Вы находитесь на странице: 1из 15

E RGO N OM ICS , 1998,




12, 1845 1858

W orking in a moving environment

TNO Human Factors Research Institute, PO Box 23, 3769 ZG Soesterberg, The
Keywords: Motion; Simulators; Work load; Task performance; Motion sickness.
The present paper provides a review of research and theories concerning the
question of how and why working in a moving environment may aect
performance. It is argued that performance decrem ents can be expected to
occur as a result of general factors or as a result of speci c impairments of
particular hum an skills. General eects happen when environm ental motion,
simulated or real, reduces motivation (due to motion sickness), increases
fatigue (due to increased energy requirements), or creates balance problem s.
Speci c eects of moving environm ents on task performance may only be
expected through biomechanical in uences on particular skills such as
perception (interference with oculom otor control) or motor skills (such as
manual tracking). There is no evidence for direct eects of motion on
performance in purely cognitive tasks.

1. Introduction
W ith the current fast rate of technological developments, the use of simulators,
especially moving-base simulators, is rapidly increasing. They are used especially for
purposes of training to carry out tasks and work in m oving environments, such as
aircraft, road vehicles and ships. This has created many challenges for human factors
researchers, who are often asked to investigate such questions as the validity of the
simulators or the e ciency of training procedures. It is surprising, however, that
much of the research carried out with simulators assum es that we know how work is
actually carried out in a moving environm ent. This is not alw ays the case. M any
investigators are only dimly aw are of research on how movement aects
performance. To help them ll this gap, the present paper presents a review of
what is currently known about the nature of human performance in moving
environments, both in simulators and real environments. It has not been the
intention to describe in detail all published material on these issues. Instead the idea
is to summ arize current knowledge, and to provide references for those who want to
obtain more detailed information.
The review is structured along a classi cation of the eects of environmental
motion on performance in terms of two categories: general eects and speci c
eects. General eects refer to any task, any performance, carried out in a
moving environment. Such eects may be of a motivational nature (motion
sickness), of an energetical nature (motion-induced fatigue caused by the
continuous muscular eort to maintain balance), or of a biomechanical nature
(interference with task performance because of a loss of balance). Speci c eects,
on the other hand, refer to interference with speci c human abilities (e.g.
cognition, perception, etc.).
0014 0139 / 98 $12.00

1998 T aylor & Francis Ltd


A. H. W ertheim

2. General eects of environm ental motion on performance

2.1. M otion sickness
There are various kinds of motion sickness, such as sea sickness, car sickness, air
sickness, space sickness, and some people are sick in trains or even in elevators. A
particularly noteworthy kind of motion sickness is simulator sickness, occurring
frequently both in xed and in moving-base simulators as well as in virtual
environm ent situations.
2.1.1. M otion sickness in general: One of the best-known phenom ena occurring in
a moving environment is motion sickness. It causes a m assive lowering of
motivation, usually resulting in a considerable slowing down of work rate, a
disruption of continuous work and often its complete abandonm ent. Sensitivity to
motion sickness varies widely am ong humans. For exam ple, tolerance to sea sickness
is very high with children a few years old, is then reduced and at old age increases
again . Furthermore, sea sickness may develop fast or slowly, depending on the
individual, while some people appear to be resistant to it. W omen are generally
somewh at m ore sensitive than m en. It is also known that motion sickness in a long
duration motion environm ent usually decreases with time (generally referred to as
adaptation). Adaptation may take a few hours as in centrifuge-induced m otion
sickness (see below) or a few days, as with sea or space sickness. However, the time
it takes for the sym ptoms to disappear may vary with circumstances, such as the type
of wave movem ents, and diers am ong individuals. W ith approxim ately 5% of
humans adaptation to sea sickness does not take place at all (see Colwell (1989) for
some reviews of these issues).
It is not really possible to test a person s general sensitivity to motion sickness.
W ithin individuals, there is no direct correlation between sensitivity to the various
forms of motion sickness (Bles et al. 1984). Nevertheless, it has long been known that
the one necessary requirement for any kind of motion sickness is a functioning
vestibular apparatus. People who do not have a functioning vestibular apparatus
(because of particular illnesses) simply cannot become motion sick (Kennedy et al.
This is not the proper place to present a detailed description of how the vestibular
apparatus works. M any good texts on the subject are availab le elsewhere (G uedry
1974, Howard 1986). Here it su ces to note that the central role of the vestibular
system is recognized in what is currently the most well-known theory of m otion
sickness (Reason and Brand 1975), usually referred to as the Theory of Intersensory
M ismatch.
According to this theory, motion sickness occurs when the vestibular apparatus
provides the brain with information about self-motion that does not m atch precisely
the sensations of self-motion generated by other sensory systems (such as the visual
or kinaesthetic system s), or what is expected from previous experience.
W e have two peripheral vestibular systems, one in each inner ear. Each one
consists of several sub-systems: linear accelerations (in the vertical gravitational
direction, in the horizontal forward / backward direction, and in sideward directions)
are picked up by two otoliths. Rotational accelerations along those three axes are
picked up by three semi-circular canals, which lie in dierent orthogonal planes.
Hence, sensory mism atches may also occur within the vestibular apparatus itself.
The mismatch theory has its shortcomings, and in recent years alternative
theories have been formulated. For exam ple, a recent view is that motion sickness

