Behavioural Brain Research 305 (2016) 164173

Contents lists available at ScienceDirect

Behavioural Brain Research

journal homepage: www.elsevier.com/locate/bbr

Research report

Modality specicity in the cerebro-cerebellar neurocircuitry during

working memory
H.B. Tommy Ng a , K.-L. Cathy Kao a , Y.C. Chan b,c , Efe Chew b,c , K.H. Chuang d ,
S.H. Annabel Chen a,e,
a
Division of Psychology, School of Humanities and Social Sciences, Nanyang Technological University, 637332, Singapore
b
Division of Neurology, University Medicine Cluster, National University Health System, Singapore
c
Department of Medicine, Yong Loo Lin School of Medicine, National University of Singapore, Singapore
d
The Queensland Brain Institute, The University of Queensland, Brisbane, Queensland, Australia
e
Centre for Research and Development in Learning (CRADLE), Nanyang Technological University, 637459, Singapore

h i g h l i g h t s

We examined hemispheric lateralization patterns in working memory.

Left lateralized IFG and IPL were activated during verbal working memory.
Right lateralized cerebellar lobular VI and VIII were also activated in verbal working memory.
A left-cortical right-cerebellar network underlies verbal working memory.
No evidence for a right-cortical left-cerebellar network in visual working memory.

a r t i c l e i n f o a b s t r a c t

Article history: Previous studies have suggested cerebro-cerebellar circuitry in working memory. The present fMRI study
Received 28 December 2015 aims to distinguish differential cerebro-cerebellar activation patterns in verbal and visual working mem-
Received in revised form 22 February 2016 ory, and employs a quantitative analysis to deterimine lateralization of the activation patterns observed.
Accepted 24 February 2016
Consistent with Chen and Desmond (2005a,b) predictions, verbal working memory activated a cerebro-
Available online 27 February 2016
cerebellar circuitry that comprised left-lateralized language-related brain regions including the inferior
frontal and posterior parietal areas, and subcortically, right-lateralized superior (lobule VI) and infe-
Keywords:
rior cerebellar (lobule VIIIA/VIIB) areas. In contrast, a distributed network of bilateral inferior frontal
Cerebellum
Cerebro-cerebellar and inferior temporal areas, and bilateral superior (lobule VI) and inferior (lobule VIIB) cerebellar areas,
Functional magnetic resonance imaging was recruited during visual working memory. Results of the study veried that a distinct cross cerebro-
Verbal working memory cerebellar circuitry underlies verbal working memory. However, a neural circuitry involving specialized
Visual working memory brain areas in bilateral neocortical and bilateral cerebellar hemispheres subserving visual working mem-
ory is observed. Findings are discussed in the light of current models of working memory and data from
related neuroimaging studies.
2016 Elsevier B.V. All rights reserved.

1. Introduction comprehension [7]. Within the theoretical framework, mecha-

Working memory may be conceptualized as a neural system that
temporarily maintains, stores, and manipulates information for
complex cognitive tasks such as reasoning, learning, and language
Corresponding author at: Division of Psychology, School of Humanities and
Social Sciences, HSS-04-19, 14 Nanyang Drive, Nanyang Technological University,
637332, Singapore.
E-mail addresses: annabelchen@ntu.edu.sg, tommymr25@yahoo.co.uk,
ENGZ0001@e.ntu.edu.sg (S.H.A. Chen).
nisms of the phonological loop and visuo-spatial sketchpad are
thought to support verbal and visual working memory, respec-
tively.
Data from neuroimaging studies have consistently demon-
strated the involvement of left inferior frontal and left inferior
parietal regions in verbal working memory, areas of the brain that
have been linked to phonological rehearsal and storage, respec-
tively [20,21,3,36,11,35,116,55]. Although cerebellar activations
were also observed in early neuroimaging studies [96,99,56], their

http://dx.doi.org/10.1016/j.bbr.2016.02.027
0166-4328/ 2016 Elsevier B.V. All rights reserved.
 H.B.T. Ng et al. / Behavioural Brain Research 305 (2016) 164173 165

Fig. 1. Event timing of a single trial of the Sternberg working memory tasks. Stimuli presented during high-load and low-load condition of (a) verbal working memory and
(b) visual working memory tasks. (c) Timing of one cycle of high-load and low-load trials interleaved by a 1.6 s rest interval.

