predicts rapid
antidepressant response to ketamine. Biol. Psychiatry
65, 289295 (2009).
34. Sanacora, G., Zarate, C. A., Krystal, J. H. & Manji, H. K.
Targeting the glutamatergic system to develop novel,
improved therapeutics for mood disorders. Nature
Rev. Drug Discov. 7, 426437 (2008).
35. Lau, C. G. & Zukin, R. S. NMDA receptor trafficking in
synaptic plasticity and neuropsychiatric disorders.
Nature Rev. Neurosci. 8, 413426 (2007).
36. Krupitsky, E. et al. Ketamine psychotherapy for heroin
addiction: immediate effects and two-year follow-up.
J. Subst. Abuse Treatment 23, 273283 (2002).
37. Moreno, F. A., Wiegand, C. B., Taitano, E. K. &
Delgado, P. L. Safety, tolerability, and efficacy of
psilocybin in 9 patients with obsessive-compulsive
disorder. J. Clin. Psychiatry 67, 17351740 (2006).
38. Brandrup, E. & Vanggaard, T. LSD treatment in a
severe case of compulsive neurosis. Acta Psychiatr.
Scand. 55, 127141 (1977).
39. Leonard, H. L. & Rapoport, J. L. Relief of obsessive
compulsive symptoms by LSD and psilocin. Am.
J. Psychiatry 144, 12391240 (1987).
40. Moreno, F. A. & Delgado, P. L. Hallucinogen-induced
relief of obsessions and compulsions. Am.
J. Psychiatry 154, 10371038 (1997).
41. Sewell, R. A., Halpern, J. H. & Pope, H. G. Jr.
Response of cluster headache to psilocybin and LSD.
Neurology 66, 19201922 (2006).
42. Gonzalez-Maeso, J. & Sealfon, S. C. Agonist-trafficking
and hallucinogens. Curr. Med. Chem. 16, 10171027
(2009).
43. Winter, J. C. Hallucinogens as discriminative stimuli
in animals: LSD, phenethylamines, and tryptamines.
Psychopharmacology (Berlin) 203, 251263 (2009).
44. Large, C. H. Do NMDA receptor antagonist models of
schizophrenia predict the clinical efficacy of antipsychotic
drugs? J. Psychopharmacol. 21, 283301 (2007).
45. Quirk, M. C., Sosulski, D. L., Feierstein, C. E.,
Uchida, N. & Mainen, Z. F. A defined network of fastspiking interneurons in orbitofrontal cortex: responses
to behavioral contingencies and ketamine
administration. Front. Syst. Neurosci. 3, 13 (2009).
46. DeRubeis, R. J., Siegle, G. J. & Hollon, S. D. Cognitive
therapy versus medication for depression: treatment
outcomes and neural mechanisms. Nature Rev.
Neurosci. 9, 788796 (2008).
47. Clark, L., Chamberlain, S. R. & Sahakian, B. J.
Neurocognitive mechanisms in depression:
implications for treatment. Annu. Rev. Neurosci. 32,
5774 (2009).
48. Geyer, M. A., Nichols, D. E. & Vollenweider, F. X.
in Encyclopedia of Neuroscience (ed. Squire, L. R.)
741748 (Academic Press, Oxford, 2009).
49. Marona-Lewicka, D., Thisted, R. A. & Nichols, D. E.
Distinct temporal phases in the behavioral
pharmacology of LSD: dopamine D2 receptormediated effects in the rat and implications for
psychosis. Psychopharmacologia (Berlin) 180,
427435 (2005).
50. Glennon, R. A., Titeler, M. & McKenney, J. D. Evidence
for 5-HT2 involvement in the mechanism of action of
hallucinogenic agents. Life Sci. 35, 25052511
(1984).
51. Aghajanian, G. K. & Marek, G. J. Serotonin induces
excitatory postsynaptic potentials in apical dendrites
of neocortical pyramidal cells.
Neuropsychopharmacology 36, 589599 (1997).

NATURE REVIEWS | NeuroscieNce

52. Aghajanian, G. K. & Marek, G. J. Serotonin, via


5-HT2A receptors, increases EPSCs in layer V
pyramidal cells of prefrontal cortex by an
asynchronous mode of glutamate release. Brain Res.
825, 161171 (1999).
53. Wing, L. L., Tapson, G. S. & Geyer, M. A. 5HT-2
mediation of acute behavioral effects of hallucinogens
in rats. Psychopharmacology 100, 417425 (1990).
54. Sipes, T. E. & Geyer, M. A. DOI disruption of prepulse
inhibition of startle in the rat is mediated by 5-HT2A
and not by 5-HT2C receptors. Behav. Pharmacol. 6,
839842 (1995).
55. Gonzalez-Maeso, J. et al. Hallucinogens recruit specific
cortical 5-HT(2A) receptor-mediated signaling
pathways to affect behavior. Neuron 53, 439452
(2007).
56. Vollenweider, F. X., Vollenweider-Scherpenhuyzen,
M. F. I., Bbler, A., Vogel, H. & Hell, D. Psilocybin
induces schizophrenia-like psychosis in humans via
a serotonin-2 agonist action. Neuroreport 9,
38973902 (1998).
57. Schmid, C. L., Raehal, K. M. & Bohn, L. M.
Agonist-directed signaling of the serotonin 2A
receptor depends on b-arrestin-2 interactions in vivo.
Proc. Natl Acad. Sci. USA 105, 10791084 (2008).
58. Puig, M. V., Celada, P., az-Mataix, L. & Artigas, F.
In vivo modulation of the activity of pyramidal neurons
in the rat medial prefrontal cortex by 5-HT2A
receptors: relationship to thalamocortical afferents.
Cereb. Cortex 13, 870882 (2003).
59. Beique, J. C., Imad, M., Mladenovic, L., Gingrich, J. A.
& Andrade, R. Mechanism of the 5-hydroxytryptamine
2A receptor-mediated facilitation of synaptic activity in
prefrontal cortex. Proc. Natl Acad. Sci. USA 104,
98709875 (2007).
60. Aghajanian, G. K. & Marek, G. J. Serotonin and
hallucinogens. Neuropsychopharmacology 21,
16S23S (1999).
61. Marek, G. J., Wright, R. A., Gewirtz, J. C. &
Schoepp, D. D. A major role for thalamocortical
afferents in serotonergic hallucinogen receptor
function in the rat neocortex. Neuroscience 105,
379392 (2001).
62. Aghajanian, G. K. Modeling psychosis in vitro by
inducing disordered neuronal network activity in
cortical brain slices. Psychopharmacology (Berlin)
206, 575585 (2009).
63. Zhang, C. & Marek, G. J. AMPA receptor involvement
in 5-hydroxytryptamine2A receptor-mediated prefrontal cortical excitatory synaptic currents and DOIinduced head shakes. Prog. Neuropsychopharmacol.
Biol. Psychiatry 32, 6271 (2008).
64. Benneyworth, M. A. et al. A selective positive
allosteric modulator of metabotropic glutamate
receptor subtype 2 blocks a hallucinogenic drug model
of psychosis. Mol. Pharmacol. 72, 477484 (2007).
65. Lambe, E. K. & Aghajanian, G. K. Hallucinogeninduced UP states in the brain slice of rat prefrontal
cortex: role of glutamate spillover and NR2B-NMDA
receptors. Neuropsychopharmacology 31, 1682
1689 (2006).
66. Celada, P., Puig, M. V., Casanovas, J. M., Guillazo, G.
& Artigas, F. Control of dorsal raphe serotonergic
neurons by the medial prefrontal cortex: Involvement
of serotonin-1A, GABA(A), and glutamate receptors.
J. Neurosci. 21, 99179929 (2001).
67. Vazquez-Borsetti, P., Cortes, R. & Artigas, F. Pyramidal
neurons in rat prefrontal cortex projecting to ventral
tegmental area and dorsal raphe nucleus express
5-HT2A receptors. Cereb. Cortex 19, 16781686
(2009).
68. Vollenweider, F. X., Vontobel, P., Hell, D. & Leenders,
K. L. 5-HT modulation of dopamine release in basal
ganglia in psilocybin-induced psychosis in man: A PET
study with [11C]raclopride.
Neuropsychopharmacology 20, 424433 (1999).
69. Jones, K. A. et al. Rapid modulation of spine
morphology by the 5-HT2A serotonin receptor
through kalirin-7 signaling. Proc. Natl Acad. Sci. USA
106, 1957519580 (2009).
70. Buckholtz, N. S., Zhou, D. F., Freedman, D. X. &
Potter, W. Z. Lysergic acid diethylamide (LSD)
administration selectively downregulates serotonin2
receptors in rat brain. Neuropsychopharmacology 3,
137148 (1990).
71. Gresch, P. J., Smith, R. L., Barrett, R. J. &
Sanders-Bush, E. Behavioral tolerance to lysergic acid
diethylamide is associated with reduced serotonin-2A
receptor signaling in rat cortex.
Neuropsychopharmacology 30, 16931702
(2005).