W orking in a moving environm ent


does not occur when there is a sensory mismatch concerning perceived self-motion,
but when there is am bivalence about the perceived vertical (Bles and De Graaf 1993).
Although this theory deviates from the traditional view, it is related to the traditional
theory, because the perceived vertical is in uenced by self-motion. There are some
other alternative theories (Storegen and Riccio 1986, Yardley 1992, Oman 1982),
and there are ideas about cognitive in uences on motion sickness (Dobie an d M ay
1994). However, here we will take sensory mismatch theory as our point of
departure, as it is still the most widely accepted theoretical approach in the literature
on motion sickness, and it explains a large variety of motion sickness phenomena.
To illustrate the theory, consider the situation of a person inside a closed cabin
on a ship at sea. The m otions of the ship are picked up by the vestibular apparatus
and they inform the brain that the head (i.e. body) moves in space. The visual
system , however, perceives a stationary environment, as the walls of the cabin do not
move across the eyes (i.e. the retinae). Consequently vestibular and visual
information about self-movement are not identical. This is a typical sensory
mismatch situation where m otion sickness may easily develop (Hettinger et al. 1990).
The theory also explains some remedies for m otion sickness. For exam ple, in the
case of a ship at sea, it is well-known that providing the visual system with an optic
pattern that remains stable relative to the world (e.g. a horizon as seen on deck or
through a large window) reduces the incidence and severeness of motion sickness. In
fact there have been some attempts (Rolnick and Bles 1989, Bles et al. 1991) to
investigate possible motion sickness reducing eects of an arti cial horizon. In these
studies the arti cial horizon consisted of a line, projected across the walls of a ship s
cabin, which moved in synchrony with the ship s pitch (forward-backward tilting)
movem ents and roll (leftward-rightward tilting) movem ents. Another option
which is still rather speculative, as it is only based on some informal observations of
the author, but which is related to the above mentioned theory about the subjective
vertical is to try and hold one s stance align ed to the real (gravitational) vertical.
In a ship s cabin, this may require a kind of balancing act that makes the body (and
therefore the head ) move quite strongly relative to the walls of the cabin, thus
providing the visual system with at least som e optic ow across the eyes, consistent
with the ship s motion.
In a fam ous series of experim ents, carried out in a Ship M otion Simulator (SM S),
M cCauley et al. (1976) suggested that it is mainly the vertical component of ego
motion (heave motion) that causes motion sickness. They found that with sinusoidal
motions of frequencies between 0.05 to 0.8 Hz and accelerations of more than

1 m s , maxim um sensitivity to m otion sickness happened at around 0.2 Hz, the
incidence of sea sickness increasing further at higher accelerations. Their
mathematical model of motion sickness has long been the most generally accepted
one within the research community concerned with motion sickness (see also
O Hanlon and M cCauley 1974).
However, more recently other mathematical models have been proposed. The
best known one was proposed by Gri n (1990), which deals not exclusively with
sinusoidal movem ents, but also with other types of motion (for a com parison of
these two models and some theoretical extensions concerning adaptation, see Colwell
The main premise of all these models is that small purely vertical accelerations

(below 1 m s ) have not m uch potential to generate motion sickness. In a series of
experiments with the SM S of the TNO Human Factors Research Institute (for


A. H. W ertheim

technical details see Bles and Vunderink 1994, Bos et al. 1995), this was found to be
true (W ertheim et al. 1995a) . However, the severity and incidence of motion sickness
(measured with a rating scale, see De Graaf et al. 1992, W ertheim et al. 1992)
increased dramatically when such small vertical movem ents were accompanied by
low frequency pitch and roll motions with am plitudes ranging between 12 and 14 8 .
In addition, separate or combined pitch and roll movem ents in the absence of any
vertical m otion appeared to be sickness provoking as well, albeit to a lesser extent.
A problem with such SM S experiments is that we do not know exactly how the
subjects move their heads when inside a sim ulator. This prevents a proper
description of how the vestibular and visual systems are stim ulated. This is
illustrated in another study (W ertheim et al. 1995b ), where subjects were not
required to sit on chairs, but had to carry out a physical task that included bending
the body (they had to move and pile crates). In this study, motion sickness incidence
was very high (reaching 50% ). This supports the view that intra-vestibular
mismatches such as coriolis eects may play an important role in the generation
of motion sickness (Reason and Brand 1975, Eyeson-Annan et al. 1996).
M otion sickness may also develop as a result of horizontal linear movem ents
(Golding and K erguelen 1992, Horii et al. 1993). In this respect it should be noted
that motion sickness is also a problem encountered in space. M any astronauts suer
from it, especially during the rst few days of a ight. Space sickness presum ably
stems from the fact that the linear acceleration sensors (the otoliths) are deprived of
their normal constant gravitational acceleration stimulus. Conversely, but in line
with this reasoning, Bles et al. (1989) observed that when subjects in a human
centrifuge are submitted to a constant hypergravity acceleration (2 or 3 g) for some
tim e (one to several hours), they m ay feel quite motion sick upon their return to
normal gravity, especially when making head movem ents. Astronauts participating
in these studies recogn ized the sym ptoms as similar to those of space sickness (Bles et
al. 1989, Ockels et al. 1990).
2.1.2. Simulator sickness: As mentioned above, a special case of m otion sickness
is sim ulator sickness. A well-known review has been published by Kennedy et al.
(1988), who de ned simulator sickness as motion sickness associated with sim ulated
movem ents that in real life are not motion sickness provoking.
Simulator sickness can be quite strong in xed-base simulators, especially if the
visual display is very large (e.g. a dome). In term s of sensory mismatch theory this
happens because the movem ents of the visual environment create strong sensations
of ego m otion that are not matched with concurrent vestibular sensations of selfmotion. Actually, this is one of the reasons why much eort has been invested to
create moving-base simulators, which, it was hoped, should provide those vestibular
sensations and thus reduce motion sickness.
H owever, in many cases these hopes appear to have been over-optimistic, as
several factors in moving-base simulators still cause sensory mismatches. For
exam ple, on the at surface of visual displays there is no real depth. It m ust be
simulated. Not only by proper perspectives that change during simulated ego m otion
but, more importantly, by concurrent relative motion between the objects in the
surroundings (motion parallax ). If motion parallax is not properly program med, it
may create impressions of self-motion that do not properly t vestibular cues from
the motion base. Another exam ple, sometimes encountered in moving-base
simulators, is that the program med point of view of the observer inside the

