doi:10.

1093/brain/aws132 Brain 2012: 135; 2192–2201 | 2192

BRAIN
A JOURNAL OF NEUROLOGY

Is the frontal dysexecutive syndrome due to

a working memory deficit? Evidence from
patients with stroke
Martine Roussel,1 Kathy Dujardin,2 Hilde Hénon2 and Olivier Godefroy1

1 Department of Neurology, Laboratory of Functional Neurosciences (EA4559), University Hospital of Amiens, 80054 Amiens Cedex, France
2 Department of Neurology, University Hospital of Lille, 59037 Lille cedex, France

Correspondence to: Martine Roussel,

Service de Neurologie,
Hopital Nord,
F-80054 Amiens,
France
E-mail: roussel.martine@chu-amiens.fr

Downloaded from by guest on October 21, 2014

Although frontal dysexecutive disorders are frequently considered to be due to working memory deficit, this has not been
systematically examined and very little evidence is available for impairment of working memory in frontal damage. The objective
of this study was to examine the components of working memory, their anatomy and the relations with executive functions in
patients with stroke involving the frontal or posterior cortex. The study population consisted of 29 patients (frontal: n = 17;
posterior: n = 12) and 29 matched controls. Phonological loop (letter and word spans, phonological store; rehearsal process),
visuospatial sketchpad (visuospatial span) and the central executive (working memory span, dual task and updating process)
were examined. The group comparison analysis showed impairment in the frontal group of: (i) verbal spans (P 5 0.03); (ii) with
a deficit of the rehearsal process (P = 0.006); (iii) visuospatial span (P = 0.04); (iv) working memory span (P = 0.001) that
disappeared after controlling for verbal span and (v) running memory (P = 0.05) unrelated to updating conditions. The clinical
anatomical correlation study showed that impairment of the central executive depended on frontal and posterior lesion.
Cognitive dysexecutive disorders were observed in 11/20 patients with central executive deficit and an inverse dissociation
was observed in two patients. Receiver operating characteristic curve analysis indicated that cognitive dysexecutive disorders
had the highest ability to discriminate frontal lesions (area under curve = 0.844, 95% confidence interval: 0.74–0.95; P = 0.0001;
central executive impairment: area under curve = 0.732, 95% confidence interval: 0.57–0.82; P = 0.006). This study reveals that
frontal lesions induce mild impairment of short-term memory associated with a deficit of the rehearsal process supporting the
role of the frontal lobe in this process; the central executive depends on lesions in the frontal lobe and posterior regions
accounting for its low frequency and the negative results of group studies. Finally, the frontal dysexecutive syndrome cannot be
attributed to central executive impairment, although it may contribute to some dysexecutive disorders.

Keywords: brain injury; ischaemic stroke; frontal lobe; executive functions; working memory