exact function in verbal working memory was not well understood
(see [56]).
In a previous fMRI study, Desmond et al. [28] identied two
cerebellar regions that were differentially activated during verbal
working memory: one located bilaterally in the superior hemi-
sphere (lobule VI/Crus I), and another in the inferior aspect (lobule
VIIB) of the right hemisphere. Based on the canonical cortico-ponto-
cerebellar network [81,59] that forms a closed loop circuitry via the
thalamus [123,2,79]; for review see [102], the authors proposed
a cerebro-cerebellar model of verbal working memory to account
for superior cerebellar contribution to phonological rehearsal and
encoding and inferior cerebellar contribution to phonological stor-
age processes. In a later event- related fMRI study, Chen and
Desmond [17] found concomitant activation of the superior cere-
bellum and left frontal regions during encoding, and concomitant
activation of the right inferior cerebellum and left parietal regions
during maintenance, supporting earlier speculations that a later-
alized left cortical-right cerebellar neurocircuitry supports verbal
working memory.
Although the laterality effect was observed in a number of
subsequent studies [18,60,61,73], a caveat was that hemispheric
lateralization was assessed via visual comparison of the number of
activated voxels and activation intensity (as indicated by the nor-
malized Z value) in homologous brain regions, without the rigour
of statistical testing. To address this, one of the objectives of the
present study was to evaluate the laterality effect of verbal working
memory by means of a laterality index measurement [10,26,29,40],
which statistically evaluates voxel beta-values of homologous
regions of interest. Specically, an unbiased threshold-free com-
putation of voxel t-values within regions of interest was used to
calculate the laterality index. This approach has been shown to
minimize computational biases that are associated with statistical
thresholding [75]; for review see [113].
Another objective of the present study was to verify the later-
ality effects of working memory processes in the visual system. In
the animal literature, an abundance of neurophysiological evidence
converged on the involvement of ventral prefrontal and occipi-
totemporal brain areas in pattern, object, and face recognition and
recall [135,83,82,101,109,106]. These results are in good concor-
166 H.B.T. Ng et al. / Behavioural Brain Research 305 (2016) 164173
dance with studies demonstrating visual recognition decits in
monkeys with ventromedial prefrontal cortex and inferior tem-
poral cortex lesions [95,85,4,54]. Findings from human lesion (for
review see [34,94,93,90,88,84] and neuroimaging [24,50,100]; for
review see [130,104] studies also provide evidence for the involve-
ment of the abovementioned brain regions in visual working
memory. However, despite the wealth of data, it remains unclear
whether or not cerebral hemispheric lateralization persists in pro-
cesses underlying visual working memory.
While there have been reports of left mid- frontal region
preponderance during object working memory tasks [115,117],
there are by far more evidence for supporting right cere-
bral dominance during visuo-spatial working memory tasks
[77,52,23,132,131,125,136]; for review see [87,120]. The incon-
sistent laterality effects observed at the neocortex prevail at the
cerebellar level, with a number of studies reporting left cerebel-
lar dominance during face [51,62], spatial [45], and abstract object
[49,57] working memory tasks, and to a lesser extent, preponder-
ance of right cerebellar activations during visuospatial working
memory tasks [97,126].
Given the inconsistency in previous ndings, the current study
aims to determine if laterality topography exists for working mem-
ory in the cerebrum and cerebellum for verbal and visual stimuli
through a quantitative analysis.
2. Materials and methods
2.1. Participants
Forty two (22 males, 20 females) neurologically healthy young-
adults took part in the study. Their age ranged from 2131 years
with a mean of 22.83 years (S.D. = 1.95 years). All the subjects were
right-handed (Laterality Quotient (LQ) >60) according to Edinburgh
Handedness Inventory (EHI LQ: Mean = 96.83, S.D. = 6.14; [92] . The
study protocol was approved by the Nanyang Technological Univer-
sity Institutional Review Board and the National Healthcare Group
(NHG) Domain Specic Review Board (DSRB)(A). The experimental
procedures were carried out in accordance with the 1964 Decla-
ration of Helsinki. Subjects were paid for their participation after
successful completion of the experiment. Informed consent was
obtained from individual subjects prior to the start of each experi-
ment.
2.2. Task description
We employed a block design for two working memory tasks
(verbal and visual), based on the Sternberg paradigm. The tasks
were presented to the subjects in alternating blocks of high- and
low- load in separate runs. Each cycle consisted of 1 block of high-
load and 1 block of low-load that had 2 trials each. Each block was
interleaved by a 1.6 s interval and block duration was 20 s. There
were 10 cycles in each run adding up to a total of 20 high-load and
20 low-load trials. Each run lasted 400 s. 2 runs of each task were
performed and task order was counterbalanced across subjects.
Stimuli were pseudo-randomly presented on a Fujitsu Lifebook
computer (Fujitsu, Tokyo, Japan) with E-Prime version 2.0 (Psychol-
ogy Software Tools, Inc., Pittsburgh, US) software. The images were
projected onto a screen that was located at the back of the scanner
bore, and subjects viewed them via a mirror mounted on the cage
coil. Before the experiment began, subjects practised the tasks to
familiarize with the experimental procedures.
In the verbal working memory (VERWM) task, a red xation
cross indicating the start of a trial was presented in the middle of the
screen for 1.2 s. Following that, a horizontal array of either six non-
repeated letters (high-load condition) or six identical letters (low-
load condition) was presented above the cross for 3 s. During this
time, subjects rehearsed the letters in sequential order subvocally
and continued to do so during a 3 s maintenance period where no
stimuli were shown on the screen. After the maintenance period,
a probe letter was presented for 2 s and the subjects determined
if it matched one of the letters in the array by pressing a button
under the right index nger for match and a button under the
right middle nger for not-match.
In the visual working memory (VISWM) task, the same xa-
tion cross preceded (1.2 s) the presentation of a horizontal array of
either three non-repeated geometrical patterns (high-load condi-
tion) or three identical geometrical pattern (low-load condition).
During a 4 s encoding phase, subjects viewed and retained the
visual patterns in their mind and continued to do so during a 1.5 s
maintenance period where no stimuli were shown on the screen.
After the maintenance period, a probe pattern was presented below
the cross for 2.5 s and the subjects determined if it matched one of
the patterns in the array by a button press in the same manner as
that in the VERWM task. Subjects were instructed not to use verbal
strategies for the VISWM task. In both the verbal and visual working
memory tasks, duration of a single trial was 9.2 s (see Fig. 1).
2.3. Behavioural data analysis
Paired-samples t-tests evaluated within-condition (high load
versus low load) differences in performance accuracy and reaction
times.
2.4. MRI data acquisition
Whole-brain MRI data were acquired on a 3 T scanner (Siemens
Trio, Erlangen, Germany) equipped with a 32-channel array coil.
To minimize movements, foam paddings were placed between the
subjects head and scanners head rest. The acquisition protocol
consisted of localizer, structural, and functional scans. High res-
olution (1 mm isotropic) T1- weighted images were acquired in
the sagittal plane with magnetization prepared rapid acquisition
gradient echo sequence. T2*-weighted echo planar images were
acquired in the oblique axial plane with the following parameters:
TR = 2500 ms; TE = 29 ms; 48 slices; ip angle = 90 ; FOV = 225 mm;
64 64 matrix giving an in-plane voxel size of 3.5 3.5 mm2 ; slice
thickness = 3.5 mm. The acquisition plane was rotated 25 with
respect to the posterior vertical axis of the brainstem to optimize
signal measurements from the cerebellum and neocortices [18].
MRI data collection was synchronized with the behavioural task via
interfacing with the E-Prime software. Prior to each run, subjects
were provided instructions via the scanners intercom system.
2.5. Functional data analysis
fMRI data were analyzed using the Statistical Parametric Map-
ping (SPM8) software package (Wellcome Department of Cognitive
Neurology, Institute of Neurology, University College London, UK,
www.l.ion.ucl.ac.uk/spm). Structural and EPI images were aligned
to the AC-PC plane. An interleaved bottom-up slice timing correc-
tion was applied to the EPI images to correct for temporal lags in
image acquisition. Head motions were corrected by aligning EPI
images of run 1 and run 2 to the rst EPI image of run 1. This rou-
tine generated a mean EPI image, which was used as a source image
for registering the realigned functional images to the individual
subjects anatomical T1 image.
The experimental effects were estimated with the general linear
model within SPM8. Two [high-load > low-load (VERWM); high-
load > low-load (VISWM)] contrast images were calculated for each
subject, yielding 42 images per task. It was assumed that these con-
trast images reected parametric recruitment of neural resources
 H.B.T. Ng et al. / Behavioural Brain Research 305 (2016) 164173
Fig. 2. For display purposes, activation maps superimposed on the canonical T1-
weighted rendered brain were thresholded at T = 4.7 [P < 0.05 (FWE corrected) at
cluster-level]. (a) During verbal working memory, left hemispheric dominance
was observed in frontal, parietal, occipital, and occipitotemporal regions. (b) Dur-
ing visual working memory, bilateral activations of the same brain regions were
observed. The colour bar refers to activation intensity expressed in T values. Lighter
colour represents increased activation intensity.
during the two load conditions. The images were then normalized
separately to t the ICBM152 whole-brain template with default
parameters in SPM8 and the Spatially Unbiased Infra-tentorial
Template (SUIT). This was done to overcome the limitations of
normalizing cerebellar activations to MNI space. Accordingly, the
MNI template, which is derived from 152 T1-weighted scans, lacks
information of important anatomical landmarks (e.g. primary and
intrabiventer ssures) crucial for aligning structures in the cere-
bellum. It has been previously suggested that such a normalization
procedure would result in substantial stretching of the cerebellum
in the z-direction, pushing activations in superior aspect of the
cerebellum to the visual cortices and obscuring those in the inferior
cerebellum [30,31].
In the SUIT normalization, individual T1- weighted images were
cropped to isolate the cerebellum from the other brain structures.
These structural images were then warped to t the SUIT space and
the resulting deformation parameters were applied to the func-
tional images for resampling at 3.5 3.5 3.5 mm3 resolution. All
contrast images were smoothed with a 6 mm Full Width Half Max-
imum (FWHM) Gaussian kernel.
After normalization, random effects analyses were conducted
on task-specic contrasts with one-sample t-tests. Cluster-level
activations that survived family-wise error (FWE) rate correction
167
with signicance level at p < 0.05 were reported. Except for cere-
bellar activations that were superimposed onto the SUIT template,
all other brain activations were superimposed onto the MNI tem-
plate. Locations of cerebral and cerebellar activation maxima were
determined using the Automated Anatomical Labeling [AAL; [129]]
software and probabilistic cerebellar atlas [31], respectively.
Lateralization of brain activations was evaluated with AveLI [75].
In this method, voxel beta-values are used to compute laterality
index (LI) of two homologous regions of interest (ROIs). Specically,
subordinate LIs (sub-LIs) are computed from task-specic positive
beta-values using the following equation:
Lt Rt
sub LI =
Lt + Rt
where Lt and Rt are the summations of voxel beta-values at and
above an arbitrary threshold in the left and right ROIs, respectively.
AveLI is subsequently calculated by averaging the sub-LIs in the
ROIs as follows:

(sub LI)
AveLI =
VN
where VN is the total number of positive t-values voxels in the ROIs.
In the current study, we used AAL template [129] in Wake For-
est University PickAtlas [71] and the SUIT cerebellum template [30]
to create bilateral ROIs for AveLI analyses in the cerebral cortex
and cerebellum, respectively. Based on previous literature, ROIs for
the VERWM task were dened in the inferior frontal gyrus (see
[17], inferior parietal lobule (see [105], cerebellar lobule VI/CrusI,
and cerebellar lobule VIIB/VIII (see [18]. For the VISWM task, ROIs
were dened in the inferior frontal gyrus (see [23], inferior tem-
poral gyrus (see [103], cerebellar lobule VI/CrusI (see [126], and
cerebellar lobule VIIB/VIII (see [49].
3. Results
3.1. Behavioural data
In the VERWM task, performance accuracy (percent) was signif-
icantly (t (40) = 5.55, P < 0.001) higher in the low-load (M = 98.90,
SD = 1.94) compared to high-load (M = 95.24, SD = 4.18) condition.
Reaction times (milliseconds) were signicantly (t (40) = 16.51,
P < 0.001) greater during high-load (M = 939.64, SD = 166.38) com-
pared to low-load (M = 653.55, SD = 114.27) condition. Similarly in
the VISWM task, subjects registered signicantly (t (40) = 12.05,
P < 0.001) higher percentage of correct responses in the low-load
(M = 97.99, SD = 2.64) compared to high-load (M = 87.44, SD = 5.85)
condition. Reaction times (milliseconds) were signicantly (t
(40) = 14.03, P < 0.001) greater during high-load (M = 864.07,
SD = 165.07) compared to low-load (M = 699.56, SD = 142.56) con-
dition.
Across tasks, subjects performed better in the verbal task com-
pared to the visual task (t (40) = 7.02, P < 0.001) although they took
a longer time to make their responses (t (40) = 4.10, P < 0.001).
3.2. Imaging data
3.2.1. Activations in MNI space
Fig. 2, Table 1, and Table 2 show regional activation proles and
loci of activation maxima during the working memory tasks.
In the VERWM task, left hemispheric dominance was observed
in inferior and middle frontal regions, posterior parietal regions,
and occipitotemporal regions. Apart from the brain regions
depicted in Fig. 2, other activated brain areas not visible from the
rendered brain included the putamen, caudate, pallidum, insular,
cingulum, and hippocampus.
168 H.B.T. Ng et al. / Behavioural Brain Research 305 (2016) 164173

Table 1 Table 4
Brain regions activated during the working memory tasks, high-load > low-load. Means and standard deviations (SDs) of the laterality indices.