W orking in a moving environm ent


simulator relative to the visual display is incorrect. During horizontal rotatory

movem ents of the simulator (e.g. when `taking a turn with a tank sim ulator), the
environment then moves in a way dierent from that which the visual system of the
observer expects on the basis of vestibular inform atio n provided by the moving base.
A similar problem occurs when the environm ent is program med to rotate with its
centre of rotation located wrongly (e.g. at a location dierent from the real point of
rotation of the rotating sim ulator cabin).
A severe coupling problem , often causing nausea with subjects wearing headmounted virtual environment displays, is that the visual image must move across the
display surface in temporal synchrony with the m ovem ents of the head. To attain
this, head movements must be recorded and on the basis of these records the
movem ents of the presented im age must be calculated. This takes time, especially
with very large and detailed visual displays. If the delay becomes longer than
something like 20 ms, the resulting visual-vestibular mism atch m ay become
extremely nauseating. In a recent experiment the gain and phase relations of visual
and vestibular information were manipulated, using an arti cial environment set-up,
mounted on a linear acceleration sled (M esland et al. 1996, 1998). The results showed
that phase dierences are much more nausea provoking than gain dierences, and
that in contradistinction to visual phase lags relative to the vestibular stimulus, small
visual phase leads are not nausea provoking.
However, even with properly program med very fast computer systems that create
close to perfect visual surrounds, visual-vestibular mismatches cannot alw ays be
prevented because of the very nature of moving-base simulators. The point is that
with most simulators strong or long duration linear accelerations cannot be
generated; they must be simulated. For instance a forward or backward linear
acceleration is usually simulated by tilting the simulator cabin backward or forward,
to create a situation where subjects feel being `pressed backward or forward into
their chairs. Although this may, on somatosensory and cognitive levels, create the
suggestion of a linear forward or backward acceleration (especially with the proper
visual display) such tiltings actually consist of rotations. Rotations are sensed by the
semi-circular canals, which normally do not react during linear accelerations. The
consequent erroneous tilting sensations are usually suppressed by the linearly
moving visual surround. W hen the motion frequencies are rather low (as with
commercial aircraft simulators) the sensory con ict is generally too weak to cause
problems, i.e. there is little risk of motion sickness. However, with higher frequency
movem ents (as with ghter aircraft or road vehicle simulators) this suppression
implies a really strong visual-vestibular mism atch, increasing the risk of motion
sickness. A sim ilar reasoning applies with large or long duration horizontal rotations
(e.g. road curves), which are usually simulated by tilting the simulator cabin
sidewards, activating the `wrong semi-circular canals. A related problem happens
when the simulator is repositioned back to its horizontal zero position to be ready
for the next manoeuvre. This repositioning should be done very slow ly, such that the
concurrent vestibular stim ulation (usually called the `wash-out ) remains below
threshold. Otherwise a strong visual-vestibular mismatch happens, as wash-out
movem ents are of course not present in the visual display. The problem with such
slow below threshold wash-outs is that they are possible only when the simulated
vehicle movem ents have very low frequency characteristics (e.g. with large civil
aircraft). W ith fast military aircraft simulators or road vehicle simulators, movements usually have much higher frequency characteristics, making slow wash-out


A. H. W ertheim

motions impossible. Thus in such simulators it is extrem ely di cult, if not

impossible, to avoid visual-vestibular mismatches caused by above threshold
wash-out movem ents of the moving base.
2.2. Balanc e problems
M otion of the platform on which one works aects postural control and this may
interfere with normal human performance and locomotion. Only quite recently has
research begun on balance problem s on ships or in ship motion simulators (Graham
1990, Baitis et al. 1994, 1996). Usually this kind of work is concerned with people
standing in upright position while instances of (near) loss of balance (usually referred
to as M otion Induced Interruptions, or M II s) are recorded.
Such M II records have been used in combination with biomechanical m odels of
the hum an body to build models that predict the frequency of M II s for a standing
person during particular ship movements or sea states (Graham 1990, Lewis and
Gri n 1995, Baitis et al. 1994, 1996). M odels like these can be used to generate
criteria as to when it is safe or dangerous to perform particular tasks on ships (e.g. on
the platform of an aircraft carrier).
Only in a few cases did these studies involve subjects who do not stand, but who
walk (W ertheim et al. 1993, 1994). Since M II models do not apply as yet to walking
humans, these data may be of relevance to an extension of current M II theorizing.
Actually, for M II research, motion is not always necessary. M II s can also be
measured on a stationary tilted oor, and there is at least one report related to
emergency procedures aboard listing ships in which such eects have been
described (W ertheim 1993).
2.3 Physical fatigue
It is a well-known fact am ong people who work on ships that they are more easily
fatigued when doing physical work at sea, than when the sam e work is done ashore.
In the scienti c literature relatively little attention has been given to this
phenom enon, known as M otion Induced Fatigue (M IF, Colwell 1989 ). Only a few
attempts have been m ade to em pirically investigate M IF. This is probably because
many members of the human factors com munity are trained as psychologists, which
might incline them to study mental rather than physical fatigue. H owever, physical
fatigue is likely to aect mental performance as well. Hence the study of M IF and its
developm ent should be quite useful.
A strong incentive to the study of M IF has come from a recent multinational
research program me, sponsored by the ABCD (Am erican, British, Canadian and
Dutch) W orking Group on Human Performance at Sea (Baitis et al. 19 95). In this
program me the method used to measure physical fatigue in quantitative term s was
taken from the eld of exercise physiology. Here physical fatigue is generally
deduced from m easuring oxygen consumption of the human body during physical
work. These are quite complex experiments in which inhaled and expired breathing
air must be analysed, together with a num ber of other physiological measures. The
am ount of oxygen consumed is then expressed as a percentage of the maxim um
capacity for oxygen consumption. That maxim um m ust be measured in a separate
test (a so-called maxim um perform ance or graded exercise test), which must be
carried out a few days prior to the actual experiment. This percentage usually
referred to as `relative physical work load appears to be mathematically related to
the maxim um time a subject can carry out the work (Bink 1962, Louhevaara et al.