Received October 20, 2011. Revised March 22, 2012. Accepted April 16, 2012. Advance Access publication May 24, 2012
ß The Author (2012). Published by Oxford University Press on behalf of the Guarantors of Brain. All rights reserved.
For Permissions, please email: journals.permissions@oup.com
Dysexecutive syndrome and working memory
Introduction
Disorders of executive functions are one of the most frequent
cognitive deficits and are observed in many brain diseases.
Following Luria’s approach (1966), the term ‘executive functions’
was coined by Lezak (1982), and was initially circumscribed to
goal setting, action initiation and inhibition, planning, shifting
and verification. This domain has been extended to include be-
havioural changes observed in frontal lesions (Baddeley and
Wilson, 1988). These higher-order functions account for cognitive
and behavioural control and depend mainly on frontal lobes and
fronto-subcortical networks. They are assessed using conventional
executive tests, such as Modified Card Sorting, Stroop, Trail
Making, verbal fluency and Tower of London tests (Godefroy
et al., 2010). Although an impressive number of studies have
documented the frequency and variability of executive disorders,
their underlying mechanisms and related anatomy remain largely
undetermined, resulting in a long list of cognitive dysexecutive
disorders (i.e. demonstrated on cognitive tests; Godefroy et al.,
2010). A better understanding of the underlying cognitive deficits
would help to clarify the dysexecutive syndrome, its assessment
and its anatomy. Three main explanations for dysexecutive dis-
orders have been proposed: a deficit of a supervisory system
involved in the control of action (Norman and Shallice, 1986), a
deficit of attention and sustained alertness (Posner and Petersen,
1990) and a disorder of working memory (Baddeley, 1986). On
the basis of animal data (Golman-Rakic, 1987), the working
memory approach has gained considerable influence (Miyake
et al., 1999; Baddeley, 2000; D’Esposito, 2007). According to
Baddeley (1986), working memory involves two modality-specific
storage components (phonological loop and visuospatial sketch-
pad) and an attentional controlling system (central executive)
that operates on temporary stored information and is involved in
task coordination, inhibition, switching and updating (Baddeley,
1996; Miyake et al., 2000). The term ‘central executive’ was
coined by Baddeley and Hitch (1974) in their initial version of
the working memory model and should be differentiated from
‘executive functions’, which was subsequently proposed by
Lezak (1982) to describe the processes individualized by Luria
(1966). The hypothesis that executive functions depend on the
intervention of the central executive is based on three lines of
argument: (i) most executive tests are complex and consequently
load working memory; (ii) the overlap between processes involved
in conventional executive tests (i.e. Modified Card Sorting, Stroop,
Trail Making, verbal fluency and Tower of London tests) and cog-
nitive operations proposed to depend on the central executive
(task coordination, inhibition, switching and updating); and (iii)
the anatomical overlap as both executive functions and central
executive involve the prefrontal cortex. The relationship between
the central executive and frontal lobe has been mainly docu-
mented by functional imaging studies showing the prominent ac-
tivation of the prefrontal cortex in conditions involving the central
executive (D’Esposito et al., 1995, 2000; Colette and Van der
Linden, 2002). Conversely, evidence for impairment of the central
executive in frontal lesions, especially stroke, remains very limited.
Frontal lesions spare digit and spatial spans (D’Esposito et al.,
Brain 2012: 135; 2192–2201 | 2193
1999), whereas contradictory results have been reported for
delayed response tasks (D’Esposito et al., 1999, 2006). Shallice
and Vallar (1990) reviewed case studies assessing mechanisms of
impairment of the phonological loop (storage versus rehearsal
mechanism) and showed prominent impairment of storage that
is usually related to a left parietal lesion. The deficit of the re-
hearsal mechanism is reported in rare cases, mostly in patients
with a left inferior frontal lesion (Belleville et al., 1992; Waters
et al., 1992; Vallar et al., 1997). Processes depending on the
central executive (task coordination, switching and updating) are
usually assessed using a dual task paradigm (requiring coordination
of the execution of diverse sets of tasks), working memory span
(requiring a simultaneous combination of online processing and
information storage) and running memory task (requiring updating
of relevant information and suppression of no-longer-relevant
data; Pollack et al., 1959; Morris and Jones, 1990). Most studies
on the central executive in focal lesions have focused on task
coordination and none have systematically examined all working
memory processes. Impairment of dual task due to a frontal lesion
has been reported in only two studies (Cowey and Green, 1996;
Leclercq et al., 2000) and only one showed that this impairment
was more marked than in lesions of the posterior cortex (Cowey
and Green, 1996). One (Baddeley et al., 1997) of the four nega-
tive studies (Vilkki et al., 1996, 2002; Baddeley et al., 1997;
Downloaded from by guest on October 21, 2014
Andres and Van der Linden, 2002) showed poor performance in
the subgroup of frontal patients with behavioural changes. Two
studies using sequential dual tasks (Godefroy et al., 1999; Stablum
et al., 2000) reported more marked response slowing in patients
with frontal lesions, but this impairment was not related to other
deficits of controlled processes (Godefroy et al., 1999). There is,
therefore, little evidence that frontal lesions impair the central ex-
ecutive as defined by Baddeley et al. (1997). This contrasts with
impairment of the central executive reported in diffuse diseases,
such as Alzheimer’s disease (Baddeley et al., 1991; Colette et al.,
1999; Huntley et al., 2010), traumatic brain injury (McDowell,
1997; Azouvi et al., 2009; Barbey et al., 2011; Hillary et al.,
2011) and Parkinson’s disease (Morris et al., 1988; Pillon et al.,
2001; Beato et al., 2008). However, no study has examined
whether working memory impairment underlies dysexecutive dis-
orders, i.e. the emergence of dysexecutive cognitive disorders is
conditioned by impairment of working memory.
The objective of this study was to examine working memory,
executive function and the relationship between the two. It was
performed in stroke, a disease that frequently impairs executive
functions (Tatemichi et al., 1994; Sachdev et al., 2004; Godefroy
et al., 2010) and which is suitable for examining associations and
dissociations between deficits (Godefroy et al., 1998; Müller and
Knight, 2006).
Materials and methods
Population
Consecutive patients aged 18–70 years referred to the Lille stroke
centre for recent stroke (51 month) visualized by brain imaging
2194 | Brain 2012: 135; 2192–2201 M. Roussel et al.

were considered for inclusion. Exclusion criteria were: (i) sensori-
motor deficit, hemineglect or aphasia precluding cognitive assess-
ment (Godefroy et al., 2002); (ii) illiteracy; (iii) alcoholism or
severe general comorbidity; (iv) previous neurological and psychiatric
diseases except for depression or anxiety; (v) recent introduction of
psychoactive or anti-epileptic medication; and (vi) absence of in-
formed consent. This study included 29 patients and 29 age and
education-matched controls (Table 1). Neuroradiological analysis
was performed using MRI scans according to a method providing
good inter-observer reliability (Godefroy et al., 1998; Supplementary
material). It determined the presence of a lesion in the 22 regions of
interest on each side and white matter abnormalities (Fazekas et al.,
1987). Lesion analysis (Fig. 1) indicated that 17 patients (seven
females; 10 males) had a lesion restricted to the anterior region
(frontal group) and 12 patients (10 females; two males) had a
lesion limited to the posterior region (posterior group; Table 1).
Demographic characteristics did not differ except for gender due
to a higher proportion of females in the control and posterior
groups. For this reason, all analyses first assessed the gender
effect; as this effect was never significant, only group comparison
analyses are reported. This study was approved by the institutional
review board.
Neuropsychological assessment
Neuropsychological assessment was performed with a mean post-stroke
interval of 17.4  10.6 days. Results of the general neuropsychological
assessment are provided in Table 1 (see Supplementary material for
detailed methods and results of general neuropsychological assessment).
Briefly, the general neuropsychological assessment showed that: (i) gen-
eral intellectual efficiency (Raven, 1965; Beauregard, 1971; Folstein
et al., 1975) was spared; (ii) the frontal group presented impairment
of digit and spatial span (Weschler, 1989), episodic memory (Signoret,
1991) and most behavioural (Godefroy et al., 1996) and cognitive
(Godefroy et al., 2010) assessments of executive functions using con-
ventional tests [including verbal fluency, Stroop (1935), Trail Making
(Reitan, 1958), Tower of London (Shallice, 1982) and Modified Card
Sorting tests (Nelson, 1976)]; and (iii) the posterior group presented
impairment of episodic memory and two executive assessments
(verbal fluency and Modified Card Sorting test).