Brain structures MNI coordinates (mm) ROIs VERWM VISWM

Nvox SPM{Z} X Y Z Mean SD Laterality Mean SD Laterality

Verbal working memory IFG 0.32 0.17 Left 0.02 0.23 Bilateral
Precentral L 2688 7.57 51 0 46 IPL 0.59 0.20 Left
Precentral L 6.56 49 2 28 ITG 0.04 0.26 Bilateral
Precentral L 6.55 41 6 30 CL VI 0.37 0.40 Right 0.06 0.36 Bilateral
Calcarine L 8351 7.07 11 91 0 CL Crus I 0.12 0.46 Bilateral 0.08 0.39 Bilateral
Occipital Inf R 6.91 21 92 4 CL VIIB 0.58 0.45 Right 0.05 0.51 Bilateral
Putamen L 2234 7.20 19 4 14 CL VIIIA 0.31 0.41 Right 0.12 0.45 Bilateral
Caudate R 6.73 17 0 18
Abbreviations: IFG: inferior frontal gyrus, IPL: inferior parietal lobule, ITG: infe-
Pallidum R 6.15 11 0 0
rior temporal gyrus, CL: cerebellar lobule, ROI: region of interest, VERWM: verbal
Parietal Sup L 966 6.98 23 64 46
working memory, VISWM: visual working memory.
Insula L 427 6.95 33 24 2
Supp Motor Area L 1181 6.88 3 12 50
Supp Motor Area L 6.78 7 16 44
Cingulum Mid R 6.75 9 20 38 In the VISWM task, brain activations were obtained bilaterally in
Insular R 342 6.13 31 22 4 the inferior and middle frontal regions, posterior parietal regions,
Hippocampus R 166 5.51 25 28 6 occipital regions, and occipitotemporal regions. Activated brain
Hippocampus L 202 5.38 27 36 4
areas not visible from the rendered brain included the cingulum,
Angular R 214 5.27 33 58 46
insular, hippocampus, and thalamus.
Activation maxima are reported using a single subject MNI template brain [129],
P < 0.05 (FWE corrected) at cluster-level. Abbreviations: Inf: inferior; L: Left; Mid:
middle; Nvox : number of voxels in a contiguous cluster; Oper: opercularis; R: right; 3.2.2. Activations in SUIT space
Sup: superior; Supp: supplementary. Fig. 3 and Table 3 show cerebellar activation proles and loci of
Table 2
activation maxima during the working memory tasks, respectively.
xxx. Although the VERWM task appeared to recruit bilateral superior
cerebellar regions, cluster-level thresholding produced activation
Brain structures MNI coordinates (mm)
maxima only in the right hemisphere. Within this region, two
Nvox SPM{Z} X Y Z activation maxima were obtained; locus of the stronger maxima
Visual working memory lay anterior, inferior, and lateral to the weaker one. In the same
Fusiform L 17637 7.61 33 52 22 hemisphere, a separate activation maxima was obtained in lob-
Occipital Mid L 7.45 25 88 10 ule VIIB/VIII of the inferior posterior cerebellum. Of note, there
Fusiform R 7.38 37 46 22
was no signicant cerebellar lobule VIIB/VIII activation in the left
Supp Motor Area L 1482 7.48 1 16 46
Cingulum Mid R 6.75 9 24 32 hemisphere.
Insula R 464 6.87 37 22 6 In the VISWM task, robust activations were obtained bilater-
Precentral L 1699 6.32 45 6 32 ally in the superior posterior region of cerebellar lobule VI, which
Precentral L 6.21 35 4 62 became a contiguous cluster through the adjoining vermal activa-
Precentral L 5.59 31 6 54
Insular L 171 5.59 33 18 4
tion in the anterior cerebellum. Smaller clusters of voxels were also
Frontal Inf Oper R 781 5.59 53 16 38 obtained in the inferior posterior aspects of both cerebellar cortices.
Frontal Inf Oper R 5.37 49 10 28 The intensity and size of the activation maxima in left cerebellar
Frontal Inf Tri R 5.33 53 38 24 lobule VIIB was greater compared to that in the right hemisphere.
Hippocampus L 43 5.45 27 36 2
Frontal Mid R 417 5.39 37 2 64
Precentral R 5.36 33 2 50 3.2.3. AveLI
Parietal Sup R 135 5.34 11 78 50 Table 4 shows the laterality effect of brain activations during the
Thalamus R 32 5.28 23 30 2
working memory tasks. According to Seghier [113] and Matsuo et al.
Activation maxima are reported using a single subject MNI template brain [129], [75], AveLI values greater than 0.2 and lesser than 0.2 represent
P < 0.05 (FWE corrected) at cluster-level. Abbreviations: Inf: inferior; L: Left; Mid:
left and right lateralization of brain activations, respectively.
middle; Nvox : number of voxels in a contiguous cluster; Oper: opercularis; R: right;
Sup: superior; Supp: supplementary; Tri: triangularis.
A robust laterality effect was observed in the VERWM task. Acti-
vations in the IFG (0.32 0.17; Mean SD) and IPL (0.59 0.20)
Table 3 were lateralized to the left cortical hemisphere whereas those in
Cerebellar brain regions activated during the working memory tasks, high-
lobule VI (0.37 0.40), lobule VIIB (0.58 0.45) and lobule VIIIA
load > low-load.
(0.31 0.41) were lateralized to the right cerebellar hemisphere.
Brain structures MNI Coordinates (mm) Activations in cerebellar lobule Crus I was bilateral.
Lobule Nvox SPM{Z} X Y Z In contrast, in the VISWM task, AveLI values approximating
zero (0.040.12) indicated that cortical (inferior frontal gyrus,
Verbal working memory
Superior posterior cerebellum VI 961 R Inf 21 63 26 inferior temporal gyrus) and cerebellar (lobule VI/CrusI and lob-
Superior posterior cerebellum VI R 6.87 11 73 23 ule VIIB/VIIIA) brain regions were quasi- equitably activated across
Inferior posterior cerebellum VIIB/VIII R 6.44 25 70 51 hemispheres.
Visual working memory
Superior posterior cerebellum VI 930 L 7.13 35 59 23
4. Discussion
Superior posterior cerebellum VI R 7.12 28 70 19
Anterior cerebellum/Vermis IX 6.75 0 56 33
Inferior posterior cerebellum VIIB 22 L 5.70 25 73 51 The present study conrms earlier ndings supporting a lateral-
Inferior posterior cerebellum VIIB 9 R 4.65 25 73 51 ized cerebro-cerebellar (left neocortical-right cerebellar) network
Activation maxima are reported using the probabilistic cerebellar atlas [31], P < 0.05 of verbal working memory. In contrast, quantitative ndings
(FWE corrected) at cluster-level. Abbreviations: L: left; Nvox : number of voxels in a showed bilateral activations in the cerebral and cerebellar regions
contiguous cluster; R: right. during visual working memory. These results did not provide evi-
 H.B.T. Ng et al. / Behavioural Brain Research 305 (2016) 164173 169

Fig. 3. Coronal section views of cerebellar activations during the working memory tasks. For display purposes, activation maps were thresholded at T = 4.7 [P < 0.05 (FWE
corrected) at cluster-level]. (a) During verbal working memory, bilateral superior cerebellar activations were obtained from slices y = 78 to y = 57. In the same y slices,
inferior cerebellar activations were obtained only in the right hemisphere. (b) During visual working memory, bilateral superior cerebellar activations were obtained from
slices y = 80 to y = 57. Bilateral inferior cerebellar activations were obtained from slices y = 78 to y = 61. The colour bar refers to activation intensity expressed in T values.
Lighter colour represents increased activation intensity.