W orking in a moving environm ent


1986, Rodgers 1997). This maxim um time can thus be used as a fatigue index: it
indicates how long one can still carry on with a particular task from the time the
measurement took place.
An early attempt to measure oxygen consumption in a moving environment
was carried out with subjects inside a ship motion sim ulation facility (Crossland
and Lloyd 1993, Crossland 1994, Baitis et al. 1994, 1996). In this study the subjects
were standing. They were not actively involved in physical work, and no maxim al
tests were carried out. The results showed a slight increase in oxygen consumption
during the sim ulated ship m ovem ents. However, the increase was m uch less than
expected, given the fact that the subjects appeared to be very tired at the end of the
experiment (Crossland 1994). In three other studies (Heus et al. 1994, W ertheim et
al. 1993, 1994, 1995b , Heus et al. 1998), oxygen consumption was measured inside
the SM S during walking on a treadmill or across the oor of the sim ulator cabin,
or during a crate stacking task. In these cases, prior to the experiments, the
subjects did perform maxim um performance tests outside the simulator. Nevertheless, the results were similar to the earlier nding: when the fatigue index was
calculated on the basis of the am ount of oxygen consum ption expressed as a
percentage of maxim um capacity for oxygen consumption again only a relatively
small increase in fatigue was found. The increase was still too small to explain that,
when they exited the SM S after experimentation, the subjects gave the impression
of being severely fatigued.
On the basis of these ndings the hyp othesis was put forward that inside a moving
environment oxygen consum ption during the work itself m ay indeed be only slightly
increased, but the body s maxim um capacity for oxyge n consumption might be
reduced. In two additional experiments (W ertheim et al. 1996a, b) this hypothesis was
tested. This time subjects performed two maxim al tests. One test was carried out inside
a stationary SM S, the other inside a moving SM S. The two maxim al tests were
separated by 1 week from each other and both were carried out 1 week or more before
the actual experiment. During the actual experiment, the task inside the moving SM S
consisted of riding a bicycle ergometer for 4 h at a xed energy level (30% of
maxim um capacity as determined prior to the experiment outside the SM S). The
results did indeed support the hyp othesis: maxim um capacity was signi cantly
reduced in the moving SM S. W hen the oxyge n consumption data from the cycling task
was expressed as a percentage of this lower m axim um capacity level, fatigue appeared
to be increased by approxim ately 100% . In other words, inside the moving SM S the
maxim um time during which subjects would be able to keep cycling until exhaustion
was halved as compared to when the task was carried out in a stationary SM S. The
nding that maxim um oxygen consumption during a maxim al test is reduced in a
moving environment was recently replicated in another SM S experiment (W ertheim et
al. 1997). Thus it is reasonable to assume that working inside a moving environment
may indeed be at least twice as fatiguing as working in stable surroundings.
3. Speci c eects of a moving environm ent on performance
So far, the possibility has been discussed that a moving environment aects
performance through general intervening m otivational, biomechanical or energetic
varia bles. Such eects apply basically to any kind of task performance. An
alternative approach is to investigate speci c eects of environm ental motion on task
performance, i.e. to ask whether environmental motion by itself speci cally aects
particular human skills.


A. H. W ertheim

3.1. Com plex tasks

W hen asked to investigate performance in a m oving environm ent, the kind of tasks
human factors researchers are most likely to be confronted with are real tasks, such
as those carried out in aircraft, or on bridges and in technical (comm and) centres of
ships. However, such tasks are usually quite complex in terms of the psychological
skills required, and this is probably why research on eects of movem ent on such
com plex tasks is hard to nd. Recently two such experiments have been carried out
in an SM S (Helsdingen 1996, W ertheim and Kistem aker 1997). The complex task
was a sim pli ed version of a real naval task, requiring decisions on the basis of
interpretations of radar images from which particular information had to be
memorized. Information had to be sam pled by clicking a mouse button (after
positioning a cursor at particular locations) to reveal hidden codes. Thus the task
consisted of an interplay of cognitive skills (decisions had to be made on the basis of
memorized inform atio n), perceptual skills (small codes had to be read next to target
locations), and ne motor co-ordination skills (precise manual positioning of the
cursor and clicking on small target sym bols). In such complex tasks, performance
cannot be analysed in terms of the classic one-dimensional param eters that are
normally used as performance indices, such as reaction times or number of correctly
detected signals. Instead a general system-analytical param eter was used, re ecting
the am ount of information transferred from the task to the hum an operator. Both
studies showed that with a moving SM S, a small but signi cant reduction of
information transfer happened.
Of course the problem with such an eect is that it cannot be explained as
motion-induced interference with any one particular human skill. Hence one should
be careful to m ake generalizations to other tasks with dierent skill structures. The
solution of this problem is not easy: for most complex tasks the skill structure is
basically unknown, which is a comm on problem in ergonomics.
The obvious way out is to use only relatively sim ple tasks in which the human
skills required are obvious. W hen taking this approach , one line of reasoning is to
distinguish three classes of tasks on the basis of their underlying skill com ponents:

cognitive tasks (e.g. attention, m emory, pattern recogn ition);

motor tasks (e.g. manual tracking, fast button press reactions); and
perceptual tasks (e.g visual or auditory detection).

In the literature there appear indeed to be several attem pts to study possible
eects of environm ental motion on task performance, attempts that take the above
distinction as their point of departure. They will be reviewed here.
3.2. Cognitive tasks
Bles and W ientjes (1988) studied the eect of a moving environment (tilting room)
on a cognitive memory com parison task. They observed no eects. In another study
(Bles et al. 1988), which consisted of a 1-day sea trial aboard a ship, the sam e
memory com parison task was used. Again no eects of ship motions were observed
(there was a small decrease in perform ance, but this was m ore likely to be the result
of sea sickness, as it disappeared when the severity of sea sickness decreased during
the day). M ore recently, Bles et al. (1991) again studied task performance aboard a
ship at sea, this time during a 2-day sea trial. Again they used the sam e memory
com parison task. Although initially it appeared that reaction times increased slightly