Table 1 Demographic characteristics, lesions and performance on tests assessing executive functions

Frontal group Posterior group Controls P

n 17 12 29

Downloaded from by guest on October 21, 2014

Age (years) 47  13 43.3  13 46.3  13.2 0.7
Education (years) 10  2.9 9.9  2.1 10.1  2.9 0.9
Stroke type: infarct/haemorrhage 12/3 9/3 –
White matter abnormalities 0/1/2/3 12/4/1/0 10/1/0/1 – 0.4
Frontal: left/right 9/9 0/0 – –
Basal/mesial/lateral 4/6/13 0/0/0 – –
Posterior cortex: left/right 1/2 6/6 – –
Temporal lateral/mesial/occipital 3/0/0 6/4/3 – –
Parietal superior/inferior lobulus 0 /0 2/1 – –
Deep structures: left/right 2/5 1/4 – –
Lenticular nucleus/caudate nucleus/thalamus 0/2/0 1/1/2 – –
Semiovale centre/internal capsule 5/0 4/2 – –
Mini-mental state examination (/30) 27.7  2 28.3  2 28.8  0.9 0.06
Progressive matrices (/36) 29.2  4.4 30.8  5.2 31.8  3.5 0.14
Premorbid IQ 104  16 105  14 110  12 0.3
Digit span forward 5.18  1.1a 5.3  1 61 0.03
Digit span backward 3.88  1.1 3.83  1.2 4.55  1.2 0.08
Behaviour
Apathy (/5) 1.24  0.4a,b 10 10 0.005
Distractibility (/5) 1.5  0.9a,b 10 1  0.1 0.01
Fluency
Animals 24.4  6.5a 24.8  12.9a 36.6  7.6 0.0001
Letter ‘P’ 13.1  6.1a 13.6  8.9a 21.5  5.8 0.0001
Modified Card Sorting Test
Category 4.9  1.6a 5.1  1.6 5.9  0.3 0.007
Perseveration 1.3  1.7 1.7  2.7a 0.4  0.7 0.04
Tower of London
Problem correct 11.1  1.1a 11.3  0.8 11.9  0.4 0.006
Trail Making test
Perseveration 1.53  3.5 0.6  0.8 0.14  0.5 0.08
Stroop
Interference index 3.6  6.3a 0.7  1.2 0.1  0.6 0.006

Results are expressed as mean  SD except lesion location (number).