VOLUME 11 | SEPTEMbER 2010 | 649


2010 Macmillan Publishers Limited. All rights reserved

PersPectives
72. Shelton, R. C., Sanders-Bush, E., Manier, D. H. &
Lewis, D. A. Elevated 5-HT 2A receptors in
postmortem prefrontal cortex in major depression is
associated with reduced activity of protein kinase, A.
Neuroscience 158, 14061415 (2008).
73. Bhagwagar, Z. et al. Increased 5-HT2A receptor binding
in euthymic, medication-free patients recovered from
depression: a positron emission study with [11C]MDL
100,907. Am. J. Psychiatry 163, 15801587
(2006).
74. Meyer, J. H. et al. Dysfunctional attitudes and 5-HT2
receptors during depression and self-harm. Am.
J. Psychiatry 160, 9099 (2003).
75. Sibille, E. et al. Antisense inhibition of
5-hydroxytryptamine2a receptor induces an
antidepressant-like effect in mice. Mol. Pharmacol.
52, 10561063 (1997).
76. Yamauchi, M., Miyara, T., Matsushima, T. &
Imanishi, T. Desensitization of 5-HT2A receptor
function by chronic administration of selective
serotonin reuptake inhibitors. Brain Res. 1067,
164169 (2006).
77. Gomez-Gil, E. et al. Decrease of the platelet 5-HT2A
receptor function by long-term imipramine treatment
in endogenous depression. Hum. Psychopharmacol.
19, 251258 (2004).
78. Cohen, H. Anxiolytic effect and memory improvement
in rats by antisense oligodeoxynucleotide to
5-hydroxytryptamine-2A precursor protein. Depress.
Anxiety. 22, 8493 (2005).
79. Weisstaub, N. V. et al. Cortical 5-HT2A receptor
signaling modulates anxiety-like behaviors in mice.
Science 313, 536540 (2006).
80. Anisman, H., Merali, Z. & Stead, J. D. Experiential and
genetic contributions to depressive- and anxiety-like
disorders: clinical and experimental studies. Neurosci.
Biobehav. Rev. 32, 11851206 (2008).
81. Lukkes, J., Vuong, S., Scholl, J., Oliver, H. & Forster, G.
Corticotropin-releasing factor receptor antagonism
within the dorsal raphe nucleus reduces social anxietylike behavior after early-life social isolation.
J. Neurosci. 29, 99559960 (2009).
82. Reul, J. M. & Holsboer, F. Corticotropin-releasing
factor receptors 1 and 2 in anxiety and depression.
Curr. Opin. Pharmacol. 2, 2333 (2002).
83. Magalhaes, A. C. et al. CRF receptor 1 regulates
anxiety behavior via sensitization of 5-HT2 receptor
signaling. Nature Neurosci. 13, 622629 (2010).
84. Frokjaer, V. G. et al. Frontolimbic serotonin 2A
receptor binding in healthy subjects is associated with
personality risk factors for affective disorder. Biol.
Psychiatry 63, 569576 (2008).
85. Amat, J. et al. Medial prefrontal cortex determines
how stressor controllability affects behavior and
dorsal raphe nucleus. Nature Neurosci. 8, 365371
(2005).
86. Kupers, R. et al. A PET [18F]altanserin study of
5-HT12A receptor binding in the human brain and
responses to painful heat stimulation. Neuroimage
44, 10011007 (2009).
87. Oye, I., Paulsen, O. & Maurset, A. Effects of ketamine
on sensory perception: Evidence for a role of
N-methyl-D-aspartate receptors. J. Pharmac. Exp.
Ther. 260, 12091213 (1992).
88. Moghaddam, B., Adams, B., Verma, A. & Daly, D.
Activation of glutamatergic neurotransmission by
ketamine: a novel step in the pathway from NMDA
receptor blockade to dopaminergic and cognitive
disruptions associated with the prefrontal cortex.
J. Neurosci. 17, 29212927 (1997).
89. Lopez-Gil, X. et al. Clozapine and haloperidol
differently suppress the MK-801-increased
glutamatergic and serotonergic transmission in the
medial prefrontal cortex of the rat.
Neuropsychopharmacology 32, 20872097 (2007).
90. Jackson, M. E., Homayoun, H. & Moghaddam, B.
NMDA receptor hypofunction produces concomitant
firing rate potentiation and burst activity reduction in
the prefrontal cortex. Proc. Natl Acad. Sci. USA 101,
84678472 (2004).
91. Homayoun, H. & Moghaddam, B. NMDA receptor
hypofunction produces opposite effects on prefrontal
cortex interneurons and pyramidal neurons.
J. Neurosci. 27, 1149611500 (2007).
92. Jodo, E. et al. Activation of medial prefrontal cortex by
phencyclidine is mediated via a hippocampo-prefrontal
pathway. Cereb. Cortex 15, 663669 (2005).
93. Moghaddam, B. & Adams, B. W. Reversal of
phencyclidine effects by a group II metabotropic
glutamate receptor agonist in rats. Science 281,
13491352 (1998).

94. Preskorn, S. H. et al. An innovative design to establish


proof of concept of the antidepressant effects of the
NR2B subunit selective N-methyl-D-aspartate
antagonist, CP-101,606, in patients with treatmentrefractory major depressive disorder. J. Clin.
Psychopharmacol. 28, 631637 (2008).
95. Maeng, S. et al. Cellular mechanisms underlying the
antidepressant effects of ketamine: role of -amino3-hydroxy-5-methylisoxazole-4-propionic acid
receptors. Biol. Psychiatry 63, 349352 (2008).
96. Anand, A. et al. Attenuation of the neuropsychiatric
effects of ketamine with lamotrigine: support for
hyperglutamatergic effects of N-methyl-D-aspartate
receptor antagonists. Arch. Gen. Psychiatry 57,
270276 (2000).
97. Jentsch, J. D., Tran, A., Taylor, J. R. & Roth, R. H.
Prefrontal cortical involvement in phencyclidineinduced activation of the mesolimbic dopamine
system: behavioral and neurochemical evidence.
Psychopharmacology (Berlin) 138, 8995 (1998).
98. Breier, A. et al. Effects of NMDA antagonism on
striatal dopamine release in healthy subjects
application of a novel PET approach. Synapse 29,
142147 (1998).
99. Vollenweider, F. X., Vontobel, P., Leenders, K. L. &
Hell, D. Effects of S-ketamine on striatal dopamine
release: a [11C] raclopride PET study of a model
psychosis in humans. J. Psych. Res. 34, 3543 (2000).
100. Krystal, J. H. et al. Interactive effects of subanesthetic
ketamine and haloperidol in healthy humans.
Psychopharmacology 145, 193204 (1999).
101. Varty, G. B., Bakshi, V. P. & Geyer, M. A. M100907, a
serotonin 5-HT2A receptor antagonist and putative
antipsychotic, blocks dizocilpine-induced prepulse
inhibition deficits in sprague-dawley and wistar rats.
Neuropsychopharmacology 20, 311321 (1999).
102. Snigdha, S. et al. Attenuation of phencyclidine-induced
object recognition deficits by the combination of
atypical antipsychotic drugs and pimavanserin (ACP
103), a 5-hydroxytryptamine(2A) receptor inverse
agonist. J. Pharmacol. Exp. Ther. 332, 622631 (2010).
103. Scruggs, J. L., Schmidt, D. & Deutch, A. Y. The
hallucinogen 1-[2,5-dimethoxy-4-iodophenyl]2-aminopropane (DOI) increases cortical extracellular
glutamate levels in rats. Neurosci. Lett. 346,
137140 (2003).
104. Muschamp, J. W., Regina, M. J., Hull, E. M.,
Winter, J. C. & Rabin, R. A. Lysergic acid diethylamide
and [-]-2,5-dimethoxy-4-methylamphetamine increase
extracellular glutamate in rat prefrontal cortex. Brain
Res. 1023, 134140 (2004).
105. Kargieman, L., Santana, N., Mengod, G., Celada, P. &
Artigas, F. Antipsychotic drugs reverse the disruption
in prefrontal cortex function produced by NMDA
receptor blockade with phencyclidine. Proc. Natl Acad.
Sci. USA 104, 1484314848 (2007).
106. Shi, W. X. & Zhang, X. X. Dendritic glutamate-induced
bursting in the prefrontal cortex: further
characterization and effects of phencyclidine.
J. Pharmacol. Exp. Ther. 305, 680687 (2003).
107. Vollenweider, F. X. et al. Metabolic hyperfrontality and
psychopathology in the ketamine model of psychosis
using positron emission tomography (PET) and [F-18]fluorodeoxyglocose (FDG). Eur.
Neuropsychopharmacol. 7, 924 (1997).
108. Vollenweider, F. X. et al. Positron emission tomography
and fluorodeoxyglucose studies of metabolic
hyperfrontality and psychopathology in the psilocybin
model of psychosis. Neuropsychopharmacology 16,
357372 (1997).
109. Vollenweider, F. X., Leenders, K. L., Oye, I., Hell, D. &
Angst, J. Differential psychopathology and patterns of
cerebral glucose utilisation produced by (S)- and
(R)-ketamine in healthy volunteers measured by
FDG-PET. Eur. Neuropsychopharmacol. 7, 2538
(1997).
110. Schreckenberger, M. et al. The psilocybin psychosis as
a model psychosis paradigma for acute schizophrenia:
a PET study with 18-FDG. Eur. J. Nucl. Med. 25, 877
(1998).
111. Gouzoulis-Mayfrank, E. et al. Neurometabolic effects
of psilocybin, 3,4-methylenedioxyethylamphetamine
(MDE) and D-methamphetamine in healthy volunteers.
A double-blind, placebo-controlled PET study with
[18F]FDG. Neuropsychopharmacology 20, 565581
(1999).
112. Walter, M. et al. The relationship between aberrant
neuronal activation in the pregenual anterior
cingulate, altered glutamatergic metabolism, and
anhedonia in major depression. Arch. Gen. Psychiatry
66, 478486 (2009).