W orking in a moving environm ent


with an increase in the magnitude of the ship s movem ents, this increase too
disappeared later on during the sea trial. Thus, from these studies it seems that there
is little or no eect of environm ental motion on cognitive performance.
Crossland and Lloyd (1993) mention similar ndings (see also Crossland 1994,
Baitis et al. 1994, 1996). They report on an experim ent in which eects of
environmental m otion were studied on a variety of cognitive paper-and-pencil tasks,
carried out inside a moving SM S. It appeared that these cognitive tasks did not suer
from the simulated ship movem ents.
Other indications that cognitive abilities are not aected by ship movem ents
stem from three separate studies performed in an SM S. In the rst experiment
(W ertheim et al. 1995a) subjects were asked to perform a digit addition task. In a
second study (W ertheim et al. 1995b ) subjects carried out a variety of cognitive and
visuo-motor tasks. M ore recently an experiment was perform ed (W ertheim and
Kistemaker 1997) in which subjects had to carry out either a visual or a cognitive
task (together with the complex task mentioned in 3.1). In none of these
experiments was any eect of (simulated) ship movem ents observed with any of the
cognitive tasks.
M ost tasks mentioned in the literature are rather short. They are carried out
typically for periods lasting from a few minutes to half an hour at m ost. However,
there is at least one report of a study about eects of sim ulated ship m otions on
performance on a long duration (several hours) radar monitoring task (M alone
1981). That task included a strong attentional, and thus cognitive, component.
However, in that study too, no motion-induced performance decrement was
Given these reports, the conclusion appears warranted that cognitive skills are
not directly aected by ship movem ents. However, it is possible that some indirect
eects of environmental motion could happen in cases where cognitive tasks require
much eort, e.g. because they make a very high demand on short-term, or working,
memory (Gaillard and W ientjes 1994, see also H ockey 1997). Perform ance on such
tasks may be aected only slightly or not at all, but when indices of mental eort are
used concurrently, such as a reduction of sinus-arithmy, they m ight show that in a
moving environment more mental eort is required than in a stationary one. If so,
this could explain the small eects of environm ental motion on the complex
cognitive task mentioned above (Helsdingen 1996, W ertheim and Kistemaker 1997).
That task seems to rely more on very short-term memory processes than the
standard mem ory comparison tests, as used in the above-m entioned sea trials or
tilting room study. The author is currently investiga ting this issue in a new series of
experiments in an SM S.
3.3. M otor tasks
In what was probably one of the rst studies carried out in a ship motion
simulator, M cLeod et al. (1980) asked subjects to carry out various motor coordination tasks involving arm , hand and nger movements. Perform ance was
degraded to the extent that the required motor activity was of a ne control rather
than of a ballistic nature. The suggestion that ship movem ents may interfere with
ne motor performance was also supported by the above-mentioned SM S study
with paper-and-pencil tests, reported by Crossland and Lloyd (1993; see also
Crossland 1994, Baitis et al. 1994), where some motion-induced performancedegrading eects appeared to occur with tests requiring ne motor control. M ore


A. H. W ertheim

recently, performance on a visuo-motor task in the above-mentioned study by

W ertheim et al. (1995b ), where a computerized tracking task was used, also
showed a decrement due to motion of the SM S.
On the other hand Bles and W ientjes (1988) had, in the above-mentioned 1-day
sea trial, also included a m an ual visuo-motor tracking task, and they reported no
eect. However, the ship motions encountered during this sea trial were rather mild.
During their later 2-day sea trial Bles et al. (1991) again included a tracking task, but
only compared performance during separate periods of particular ship movem ents.
They found no dierences in perform ance, but comparisons with baseline
perform ance data (obtained in calm waters) could not be made, as these were lost.
Hence, their conclusion was only that if there is a degrading eect of ship movem ents
on manual tracking, it did not vary with the kinds of ship movem ents encountered
during their sea trial.
In conclusion, it seems that ship movem ents interfere to some extent with ne
motor control, but not necessarily alw ays. It is reasonable to assume that when these
interfering eects happen, they are caused by biomechanical factors.
3.4. Perceptual tasks
As mentioned before, M alone (1981) with his long duration radar m onitoring task
(which of course was not only cognitive, but also to a large extent perceptual)
observed no m otion-induced performance decrement. This suggests that perception
is not aected by ship movem ents.
H owever, although perception itself may not be aected, biomechanical eects
may indirectly impair perceptual performance. For exam ple, in an experiment with
subjects seated in a rotating chair, W ientjes and Bles (1989) studied the eects of
passive body rotation on perform ance in a visual search task (the visual display was
attached to the chair and thus remained head-stationary). A decrease in performance
was observed. H ow ever, the authors attribute the eect to the fact that su ciently
strong rotatory accelerations of the head, as used in their exp eriment, induce
re exive nystagm oid eye movem ents, which blur the image on the stimulus display.
The sam e explanation may hold for an eect observed by W ertheim and
Kistemaker (1997), who used a visual performance task in their SM S study, in which
subjects had to identify a particular target letter presented within brie y visible
arrays of letters on a computer monitor. W ith large letters there was no eect, but
with small letters a signi cant performance decrement was observed.
Thus, eects of (simulated) ship movem ents on perceptual skills may stem from
visual blur, caused by the re exive nystagm oid eye movem ents that are known to
accompany vestibular stimulation. However, such re exive eye movem ents need not
be the only biomechanical factor aecting visual perception. Small high-frequency
vibrations, such as are known to happen in aircraft (most notably helicopters), may
cause the eyes to slightly vibrate in their sockets. If so, dashboard displays or control
panels vibrate across the eyes. This blurs the visual image. Such vibrations may also
occur aboard particular ships or in ship motion simulators (for som e reviews on such
vibratory eects see Guedry 1974, M oseley 1986, M oseley and Gri n 1986, Von
Gierke et al. 1991, G ri n and Hayw ard 1994).
In conclusion then, from the literature on speci c eects of a moving
environm ent on task performance it appears that such speci c eects m ay be
expected only to the extent that biomechanical factors are of relevance. Hence tasks
requiring good (oculo) motor control may suer, but one should not expect speci c