a Significant difference with controls.
b Significant difference between both patient groups.
Dysexecutive syndrome and working memory
Figure 1 Lesion overlaps in MNI space ( 15, 0, 30 and 45 mm level). Right side corresponds to right hemisphere.
Examination of working memory
components
According to the model of Baddeley (1996), we examined the phono-
logical loop (phonological short-term store and the rehearsal process),
the visuospatial sketchpad and the central executive (working memory
span, dual task test and running memory span) by means of compu-
terized tests. Auditory stimuli were presented through loudspeakers
and visual stimuli were displayed in the centre of the monitor
screen. Responses were provided either verbally or using a
touch-sensitive screen. To control for a possible decrease of storage
capacity, tasks assessing the phonological loop and central executive
used list length at each subject’s span level.
Assessment of the phonological loop
According to Belleville’s procedure (1996), the following functions
were assessed: (i) phonological abilities using rhyme judgement and
homophony judgement tasks; (ii) overall capacity using span tasks
(consonant, word); (iii) short-term store using phonological similarity
effect; and (iv) the rehearsal process using the word length effect.
Except for phonological abilities, performance was analysed using sti-
muli delivered according to both auditory and visual modalities.
In the rhyme judgement and homophony judgement tasks, refer-
ence words were displayed one by one on the monitor screen
(Supplementary material). Four stimulus words (one target and three
distractors) were displayed. The subject had to indicate which word
rhymed with (rhyme judgment test) or sounded like (homophony
judgment test) the reference word.
Verbal span tests used dissimilar consonants (‘f, h, l, j, r, k, m’) and
monosyllabic words [‘sac’, ‘lune’, ‘train’, ‘mur’, ‘jambe’, ‘neige’,
‘chien’, ‘clef’, ‘bois’ (bag, moon, train, wall, leg, snow, dog, key,
wood)]. Five series of various lengths (from two to seven) were pre-
sented. In both auditory and visual modalities, items were displayed at
a rate of one item per 1.5 s. Span was determined as the longest series
length allowing correct recall of at least three out of five series.
The phonological similarity effect was determined by comparison of
recall of dissimilar (‘f, h, l, j, r, k, m’) and similar sounding consonants
(‘b, c, d, g, p, t, v’). List length corresponded to each subject’s letter
span. In both auditory and visual modalities, letters were displayed at a
rate of 1.5 s. Performance was determined by the number of correctly
recalled lists of similar (n = 5 trials) and dissimilar (n = 5 trials) conson-
ants on auditory and visual modalities. The dependant variable was the
similarity index (number of correctly recalled lists of similar consonants
minus dissimilar consonants) computed for both auditory and visual
modalities.
Brain 2012: 135; 2192–2201 | 2195
The word length effect was determined by comparison of recall of
lists of monosyllabic and four-syllable words [‘ordinateur’, ‘bib-
liothèque’, ‘photographie’, ‘anniversaire’, ‘appartement’, ‘médicament’,
‘dessinateur’, ‘télévision’, ‘publicité’ (computer, library, photography,
birthday, apartment, drug, draughtsman, television, publicity)]. List
length was adjusted to each subject’s word span. Word length effect
was assessed for both auditory and visual modalities. Words were dis-
played at a rate of 1.5 s. Performance was assessed by the number of
correctly recalled lists of monosyllabic (n = 5 trials) and four-syllable
(n = 5 trials) words on auditory and visual modalities. The dependent
variable was the word length index (number of correctly recalled lists
Downloaded from by guest on October 21, 2014
of monosyllabic words minus four-syllable words) computed for both
auditory and visual modalities.
Assessment of the visuospatial sketchpad
The spatial span test procedure was similar to that described by
Owen et al. (1990). The subject was required to reproduce a se-
quence of illuminated squares (rate: one square per second) within
an array of nine blocks by pointing to each location in the same
order. Five series of different lengths were presented. Span was deter-
mined as the longest series length allowing correct recall of at least
3/5 series.
Central executive
Assessment of the central executive included working memory span,
dual task coordination and a running memory task. Working memory
span used three-syllable non-words as stimuli. Non-words were con-
structed by adding a syllable between the first and last syllables of
French disyllabic words [e.g. ‘maison’ (‘house’) led to ‘mai/ta/son’]
matched according to occurrence frequency and imagery. Stimuli
were produced aloud (one item/2 s). Subjects had to identify each
word by deleting the middle syllable (e.g. ‘mai/ta/son’ = ‘maison’)
and then remember them. The efficiency of syllabic deletion was
checked before testing. Series of increasing length from two to five
words were presented with the same procedure as the verbal span.
The computerized dual task used a visuomotor tracking task and an
auditory-verbal memory task. Each task (lasting 2 min) was performed
first as a single task condition and then as a combined task condition
(Baddeley et al., 1986). The visuomotor tracking task required the
subject to track with the index finger a square (2 cm) unpredictably
moving on the centre of the touch-sensitive screen. For each subject,
the target speed was determined to achieve a time on target between
40 and 60%. The verbal memory task required the subject to recall
series of dissimilar consonants at lengths corresponding to each sub-
ject’s span level. The mu index was used to quantify the dual task
decrement (Baddeley et al., 1997). This index was computed as
2196 | Brain 2012: 135; 2192–2201
follows: Mu = [1 (pm + pt)/2]  100 where pm is the proportional
loss of memory performance under dual-task conditions and pt is the
proportional loss in tracking score.
The running memory task (Pollack et al., 1959) was derived from
that developed by Morris and Jones (1990). It used strings of alpha-
betical series, dissimilar consonants and abstract drawings. Strings of
alphabetical series (‘b, c, d, e, f, g’) were used to assess the updating
process per se, as they represented ‘overlearned’ material, hence not
requiring short-term storage. The procedure was similar across condi-
tions: strings of items (alphabetical series, consonant, abstract draw-
ings) were presented one by one at a 1.5 s rate in the centre of the
computer screen. The subject was instructed to recall the last N items,
where N corresponds to the subject’s span. Four updating conditions
were used: a null update condition (list length = subject span) and
conditions with 2, 4, 6 updating (list length = subject span + 2, 4 or
6 items). For each material (alphabetical series, consonant, abstract
drawings), 20 lists (five per updating condition 0, 2, 4, 6) were pre-
sented after 10 practice trials. Performance was defined by the number
of lists correctly recalled (/5) in each updating condition (n = 4) for
each material (n = 3).
Statistics
Group comparisons were performed by one-way and repeated
ANOVA with group (frontal, posterior and control groups) as
between-subject factor and task conditions as within-subject factor.
Response time was log-transformed. Post hoc analyses used the
Bonferroni test for the factor ‘Group’ and, when relevant, contrast
analyses for the factor ‘Condition’.
Individual performance was analysed using cut-off scores at the
5% level. The presence of dysexecutive disorders was determined
using performance on conventional tests of executive functions and
behavioural disorders. According to a previously validated method
(Godefroy et al., 2010), cognitive dysexecutive disorders were con-
sidered to be present when at least two of the following five indices
were impaired: (i) 51 fluency test impaired; (ii) 51 impaired com-
pletion time on Trail Making test, naming and reading subtests of the
Stroop; (iii) 51 impaired perseveration rate on Trail Making test B
and Modified Card Sorting test; (iv) Stroop interference index; (v)
category achieved on the Modified Card Sorting test. A central ex-
ecutive index was computed by summing the impairment on working
memory span, dual task and total scores of the running memory span
using consonants and drawings. Based on control data, central ex-
ecutive index 51/3 was considered to be impaired. A clinical–ana-
tomical correlation study was performed to identify the lesions
responsible for central executive impairment. This method has been
previously validated (Godefroy et al., 1998) and is similar to that
used in voxel-based lesion–symptom mapping procedures (Bates
et al., 2003), although the anatomical grain used is much coarser.
It used classification and regression tree analysis with impairment of
the central executive index as dependent variable and the following
independent variables submitted to analysis: age (years), education
(years of schooling) and presence of a lesion (0 = no, 1 = yes) in
the 22 paired regions of interest. Factors selected by regression tree
analysis were then entered into a logistic regression analysis. The
relationship between dysexecutive disorders and central executive im-
pairment was examined using Fisher’s exact test. A dissociated deficit
was defined by impairment of one set of tests and preservation of
the other set (i.e. impairment of conventional tests of executive func-
tions with sparing of central executive index or vice versa). The dis-
sociation was considered significant only when a set of tests was
performed within 1 SD of controls and the other set scored 51.98
M. Roussel et al.
SD below their mean, which corresponds to the 0.05 level (Anastasi,
1988; Godefroy et al., 1999). Receiver operating characteristic (ROC)
and areas under curves (AUC) were generated to examine the ability
to discriminate patients with frontal lesions. This analysis used central
executive index and conventional tests of executive functions. The
alpha level was set at 0.05 for all analyses. Statistical analyses were
performed using SPSS (SPSS Inc.).
Results
Phonological loop
Rhyme judgement and homophony judgement tasks
Performance was analysed by two one-way ANOVAs using the
group factor (frontal, posterior, controls) as between-subject
factor. In the rhyme judgement task, a significant group effect
[F(2,51) = 3.5, P = 0.037] was observed due to impairment of
the frontal group (P = 0.03; Table 2). The homophony judgement
task did not differ between groups (P = 0.2). This analysis indicates
that rhyme judgement was impaired in the frontal group.
Consonant and word spans
Downloaded from by guest on October 21, 2014
Consonant and word spans were analysed by two ANOVA’s using
the group factor (frontal, posterior and control groups) as
between-subject factor with repeated measures on modality (audi-
tory, visual).
Consonant span analysis (Table 2) showed a significant group
effect [F(2,54) = 4.92, P = 0.039] due to lower span in the frontal
group (P = 0.03) and a modality effect [F(1,54) = 14.92, P =
0.0005] was due to shorter span on visual presentation
(group  modality: P = 0.75).
Word span analysis showed a significant group effect [F(2,55) =
5.72, P = 0.006] due to shorter span in the frontal group
(P = 0.034) with no modality effect (P = 0.11; group  modality:
P = 0.55).
These analyses show that word and consonant spans were
shorter in the frontal group, indicating decreased verbal short-term
memory in the frontal group.
Phonological similarity effect
Performance (similarity index) was analysed by ANOVA using the
group factor (frontal, posterior and control groups) as between-
subject factor with repeated measures on modality (auditory,
visual). No factor was found to be significant (group: P = 0.15;
modality: P = 0.8; modality  group: P = 0.5; Table 2;
Supplementary material). This analysis indicates that the phono-
logical similarity effect did not differ across groups.
Word length effect
Performance (word length index) was analysed by ANOVA
using the group factor (frontal, posterior and control groups)
as between-subject factor with repeated measures on modality
(auditory, visual). The significant group effect [F(2,51) = 6.8
P = 0.002] consisted of a lower word length effect in the
frontal group (frontal = 1.77  1.2; posterior = 2.88  1.3;
Dysexecutive syndrome and working memory Brain 2012: 135; 2192–2201 | 2197