650 | SEPTEMbER 2010 | VOLUME 11

113. Hasler, G. et al. Reduced prefrontal glutamate/


glutamine and gamma-aminobutyric acid levels in
major depression determined using proton magnetic
resonance spectroscopy. Arch. Gen. Psychiatry 64,
193200 (2007).
114. Bishop, S. J. Trait anxiety and impoverished prefrontal
control of attention. Nature Neurosci. 12, 9298
(2009).
115. Bishop, S. J. Neural mechanisms underlying selective
attention to threat. Ann. NY Acad. Sci. 1129,
141152 (2008).
116. Johnstone, T., van Reekum, C. M., Urry, H. L.,
Kalin, N. H. & Davidson, R. J. Failure to regulate:
counterproductive recruitment of top-down prefrontalsubcortical circuitry in major depression. J. Neurosci.
27, 88778884 (2007).
117. Chen, C. H. et al. Functional coupling of the amygdala
in depressed patients treated with antidepressant
medication. Neuropsychopharmacology 33,
19091918 (2008).
118. Fu, C. H. et al. Attenuation of the neural response to
sad faces in major depression by antidepressant
treatment: a prospective, event-related functional
magnetic resonance imaging study. Arch. Gen.
Psychiatry 61, 877889 (2004).
119. Sheline, Y. I. et al. Increased amygdala response
to masked emotional faces in depressed
subjects resolves with antidepressant treatment:
an fMRI study. Biol. Psychiatry 50, 651658 (2001).
120. Martinowich, K., Manji, H. & Lu, B. New insights into
BDNF function in depression and anxiety. Nature
Neurosci. 10, 10891093 (2007).
121. Krystal, J. H. et al. Neuroplasticity as a target for the
pharmacotherapy of anxiety disorders, mood
disorders, and schizophrenia. Drug Discov. Today 14,
690697 (2009).
122. Machado-Vieira, R., Salvadore, G., DiazGranados, N.
& Zarate, C. A. Jr. Ketamine and the next
generation of antidepressants with a rapid onset
of action. Pharmacol. Ther. 123, 143150 (2009).
123. Vaidya, V. A., Marek, G. J., Aghajanian, G. K. &
Duman, R. S. 5-HT2A receptor-mediated regulation
of brain-derived neurotrophic factor mRNA in the
hippocampus and the neocortex. J. Neurosci. 17,
27852795 (1997).
124. Cavus, I. & Duman, R. S. Influence of estradiol, stress,
and 5-HT2A agonist treatment on brain-derived
neurotrophic factor expression in female rats. Biol.
Psychiatry 54, 5969 (2003).
125. Garcia, L. S. et al. Ketamine treatment reverses
behavioral and physiological alterations induced by
chronic mild stress in rats. Prog.
Neuropsychopharmacol. Biol. Psychiatry 33,
450455 (2009).
126. Studerus, E., Kometer, M., Hasler, F. &
Vollenweider, F. X. Acute, subacute and long-term
subjective effects of psilocybin in healthy humans: a
pooled analysis of experimental studies.
J. Psychopharmacology (in the press).
127. Perry, E. B. Jr et al. Psychiatric safety of ketamine in
psychopharmacology research. Psychopharmacology
(Berlin) 192, 253260 (2007).
128. Savage, C., Savage, E., Fadiman, J. & Harman, W. W.
LSD: Therapeutic effects of the psychedelic experience.
Psychol. Rep. 14, 111120 (1964).
129. Pahnke, W. N., Kurland, A. A., Unger, S., Savage, C. &
Grof, S. The experimental use of psychedelic (LSD)
psychotherapy. JAMA 212, 18561863 (1970).
130. Kurland, A. A., Grof, S. & Panke, W. N. G. L. E. LSD in
the treatment of alcoholics. Pharmakopsychiatr.
Neuropsychopharmakol. 4, 8394 (1971).
131. Griffiths, R. R., Richards, W., Johnson, M., McCann, U.
& Jesse, R. Mystical-type experiences occasioned by
psilocybin mediate the attribution of personal
meaning and spiritual significance 14 months later.
J. Psychopharmacol. 22, 621632 (2008).
132. Griffiths, R. R., Richards, W. A., McCann, U. & Jesse, R.
Psilocybin can occasion mystical-type experiences
having substantial and sustained personal meaning
and spiritual significance. Psychopharmacology
(Berlin) 187, 268283 (2006).
133. Dittrich, A. The standardized psychometric
assessment of altered states of consciousness
(ASCs) in humans. Pharmacopsychiatry 31, 8084
(1998).
134. Vollenweider, F. X. Advances and pathophysiological
models of hallucinogen drug actions in humans: a
preamble to schizophrenia research.
Pharmacopsychiatry 31, 92103 (1998).
135. Fischer, R. A cartography of the ecstatic and
meditative states. Science 174, 897904 (1971).

www.nature.com/reviews/neuro
2010 Macmillan Publishers Limited. All rights reserved

PersPectives
136. Osmond, H. A review of the clinical effects of
psychotomimetic agents. Ann. NY Acad. Sci. 66,
418434 (1957).
137. Kurland, A. A. LSD in the supportive care of the
terminally ill cancer patient. J. Psychoactive Drugs
17, 279290 (1985).
138. Abramson, H. A. The Use of LSD in Psychotherapy
and Alcoholism (Bobbs-Merrill, Indianapolis, 1967).
139. Hollister, L. E., Shelton, J. & Krieger, G. A controlled
comparison of lysergic acid diethylamide (LSD) and
dextroamphetmine in alcoholics. Am. J. Psychiatry
125, 13521357 (1969).
140. Savage, C. & McCabe, O. L. Residential psychedelic
(LSD) therapy for the narcotic addict. A controlled
study. Arch. Gen. Psychiatry 28, 808814 (1973).
141. Grof, S., Goodman, L. E., Richards, W. A. & Kurland,
A. A. LSD-assisted psychotherapy in patients with
terminal cancer. Int. Pharmacopsychiatry 8,
129144 (1973).
142. Pahnke, W. N. Psychedelic drugs and mystical
experience. Int. Psychiatry Clin. 5, 149162
(1969).
143. Grinspoon, L. & Bakalar, J. B. Psychedelic Drugs
Reconsidered (Basic Books., New York, 1979).
144. Crocket, R., Sandison, R. A. & Walk, A. in Proc.
R. MedPsychol. Assoc. (Lewis & Co., London,
1963).
145. Leuner H. in Ethnopsychotherapie (eds Dittrich, A. &
Scharfetter, C.) 151161 (Enke, Stuttgard, 1987)
146. Geert-Jorgensen, E. Further observations regarding
hallucinogenic treatment. Acta Psychiatr. Scand. 203
(Suppl.), 195200 (1968).

147. Khorramzadeh, E. & Lotfy, A. O. The use of ketamine


in psychiatry. Psychosomatics 14, 344346 (1973).
148. Mascher, E. in Neuro-Psychopharmacology (eds Brill, H.,
Cole, J. O., Denker, P., Hippins, H. & Bradley, P. B.)
441444 (Excerpta-Medica, Amsterdam, 2010).
149. Vollenweider, F. X. Brain mechanisms of hallucinogens
and entactogens. Dialogues Clin. Neurosci. 3,
265279 (2001).

Acknowledgements
The authors would like to acknowledge the financial support
of the Swiss Neuromatrix Foundation (to F.X.V. and M.K.),
and of the Heffter Research Institute (to F.X.V.). The authors
thank D. Nichols for critical comments on the manuscript.

Competing interests statement


The authors declare no competing financial interests.

DATABASES
clinicaltrials.gov: http://clinicaltrials.gov
NcT00302744 | NcT00465595 | NcT00920387 |
NcT00947791 | NcT00957359
UniProtKB: http://www.uniprot.org
b-arrestin 2 | eGR1 | eGR2 | mGluR2 | mGluR3

FURTHER inFORMATiOn
University of Zurich Neuropsychopharmacology and Brain
imaging Groups homepage: http://www.dcp.uzh.ch/
research/groups/neuropsychopharmacology.html
All liNks Are Active iN the oNliNe pdf

SCiEnCE AnD SOCiETy

Socioeconomic status and the brain:


mechanistic insights from human
and animal research
Daniel A. Hackman, Martha J. Farah and Michael J. Meaney

Abstract | Human brain development occurs within a socioeconomic context and


childhood socioeconomic status (SeS) influences neural development
particularly of the systems that subserve language and executive function.
Research in humans and in animal models has implicated prenatal factors,
parentchild interactions and cognitive stimulation in the home environment in
the effects of SeS on neural development. These findings provide a unique
opportunity for understanding how environmental factors can lead to individual
differences in brain development, and for improving the programmes and policies
that are designed to alleviate SeS-related disparities in mental health and
academic achievement.
As the field of human neuroscience has
matured, it has progressed from describing
the typical or average human brain to
characterizing individual differences in
brain structure and function, and identifying their determinants. Socioeconomic status (SES), a measure of ones overall status
and position in society, strongly influences
an individuals experiences from childhood and through adult life. Research is
beginning to shed light on the mechanisms

through which experiences in the social


world during early childhood affect the
structure and function of the brain.
Growing up in a family with low SES is
associated with substantially worse health
and impaired psychological well-being, and
impaired cognitive and emotional development throughout the lifespan16. In contrast to sociological and epidemiological
approaches, neuroscience can identify the
underlying cognitive and affective systems