dence to suggest a distinct right neocortical-left cerebellar network
for visual working memory.
The activations observed in the region representative of the
Brocas area and posterior aspect of the left parietal cortex during
verbal working memory are in concordance with previous nd-
ings [17,18,96,21,105]. Specically, the left inferior parietal lobule
(BA 40) has been linked to processes related to phonological stor-
age. Although cluster maxima was obtained in the proximity of
the Brocas area (see Fig. 2, Table 1), the contrast images in the
present study did not reveal any cluster maximum in the left infe-
rior parietal lobule. However, based on laterality effect analysis
that showed strong dominance of the left inferior parietal lobule
(see Table 4), it is likely that this brain region might be part of
a larger contiguous node that was localized to the superior pari-
etal lobule (see Table 1). A number of previous investigations have
observed superior parietal activations during verbal working mem-
ory [99,128,25,89,74]. While its exact function is unclear, the brain
region is thought to be involved in the storage of verbal informa-
tion [3], memory retrieval [15], and processes related to attentional
demands [118]. In the current Sternberg paradigm, it is likely that
the subjects iteratively attended to the alphabets on the screen
to subvocally rehearse the visual information for memory main-
tenance and storage. In general, our data are in agreement with
current models of verbal working memory that hypothesize the
contribution of the Brocas area to articulatory rehearsal and the
posterior parietal region to phonological storage [28,17], processes
of which sustain the phonological loop assumed to be fundamental
to the perception and production of language [6,86,8].
Hellige [53] and Banich [9] proposed that hemispheric spe-
cialization of cognitive functions is innate in the human brain. A
number of fMRI studies have reported dominance of the left cere-
bral hemisphere during verbal working memory tasks [16,58,44]
and right cerebral hemisphere during nonverbal working mem-
ory tasks [124,72]. In these studies, hemispheric specialization was
determined by either contrasting the activation maps of working
memory versus control tasks [16,124], spatial working memory
versus object working memory tasks [72], or verbal working mem-
ory versus nonverbal working memory tasks [72]. The results of
these between-task contrasts informed us about the neural sub-
strates that were uniquely activated or brain regions that were
activated more in extent and intensity in one task relative to the
other. In contrast, a within-task (high-load minus low-load con-
dition) parametric contrast reects neural underpinnings of the
task that was performed. Previously, Thomason et al. [125] had
employed laterality index measurement to determine hemispheric
specialization of phonological loop processes and found left cere-
bral lateralization in prefrontal and parietal regions. In the present
study, we compared high- and low-load verbal working memory
within-subjects and found not only left lateralization of the afore-
mentioned neocortical brain regions but also right lateralization
of cerebellar substrates in lobule VI and lobule VIIB/VIII. To our
knowledge, this is the rst quantitative demonstration of a cerebro-
cerebellar network in verbal working memory, conrming ndings
of earlier studies that implicated right-sided lateralization of cere-
bellar involvement in phonological loop routines [17,18,60,61,73].
Studies of primates have shown that neural connections orig-
inating from neocortical frontal regions terminate predominantly
along medial aspects of the pontine nucleus [134,111,112], which
in turn project to superior portions of the cerebellar cortex
including lobule VI and Crus I [47,80,81,32,59]. A distinct cir-
cuitry originating from the parietal and temporal regions has
also been identied to synapse along lateral portions of the
pontine nucleus [134,67,133,76,14], which in turn project to
inferior aspects of the paramedian lobule [(hemispheric lobule
VIIB); [12,13,43]]. Taking together these ndings and our cur-
rent data, they form a strong neuroanatomical and functional
basis for implicating a crossed coupling between left inferior
frontal region and right superior cerebellar region in articulatory
rehearsal and a separate crossed coupling between left inferior
parietal region and right inferior cerebellar region in phonological
store [17,18,27,96,108,107,3,28,118]. Future research employing
effective connectivity analysis should provide insights into the inte-
gration between functionally specialized brain areas.
Of note, our ndings are not consistent with that of two recent
neuroimaging studies that reported the absence of cerebellar hemi-
spheric lateralization in verbal working memory [49,126]. In the
earlier study that employed an n-back paradigm [49], subtrac-
tion analyses of 2-back minus 0-back yielded bilateral cerebellar
170 H.B.T. Ng et al. / Behavioural Brain Research 305 (2016) 164173
activation in lobule VI and lobule VIIIA/VIIB. It was argued that
the lack of cerebellar lateralization could be due to the nature of
the n-back task that emphasized manipulation of information and
executive control in excess of stimulus maintenance, thus super-
seding processes of the phonological loop that might have been
induced. Although reasonable, the assumption did not prevail when
a Sternberg paradigm emphasizing item encoding and mainte-
nance was used in a subsequent fMRI study [126]. We suggest that
the lack of cerebellar hemispheric lateralization might be due to
inequitable articulatory rehearsal between the load conditions. In
that study, high-load condition comprised a matrix of six letters
and low-load condition comprised a matrix of one letter and ve
# symbols. In the latter condition, it is not reported whether the
subjects rehearsed subvocally only the letter or every item in the
matrix including the # symbols, or if they did rehearse at all dur-
ing the maintenance phase of the task. In the present study, the
number of items presented was identical across conditions and
all items were alphabets (see Fig. 1) and subjects were asked to
continue subvocal rehearsal even during the maintenance phase.
This, we reasoned has the effect of equating motoric processes
underlying articulatory rehearsal for encoding and maintenance,
without which subtraction analyses might elicit additional cere-
bellar resources for motoric processes in the high-load condition.
Alternatively, the additional cerebellar activations might reect
greater motoric demands for rehearsing different alphabets under
the high-load condition or response-time variability between the
two conditions [114].
While brain regions associated with motoric and cognitive
functions may be topographically located in the same cerebellar
lobules (i.e. VI and VIIb/VIII), meta-analytic data of Stoodley and
Schmahmann [120] suggests a dissociation of functional locality.
Accordingly, lobular VI and VIIb/VIII activations associated with
motoric processes are located anterior (y coordinate between 60
to 70) to those that are associated with working memory (y
coordinate between 70 to 80). Our data concur well with their
ndings in that two distinct regions of cerebellar lobule VI were
activated during the verbal working memory task, i.e. one in the
anterior aspect and another in the posterior aspect of the lobule
(Table 3). Recently, Peterburs et al. [98] were able to attribute a
motoric cerebellar network to oculomotor demands and a separate
load effect cerebellar network to memory demands during stim-
ulus encoding. In the case of lobule VIIb/VIII, the activation was
localized to the posterior aspect of the brain structure.
Apart from the phonological loop, working memory models
have also envisaged a separate system (visuo-spatial sketchpad)
that enables the human brain to hold and manipulate non-verbal
information for deciphering semantic and geographical properties
of objects in the environment [1,22]. As proposed by Baddeley
and colleagues [7,5,68], the visuo-spatial sketchpad comprises two
mechanisms: an inner scribe for image rehearsal and a visual cache