Table 2 Phonological loop and visuospatial assessment

Frontal Posterior Controls P

n 17 12 29
Homophony judgement (/10) 9.3  1.21 9.4  1.07 9.7  0.52 0.2
Rhyme judgement (/10) 7.9  2.75a 8.5  2.01 9.5  0.85 0.04
Consonant span
Auditory modality 4.1  1.1a 4.7  1.1 5  1.2 –
Visual modality 3.8  0.8a 4.2  1.05 4.5  0.9 –
Word span
Auditory modality 3.9  0.7a 4.1  1 4.6  0.7 –
Visual modality 3.7  0.8a 3.8  1.2 4.6  1.2 –
Phonological similarity effect
Auditory modality 0.75  1.3 0.5  1.0 1.2  1.2 –
Visual modality 1.00  1.6 0.4  1.2 0.9  1.2 –
Word length effect
Auditory modality 1.8  1.1a 2.7  1.2 2.1  1.2 –
Visual modality 2.3  1.2 3.3  1.5 2.9  1.2 –
Spatial span 4.6  1.06a 51 5.5  1.09 0.04

a Significant difference compared with controls.

controls = 2.67  1.2) compared with controls (P = 0.006) and the of a span task and a concurrent processing task. This issue was
posterior group (P = 0.007; Table 2; Supplementary material). The addressed by calculating the span decrement due to the concur-

Downloaded from by guest on October 21, 2014

modality effect was significant [F(1,51) = 9.26 P = 0.004] due to a
lower word length effect on the verbal modality than on the visual
modality (verbal = 2.1  1.3; visual = 2.8  1.3; group  modality
interaction: P = 0.9). This analysis indicates that the word length
effect was higher for visually presented words and lower in the
frontal group.
In summary, assessment of the phonological loop was impaired
in the frontal group with lower letter and word spans, a mild but
significant deficit of rhyme judgement, and a lower effect of word
length, suggesting a deficit of the rehearsal component. Inversely,
the storage component was spared as shown by analysis of the
similarity effect.
Visuospatial sketchpad
rent processing task (span decrement = word span in visual pres-
entation – working memory span). The span decrement (frontal:
1.65  0.8; posterior: 1.67  0.8; controls: 1.52  1.2) did not
differ between groups (P = 0.9). This analysis indicates that both
patient groups had lower working memory span and that this
impairment disappeared after controlling for short-term memory
capacity.
Dual task test
Mu index was analysed by ANOVA using a group factor (frontal,
posterior and control groups). No difference was observed be-
tween groups [F(2,51) = 1.34, P = 0.3] (frontal = 88.2  16.8;
posterior = 92.5  17.7; controls = 95.3  10.1). Dual task per-
formance did not differ between groups.