NATURE REVIEWS | NeuroscieNce

that are influenced by SES (BOX 1). In addition, neuroscience research in animals
and in humans has provided candidate
mechanisms for the causeeffect relationships between SES and neural development.
This research has also demonstrated that at
least some of these effects are reversible. Such
a mechanistic understanding will enable the
design of more specific and powerful interventions to prevent and remediate the effects
of low childhood SES79.
Other recent reviews have discussed
research on SES-related differences in
neurocognitive development 79. In this
Perspective, we focus on the candidate
mechanisms by which SES influences brain
development, drawing from research in
humans and in animal models. We first
describe studies in humans that show that
SES influences cognitive and affective function in children, adolescents and young
adults. We then discuss studies in human
populations that have identified possible
mediators of the effects of SES, and review
research in animals in which these factors
were directly manipulated to assess their
effect on offspring outcomes.
SES effects on mental health and cognition
SES is a complex construct that is based
on household income, material resources,
education and occupation, as well as related
neighbourhood and family characteristics,
such as exposure to violence and toxins,
parental care and provision of a cognitively
stimulating environment 2,5,10,11 (for controversies regarding the measurement and
defining levels of SES see REFS 1,10,11). Not
only the lowest stratum but all levels of SES
affect emotional and cognitive development
to varying degrees1,1214. This implies that the
effects of SES that are reviewed here are
relevant to the entire population, although
it should be noted that the strongest effects
are often seen in people with the lowest
levels of SES.
Compared with children and adolescents
from higher-SES backgrounds, children
and adolescents from low-SES backgrounds
show higher rates of depression, anxiety,
attention problems and conduct disorders12,1518, and a higher prevalence of internalizing (that is, depression- or anxiety-like)
and externalizing (that is, aggressive and
impulsive) behaviours6,1921, all of which
increase with the duration of impoverishment 12,21. In addition, childhood SES influences cognitive development; it is positively
correlated with intelligence and academic
achievement from early childhood and
through adolescence2,3,6,14,19,22,23.

VOLUME 11 | SEPTEMbER 2010 | 651


2010 Macmillan Publishers Limited. All rights reserved

The Cinderella of Psychology


The Neglect of Motor Control in the Science of Mental Life and Behavior
David A. Rosenbaum
Pennsylvania State University

One would expect psychologythe science of mental life


and behaviorto place great emphasis on the means by
which mental life is behaviorally expressed. Surprisingly,
however, the study of how decisions are enactedthe focus
of motor control research has received little attention in
psychology. This article documents the neglect and considers possible reasons for it. The hypotheses considered
include three that are raised and then rejected: (a) no
famous psychologists have studied motor control, (b) cognitive psychologists are mainly interested in uniquely human functions, and (c) motor control is simply too hard to
study. Three other hypotheses are more viable: (d) cognitive psychologists have been more interested in epistemology than in action, (e) psychologists have disfavored motor
control because overt responses were the only admissible
measure in behaviorism, and (f) psychologists have felt that
neuroscientists have the market cornered when it comes to
motor control research. There are signs that motor controls Cinderella status is changing.
Keywords: motor control, history of psychology, perception and action, cognitive control of movement, cognitive
neuroscience

newcomer to psychologyfor example, a student


taking an introductory psychology course
would probably be unsurprised to learn that a
major aim of psychological research is to understand how
decisions are enacted. Psychology, this student would
learn, is the science of mental life and behavior, so he or
she would find it natural that many psychologists study
how people reach for and manipulate objects, walk around
obstacles, and control movements required for speaking,
writing, smiling, and gesturing. The work of such investigators, presented under the rubric of motor control, would
complement other topics in psychology such as perception,
learning, emotion, and development.
This scenario is nothing like the reality of what is
taught in introductory psychology courses, nor, for that
matter, other psychology classes. Ironically, psychology
has paid little attention to motor control, which may be
defined as the set of processes that enables creatures
(living or artificial) to stabilize or move the body or physical extensions of the body (tools) in desired ways (Rosenbaum, 2002, p. 315). What accounts for this neglect? Why
have psychologists given short shrift to the translation of
intentions into overt behaviors? In this article, I consider
308

several possibilities. Other authors have commented on the


neglect of motor control in behavioral science (Gazzaniga,
Ivry, & Mangun, 1998; Jeannerod, 1985; Schmidt & Lee,
1999; Wiesendanger, 1997), but no one, to my knowledge,
has provided a full-length treatment of the source of the
neglect in psychology. The main contribution of this article
is to ask why motor control has had the status of a Cinderella in psychological research. The answers to which I
have been led are informative about the factors that motivate psychologists.
The article has four parts. First, I briefly present the
main questions that are pursued by psychologists interested
in motor control. Second, I document psychologys neglect
of motor control. Third, I consider the possible reasons for
the neglect. Fourth, I comment on the future of motor
control in psychology.

Psychological Issues in Motor Control


To describe psychologys neglect of motor control, I will
first clarify the aims of research in this domain. The main
question concerns the necessary and sufficient conditions
for the generation of voluntary movements. One way of
posing the question is to ask how, of all the movements that
could possibly be performed given an actors intentional
state, a particular movement is carried out? Consider the
act of picking up a pen. Although this act is simple, it
requires many decisions. Where along the length of the pen
should the pen be grasped? Which hand should be used?
What should the orientation of the hand be? How should
the grasp depend on what will be done with the pen?
Should the pen be grasped the same way if it will be used
for writing or for poking a hole?
Distinguishing between movement possibilities and
intentional states implies a distinction between asking how
and why a task is performed. The separation of the two
This research was supported by grants from the National Science Foundation, a Research Scientist Development Award from the National Institute of Mental Health, and a grant from the Research Graduate Studies
Office of Pennsylvania State University. I thank Dan Anderson, Jason
Augustyn, Richard Carlson, Rajal Cohen, Rachel Clifton Kean, Martin
Fischer, Steven Jax, Judith Kroll, Elizabeth Loftus, Ruud Meulenbroek,
Dagmar Sternad, and Jonathan Vaughan for helpful comments during
formulation of the article and Richard Ivry and Joe Martinez Jr. for useful
feedback on the submitted manuscript.
Correspondence concerning this article should be addressed to David
A. Rosenbaum, Department of Psychology, Pennsylvania State University, University Park, PA 16802. E-mail: dar12@psu.edu

MayJune 2005 American Psychologist


Copyright 2005 by the American Psychological Association 0003-066X/05/$12.00
Vol. 60, No. 4, 308 317
DOI: 10.1037/0003-066X.60.4.308

David A.
Rosenbaum

need not imply that intention formation and intention enactment are unrelated. Knowledge of how one can perform
affects what one intends to do, just as what one intends to
do affects how one acts. Still, it has proven useful to take
for granted that when an actor performs some voluntary
action, he or she has some goal in mind. Insofar as there are
different means of achieving that goal, the question of
interest for students of motor control is how the performed
movements are selected and controlled.
Researchers interested in the translation of intentions
into physical actions have largely focused on anticipatory
phenomena. The logic of the approach is straightforward. If
the activity of the nervous system prior to the performance
of some motor act differs from the activity of the nervous
system prior to the performance of some other motor act, it
is reasonable to suppose that the state of the nervous system
played a role in differentiating the two acts. Said another
way, the state of the nervous system is a necessary, if not
sufficient, condition for performing any particular act. By
this way of thinking, changes in the nervous system prior to
performance of a motor act reflect the history of the acts
genesis (e.g., Jeannerod, 1988).
The analysis of the precursors of voluntary motor acts
is not restricted to studies of neural activity. Behaviors, too,
betray their histories. Errors in performance provide clues
into the nature of plans for forthcoming actions, whether
for speech (e.g., Dell, 1986; Fromkin, 1973, 1980; Lashley,
1951), typewriting (Cooper, 1983; Rosenbaum, 1991, chap.
8), or other kinds of performance (Norman, 1981; Reason,
1990). Reaction times to begin production of motor sequences also provide information about the processes underlying movement generation (Henry & Rogers, 1960;
Klapp, 1977; Rosenbaum, 1987; S. Sternberg, Monsell,
Knoll, & Wright, 1978).
MayJune 2005 American Psychologist

Studies of motor control have also focused on learning. Through learning, motor acts are performed more
quickly, automatically, and consistently (Schmidt & Lee,
1999). An insight from the study of motor learning is that,
as practice continues, actors achieve greater flexibility in
the way they perform. One way they do so is by unlocking
biomechanical degrees of freedom. Thus, novice pistol
shooters tend to keep their elbows and wrists locked, but
with practice they allow these joints to counterrotate so
extension of one joint compensates for flexion of the other
(Arutyunyan, Gurfinkel, & Mirsky, 1969). Learners can
also acquire the ability to decouple joint motions. Thus,
skilled pianists can achieve greater independence of the
two hands than can novice piano players (Shaffer, 1976).
Similarly, experienced percussionists can generate more
complex polyrhythms than can new drummers (Pressing,
Summers, & Magill, 1996).
Studies of motor control have also focused on the
connection between perception and performance. What one
perceives affects how one acts, just as how one acts affects
what one perceives. Thus, ones capacity to tune ones
actions to the perceptual environment depends on ones
opportunity to actively explore the relations between ones
actions and the perceptions those actions afford (Held,
1965). Such exploration permits prediction of the perceptual consequences of behavior, and such predictions help
one determine whether perceptual changes originate from
changes in the external environment or from ones activity
in the environment. As a result, if the image of the visual
world shifts across the retina and the eye has been commanded to move, the retinal shift can be ascribed to motion
of the eye rather than to motion of the external environment
(von Helmholtz, 1909/1911).
Predicting perceptual consequences of motor acts
plays a key role in motor planning. As James (1890) wrote,
If I will to write Peter rather than Paul, it is the thought
of certain digital sensations, of certain alphabetic sounds,
of certain appearances on the paper, and of no others,
which immediately precedes the motion of my pen
(p. 500).
The idea that plans for motor action include perceptual
goals has received a great deal of support in recent years
(for reviews, see Hommel, Musseler, Aschersleben, &
Prinz, 2001). That plans for motor activity are partly perceptual makes sense from the perspective of feedback
processing. To respond adaptively to movement-related
feedback, one needs to have a goal against which the
feedback can be compared.