for storage, processes of which operate in tandem to support visual
working memory. In a review of data from lesion and neuroimag-
ing studies, Baddeley [8] proposed candidate brain structures that
might be involved in visuo-spatial working memory: for inner
scribe, inferior frontal gyrus (BA 47), middle frontal gyrus, and a
region overlapping the superior parietal lobule (BA 7) and ante-
rior extrastriate occipital cortex (BA 19); for visual cache, inferior
aspect of the occipitotemporal region and the cuneus (BA 17). The
current activations obtained in the aforementioned brain areas (see
Table 2) support Baddeleys conjecture and concur with data from
animal [4,54], human lesion [94,93,90] and neuroimaging [50,100]
studies of visual working memory.
Although bilateral parietal activations were observed during
visual working memory (Fig. 2b), only a maximum in the right supe-
rior parietal lobule was revealed under the current thresholding
(Table 2). Interestingly, this brain area lies near the occipitopari-
etal region postulated by Baddeley [8] to be involved in inner
scribe processes. While mechanisms of these processes are not
well understood, a parsimonious prediction would be that visual
imagery plays a part in keeping the stimuli in working memory.
A number of neuroimaging studies have demonstrated precuneal
activations, a component brain region of the superior parietal lob-
ule, during visual mental imagery [63,37,42,41]. Lundstrom et al.
[69] further showed that these brain activities are related to accu-
rate item recognition and recall, coherent with the high accuracy
scores in the current study.
Apart from these brain regions, additional activations were
also obtained in the fusiform gyrus of the inferior temporal cor-
tex (see Table 2), as reported in a number of other investigations
[38,83,82,100]. While its function is not as clear-cut as those of
the phonological loop, there is evidence to show that the inferior
temporal cortex is centrally involved in visual working memory,
as lesion or cooling of this brain region signicantly impaired per-
formance of delayed matched to sample working memory tasks
[54,39,90].
Unlike the distinct laterality effect that was observed in the
verbal working memory task, we failed to obtain evidence for hemi-
spheric lateralization in visual working memory, as the contrast
images revealed activations in bilateral frontal and inferior tempo-
ral regions (see Fig. 2), and subcortically in bilateral superior and
inferior cerebellar regions (see Fig. 3). The zero-centering laterality
indices conrmed that visual working memory engaged specialized
brain areas in both neocortical and cerebellar hemispheres (see
Table 4). Previous neuroimaging studies have demonstrated the
involvement of bilateral frontal and inferior temporal brain regions
in working memory of novel faces [52,24], and more recently
with event-related fMRI, Ranganath et al. [103] showed that active
maintenance of visual information is supported by activation of
object representations in inferior temporal cortices, and that mem-
ory retrieval is effected by top-down signals originating from the
frontal cortices. Therefore, lesions in and along the pathway con-
necting these two brain regions would severely impair visual recall
[33,46,48,127].
Consistent with our nding, two recent fMRI studies reported,
among other cerebellar areas, bilateral activation of lobule VI and
lobule VIIB during working memory of abstract object [49,126].
Of note, in the Thurling et al. [126] study, stronger lobule VI
and lobule VIIB activations were obtained in the right cerebel-
lar hemisphere. Through post- experiment feedback, the authors
posited that the use of verbal strategies might have activated the
phonological loop, thereby reducing demand of the visuo-spatial
sketchpad. This may offer a plausible explanation for the ndings
in our study. In a recent event-related fMRI study, Sobzak-Edmans
et al. [119] reported that stored visual representations could pos-
sibly involve maintenance mechanisms that employ verbal coding.
Alternatively, the quasi-equitable cerebellar activations obtained
in lobule VI and VIIB might not be the result of phonological loop
recruitment but rather a reection that a distributed network is
needed to support processes of the visuo-spatial sketchpad. Cere-
bellar activations obtained in the midline vermal regions could be
associated with online oculomotor control [121,78,64], as subjects
in the current study might have made saccadic eye movements
from one geometrical pattern to another during encoding.
Data from neurophysiological and behavioural experiments in
monkeys have shown that object recognition involves a multistage
process leading from localized feature analysis in the primary visual
cortex, through a sequence of association areas, to more global
object recognition in the inferior temporal cortex [70]. This might
explain why an extensive recruitment of brain resources is needed
to support processes of not only object recognition but also mem-
ory retrieval. Alternatively, we could suppose that if subjects in the
present study had chosen to remember the letter array in the verbal
 H.B.T. Ng et al. / Behavioural Brain Research 305 (2016) 164173
working memory task as a visual pattern, the same way visual stim-
uli were maintained in visual working memory, it is possible that
an equally extensive bilateral network might have been activated.
However, this process is cost intensive and the brain knowing of a
more efcient system would convert the graphemes to phonemes
before entering them into the phonological loop for encoding and
maintenance, thus recruiting the left cerebro-right cerebellar net-
work. We speculate that the lack of an alternative cost-effective
system in visual working memory might have resulted in the exten-
sive recruitment of bilateral neocortical and cerebellar structures.
The present study is not without limitations. First, similar to the
trial setup of Thurling et al.s [126] study, timing and item param-
eters were differentiated across tasks to match for performance
accuracy. Results from our pilot study showed that by increas-
ing encoding and reducing maintenance timing, and reducing the
number of items in the visual task, accuracy rate was compara-
ble between the two tasks (t (9) = 1.53, P = .16). Of note, subjects in
our pilot study did not receive practice prior to the behavioural
experiment. In contrast, subjects practised the tasks until they
attained eighty percent accuracy prior to fMRI scanning. We posit
that additional exposure to the stimuli might have aided, to a
greater extent, performance of the verbal task compared to the
visual task, biasing the interpretation that the observed laterality
effect was a consequence of task difculty. While that could be a
plausible explanation, it is noteworthy to highlight that behavioural
confounds of disparate task performance have been reported in
previous studies [91,45,19,126], implicating the difculty of match-
ing performance accuracy across different modality of working
memory.
Second, while the current analyses are sufcient for answer-
ing questions about hemispheric lateralization in working memory,
they lack the temporal resolution for evaluating specic processes
of working memory. Future research employing fast event-related
fMRI will help us better understand underlying cerebro-cerebellar
circuitry of encoding and maintenance in human working memory.
In summary, the results of the present study provide quan-
titative evidence for an underlying cerebro-cerebellar circuitry
in verbal working memory. As postulated in earlier studies
[18,60,61,73], the phonological loop recruited left-lateralized
language-related brain areas in the inferior frontal and posterior
parietal regions, and right-lateralized cerebellar areas in lobular VI
and VIIB/VIII regions. In contrast, a lack of hemispheric dominance
in visual working memory suggests that a distributed network of
brain areas in the cerebral and cerebellar hemispheres is needed to
sustain this modality of working memory.
Conict of interest
All the authors in the manuscript declare no conict of interest.
Funding