Spatial span was analysed by ANOVA using the group factor
(frontal, posterior and control groups) as between-subject factor.
The group effect [F(2,54) = 5.2, P = 0.036] was significant due to
a shorter span (frontal = 4.6  1.06; posterior = 5  1; con-
trols = 5.5  1.06) in the frontal group (P = 0.035; Table 2). This
analysis indicates that the frontal group had shorter visuospatial
spans.
Central executive
Working memory span
Working memory span was analysed by ANOVA using a group
factor (frontal, posterior and control groups) and showed a sig-
nificant group effect [F(2,51) = 11.2; P = 0.0001] due to shorter
span in both the frontal (P = 0.0001) and posterior (P = 0.001)
groups (frontal = 2.1  0.7; posterior = 2.3  1.1; con-
trols = 3.1  0.7). This deficit could be due to lower short-term
memory capacity or may be specifically due to the combination
Running memory task
The number of correct recalls was analysed by ANOVA using a
group factor (frontal, posterior and control groups) with repeated
analyses on material factors (alphabetical series, drawings, conson-
ants) and number of updates (0, 2, 4, 6). It showed a significant
effect of (i) group [F(2,48) = 3.61, P = 0.03] due to lower per-
formance in the frontal group (frontal = 2.96  1.6; poster-
ior = 3.2  1.5; controls = 3.47  1.6; P = 0.05); (ii) material
[F(2,48) = 51.4, P = 0.0001], due to lower recall of consonants
compared to alphabetical series (P = 0.001) and drawings
(P = 0.04; alphabetical series = 4.3  1.1; drawings = 2.97  1.6;
consonants = 2.6  1.6); (iii) update number [F(3,48) = 57, P =
0.0001], due to lower recall in six update conditions compared
with the other conditions (zero update = 4.2  1.1; two up-
dates = 3.44  1.5; four updates = 2.85  1.6; six updates =
2.57  1.7; P = 0.01; Table 3). The material  update interaction
was significant due to a more marked decrease of recall in the six
update conditions for consonants than for drawings (P = 0.04) and
2198 | Brain 2012: 135; 2192–2201 M. Roussel et al.

alphabetical series (P = 0.0001). Finally, the group  mater- group  update number: P = 0.25). Patient recall was therefore
ial  update number interaction (P = 0.001) was significant (Fig. not disproportionately sensitive to the number of updates.
2) due to a less marked reduction of recall with increasing updates However, in contrast with controls, patients did not benefit from
in the frontal group compared with controls (P = 0.005) for draw- over-learned alphabetical series.
ings (frontal: 0.7  1.4; posterior: 1.2  1.2; controls: This analysis indicates a global recall deficit in the frontal group
2.76  1.5). The reverse pattern was observed for alphabetical with no disproportionate sensitivity to the number of updates and
series for which increasing updates did not reduce recall in controls the lack of benefit for the over-learned alphabetical series. In sum-
( 0.3  0.9) but did in patients (frontal: 1.2  1.3; posterior: mary, central executive assessment was impaired in both patient
1.4  1.4; P = 0.04, both; group  material: P = 0.7; groups for working memory span, but this disappeared after con-
trolling for short-term memory capacity, and in the frontal group
for the running memory task without disproportionate sensitivity
to the number of updates. The dual task was not impaired.
Table 3 Running memory: mean number of correct
responses at span according to update numbers

Frontal group Posterior group Controls Clinical–anatomical correlation

Alphabetic series Impairment of the central executive index was more frequent in
0 updates 4.87  0.35 4.67  0.49 4.62  1.01
the frontal (9/17) and posterior (4/12) groups than in controls
2 updates 4.07  1.28 4.42  1.00 4.55  0.91
(0/29; P = 0.001, both). Bivariate analyses (Supplementary Table
4 updates 3.80  1.21 3.83  1.03 4.14  1.22
6 updates 3.67  1.29 3.25  1.14 4.34  1.23
3) showed that it was more frequent in right lesions of the frontal
Consonant series basal region and of the posterior part of the centrum semiovale. A
0 updates 3.40  1.35 4.00  1.28 4.45  0.83 strong tendency was observed for left lesions of frontomesial,
2 updates 2.07  1.33 2.67  1.67 3.41  1.30 frontobasal, temporal lateral areas and right lesions of the
4 updates 1.73  1.10 1.67  1.37 1.93  1.36 middle frontal gyrus. Multivariate analyses selected right lesions

Downloaded from by guest on October 21, 2014

6 updates 1.53  1.51 1.83  1.59 1.83  1.26 of frontobasal [odds ratio (OR): 18.4, 95% confidence interval
Drawings series (CI): 1.7–203; P = 0.02] and posterior centrum semiovale (OR:
0 updates 3.20  1.42 3.58  1.00 4.52  0.74
18.4, 95% CI: 1.7–203; P = 0.02) regions. These lesions correctly
2 updates 2.67  1.50 3.00  1.60 3.31  1.51
classified 20 (69%) patients (2 = 12.2; P = 0.002). This analysis
4 updates 2.40  1.68 3.08  1.24 2.79  1.68
6 updates 2.47  1.46 2.42  0.90 1.76  1.48 indicates that central executive impairment depends on frontal and
posterior lesions.