Documenting the Neglect


Having summarized some of the concerns of motor control
research, I turn to psychologys neglect of this field. I
document the neglect with reference to two main sources:
(a) coverage in textbooks and (b) coverage in journals.
Coverage in Textbooks
Insofar as textbooks reflect the paradigm of a field (Kuhn,
1970), the coverage of motor control in psychology text309

books gives an indication of the level of interest in this area


of study. It is fitting in this connection to look at the table
of contents of one of the most influential textbooks in
behavioral science, Neissers (1967) Cognitive Psychology.
This volume helped establish cognitive psychology as the
preeminent approach to experimental psychology. Scanning the contents of Neissers book (see Appendix A), one
sees what cognitive psychology entailed for Neisser, at
least in 1967: perceiving, attending, and remembering visual and auditory information, including verbal
information.
Surprisingly, the picture has not changed much in the
ensuing years. Appendix B lists the table of contents of
another successful textbook in cognitive psychology, Ashcrafts (2002) Cognitive Psychology (3rd ed.). The organization of Ashcrafts book is typical of cognitive psychology textbooks on the market today (e.g., Anderson, 2005;
Matlin, 2002; Medin, Ross, & Markman, 2005; R. J. Sternberg, 2003). The main difference between Ashcrafts and
Neissers (1967) tables of contents is that Ashcrafts contents include a chapter on decisions, judgments, and reasoning as well as a chapter on problem solving. The presence of these chapters highlights the growth of knowledge
about higher-level aspects of cognition in the last 40 years
or so.
Coverage in Journals
Because textbooks may not be the best indicators of research activity, it is useful to ask whether research journals
provide a different picture of the level of activity in motor
control research. To find out, I used Web of Science to
obtain counts of articles on selected topics from the Social
Sciences Citation Index. The period I used was as inclusive
as possible on the date of the inquiry, August 11, 2004. The
topics I searched for were the ones Ashcraft (2002) included in his table of contents, although I trimmed or
combined words for purposes of the search (see Table 1).
The one term I queried from Web of Science that was not
included in Ashcrafts list was motor. I did not use the term
movement because this can include studies of visual move-

Table 1
Citations of Selected Topics in the Social Science
Citation Index From 1986 to 2004
Topic

Attention
Cognitive
Decision or judgment or reasoning
Language
Memory
Motor
Perception or pattern recognition

Occurrence

51,946
65,039
54,367
42,205
48,867
17,424
34,328

Note. All topics except for Motor are referred to in Ashcrafts (2002) Cognitive Psychology (3rd ed.; see Appendix B).

310

ment perception as well as political or social tides. As


shown in Table 1, coverage of motor-related topics was
lower than coverage of topics that comprise the standard
fare of cognitive psychology textbooks. This outcome indicates that journals are similar to textbooks in the extent to
which they reflect research activity.

Possible Reasons for the Neglect


Why has motor control been neglected in psychology?
Several explanations come to mind.
The No-Celebrity Hypothesis
One possibility is that no famous psychologists have studied motor control. This hypothesis is worth considering
because luminaries attract acolytes, and if no psychologists
of note have studied movement, it stands to reason that few
psychologists, famous or otherwise, have gravitated to this
topic. In fact, the list of psychologists who have studied
motor control is notable for its celebrity. Robert Woodworth, the author of some of the most masterful reviews of
experimental psychology (Woodworth, 1938; Woodworth
& Schlosberg, 1954), wrote his doctoral dissertation on
manual aiming (Woodworth, 1899). This work became a
classic (for a review, see Elliott, Helsen, & Chua, 2001)
and inspired subsequent influential studies of eye hand
coordination by other well-known psychologists, including
Paul Fitts (1954), Steven Keele and Michael Posner (1968),
and David Meyer, Keith Smith, Sylvan Kornblum, Richard
Abrams, and Charles Wright (1990). Other well-known
psychologists have also made major contributions to the
study of motor control. Among them are Frederic Bartlett
(1932); Karl Lashley (1951); Saul Sternberg, Stephen
Monsell, Ronald Knoll, and Charles Wright (1978); and
Michael Turvey (1990). As this list of names indicates,
established investigators have indeed contributed to the
study of motor control.
The Only-Human Hypothesis
Another hypothesis is that psychologistsand especially
cognitive psychologistsare mainly interested in human
mental life and behavior. Motor control is not very interesting, according to the only-human hypothesis, because
the way humans move does not seem very different from
the way animals move. Thought and language are what
distinguish humans from animals. Consequently, if a cognitive psychology textbook discusses any form of motor
control in detail, it is usually speech production.
Counterarguments can be made to the only-human
hypothesis. One is that a number of uniquely human forms
of action receive little or no coverage in most cognitive
psychology texts. Handwriting and typing are rarely mentioned, for example, though both topics are distinctly human and have received quite a bit of attention from psychologists (Cooper, 1983), including famous ones (Logan,
1983, 2003; Rumelhart & Norman, 1982). A second counterargument is that aspects of perception and cognition that
are not uniquely human are covered in considerable detail
in cognitive psychology textbooks as well as in contempoMayJune 2005 American Psychologist

rary psychology journals (e.g., pattern recognition). Third,


plenty of psychological research is done with nonhuman
species.
The Dumb-Jock Hypothesis
Another possible reason for the neglect of motor control in
psychology is that motor activity does not appear to reflect
much intelligence. According to the dumb-jock hypothesis,
one does not have to be highly intelligent, as measured by
IQ tests, to move well. Hence motor control is not very
interesting.
The dumb-jock hypothesis may have more to it than
first meets the eye, or at least reviewing evidence bearing
on it provides information about previous findings that may
be unfamiliar to many psychologists. It turns out that
several well-known studies have shown that when it is
plausible to expect motor factors to be a limiting factor in
some task or other, intellectual factors actually limit performance. One example is Tolmans (1948) classic demonstration of cognitive maps. Tolman coined the term
cognitive maps after finding that animals learned the spatial
layouts of the mazes they occupied, not the movements
they made within the mazes. Similarly, studies of stimulus
response compatibility have shown that the relation between locations of stimuli and responses mainly determine
reaction times, not which hand makes the response (Proctor
& Reeve, 1990). Mirror neurons, which fire both when
animals perform actions or observe the same actions performed by other individuals (Gallese, Fadiga, Fogassi, &
Rizzolatti, 1996), appear to code the goals of actions rather
than the movements that are made (Rizzolatti, Fogassi, &
Gallese, 2001). Even the interactions between two hands
that arise when one tries to carry out different movements
with the two hands simultaneously (e.g., drawing a curve
with one hand while drawing a straight line with the other;
Franz, Zelaznik, & McCabe, 1991) turn out to be due
primarily to the difficulty of perceiving the positions of the
two hands, not the difficulty of moving the two hands at the
same time (Mechsner, Kerzel, Knoblich, & Prinz, 2001).
These examples suggest that motor behavior is not, so to
speak, where the action is.
Despite these examples, I doubt the dumb-jock hypothesis. One source of doubt is that psychologists have
devoted a great deal of attention to many functions that are
not intellectually intensive. Hunger, thirst, sex, and the
visual perception of texture are examples. The second
source of doubt is that although movement may not take
much intelligence in the conventional sense, current technology, as reflected in robotics, is unable to achieve what
most normal 4-year-olds can do quite easilypeel a banana, climb a tree, or put on a shirt. It may be that
movement does not require the kind of intelligence that is
required to run a company or play chess, but our ignorance
of the intelligence underlying everyday movement is more
vast than our ignorance of white-collar forms of intelligence. This disparity was apparent when IBMs Big Blue
computer beat Gary Kasparov, the worlds best chess
player, in a highly touted match. A human operator moved
the pieces for Big Blue.
MayJune 2005 American Psychologist

The Too-Hard-to-Study Hypothesis


Perhaps motor control is the Cinderella of psychology
because it is too hard to study. This hypothesis has been
publicized by at least two notable contributors to behavioral and neural science. Donald Broadbent (1993), a pioneer in applied and experimental psychology, wrote that
motor performance has always been a neglected and deprived area
of psychology. There are technical reasons for that; it is much
harder to control what a person does than what stimulates them,
and therefore harder to produce scientific laws of the type A is
followed by B. (p. 864)

Edward Evarts (1973), a pioneer in neurophysiology, made


a comparable claim in his description of early neural recording techniques:
One difficulty in studying volitional movement arose from the
necessity of having the active participation of the experimental
subject; that precluded the use of an anesthetized animal. Research on sensory processes moved ahead rapidly because sensory
functions could be tested in such an animal. For example, the
physiology of visual receptors could be studied in anesthetized
animals, but the physiology of eye movement could not since such
studies required animals capable of perception, attention, and
coordinated motor function. (p. 96)