This work was supported by a Tier 2 Academic Research Fund
(MOE2011-T2-1-031) that was approved by the Ministry of Educa-
tion, Singapore.
Ethical approval
All procedures performed in studies involving human partici-
pants were in accordance with the ethical standards of Nanyang
Technological University Institutional Review Board and National
Healthcare Group (NHG) Domain Specic Review Board (DSRB)(A),
as well as, with the 1964 Helsinki declaration and its later amend-
ments or comparable ethical standards.
171
Consent
Informed consent was obtained from all individual participants
included in the study.
References
[1] J. Andrade, Working Memory in Context, Psychology Press, Hove, Sussex,
2001.
[2] C. Asanuma, W.T. Thach, E.G. Jones, Distribution of cerebellar terminations
and their relation to other afferent terminations in the ventral lateral
thalamic region of the monkey, Brain Res. 286 (1983) 237265.
[3] E. Awh, J. Jonides, E.E. Smith, E.H. Schumacher, R.A. Koeppe, S. Katz,
Dissociation of storage and rehearsal in verbal working memory: evidence
from positron emission tomography, Psychol. Sci. 7 (1996) 2531.
[4] J. Bachevalier, M. Mishkin, Visual recognition impairment follows
ventromedial but not dorsolateral prefrontal lesions in monkeys, Behav.
Brain Res. 20 (1986) 249261.
[5] A.D. Baddeley, Working Memory, Oxford University Press, Oxford, UK, 1986.
[6] A.D. Baddeley, S.E. Gathercole, C. Papagno, The phonological loop as a
language learning device, Psychol. Rev. 105 (1998) 158183.
[7] A.D. Baddeley, G.J. Hitch, Working memory, in: G.A. Bower (Ed.), The
Psychology of Learning and Motivation: Advances in Research and Theory,
Academic, New York, 1974, pp. 4789.
[8] A.D. Baddeley, Working memory: looking back and looking forward, Nat.
Rev. Neurosci. 4 (2003) 829839.
[9] M.T. Banich, Neuropsychology: The Neural Bases of Mental Function,
Houghton-Mifin., Boston, 1997.
[10] J.R. Binder, S.M. Rao, T.A. Hammeke, J.A. Frost, P.A. Bandettini, A.
Jesmanowicz, J.S. Hyde, Lateralized human brain language systems
demonstrated by task subtraction functional magnetic resonance imaging,
Arch. Neurol. 52 (1995) 593601.
[11] T.S. Braver, J.D. Cohen, L.E. Nystrom, J. Jonides, E.E. Smith, D.C. Noll, A
parametric study of prefrontal cortex involvement in human working
memory, Neuroimage 5 (1997) 4962.
[12] P. Brodal, The ponto-cerebellar projection in the rhesus monkey: an
experimental study with retrograde axonal transport of horseradish
peroxide, Neuroscience 4 (1979) 193208.
[13] P. Brodal, The cerebropontocerebellar pathway: salient features of its
organization, Exp. Brain Res. 6 (1982) 108133.
[14] P. Brodal, J.G. Bjaalie, Salient anatomic features of the
cortico-ponto-cerebellar pathway, Prog. Brain Res. 114 (1997) 227249.
[15] J.H. Callicot, V.S. Mattay, A. Bertolino, K. Finn, R. Coppola, J.A. Frank, T.E.
Goldberg, D.R. Weinberger, Physiological characteristics of capacity
constraints in working memory as revealed by functional MRI, Cereb. Cortex
9 (1999) 2026.
[16] B.J. Casey, J.D. Cohen, P. Jezzard, R. Turner, D.C. Noll, R.J. Trainor, J. Giedd, D.
Kaysen, L. Hertz-Pannier, J.L. Rapaport, Activation of prefrontal cortex in
children during nonspatial working memory task with functional MRI,
Neuroimage 2 (1995) 221229.
[17] S.H.A. Chen, J.E. Desmond, Temporal dynamics of cerebro-cerebellar
network recruitment during a cognitive task, Neuropsychologia 43 (2005)
12271237.
[18] S.H.A. Chen, J.E. Desmond, Cerebrocerebellar networks during articulatory
rehearsal and verbal working memory tasks, Neuroimage 24 (2005)
332338.
[19] Y.N. Chen, S. Mitra, Distinctions between spatial and verbal working
memory: a study using event-related potential, Chang Gung Med. J. 32
(2009) 380389.
[20] J.D. Cohen, S.D. Forman, T.S. Braver, B.J. Casey, D. Servan-Schreiber, D.C. Noll,
Activation of the prefrontal cortex in a nonspatial working memory task
with function MRI, Hum. Brain Mapp. 1 (1994) 293304.
[21] J.D. Cohen, W.M. Perlstein, T.S. Braver, L.E. Nystrom, D.C. Noll, J. Jonides, E.E.
Smith, Temporal dynamics of brain activation during working memory task,
Nature 386 (1997) 604608.
[22] A.R.A. Conway, C. Jarrold, M.J. Kane, A. Miyake, J.N. Towse, Variation in
Working Memory, Oxford University Press, New York, 2007.
[23] S.M. Courtney, L.G. Ungerleider, K. Keil, J.V. Haxby, Object and spatial visual
working memory activate separate neural systems in human cortex, Cereb.
Cortex 6 (1996) 3949.
[24] S.M. Courtney, L.G. Ungerleider, K. Keil, J.V. Haxby, Transient and sustained
activity in a distributed neural system for human working memory, Nature
386 (1997) 608611.
[25] S. Crottaz- Herbette, R.T. Anagnoson, V. Menon, Modality effects in verbal
working memory: differential prefrontal and parietal responses to auditory
and visual stimuli, Neuroimage 21 (2004) 340351.
[26] C.A. Cuenod, S.Y. Bookheimer, L. Hertz-Pannier, T.A. Zefro, W.H. Theodore,
D. Le Bihan, Functional MRI during word generation, using conventional
equipment: a potential tool for language localization in the clinical
environment, Neurology 45 (1995) 18211827.
[27] J.E. Desmond, J.A. Fiez, Neuroimaging studies of the cerebellum: language,
learning and memory, Trends Cogn. Sci. 2 (1998) 355362.
[28] J.E. Desmond, J.D. Gabrieli, A.D. Wagner, B.L. Ginier, G.H. Glover, Lobular
patterns of cerebellar activation in verbal working- memory and
172 H.B.T. Ng et al. / Behavioural Brain Research 305 (2016) 164173
nger-tapping tasks as revealed by functional MRI, J. Neurosci. 17 (1997)
96759685.
[29] J.E. Desmond, J.M. Sum, A.D. Wagner, J.B. Demb, P.K. Shear, G.H. Glover, J.D.E.
Gabrieli, M.J. Morrell, Functional MRI measurement of language
lateralization in Wada- tested patients, Brain 118 (1995) 14111419.
[30] J. Diedrichsen, A spatially unbiased atlas template of the human cerebellum,
Neuroimage 33 (2006) 127138.
[31] J. Diedrichsen, J.H. Balsters, J. Flavell, E. Cussans, N. Ramnani, A probabilistic
atlas of the human cerebellum, Neuroimage 46 (1) (2009) 3946.
[32] R.P. Dum, P.L. Strick, An unfolded map of the cerebellar dentate nucleus and
its projection to the cerebral cortex, J. Neurophysiol. 89 (2003) 634639.
[33] M.J. Eacott, D. Gaffon, Inferotemporal- frontal disconnection: the uncinate
fascicle and visual associative learning in monkeys, Eur. J. Neurosci. 4 (1992)
13201332.
[34] M.J. Farah, Visual Agnosia: Disorders of Object Recognition and What They
Tell Us About Normal Vision, MIT Press, Cambridge, MA, 1990.
[35] J.A. Fiez, M.E. Raichle, Linguistic processing, in: J. Schmahmann (Ed.), The
Cerebellum and Cognition International Review of Neurobiology, vol. 41,
Academic Press, San Diego, CA, 1997, pp. 233254.
[36] J.A. Fiez, E.A. Raife, D.A. Balota, J.P. Schwarz, M.E. Raichle, S.E. Petersen, A
positron emission tomography study of the short-term maintenance of
verbal information, J. Neurosci. 16 (1996) 808822.
[37] P.C. Fletcher, C.D. Frith, S.C. Baker, T. Challice, R.S.J. Frackowiak, R.J. Dolan,
The minds eye-precuneus activation in memory-related imagery,
Neuroimage 2 (1995) 195200.
[38] J.M. Fuster, J.P. Jervey, Neuronal ring in the inferotemporal cortex of the
monkey in a visual memory task, J. Neurosci. 2 (1982) 361375.
[39] D. Gaffan, E.A. Murray, Monkeys (Macaca fascicularis) with rhinal cortex
ablations succeed in object discrimination learning despite 24-hr intertrial
intervals and fail at matching to samples despite double sample
presentations, Behav. Neurosci. 106 (1992) 3038.
[40] W.D. Gaillard, S.Y. Bookheimer, L. Hertz- Pannier, T.A. Blaxton, The
noninvasive identication of language function. Neuroimaging and rapid
transcranial magnetic stimulation, Neurosurg. Clin. N. Am. 8 (1997)
321335.
[41] G. Ganis, W.L. Thompson, S.M. Kosslyn, Brain areas underlying visual mental
imagery and visual perception: an fMRI study, Cognitive Brain Res. 20
(2004) 226241.
[42] O. Ghaem, E. Mellet, F. Crivello, N. Tzourio, B. Mazoyer, A. Berthoz, M. Denis,
Mental navigation along memorized routes activates the hippocampus,
precuneus, and insula, Neuroreport 8 (1997) 739744.
[43] M. Glickstein, N. Gerrits, I. Kralj- Hans, B. Mercier, J. Stein, J. Voogd, Visual
pontocerebellar projections in the macaque, J. Comp. Neurol. 349 (1994)
5172.
[44] A.J. Golby, R.A. Poldrack, J.B. Brewer, D. Spencer, J.E. Desmond, A.P. Aron,