5
Frontal alphabetic
Number of correct recall at span

Posterior alphabetic
4
Control alphabetic

Frontal consonant

3 Posterior consonant

Control consonant

Frontal drawings
2
Posterior drawings

Control drawings

0
0 update 2 updates 4 updates 6 updates
Figure 2 Running memory task: effect of updating number (slope) on correct response for alphabetic series (triangle), consonants
(square) and drawings (circle) in frontal (black symbols and lines), posterior (grey symbols and lines) and control (white symbols and dotted
lines) groups.
Dysexecutive syndrome and working memory
Relationship with conventional
executive tests
Cognitive dysexecutive disorders determined on conventional ex-
ecutive tests were more frequent in the frontal (11/17) and pos-
terior (9/12) groups than in controls (0/29; P = 0.0001, both).
Impairment of central executive index was more frequent
(P = 0.0001) in patients with cognitive dysexecutive disorders: 11
patients presented impairment of both functions. However, 11
patients had a dissociated deficit: nine had dysexecutive disorders
without central executive impairment and two presented the re-
verse pattern. Thus, there were an equal number of cases with
associated and dissociated deficits (n = 11, both). The ROC curve
analysis (Supplementary Fig. 1) indicated that cognitive dysexecu-
tive disorders had the highest area under curve (AUC = 0.844,
95% CI: 0.74–0.95; P = 0.0001) suggesting a good to excellent
ability to discriminate frontal lesions from non-frontal lesions. The
AUC of central executive disorders was lower (AUC = 0.732, 95%
CI: 0.57–0.82; P = 0.006).
This analysis indicates that about one-half of patients with cog-
nitive dysexecutive disorders on conventional tests had no central
executive impairment and reciprocally that central executive im-
pairment may be observed in a few patients with no cognitive
dysexecutive disorders.
Discussion
This study shows that frontal damage was associated with a mild
deficit of verbal spans, a decreased word length effect suggesting
impairment of the rehearsal process of the phonological loop and
a mild deficit of spatial span suggesting an alteration of the visuo-
spatial sketchpad. Results also showed that frontal damage was
associated with impairment of some central executive tests, im-
pairment of working memory span that disappeared after control
for short-term capacity and impairment of running memory with-
out a disproportionate decrease of performance with updating
number. The clinical–anatomical correlation study showed that
central executive impairment depended on frontal and posterior
lesions. Finally, one-half of patients with cognitive dysexecutive
disorders on conventional tests had no central executive impair-
ment and a reverse pattern (impaired central executive without
cognitive dysexecutive disorders) was observed in a few patients.
The demonstration of impairment in verbal and spatial spans in
the frontal group was unexpected, as pooled analysis of previous
studies showed that these functions are usually spared in patients
with frontal lesions (D’Esposito et al., 1999, 2006) and that severe
deficits are observed in the context of parietal lesions (Warrington
and Shallice, 1969; Shallice and Vallar, 1990; Godefroy et al.,
1999). Accordingly, the present impairment of the frontal group
is characterized by a mild decrement of letter, word and spatial
spans. This contrasts sharply with the major impairment of
short-term memory capacity observed in parietal lesions, in
which span is usually 42 (Warrington and Shallice, 1969). Both
lines of results suggest that deficits of short-term memory are
mainly due to parietal lesions and that frontal damage induces a
mild decrement of spans.
Brain 2012: 135; 2192–2201 | 2199
The decrement of consonant and word spans is likely to be due
to impairment of the phonological loop. This is supported by the
relative sparing of phonological processing: the sole deficit con-
cerned rhyme judgement and although significant, was very mod-
erate. More importantly, the spared similarity effect with a
reduced word length effect indicates preservation of the phono-
logical store and impairment of the rehearsal process. These results
converge to indicate that frontal lesions induce a deficit of the
rehearsal process of the phonological loop. This supports the
case studies that show deficits of rehearsal process in patients
with lesions of Broca’s area (Shallice and Vallar, 1990) and sub-
cortical and premotor regions (Vallar et al., 1997). This is consist-
ent with functional activation studies showing that activation of
Broca’s area is related to the rehearsal component and that the left
inferior parietal region is related to the phonological store (Paulesu
et al., 1993; Ahw et al., 1996).
Although some tests assessing the central executive were im-
paired in patient groups, these impairments cannot be attributed
to a disorder of the central executive per se. The working memory
span was impaired in both patient groups; however, the impair-
ment completely disappeared after adjustment for short-term
memory capacity. This indicates that impairment of working
memory span is simply due to decreased short-term memory cap-
acity and therefore cannot be attributed to impairment of the
Downloaded from by guest on October 21, 2014
combination of memory task and a concurrent processing task.
Dual task paradigm was spared at the group level. This finding
is consistent with most previous group studies of focal lesions
(Vilkki et al., 1996, 2002; Baddeley et al., 1997; Andres and
Van der Linden, 2002). It is likely to be due to the low frequency
of deficits on these tasks (Godefroy et al., 2010). The running
memory test was impaired in the frontal group; however, this
impairment cannot be attributed to a deficit of updating processes,
as it would have induced a disproportionate decrement of per-
formance with increasing number of updates. The reverse pattern
was observed, indicating that frontal patients do not benefit from
the easiest condition requiring no updating. This inability to benefit
from easiest conditions was also observed for alphabetical series,
in which a disproportionate decrement with increasing update
number was observed in patients. The deficit of the frontal