The methodological difficulties that Evarts (1973) described were later overcome, thanks in part to his own
refinement of microelectrode technology. A great deal of
research has subsequently been done on the brain activity
of awake, moving animals. Still, the range of movements
that is possible while brain activity is being recorded remains small compared with what is possible in the everyday environment because the technology used for recording
brain activity in awake subjects (e.g., functional magnetic
resonance imagery) requires the use of scanners that significantly limit mobility.
In addition to the technical difficulties of studying
motor control, the processes underlying movement planning and generation are relatively immune to conscious
inspection. As James (1890) wrote, For many actions, we
are aware of nothing between the conception and the execution. All sorts of neuromuscular processes come between
. . . but we know absolutely nothing of them. We think the
act, and it is done (p. 790).
The fact that psychologists seem to have no sense of
motor innervationa topic discussed at length by James
(1890)may have put the study of motor control at a
disadvantage, compared, say, with the study of visual perception, where visual images can be formed in ones mind.
There are problems with the too-hard-to-study-hypothesis, however. One is that much of what psychologists
know about motor control was discovered with techniques
that are no more complex than the techniques used to study
other psychological phenomena. Psychologists have
learned about the planning and control of movements by
recording mistakes that people make, by recording how
long it takes to initiate and perform predetermined motor
sequences, and by using other methods that are common in
experimental psychology. For example, using a simple
311

video recording system, Cohen and Rosenbaum (2004)


showed that when people took hold of a vertically oriented
rod to move it from one position to another, they took hold
of the rod at a height that was inversely related to the height
to which it would be brought. This result corroborates the
hypothesis that people anticipate final body positions when
carrying out voluntary movements (Rosenbaum, Meulenbroek, Vaughan, & Jansen, 2001). The fact that this principle could be supported with little more than a stick and a
camcorder shows that motor control is not too hard to study
so long as one is interested in studying it.
The other problem with the too-hard-to-study hypothesis is that problems of measurement rarely dissuade people from developing solutions to those problems when they
are curious enough. An example is the movie camera,
whose invention was spurred by the desire to see individual
frames of rapid movements so that, among other things,
one could determine whether a galloping horse had all its
hooves off the ground at any timea hotly debated topic in
the 19th century (Muybridge, 1887/1957). The commercial
potential of cinema only came to be recognized later (Newhall, 1999).
The Think-Before-You-Act Hypothesis
All the hypotheses considered so far were ones I raised and
then dismissed. Now I consider three hypotheses that strike
me as more viable. The first is the think-before-you-act
hypothesis. The idea is that the core question in cognitive
psychologyWhat is knowledge?is not one that naturally inspires work on the question, How do people move?
Scientific psychology originated in philosophy, many of
whose long-standing questions had to do with epistemology: How do people come to know the world? Can people
know the world as it really is or only as they imagine it?
and so on. Inheriting these concerns, psychologists were
naturally inclined to investigate the topics listed in most
cognitive psychology textbooks today: perception, attention, learning, and memory. Reasoning, decision making,
and problem solving also fit in because they may illuminate
how and what people learn.
If perceiving and knowing have higher priority than
motor control, one would expect a boost in the study of
movement when such study can provide new insights into
perceiving and knowing. Just such a boost occurred in the
past few years because of research concerning a woman
who suffered damage to the ventral pathway of her visual
system (Milner & Goodale, 1995). When asked to report
the seen size or orientation of a bar, the woman typically
performed at chance, but when she reached for the bar, her
hand approached it as if the bars visual size and orientation
were available to her. Her inability to use vision for recognition contrasted sharply with her ability to use vision for
movement. The disparity suggested that she could see for
the sake of action but not for the sake of identification.
On the basis of these and other observations, Milner
and Goodale (1995) proposed that there is a how visual
system and a what visual system. For students of perception, this provocative claim made the study of hand
movements attractive for investigating visual perception.
312

Consequently, many studies have since been done on visual


factors affecting manual control (for a review, see Glover,
2002). Such studies have been useful not just for shedding
light on Milner and Goodales howwhat distinction; they
have also permitted new insights into the control of hand
movements per se (e.g., Glover & Dixon, 2001).
Other studies of motor control have been similarly
inspired by perceptual questions. For example, to explore
the possibility that spatial attention depends on the actions
carried out in space, Tipper, Lortie, and Baylis (1992)
showed that hand movements directed to visual targets
were affected by the presence of visual distractors. This
work inspired further research on action-related attention
(Humphreys & Riddoch, 2003). Similarly, to test Piagets
(1936/1952) assertion that, for babies, out of sight means
out of mind, Clifton, Rochat, Litovsky, and Perris (1991)
studied babies reaches to sounding objects that first were
visible but then were plunged into darkness. The babies
reaches were well directed to the objects, including to parts
of the objects that were removed from the sound source,
indicating that out of sight does not mean out of mind,
contra Piaget. In much the same vein, Smith, Thelen,
Titzer, and McLin (1999) challenged Piagets claims about
the immaturity of infants understanding of objects as permanent entities. The challenge was made possible by carefully investigating babies reaches to seen and hidden
objects.
The Baby-With-the-Bathwater Hypothesis
I turn now to the penultimate hypothesis concerning psychologys neglect of actionthe baby-with-the-bathwater
hypothesis. According to this hypothesis, when mainstream
psychology rejected behaviorisman approach that treated
the response as the only admissible source of psychological
datait eschewed response measures more sweepingly
than would have occurred otherwise. The study of motor
control was guilty by association. Motor behavior was
associated with mindlessness, and a mindless responsecentered program of research was anathema to psychologists basking in the glow of cognitivism.
If the baby-with-the-bathwater hypothesis is correct,
one would expect the study of action to fare better when
and where behaviorism has not held sway than when and
where it has. Consistent with this expectation, the study of
movement prospered in America at the end of the 19th
century and early in the 20th century, before the advent of
John B. Watson. Important contributions to motor control
research were made in this period. For a review, see
Schmidt and Lee (1999).
Concerning the where of action research, in England
and continental Europe, where behaviorism never took
hold as it did in America, there have been long-standing
programs of research on movement. A reflection of the
healthy state of motor control research in England is that a
textbook by a team of British cognitive psychologists
(Smyth, Collins, Morris, & Levy, 1994) entitled Cognition
in Action has chapters with names like Reaching for a
Glass of Beer (chap. 4) and Tapping Your Head and
Rubbing Your Stomach (chap. 5). In Germany and HolMayJune 2005 American Psychologist

land, university students majoring in psychology are invited to take courses in motor control. Such classes are rare
in America.
These points notwithstanding, a predilection of Americas foremost behaviorist had the surprising effect of turning psychologists away from movement. When B. F. Skinner promoted operant conditioning, he downplayed the
importance of body movements per se, stressing instead the
instrumental effects of muscle activity. Thus, whether a
pigeon pecked a key or kicked the key did not much matter
to Skinner, though it mattered much to the pigeon. Paradoxically, Skinners pooh-poohing of movements appears
to have struck a chord with psychologists. With few exceptions, the analysis of body movements has received
little attention in psychology except as a means of addressing other questions. Thus, facial expressions have been
used to study emotion (Ekman & Oster, 1979), eye movements have been used to study reading (Rayner, 1983),
eyeblinks have been used to study memory (McCormick &
Thompson, 1984), and hand gestures have been used to
study language (Goldin-Meadow, 1999). The movements
people make to control external devices such as buttons or
joysticks have generally been less studied than the participants disposition to use the devices, as measured by
response probabilities or reaction times. Generally, the
question of how movements are controlled has been
ignored.
The Neuroscientists-Have-It-Covered
Hypothesis
The final viable hypothesis about the cause of psychologys
neglect of motor control is that motor control has long been
a forte of neuroscience. Why study a topic when another
group of researchers handles it well? To evaluate the neuroscientists-have-it-covered hypothesis, I used Web of Science to access the Science Citation Index in order to count
articles on the same topics as I had checked earlier in the
Social Science Citation Index. To limit the journals to ones
that were relevant, I restricted the search to journals that
had the word brain or the letters neur in their title. The
results appear in Table 2, where it is seen that the topic that
yielded the most citations was motor. Figure 1 shows the
relation between number of citations for the same set of
topics in Social Science Citation Index and Science Citation Index. Overall, there was a negative relation between
the number of citations in the two sources, with motor
being the topic with the greatest disparity in citation counts.
Why has motor-control research prospered in neuroscience? Apart from the fact that motor disorders cry out
for medical research, motor neurophysiologists can precisely stimulate different parts of the nervous system and
record the ensuing motor effects. Thus, motor neurophysiologists can escape the problem Broadbent (1993, p. 863)
identified: [I]t is much harder to control what a person
does than what stimulates them.
Does the success of motor-control research in neuroscience account for psychologys neglect of this topic? I
doubt it is the only source of the neglect, but I think it has
been an important one. Psychologists, like professionals in
MayJune 2005 American Psychologist

Table 2
Citations of Selected Topics in the Science Citation
Index for Journals Containing Brain or Neur in
Their Titles From 1986 to 2004
Topic

Attention
Cognitive
Decision or judgment or reasoning
Language
Memory
Motor
Perception or pattern recognition

Occurrence

3,747
6,049
1,177
1,833
8,537
10,913
2,791

Note. All topics except for Motor are referred to in Ashcrafts (2002) Cognitive Psychology (3rd ed.; see Appendix B).

any field, are less prone to pursue topics that are well
covered in other disciplines, particularly when they feel
they may have nothing special to offer. Psychologists do in
fact have something special to offer the study of motor
control: They can analyze macroscopic as well as microscopic aspects of behavior, and they can exploit their
knowledge of experimental design to reveal functional
principles that might otherwise go unnoticed. Still, if psychologists come from a tradition that is epistemologically
rather than action-based (the think-before-you-act hypothesis), if their tradition has made motor behavior a pariah
rather than an attractive research target (the baby-with-thebathwater hypothesis), and if grant money and other
sources of recognition are more liberally doled out to
scientists in other fields (the neuroscientists-have-it-covered hypothesis), there is not much incentive for psychologists to get on the move.