W orking in a moving environm ent


eects on performance characterized by other components (Rolnick and Gordon

Seen in the light of this conclusion one may assum e that the motion-induced
performance degradation in the above-mentioned complex task used by Helsdingen
(1996) and W ertheim and Kistemaker (1997), could apart from a possible loading
of short-term memory also have stemmed from biomechanical interference with
visual perception (reading the sm all codes in the target locations) and with ne
motor control (cursor steering).
4. Conclusions
Research on motion-induced performance decrements shows that such decrements
can be expected to occur when motion creates gen eral artefacts such as reduced
motivation (due to motion sickness), balance problems, or increased fatigue (due to
increased energy requirements). On the other hand, speci c eects of moving
environments on cognitive performance have as yet to be demonstrated. If one looks
at concurrent indices of mental load, however, one might have a chance to nd some
evidence for motion-induced interference. In contrast with this apparent absence of
motion eects on cognitive performance, it is quite likely that, through
biomechanical factors, motion interferes with ne motor control or with the
perception of small visual detail. These factors should be taken into account when
considering human performance in m oving environments, such as ships, aircraft or
moving-base simulators.

B AITIS, A. E., C OW ELL , J. L., C ROSSLA N D , P., D AN IEL , A., H OLCOM BE , F. D., K R IK KE , M.,
P ATTISO N , J. H., P IGEAU , R. A., R ICH , K. J. and W ER TH EIM , A. H. 1995, Generating and
using human performance simulation data to guide designers and operators of navy
ships: two large multinational program s, Proceedings of RINA International Conference
on Seakeeping and Weather, February /March, London (London: Royal Institute of
Naval Architects), 1 8.
B AITIS, A. E., H O LCOM BE , F. D., C OLW ELL , S. L., C RO SSLAND , P., C OLWELL , J. and P ATTISO N , J. H.
1994, 1991 1992 Motion Induced Interruptions (MII) and Motion Induced Fatigue
(MIF) experiments at the Naval Biodynamics Laboratory. Technical report
CRDKNSW C-HD-1423-01, Carderock Division, Naval Surface Warfare Center,
Bethesda, MD, USA.
B AITIS, A. E., H O LCOM BE , F. D., C OLW ELL , S. L., C R OSSLAN D , P., C O LW ELL , J., P ATTISO N , J. and
S TRONG , R. 1996, 1991 1992 Motion Induced Interruptions (MII) and Motion Induced
Fatigue (MIF) experiments at the Naval Biodynamics Laboratory. Technical report for
the ABCD Working Group on Human Factors at Sea, Carderock Division, Naval
Surface Warfare Center, Bethesda, MD, USA.
B IN K , B. 1962, The physical working capacity in relation to working time and age, Ergonomics,
5, 25 28,
B LES , W. and D E G R AAF , B. 1993, Postural consequences of long duration centrifugation,
Journal of Vestibular Research, 3, 88 95.
B LES , W. and V UN DERIN K , R. A. F. 1994, The TNO ship motion simulator: a tool to investigate
the eects of ship motions on human performance. TNO-TM report 1994 A-40, TNO
Human Factors Research Institute, Soesterberg, The Netherlands.
B LES , W. and W IEN TJES , C. J. E. 1988, Welbevinden, taakprestatie en hyperventilatie in de
kantelkamer: invloed van visueel referentiekader en kunstm atige horizon (Well-being,
task performance and hyperventilation in the tilting room : in uence of a visual frame of
reference and arti cial horizon). TNO-TM report IZF 1988-30, TNO Human Factors
Research Institute, Soesterberg, The Netherlands.