J.D.E. Gabrieli, Material-specic lateralization in the medial temporal lobe
and prefrontal cortex during memory encoding, Brain 124 (2001)
18411854.
[45] O. Gruber, D. Yves von Cramon, The functional neuroanatomy of human
working memory revisited. Evidence from 3-T fMRI studies using classical
domain-specic interference tasks, Neuroimage 19 (2003) 797809.
[46] S.A. Gutnikov, Y.Y. Ma, D. Gaffan, Temporo-frontal disconnection impairs
visualvisual paired association learning but not congural learning in
Macaque monkeys, Eur. J. Neurosci. 9 (1997) 15241529.
[47] D.E. Haines, E. Dietrichs, G.A. Mihailoff, E.F. McDonald, Cerebellar-
hypothalamic axis: basis circuits and clinical observations, Int. Rev.
Neurobiol. 41 (1997) 83107.
[48] I. Hasegawa, T. Fukushima, T. Ihara, Y. Miyashita, Callosal window between
prefrontal cortices: cognitive interaction to retrieve long-term memory,
Science 281 (1998) 814818.
[49] H. Hautzel, F.M. Mottaghy, K. Specht, H.-W. Muller, B.J. Krause, Evidence of a
modality-dependent role of the cerebellum in working memory? An fMRI
study comparing verbal and abstract n-back tasks, Neuroimage 47 (2009)
32082.
[50] J.V. Haxby, L. Petit, L.G. Ungerleider, S.M. Courtney, Distinguishing the
functional roles of multiple regions in distributed neural systems for visual
working memory, Neuroimage 11 (2000) 380391.
[51] J.V. Haxby, L.G. Ungerleider, B. Horwitz, J.M. Maisog, S.I. Rapoport, C.L.
Grady, Face encoding and recognition in the human brain, Proc. Natl. Acad.
Sci. U. S. A. 93 (1996) 922927.
[52] J.V. Haxby, L.G. Ungerleider, B. Horwitz, S.I. Rapoport, C.L. Grady,
Hemispheric differences in neural system for face working memory: a
PET-rCBF study, Hum. Brain Mapp. 3 (1995) 6882.
[53] J.B. Hellige, Hemispheric Asymmetry: Whats Right and Whats Left, Harvard
University Press, Cambridge, MA, 1993.
[54] J.A. Horel, D.E. Pytko-Joiner, M.L. Voytko, K. Salsbury, The performance of
visual tasks while segments of the inferotemporal cortex are suppressed by
cold, Behav. Brain Res. 23 (1987) 2942.
[55] J. Jonides, E.H. Schumacher, E.E. Smith, R.A. Koeppe, E. Awh, P.A.
Reuter-Lorenz, C. Marshuetz, C.R. Willis, The role of the parietal cortex in
verbal working memory, J. Neurosci. 18 (1998) 50265034.
[56] J. Jonides, E.H. Schumacher, E.E. Smith, E.J. Lauber, E. Awh, S. Minoshima,
R.A. Koeppe, Verbal working memory load affects regional brain activation
as measured by PET, J. Cognitive Neurosci. 9 (1997) 462475.
[57] S.S. Kang, S.R. Sponheim, M.V. Chafee, A.W. MacDonald III, Disrupted
functional connectivity for controlled visual processing as a basis for
impaired spatial working memory in schizophrenia, Neuropsychologia 49
(2011) 28362847.
[58] W.M. Kelley, F.M. Miezin, K.B. McDermott, R.L. Buckner, M.E. Raichle, N.J.
Cohen, J.M. Ollinger, E. Akbudak, T.E. Conturo, A.Z. Snyder, S.E. Petersen,
Hemispheric specialization in human dorsal frontal cortex and medial
temporal lobe for verbal and nonverbal memory encoding, Neuron 20
(1998) 927936.
[59] R.M. Kelly, P.L. Strick, Cerebellar loops with motor cortex and prefrontal
cortex of a nonhuman primate, J. Neurosci. 23 (2003) 84328444.
[60] M.P. Kirschen, S.H. Chen, P. Schraedley- Desmond, J.E. Desmond, Load- and
practice-dependent increases in cerebro-cerebellar activation in verbal
working memory, Neuroimage 24 (2005) 462472.
[61] M.P. Kirschen, S.H.A. Chen, J.E. Desmond, Modality specic
cerebro-cerebellar activations in verbal working memory: an fMRI study,
Behav. Neurol. 23 (2010) 5163.
[62] H. Koshino, R.K. Kana, T.A. Keller, V.L. Cherkassky, N.J. Minshew, M.A. Just,
fMRI investigation of working memory for faces in autism: visual coding and
underconnectivity with frontal areas, Cereb. Cortex 18 (2008) 289300.
[63] S.M. Kosslyn, N.M. Alpert, W.L. Thompson, V. Maljkovic, S.B. Weise, C.F.
Chabris, S.E. Hamilton, S.L. Rauch, F.S. Buonanno, Visual mental imagery
activates topographically organized visual cortex: PET investigations, J.
Cognitive Neurosci. 5 (1993) 263287.
[64] R.J. Krauzlis, The control of voluntary eye movements: new perspectives,
Neuroscientist 11 (2005) 124137.
[67] G.R. Leichnetz, D.J. Smith, R.F. Spencer, Cortical projections to the
paramedian tegmental and basilar pons in the monkey, J. Comp. Neurol. 228
(1984) 388408.
[68] R.H. Logie, Visuo-spatial Working Memory, LEA Publishers, Hove, Sussex,
1995.
[69] B.N. Lundstrom, M. Ingvar, K.M. Petersson, The role of precuneus and left
inferior frontal cortex during source memory episodic retrieval, Neuroimage
27 (2005) 824834.
[70] R. Malach, J.B. Reppas, R.R. Benson, K.K. Kwong, H. Jiang, W.A. Kennedy, P.J.
Ledden, T.J. Brady, B.R. Rosen, R.B.H. Tootell, Object-related activity revealed
by functional magnetic resonance imaging in human occipital cortex, Proc.
Natl. Acad. Sci. U. S. A. 92 (1995) 81358139.
[71] J.A. Maldjian, P.J. Laurienti, J.B. Burdette, R.A. Kraft, An automated method
for neuroanatomic and cytoarchitectonic atlas-based interrogation of fMRI
data sets, Neuroimage 19 (2003) 12331239.
[72] D.S. Manoach, M.S. Cain, M.G. Vangel, A. Khurana, D.C. Goff, R. Stickgold, A
failure of sleep-dependent procedural learning in chronic, medicated
schizophrenia, Biol. Psychiatry 56 (2004) 951956.
[73] C.L. Marvel, J.E. Desmond, The contributions of cerebro-cerebellar circuitry
to executive verbal working memory, Cortex 46 (2010) 880895.
[74] C.L. Marvel, J.E. Desmond, From storage to manipulation: how the neural
correlates of verbal working memory reect varying demands on inner
speech, Brain Lang. 120 (2012) 4251.
[75] K. Matsuo, S.H.A. Chen, W.Y.I. Tseng, AveLI: a robust lateralization index in
functional magnetic resonance imaging using unbiased threshold-free
computation, J. Neurosci. Methods 205 (2012) 119129.
[76] J.G. May, R.A. Andersen, Different patterns of corticopontine projections
from separate cortical elds within the inferior parietal lobule and dorsal
prelunate gyrus of the macaque, Exp. Brain Res. 63 (1986) 265278.
[77] G. McCarthy, A.M. Blamire, A. Puce, A.C. Nobre, G. Bloch, F. Hyder, P.
Goldman- Rakic, R.G. Shulman, Functional magnetic resonance imaging of
human prefrontal cortex activation during a spatial working memory task,
Proc. Natl. Acad. Sci. U. S. A. 91 (1994) 86908694.
[78] R.C. Miall, H. Imamizu, S. Miyauchi, Activation of the cerebellum in
co-ordinated eye and Hand tracking movements: an fMRI study, Exp. Brain
Res. 135 (2000) 2233.
[79] F.A. Middleton, P.L. Strick, Cerebellar output channels, Int. Rev. Neurobiol.
(1997) 41614182.
[80] F.A. Middleton, P.L. Strick, Dentate output channels: motor and cognitive
components, Prog. Brain Res. 114 (1997) 553566.
[81] F.A. Middleton, P.L. Strick, Cerebellar projections to the prefrontal cortex of
the primate, J. Neurosci. 21 (2001) 700712.
[82] E.K. Miller, C.A. Erickson, R. Desimone, Neural mechanisms of visual working
memory in prefrontal cortex of the Macaque, J. Neurosci. 16 (1996)
51545167.
[83] E.K. Miller, L. Li, R. Desimone, A neural mechanisms for working and
recognition memory in inferior temporal cortex, Science 254 (1991)
13771379.
[84] B. Milner, Visual recognition and recall after right temporal-lobe excision in
man, Epilepsy Behav. 4 (2003) 799812.
[85] M. Mishkin, F. Manning, Non-spatial memory after selective prefrontal
lesions in monkeys, Brain Res. 143 (1978) 313323.
[86] A. Miyake, P. Shah, Models of Working Memory: Mechanisms of Active
Maintenance and Executive Control, University Press, New York:
Cambridge, 1999.
[87] F.M. Mottaghy, Interfering with working memory in humans, Neuroscience
139 (2006) 8590.
[88] N.G. Muller, L. Machado, R.T. Knight, Contributions of subregions of the
prefrontal cortex to working memory: evidence from brain lesions in
humans, J. Cognitive Neurosci. 14 (2002) 673686.
[89] N.S. Narayanan, V. Prabhakaran, S.A. Bunge, K. Christoff, E.M. Fine, J.D.E.
Gabrieli, The role of the prefrontal cortex in the maintenance of verbal
 H.B.T. Ng et al. / Behavioural Brain Research 305 (2016) 164173
working memory: an event-related fMRI analysis, Neuropsychology 19
(2005) 223232.
[90] F. Newcombe, G. Ratcliff, H. Damasio, Dissociable visual and spatial
impairments following right posterior cerebral lesions: clinical,