The Future
In the story of Cinderella, a modest chamber maid, abandoned by her wicked stepmother and stepsisters, is rescued
by a handsome prince. My aim in likening motor control to
Cinderella has not been to equate research domains to
wicked relatives, nor to equate myself with a prince, handsome or otherwise. Instead, my aim has been to point out
that motor control, which one may argue lies at the heart of
the science of mental life and behavior because it joins the
two, has had a surprisingly modest presence in psychology.
The reasons, I have suggested, are intellectual and economic. Intellectually, psychology grew out of philosophy,
where questions of knowing were taken to be quintessential
to epistemology. Only recently have psychologists come to
appreciate that acting and knowing are inseparable (Carlson, 1997), and only recently have psychologists come to
appreciate that purposeful movement helps initiate or sustain perceptionaction cycles rather than just being a response to input (for a particularly eloquent, early statement
of this position, see Weimer, 1977). Economically, psychologists have been inclined to work on problems for
313

Figure 1
Number of Articles in the Social Science Citation Index (Abscissa) and Science Citation Index (Ordinate)
Pertaining to Each Topic Listed in the Graph

Note. Values on the abscissa are 1/10,000 the number of reported values. Values on the ordinate are 1/1,000 the number of reported values. For the Science
Citation Index, the only journals included had the word brain or the letters neur in their titles.

which they were especially well equipped. Thus, motor


control, long a jewel in the crown of neuroscience, became
less attractive than other topics for which psychologists felt
they could make more distinctive contributions.
Will psychologists pay more attention to motor control in the future? There are reasons to think they will. One
is that the division between neuroscience and psychology is
blurring. Neuroscientists are becoming more interested in
the insights that psychologists can provide and vice versa.
As more neuroscientists identify with psychologists and as
more psychologists identify with neuroscientists, motor
control is becoming an interdisciplinary topic to which
psychologists are being invited. One sign that such a
change is occurring is the recent appearance of the first
American cognitive psychology textbook with an entire
chapter devoted to motor control, Cognition: The Thinking
Animal, by Willingham (2004), a cognitive neuroscientist.
Another reason to expect motor control to become
more popular in psychology is the emergence of ecological
psychology and dynamical systems analysis. Advocates of
ecological psychology argue that the primary function of
perception is to guide action (Gibson, 1979) and that the
control of action enlists rather than resists physical properties of actor environment couplings (Bernstein, 1967).
Advocates of dynamical systems analysis seek to describe
ongoing cycles of perceiving and acting in the form of
differential equations (e.g., Sternad, Duarte, Katsumata, &
Schaal, 2001). The advent of the ecological and dynamical
314

systems perspectives has fostered the analysis of classes of


behavior that were left out of the research portfolio of
traditional cognitive psychological research, which focused
on internal representations and computations to the exclusion of embodied cognition (Clark, 1997; Glenberg, 1997).
Newly studied topics include walking and jumping (Goldfield, Kay, & Warren, 1993; Thelen, 1995), juggling (Beek
& Turvey, 1992), skiing (Vereijken, Whiting, & Beek,
1992), pistol shooting (Arutyunyan et al., 1969), wielding
objects (Carello & Turvey, 2004), bouncing a ball on a
tennis racquet (Sternad et al. 2001), swinging two handheld pendulums of different lengths and weights (Turvey,
1990), and oscillating two index fingers at different frequencies and relative phases (Zanone & Kelso, 1997).
A third reason to expect a growth of interest in motor
control is that there is an expanding appreciation of the
computational challenges of skilled movement. Although
humanoid robots can walk in controlled environments
(Sony QRIO Honda Asimo), can vocalize (KRT-v.3; Kagawa University, Takamatsu, Japan), can smile and frown
(WE-4R; Waseda University, Tokyo, Japan), can play the
trumpet (Toyotas Partner robot), and can hit baseballs
(University of Tokyo), they are poor at performing in
open-ended situations where novel movements are required
(see http://informatiksysteme.pt-it.de/mti-2/cd-rom/index
.html). Thus, robots cannot clear tables at restaurants, make
beds in hotel rooms, or open and inspect luggage at airports. Engineers interested in building better robots have
MayJune 2005 American Psychologist

become interested in biological perception and action to


improve robot design. Their interest in this topic may spur
more psychological research on motor control and, from
there, the connections between mental life and behavior.
REFERENCES
Anderson, J. R. (2005). Cognitive psychology and its implications (6th
ed.). New York: Worth.
Arutyunyan, G. H., Gurfinkel, V. S., & Mirsky, M. L. (1969). Investigation of aiming at a target. Biophysics, 13, 536 538.
Ashcraft, M. H. (2002). Cognitive psychology (3rd ed.). Upper Saddle
River, NJ: Prentice Hall.
Bartlett, F. C. (1932). Remembering. London: Cambridge University
Press.
Beek, P. J., & Turvey, M. T. (1992). Temporal patterning in cascade
juggling. Journal of Experimental Psychology: Human Perception and
Performance, 18, 934 947.
Bernstein, N. (1967). The coordination and regulation of movements.
London: Pergamon Press.
Broadbent, D. E. (1993). A word before leaving. In D. E. Meyer & S.
Kornblum (Eds.), Attention and performance XIV: Synergies in experimental psychology, artificial intelligence, and cognitive neuroscience
(pp. 863 879). Cambridge, MA: MIT Press.
Carello, C., & Turvey, M. T. (2004). Physics and psychology of the
muscle sense. Current Directions in Psychological Science, 13, 2528.
Carlson, R. A. (1997). Experienced cognition. Mahwah, NJ: Erlbaum.
Clark, A. (1997). Being there: Putting brain, body, and world together
again. Cambridge, MA: MIT Press.
Clifton, R. K., Rochat, P., Litovsky, R. Y., & Perris, E. E. (1991). Object
representation guides infant reaching in the dark. Journal of Experimental Psychology: Human Perception and Performance, 17, 323329.
Cohen, R. G., & Rosenbaum, D. A. (2004). Where objects are grasped
reveals how grasps are planned: Generation and recall of motor plans.
Experimental Brain Research, 157, 486 495.
Cooper, W. E. (Ed.). (1983). Cognitive aspects of skilled typewriting. New
York: Springer-Verlag.
Dell, G. S. (1986). A spreading activation theory of retrieval in sentence
production. Psychological Review, 93, 283321.
Ekman, P., & Oster, H. (1979). Facial expressions of emotion. Annual
Review of Psychology, 30, 527554.
Elliott, D., Helsen, W. F., & Chua, R. (2001). A century later: Woodworths (1899) two-component model of goal-directed aiming. Psychological Bulletin, 127, 342357.
Evarts, E. (1973). Brain mechanisms in movement. Scientific American,
229, 96 103.
Fitts, P. M. (1954). The information capacity of the human motor system
in controlling the amplitude of movement. Journal of Experimental
Psychology, 47, 381391.
Franz, E. A., Zelaznik, H. N., & McCabe, G. (1991). Spatial topological
constraints in a bimanual task. Acta Psychologica, 77, 137151.
Fromkin, V. A. (Ed.). (1973). Speech errors as linguistic evidence. The
Hague, the Netherlands: Mouton.
Fromkin, V. A. (Ed.). (1980). Errors in linguistic performance. New
York: Academic Press.
Gallese, V., Fadiga, L., Fogassi, L., & Rizzolatti, G. (1996). Action
recognition in the premotor cortex. Brain, 119, 593 609.
Gazzaniga, M. S., Ivry, R. B., & Mangun, G. R. (1998). Cognitive
neuroscience: The biology of the mind. New York: Norton.
Gibson, J. J. (1979). The ecological approach to visual perception. Boston: Houghton Mifflin.
Glenberg, A. M. (1997). What memory is for. Behavioral and Brain
Sciences, 20, 155.
Glover, S. R. (2002). Visual illusions affect planning but not control.
Trends in Cognitive Sciences, 6, 288 292.
Glover, S. R., & Dixon, P. (2001). Dynamic illusion effects in a reaching
task: Evidence for separate visual representations in the planning and
control of reaching. Journal of Experimental Psychology: Human Perception and Performance, 27, 560 572.
Goldfield, E. C., Kay, B. A., & Warren, W. H. (1993). Infant bouncing:

MayJune 2005 American Psychologist

The assembly and tuning of action systems. Child Development, 64,


1128 1142.
Goldin-Meadow, S. (1999). The role of gesture in communication and
thinking. Trends in Cognitive Sciences, 3, 419 429.
Held, R. (1965). Plasticity in sensory-motor systems. Scientific American,
213(5), 84 94.
Henry, F. M., & Rogers, D. E. (1960). Increased response latency for
complicated movements and a memory drum theory of neuromotor
reaction. Research Quarterly, 31, 448 458.
Hommel, B., Musseler, J., Aschersleben, G., & Prinz, W. (2001). The
theory of event coding (TEC): A framework for perception and action
planning. Behavioral and Brain Sciences, 24, 849 937.
Humphreys, G. W., & Riddoch, M. J. (2003). From vision to action and
action to vision: A convergent route approach to vision, action, and
attention. In D. E. Irwin & B. H. Ross (Eds.), Psychology of learning
and motivation: Advances in research and theory (Vol. 42, pp. 225
264). New York: Academic Press.
James, W. (1890). The principles of psychology (Vol. 2). New York: Holt.
Jeannerod, M. (1985). The brain machine: The development of neurophysiological thought. Cambridge, MA: Harvard University Press.
Jeannerod, M. (1988). The neural and behavioral organization of goaldirected movements. Oxford, England: Oxford University Press.
Keele, S. W., & Posner, M. I. (1968). Processing of visual feedback in
rapid movements. Journal of Experimental Psychology, 77, 155158.
Klapp, S. T. (1977). Reaction time analysis of programmed control.
Exercise and Sport Sciences Reviews, 5, 231253.
Kuhn, T. S. (1970). The structure of scientific revolutions. Chicago:
University of Chicago Press.
Lashley, K. S. (1951). The problem of serial order in behavior. In L. A.
Jeffress (Ed.), Cerebral mechanisms in behavior (pp. 112131). New
York: Wiley.
Logan, G. D. (1983). Time, information, and the various spans in typewriting. In W. E. Cooper (Ed.), Cognitive aspects of skilled typewriting
(pp. 197224). New York: Springer-Verlag.
Logan, G. D. (2003). Simon-type effects: Chronometric evidence for
keypress schemata in typewriting. Journal of Experimental Psychology:
Human Perception and Performance, 29, 741757.
Matlin, M. W. (2002). Cognition (5th ed.). New York: Wiley.
McCormick, D. A., & Thompson, R. F. (1984, January 20). Cerebellum:
Essential involvement in the classically conditioned eyelid response.
Science, 223, 296 299.
Mechsner, F., Kerzel, D., Knoblich, G., & Prinz, W. (2001, November 1).
Perceptual basis of bimanual coordination. Nature, 414, 69 73.
Medin, D. L., Ross, B. H., & Markman, A. B. (2005). Cognitive psychology (4th ed.). New York: Wiley.
Meyer, D. E., Smith, J. E. K., Kornblum, S., Abrams, R. A., & Wright,
C. E. (1990). Speedaccuracy tradeoffs in aimed movements: Toward a
theory of rapid voluntary action. In M. Jeannerod (Ed.), Attention and
performance XIII: Motor representation and control (pp. 173226).
Hillsdale, NJ: Erlbaum.
Milner, A. D., & Goodale, M. A. (1995). The visual brain in action. New
York: Oxford University Press.
Muybridge, E. (1957). Animals in motion. New York: Dover. (Original
work published 1887)
Neisser, U. (1967). Cognitive psychology. New York: AppletonCentury-Crofts.
Newhall, B. (1999). The history of photography: From 1839 to the present
(5th ed.). New York: Museum of Modern Art.
Norman, D. A. (1981). Categorization of action slips. Psychological
Review, 88, 115.
Piaget, J. (1952). The origins of intelligence in children (M. Cook, Trans.).
New York: International Universities Press. (Original work published
1936)
Pressing, J., Summers, J., & Magill, J. (1996). Cognitive multiplicity in
polyrhythmic pattern performance. Journal of Experimental Psychology: Human Perception and Performance, 22, 11271148.
Proctor, R. W., & Reeve, G. (Eds.). (1990). Stimulusresponse compatibility. Amsterdam: North-Holland.
Rayner, K. (Ed.). (1983). Eye movements in reading. New York: Academic Press.
Reason, J. (1990). Human error. Cambridge, England: Cambridge University Press.

315

Rizzolatti, G., Fogassi, L., & Gallese, V. (2001). Neurophysiological


mechanisms underlying the understanding and imitation of action.
Nature Reviews Neuroscience, 2, 661 670.
Rosenbaum, D. A. (1987). Successive approximations to a model of
human motor programming. In G. H. Bower (Ed.), Psychology of
learning and motivation (Vol. 21, pp. 153182). Orlando, FL: Academic Press.
Rosenbaum, D. A. (1991). Human motor control. San Diego: Academic
Press.
Rosenbaum, D. A. (2002). Motor control. In H. Pashler (Series Ed.) & S.
Yantis (Vol. Ed.), Stevens handbook of experimental psychology: Vol.
1. Sensation and perception (3rd ed., pp. 315339). New York: Wiley.
Rosenbaum, D. A., Meulenbroek, R. G., Vaughan, J., & Jansen, C. (2001).
Posture-based motion planning: Applications to grasping. Psychological Review, 108, 709 734.
Rumelhart, D. E., & Norman, D. A. (1982). Simulating a skilled typist: A
study of skilled cognitive-motor performance. Cognitive Science, 6,
136.
Schmidt, R. A., & Lee, T. D. (1999). Motor control and learning: A
behavioral emphasis (3rd ed.). Champaign, IL: Human Kinetics.
Shaffer, L. H. (1976). Intention and performance. Psychological Review,
83, 375393.
Smith, L. B., Thelen, E., Titzer, R., & McLin, D. (1999). Knowing in the
context of acting: The task dynamics of the A-not-B error. Psychological Review, 106, 235260.
Smyth, M. L., Collins, A. F., Morris, P. E., & Levy, P. (1994). Cognition
in action (2nd ed.). East Sussex, England: Psychology Press.
Sternad, D., Duarte, M., Katsumata, H., & Schaal, S. (2001). Bouncing a
ball: Tuning into dynamic stability. Journal of Experimental Psychology: Human Perception and Performance, 27, 11631184.
Sternberg, R. J. (2003). Cognition (3rd ed.). Belmont, CA: Wadsworth.
Sternberg, S., Monsell, S., Knoll, R. L., & Wright, C. E. (1978). The
latency and duration of rapid movement sequences: Comparisons of
speech and typewriting. In G. E. Stelmach (Ed.), Information processing in motor control and learning (pp. 117152). New York: Academic
Press.

316

Thelen, E. (1995). Motor development: A new synthesis. American Psychologist, 50, 79 95.
Tipper, S. P., Lortie, C., & Baylis, G. C. (1992). Selective reaching:
Evidence for action-centered attention. Journal of Experimental Psychology: Human Perception and Performance, 18, 891905.
Tolman, C. E. (1948). Cognitive maps in rats and man. Psychological
Review, 55, 189 208.
Turvey, M. T. (1990). Coordination. American Psychologist, 45,
938 953.
Vereijken, B., Whiting, H. T. A., & Beek, W. J. (1992). A dynamicsystems approach to skill acquisition. Quarterly Journal of Experimental Psychology: Human Experimental Psychology, 45(A), 323344.
von Helmholtz, H. (1911). Treatise on physiological optics (J. P. Southall,
Ed. & Trans.; 3rd ed., Vol. 3). Rochester, NY: Optical Society of
America. (Original work published 1909)
Weimer, W. B. (1977). A conceptual framework for cognitive psychology: Motor theories of the mind. In R. Shaw & J. Bransford (Eds.),
Perceiving, acting, and knowing: Toward an ecological psychology
(pp. 267311). Hillsdale, NJ: Erlbaum.
Wiesendanger, M. (1997). Paths of discovery in human motor control. In
M. C. Hepp-Reymond & G. Marini (Eds.), Perspectives of motor
behavior and its neural basis (pp. 103134). Basel, Switzerland:
Karger.
Willingham, D. T. (2004). Cognition: The thinking animal (2nd ed.).
Upper Saddle River, NJ: Prentice Hall.
Woodworth, R. S. (1899). The accuracy of voluntary movement. Psychological Review Monograph Supplements, 3(3, Suppl. 13), 1119.
Woodworth, R. S. (1938). Experimental psychology. New York: Holt.
Woodworth, R. S., & Schlosberg, H. (1954). Experimental psychology
(2nd ed.). New York: Holt.
Zanone, P. G., & Kelso, J. A. S. (1997). Coordination dynamics of
learning and transfer: Collective and component levels. Journal of
Experimental Psychology: Human Perception and Performance, 23,
1454 1480.

MayJune 2005 American Psychologist

Appendix A
Brief Contents of Neissers (1967) Cognitive Psychology
1. The Cognitive Approach
2. Iconic Storage and Verbal Coding
3. Pattern Recognition
4. Focal Attention and Figural Synthesis
5. Words as Visual Patterns
6. Visual Memory
7. Speech Perception
8. Echoic Memory and Auditory Attention
9. Active Verbal Memory
10. Sentences
11. A Cognitive Approach to Memory and Thought

Appendix B
Brief Contents of Ashcrafts (2002) Cognitive Psychology
(3rd ed.)
1. Cognitive Psychology: An Introduction
2. The Cognitive Science Approach
3. Perception and Pattern Recognition
4. Attention
5. Short-Term Working Memory
6. Episodic Long-Term Memory
7. Semantic Long-Term Memory
8. Interactions in Long-Term Memory
9. Language
10. Comprehension: Written and Spoken Language
11. Decisions, Judgments, and Reasoning
12. Problem Solving

MayJune 2005 American Psychologist

317