A. H. W ertheim

B LES , W., B OER , L. C., K EUN IN G , J. A., V ERM EY , P. and W IENTJES , C. J. E. 1988, Zeeziekteonderzoek: dosis-eect metingen aan boord van Hr. Ms. Makkum (Dose-eect relations
in sea sickness research aboard the Hr. Ms. Makkum). TNO-TM report IZF 1988-5,
TNO Human Factors Research Institute, Soesterberg, The Netherlands.
B LES , W., B OS, J. E., F UR RER , R., D E G RAAF , B., H OSM AN , R. J. A. W., K ORTSCH OT , H. W.,
W. J., S M IT , J., W ERTHEIM , A. H. and W IEN TJES , C. J. E. 1989, Space Adaptation
Syndrome induced by a long duration + 3Gx centrifuge run. Report IZF-1989-25, TNO
Human Factors Research Institute, Soesterberg, The Netherlands.
B LES , W., D E G RA AF , B., K EU NING , J. A., O OMS , J., D E V R IES, J. and W IENTJES , C. J. E. 1991,
Experim ents on motion sickness aboard the M.V. Zeefakkel. Report IZF-1991-A-34,
TNO Human Factors Research Institute, Soesterberg, The Netherlands.
B LES , W., D E J ON G , H. A. A. and O OSTERV ELD , W. J. 1984, Prediction of sea sickness
susceptibility. AGARD Conference Proceedings No 372: Motion Sickness: Mechanisms,
Prediction, Prevention and treatment, North Atlantic Treaty Organization, Neuilly Sur
Seine, France.
B OS , J. E., D E V RIES , S. C. and V UN DERINK , R. A. F. 1995, Motion control of the TNO ship
motion simulator. TNO-TM report 1995 A-32, TNO Human Factors Research
Institute, Soesterberg, The Netherlands.
C OLWELL , J. L. 1989, Human factors in the naval environment: a review of motion sickness and
biodynamic problems. DREA Technical Memorandum 89 / 220, Canadian National
Defence Research Establishment Atlantic, Dartmouth.
C OLWELL , J. L. 1994, Motion sickness habituation in the naval environm ent. DREA Technical
Memorandum 94 / 211, Canadian National Defence Research Establishment Atlantic,
C ROSSLA ND , P. 1994, Experim ents to quantify the eects of ship motions on crew task
performance Phase II, assessment of cognitive performance. Technical report DRA /
AW /AWH /TR94001, Defence Research Agency, Farnborough, UK.
C ROSSLA ND , P. and L LO YD , A. R. J. M. 1993, Experim ents to quantify the eects of ship
motions on crew task performance Phase I, motion induced interruptions and motion
induced fatigue. Technical report DRA / AWMH /TR / 93025, Defence Research Agency,
Farnborough, UK.
D E G R AAF , B., B LES , W., O OM S, J. and D OU WE , E. 1992, Exploratory study with a sea sickness
questionnaire on Hr. Ms. Frigate Trom p. Report IZF-1992-A-1, TNO Human Factors
Research Institute, Soesterberg, The Netherlands.
D OBIE , T. G. and M AY , J. G. 1994, Cognitive-behavioral management of motion sickness,
Aviation, Space and Environmental Medicine, 65 (10), section II, C1 C20.
E YESON -A NN AN , M., P ETER KEN , C., B ROW N , B. and A TCHISO N , D. 1996, Visual and vestibular
com ponents of motion sickness, Aviation, Space and Environmental Medicine, 67, 955
962. A. W. K. and Wienties, C. J. E. 1994, Mental work load and stress as two types of
energy mobilization, Work and Stress, 8, 141 152.
G AILLAR D , A. W. K. and W IEN TJES C. J. E. 1994, Mental work load and stress as two types of
energy mobilization, Work and Stress, 8, 141 152.
G OLDING , J. F. and K ER GU ELEN , M. 1992, A comparison of the nauseogenic potential of low
frequency vertical versus horizontal linear oscillation, Aviation, Space and Environmental Medicine, 63, 491 497.
G RAH AM , R. 1990, Motion-induced interruptions as ship operability criteria, Naval Engineers
Journal, March, 65 72.
G RIFFIN , M. J. 1990, Handbook of Human Vibration (London: Academ ic Press).
G RIFFIN , M. J. and H AYW ARD , R. A. 1994, Eects of horizontal whole-body vibration on
reading, Applied Ergonomics, 25, 165 169.
G U ED RY , F. E. 1974, Psychophysics of vestibular sensation, in H. H. Kornhuber (ed.),
Handbook of Sensory Physiology, vol. 6 /2 (New York: Springer), 3 154.
H ELSDIN GEN , A. S. 1996, Zeebew egingen en taakprestatie (Sea movem ents and task
performance). Technical report, Department of Psychology, Free University of
Amsterdam, The Netherlands.
H ETTIN GER , L. J., B ERBAUM , K. S., K ENN ED Y , R. S., D U NLAP , W. P. and N OLA N , M. D. 1990,
Vection and simulator sickness, Military Psychology, 2, 171 181.

W orking in a moving environm ent


H EU S, R., W ER TH EIM , A. H. and H AVEN ITH , G. 1998, Human energy expenditure when walking
on a moving platform, European Journal of Applied Physiology, in press.
H EU S, R., W ERTHEIM , A. H. and V R IJK OTTE , T. G. M. 1994, Energy expenditure, physiological
workload and postural control during walking on a moving platform, in J. Frim, M.B.
Ducharm e and P. Tikusis (eds), Proceedings of the Sixth International Conference on
Environmental Ergonomics, Montobello, Canada, Septem ber (North York, Ontario,
Canada: Scienti c Information Center, Defence and and Civil Institute of Environmental Medicine), 274 275.
H O CKEY , G. R. J. 1997, Compensatory control in the regulation of human performance under
stress and high work load: a cognitive-energetical framework, Biological Psychology, 45,
73 93.
H O RII , A., T AKEDA , N., M ORITA , M., K U BO , T. and M ATSU NA GA , T. 1993, Acta Otolaryngologica,
Suppl. 501, 31 33
H O WAR D , I. P. 1986, The vestibular system, in K. R. Bo, L. Kaufman. and J. P. Thom as (eds),
Handbook of Perception and Human Performance, Vol. I: Sensory Processes and
Perception (New York: John Wiley).
K EN NEDY , R. S., B ERBA UM , K. S, A LLG OOD , G. O., L A NE , N. E., L ILIEN TH AL , M. G. and
B ALTZLEY , D. R. 1988, Etiological signi cance of equipm ent features and pilot history in
simulator sickness, in AGARD Conference Proceedings No. 433, Motion Cues in Flight
Simulation and Simulator Induced Sickness, vol. 1, (Neuilly Sur Seine, France: North
Atlantic Treaty Organization), 1 22.
K EN NEDY , R. S., G R AYBIEL , A., M C D ON OUG H , R. C. and B ECKW ITH , F. D. 1968, Symptomology
under storm conditions in the North Atlantic in control subjects and in persons with
bilateral labyrinthine defects, Acta Otolaryngologica, 66, 533 540.
L EW IS , C. H. and G RIFFIN , M. J. 1995, Modelling the eects of deck motion on postural
stability. ISVR Contract report 95 /12, May, Institute of Sound and Vibration Research,
University of Southam pton, UK.
L OUH EVAA RA , V., S M OLEA ND ER , J., K ORHO NEN , O. and T U OM I, T. 1986, Maximal working times
with a self containing breathing apparatus, Ergonomics, 29, 77 85.
M ALO NE , W. L. 1981, Eects of simulated surface eect ship motion on crew habitability
phase II, Vol. 1: Sum mary report and comm ents. PM S-304 Technical report no. 1070,
Naval Sea Systems Command, Bethesda, MD, USA.
M C C AULEY , M. E., R O YAL , J. W., W YLIE , C. D., O H AN LON , J. F. and M ACKIE , R. R. 1976,
Motion sickness incidence: exploratory studies of habituation, pitch and roll, and the
re nement of a mathematical model. Technical report 1733-2, Human Factors Research
Inc., Santa Barbara Research Park, Goleta, California, USA.
M C L EOD , P., P O ULTON , C., D U R OSS , H. and L EWIS , W. 1980, The in uence of ship motion on
manual control skills, Ergonomics, 23, 623 634.
M ESLAND , B. S., B LES , W. and W ERKH OVEN , P. J. 1996, Eects of dissociating optic inputs from
vestibular inputs on self-motion perception during virtual-environm ent immersion,
Journal of Vestibular Research, 6, 4S, S13.
M ESLAND , B. S., B LES , W., W ER KHO VEN , P. J., and W ERTHEIM , A. H. 1998, How exible is the
self-motion perceptual system?: Introducing phase and amplitude dierences between
visual and proprioceptive passive linear horizontal self-motion stimuli, in B. S. Mesland,
About self-motion perception. Ph.D . thesis, University of Utrecht, The Netherlands,
99 142.
M OSELEY , M. J. 1986, Eects of vibration on visual performance and display legibility. Ph.D .
thesis, University of Southampton, UK.
M OSELEY , M. J. and G RIFFIN , M. J. 1986, Eects of display vibration and whole body vibration
on visual performance, Ergonomics, 29, 977 983.
O CK ELS , W. J. R., F U RRER , R. and M ESSERSC H MID , E. 1990, Space sickness on earth,
Experimental Brain Research, 79, 661 663.
O H A NLON , J. F. and M C C AU LEY , M. E. 1974, Motion sickness incidence as a function of the
frequency and acceleration of vertical sinusoidal motion, Aerospace Medicine, 45, 366
O MAN , C. M. 1982, A heuristic mathematical model for the dynam ics of sensory con ict and
motion sickness, Acta Otolaryngologica, Suppl. 392.