neuropsychological and anatomical evidence, Neuropsychologia 25 (1987)
149161.
[91] L.E. Nystrom, T.S. Braver, F.W. Sabb, M.R. Delgado, D.C. Noll, J.D. Cohen,
Working memory for letter, shapes, and locations: fMRI evidence against
stimulus-based regional organization in human prefrontal cortex,
Neuroimage 11 (1999) 424446.
[92] R.C. Oldeld, The assessment and analysis of handedness: the Edinburgh
inventory, Neuropsychologia 9 (1971) 97113.
[93] A.M. Owen, R.G. Morris, B.J. Sahakian, C.E. Polkey, T.W. Robbins, Double
dissociations of memory and executive functions in working memory tasks
following frontal lobe excisions, temporal lobe excisions or
amygdalo-hippocampectomy in man, Brain 119 (1996) 15971615.
[94] A.M. Owen, B.J. Sahakian, J. Semple, C.E. Polkey, T.W. Robbins, Visuo-spatial
short-term recognition memory and learning after temporal lobe excisions,
frontal lobe excisions or amygdalo-hippocampectomy in man,
Neuropsychologia 33 (1995) 124.
[95] R. Passingham, Delayed matching after selective prefrontal lesions in
monkeys, Brain Res. 92 (1975) 89102.
[96] E. Paulesu, C.D. Frith, R.S.J. Frackowiak, The neural correlates of the verbal
component of working memory, Nature 362 (1993) 342345.
[97] L. Pessoa, E. Gutierrez, P.A. Bandettini, L.G. Ungerleider, Neural correlates of
visual working memory: fMRI amplitude predicts task performance, Neuron
35 (2002) 975987.
[98] J. Peterburs, D.T. Cheng, J.E. Desmond, The association between eye
movements and cerebellar activation in a verbal working memory task,
Cereb. Cortex (2016), http://dx.doi.org/10.1093/cercor/bhv187.
[99] M.E. Petrides, B. Alivisatos, E. Meyer, A.C. Evans, Functional activation of the
human frontal cortex during the performance of verbal working memory,
Proc. Natl. Acad. Sci. U. S. A. 90 (1993) 878882.
[100] B.R. Postle, C.E. Stern, B.R. Rosen, S. Corkin, AN fMRI investigation of cortical
contributions to spatial and nonspatial visual working memory,
Neuroimage 11 (2000) 409423.
[101] G. Rainer, W.F. Asaad, E.K. Miller, Memory elds of neurons in the primate
prefrontal cortex, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 1500815013.
[102] N. Ramnani, The primate cortico-cerebellar system: anatomy and function,
Nat. Rev. 7 (2006) 511522.
[103] C. Ranganath, M.X. Cohen, C. Dam, M. DEsposito, Inferior temporal,
prefrontal, and hippocampal contributions to visual working memory
maintenance and associative memory retrieval, J. Neurosci. 24 (2004)
39173925.
[104] C. Ranganath, M. DEsposito, Directing the minds eye: prefrontal, inferior
and medial temporal mechanisms for visual working memory, Curr. Opin.
Neurobiol. 15 (2005) 175182.
[105] S.M. Ravizza, M.R. Delgado, J.M. Chein, J.T. Becker, J.A. Fiez, Functional
dissociations within the inferior parietal cortex in verbal working memory,
Neuroimage 22 (2004) 562573.
[106] E.T. Rolls, Functions of the primate temporal lobe cortical visual areas in
invariant visual object and face recognition, Neuron 27 (2000) 205218.
[107] B. Rypma, V. Prabhakaran, J.E. Desmond, G.H. Glover, J.D. Gabrieli,
Load-dependent roles of frontal brain regions in the maintenance of
working memory, Neuroimage 9 (1999) 216226.
[108] E. Salmon, M. Van der Linden, F. Collette, G. Delore, P. Maquet, C.
Degueldre, A. Luxen, G. Franck, Regional brain activity during working
memory tasks, Brain 119 (1996) 16171625.
[109] S.P. Scalaidhe, F.A.W. Wilson, P.S. Goldman- Rakic, Face-selective neurons
during passive viewing and working memory performance Rhesus
monkeys: evidence for intrinsic specialization of neuronal coding, Cereb.
Cortex 9 (1999) 459475.
[111] J.D. Schmahmann, D.N. Pandya, Anatomic organization of the basilar pontine
projections from prefrontal cortices in rhesus monkey, J. Neurosci. 17 (1997)
438458.
[112] J.D. Schmahmann, D.N. Pandya, The cerebrocerebellar system, Int. Rev.
Neurobiol. 41 (1997) 3160.
[113] M.L. Seghier, Laterality index in functional MRI: methodological issues,
Magn. Reson. Imaging 26 (2008) 594601.
173
[114] D.J. Simmonds, S.G. Fotedar, S.J. Suskauer, J.J. Pekar, M.B. Denckla, S.H.
Mostofsky, Functional brain correlates of response time variability in
children, Neuropsychologia 45 (2007) 21472157.
[115] E.E. Smith, J. Jonides, M. Gazzaniga, Working memory in humans:
neuropsychological evidence, in: The Cognitive Neurosciences, MIT Press,
Cambridge, MA, 1995, pp. 10091020.
[116] E.E. Smith, J. Jonides, Neuroimaging analyses of human working memory,
Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 1206112068.
[117] E.E. Smith, J. Jonides, R.A. Koeppe, E. Awh, E.H. Schumacher, S. Minoshima,
Spatial vs. object working memory: PET investigations, J. Cognitive
Neurosci. 7 (1995) 337356.
[118] E.E. Smith, J. Jonides, C. Marshuetz, R.A. Koeppe, Components of verbal
working memory: evidence from neuroimaging, Proc. Natl. Acad. Sci. U. S. A.
95 (1998) 876882.
[119] M. Sobczak- Edmans, T.H.B. Ng, Y.C. Chan, E. Chew, K.H. Chuang, S.H.A. Chen,
Temporal dynamics of visual working memory, Neuroimage 124 (2016)
10211030.
[120] C.J. Stoodley, J.D. Schmahmann, Functional topography in the human
cerebellum: a meta-analysis of neuroimaging studies, Neuroimage 44
(2009) 489501.
[121] M. Takagi, D.S. Zee, R.J. Tamargo, Effects of lesions of the oculomotor vermis
on eye movement in primate: saccades, J. Neurophysiol. 80 (1998)
19111931.
[123] W.T. Thach, E.G. Jones, The cerebellar dentatothalamic connection: terminal
eld, lamellae, rods and somatotopy, Brain Res. 169 (1979) 168172.
[124] K.M. Thomas, S.W. King, P.L. Franzen, T.F. Welsh, A.L. Berkowitz, D.C. Noll, V.
Birmaher, B.J. Casey, A developmental functional MRI study of spatial
working memory, Neuroimage 10 (1999) 327338.
[125] M.E. Thomason, E. Race, B. Burrows, S. Whiteld-Gabrieli, G.H. Glover, J.D.E.
Gabrieli, Development of spatial and verbal working memory capacity in the
human brain, J. Cognitive Neurosci. 21 (2009) 316332.
[126] M. Thurling, H. Hautzel, M. Kuper, M.R. Stefanescu, S. Maderwald, M.E. Ladd,
D. Timmann, Involvement of the cerebellar cortex and nuclei in verbal and
visuo-spatial working memory: a 7 T fMRI study, Neuroimage 62 (2012)
15371550.
[127] H. Tomita, M. Ohbayashi, K. Nkahara, I. Hasegawa, Y. Miyashita, Top-down
signal from prefrontal cortex in executive control of memory retrieval,
Nature 401 (1999) 699703.
[128] T. Tsukiura, T. Fujii, T. Takahashi, R.T. Xiao, M. Inase, T. Iijima, A. Yamadori, J.
Okuda, Neuroanatomical discrimination between manipulating and
maintaining processes involved in verbal working memory: a functional
MRI study, Cognitive Brain Res. 11 (2001) 1321.
[129] N. Tzourio- Mazoyer, B. Landeau, D. Papathanassiou, F. Crivello, O. Etard, N.
Delcroix, B. Mazoyer, M. Joliot, Automated anatomical labeling of activations
in SPM using a macroscopic anatomical parcellation of the MNI MRI
single-subject brain, Neuroimage 15 (2002) 273289.
[130] L.G. Ungerleider, S.M. Courtney, J.V. Haxby, A neural system for human
visual working memory, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 883890.
[131] M. van Asselen, R.P.C. Kessels, S.F.W. Neggers, L.J. Kappelle, C.J.M. Frijns, A.
Postma, Brain areas involved in spatial working memory, Neuropsychologia
44 (2006) 11851194.
[132] H. Walter, V. Bretschneider, G. Gron, B. Zurowski, A.P. Wunderlich, R.
Tomczak, M. Spitzer, Evidence for quantitative domain dominance for
verbal and spatial working memory in frontal and parietal cortex, Cortex 39
(2003) 897911.
[133] J.T. Weber, C.T. Yin, Subcortical projections of the inferior parietal cortex
(area 7) in the stump-tailed monkey, J. Comp. Neurol. 224 (1984) 206230.
[134] R. Wiesendanger, M. Wiesendanger, D.G. Ruegg, An anatomical
investigation of the corticopontine projection in the primate
(Macaca-fascicularis and Saimiri sciureus)-II. The projection from frontal
and parietal association areas, Neuroscience 4 (1979) 747766.
[135] F.A.W. Wilson, S.P. Scalaidhe, P.S. Goldman- Rakic, Dissociation of object and
spatial processing domains in primate prefrontal cortex, Science 260 (1993)
19551958.
[136] K. Yamanaka, B. Yamagata, H. Tomioka, S. Kawasaki, M. Mimura,
Transcranial magnetic stimulation of the parietal cortex facilitates spatial
working memory: near-infrared spectroscopy study, Cerebal. Cortex 20
(2010) 10371045.