A. H. W ertheim

R EASON , J. T. and B R AND , J. J. 1975, Motion Sickness (London: Academ ic Press).

R OD GERS , S. H. 1997, Work physiology fatigue and recovery, in G. Salvendy (ed.), Handbook
of Human Factors and Ergonomics, 2nd edn (New York: John Wiley), 268 297.
R OLNICK , A. and B LES , W. 1989, Performance and well-being under tilting conditions: the
eects of visual reference and arti cial horizon, Aviation, Space and Environmental
Medicine, 60, 779 785.
R OLNICK , A. and G OR DON , C. R. 1991, The eects of motion induced sickness on military
performance, in R. Gal and A.D. Mangelsdor(eds), Handbook of Military Psychology
(New York: John Wiley), 279 293.
S TOFFRE GEN , T. A. and R ICCIO , G. E. 1986, Out of control: an ecological perspective on motion
sickness. Paper presented at the Fall meeting of the International Society for Ecological
Psychology, October, Philadelphia, PA.
V ON G IERK E , H. E., M C C LOSK EY . K. and A LBERY , W. B. 1991, Military performance in
sustained acceleration and vibration environm ents, in R. Gal and A.D. Mangelsdor
(eds), Handbook of Military Psychology (New York: John Wiley), 352 364.
W ERTHEIM , A. H. 1993, Postural stability and locom otion, in L.C.Boer (ed.), Ship listing;
reactions of passengers. Report IZF-1993-C-26, TNO Human Factors Research
Institute, Soesterberg, The Netherlands, 13 17.
W ERTHEIM , A. H. and K ISTEM AKER , J. A. 1997, Task performance during simulated ship
movem ents. TNO-TM report TM -97-A014, TNO Human Factors Research Institute,
Soesterberg, The Netherlands.
W ERTHEIM , A. H., H EUS , R. and K ISTEM AKER , J. A. 1996a, Maximum capacity for human
energy expenditure in a moving environm ent. Report TNO-TM-1996 C-052, TNO
Human Factors Research Institute, Soesterberg, The Netherlands.
W ERTHEIM , A. H., H EUS , R. and K ISTEMA K ER , J. A. 1996b, Motion induced fatigue during
simulated ship movem ents. Report TNO-TM-96-C-073, TNO Human Factors Research
Institute, Soesterberg, The Netherlands.
W ERTHEIM , A. H., H EUS , R., K ISTEMA K ER , J. A. and H AVEN ITH , G. 1997, Human energy
expenditure in a moving environm ent during graded exercise tests; a replication study.
Report TNO-TM-97-C045, TNO Human Factors Research Institute, Soesterberg, The
W ERTHEIM , A. H., H EUS , R. and V RIJKOT TE , T. G. M. 1994, Energy expenditure, physical work
load and postural control during walking on a moving platform. TNO report TM -1994B-8, TNO Human Factors Research Institute, Soesterberg, The Netherlands.
W ERTHEIM , A. H., H EU S, R. and V RIJKO TTE , T. G. M. 1995b, Human energy expenditure, task
performance and sea sickness during simulated ship movem ents. TNO report TM -1995C-29, TNO Human Factors Research Institute, Soesterberg, The Netherlands.
W ERTHEIM , A. H., H EUS , R., V RIJKO TTE , T. G. M. and M AR CU S , J. T. 1993, The eect of
platform motion on hum an energy expenditure during walking: an exploratory
experiment. TNO report IZF-1993-B-10, TNO Human Factors Research Institute,
Soesterberg, The Netherlands.
W ERTHEIM , A. H., O O MS , J., D E R EGT , G. P. and W IENTJES , C. J. E. 1992, Incidence and
severeness of sea sickness: validation of a rating scale. Report IZF-1992-A-41, TNO
Human Factors Research Institute, Soesterberg, The Netherlands.
W ERTHEIM , A. H., W IENTJES , C. J. E., B LES , W. and B OS , J. E. 1995a, Motion sickness studies in
the TNO-TM Ship Motion Simulator (SMS). TNO report TM -1995 A-57, TNO
Human Factors Research Institute, Soesterberg, The Netherlands.
W IEN TJES C. J. E. and B LES W. 1989, Situationeel bewustzijn en vestibulaire stimulatie: de
invloed van draaibewegingen op de taakprestatie (Situational awareness and vestibular
stimulation: The in uence of rotational movement on task performance). TNO-TM
report IZF 1989-14, TNO Human Factors Research Institute, Soesterberg, The
Y AR DLEY , L. 1992, Motion sickness and perception: a reappraisal of the sensory con ict
approach, British Journal of Psychology, 83, 449 471.