group is likely to be due to their inability to take advantage of
an easy condition as observed in a simple reaction time test
(Godefroy et al., 2002). The running memory impairment there-
fore cannot be attributed to a deficit of updating processes.
Overall, these results do not clearly demonstrate that frontal
patients have a deficit of combination of online processing and
information storage that depends on the central executive.
Accordingly, the clinical–anatomical correlation study shows that
central executive impairment depends on lesions involving the
frontal and posterior regions.
Finally, this study showed that about one-half of patients with
cognitive dysexecutive disorders on conventional tests have no
central executive impairment. This result strongly argues against
the central executive theory of dysexecutive disorders, which
would have predicted that all dysexecutive patients would present
central executive impairment. In addition, a double dissociation
was observed as a few patients with impaired central executive
had no dysexecutive disorders. This double dissociation rules out a
2200 | Brain 2012: 135; 2192–2201
potential difference in sensitivity between working memory tasks
and conventional executive tasks, as this would result in a simple
dissociation (i.e. all patients with impairment on low sensitivity
tasks are impaired on highly sensitive tasks). This double dissoci-
ation therefore supports the relative independence of central ex-
ecutive and cognitive dysexecutive disorders. Although working
memory is certainly involved in executive tests, there is no evi-
dence to suggest that the critical point is the ability to manipulate
temporarily stored information, even in patients with well-charac-
terized dysexecutive syndrome. The main conclusion of this study
is that central executive impairment contributes to certain dysex-
ecutive disorders but that it cannot be regarded to be the sole
origin of dysexecutive disorders.
Supplementary material
Supplementary material is available at Brain online.
References
Andrés P, Van der Linden M. Are central executive functions working in
patients with focal frontal lesions? Neuropsychologia 2002; 40:
835–45.
Anastasi A. Psychological testing. 6th edn. New York: Macmillan
Publishing Company; 1988.
Awh E, Jonides J, Smith EE, Schumacher EH, Koeppe RA, Katz S.
Dissociation of storage and rehearsal in verbal working memory: evi-
dence from positron emission tomography. Psychol Sci 1996; 7:
25–31.
Azouvi P, Vallat-Azouvi C, Leclercq M, Belmont A. Cognitive deficits
after traumatic coma. Prog Brain Res 2009; 177: 89–110.
Baddeley A. Working memory. Oxford, UK: Clarendon Press; 1986.
Baddeley A. Exploring the central executive. Q J Exp Psychol A 1996; 49:
5–28.
Baddeley A. The episodic buffer: a new component of working memory?
Trends Cogn Sci 2000; 4: 417–23.
Baddeley A, Hitch G. Working memory. In: Bower GA, editor. Recent
advances in learning and motivation. New York: Academic Press;
1974. p. 47–90.
Baddeley AD, Wilson B. Frontal amnesia and the dysexecutive syndrome.
Brain Cogn 1988; 7: 212–30.
Baddeley A, Della Sala S, Papagno C, Spinnler H. Dual-task performance
in dysexecutive and non dysexecutive patients with a frontal lesion.
Neuropsychology 1997; 11: 187–94.
Baddeley A, Logie RH, Bressi S, Della Sala S, Spinnler H. Dementia and
working memory. Q J Exp Psychol A 1986; 38: 603–18.
Baddeley A, Bressi S, Della Sala S, Logie RH, Spinnler H. The decline of
working memory in Alzheimer’s disease: a longitudinal study. Brain
1991; 114: 2521–42.
Barbey AK, Koenigs M, Grafman J. Orbitofrontal contributions to human
working memory. Cereb Cortex 2011; 21: 789–95.
Bates E, Wilson SM, Saygin AP, Dick F, Sereno MI, Knight RT, et al.
Voxel-based lesion-symptom mapping. Nat Neurosci 2003; 6: 448–50.
Beato R, Levy R, Pillon B, Vidal C, du Montcel ST, Deweer B, et al.
Working memory in Parkinson’s disease patients: clinical features
and response to levodopa. Arq Neuropsiquiatr 2008; 66: 147–51.
Beauregard J. Le test des automatismes verbaux. Issy les Moulineaux:
Editions scientifiques et psychotechniques; 1971.
Belleville S, Peretz I, Arguin M. Contribution of articulatory rehearsal to
short-term memory: evidence from a case of selective disruption. Brain
Lang 1992; 43: 713–46.
M. Roussel et al.
Belleville S, Peretz I, Malenfant D. Examination of the working memory
components in normal aging and in dementia of the Alzheimer type.
Neuropsychologia 1996; 34: 195–207.
Colette F, Van der Linden M. Brain imaging of the central executive
component of working memory. [Review]. Neurosci Biobehav Rev
2002; 26: 105–25.
Colette F, Van der Linden M, Bechet S, Salmon E. Phonological loop and
central executive functioning in Alzheimer’s disease. Neuropsychologia
1999; 37: 905–18.
Cowey CM, Green S. The hippocampus: a “working memory” structure?
The effect of hippocampal sclerosis on working memory. Memory
1996; 4: 19–30.
De Renzi E, Nichelli P. Verbal and non-verbal short-term memory
impairment following hemispheric damage. Cortex 1975; 11: 341–54.
D’Esposito M. From cognitive to neural models of working memory.
Philos Trans R Soc Lond B Biol Sci 2007; 362: 761–72.
D’Esposito M, Postle BR. The dependence of span and delayed-response
performance on prefrontal cortex. Neuropsychologia 1999; 37:
1303–15.
D’Esposito M, Cooney JW, Gazzaley A, Gibbs SE, Postle BR. Is the
prefrontal cortex necessary for delay task performance?
Evidence from lesion and FMRI data. J Int Neuropsychol Soc 2006;
12: 248–60.
D’Esposito M, Detre JA, Alsop DC, Shin RK, Atlas S, Grossman M. The
neural basis of the central executive system of working memory.
Nature 1995; 378: 279–81.
D’Esposito M, Postle BR, Rypma B. Prefrontal cortical contributions to
working memory: evidence from event-related fMRI studies. Exp Brain
Res 2000; 133: 3–11.
Downloaded from by guest on October 21, 2014
Eslinger PJ, Damasio AR. Severe disturbance of higher cognition follow-
ing bilateral frontal lobe ablation: patient EVR. Neurology 1985; 35:
1731–41.
Fazekas F, Chawluk JB, Alavi A, Hurtig HI, Zimmerman RA. MR signal
abnormalities at 1.5 T in Alzheimer’s dementia and normal aging. AJR
Am J Roentgenol 1987; 149: 351–6.
Folstein MF, Folstein SE, McHugh PR. “Mini Mental State”: a practical
method for grading the cognitive state of patients for the clinicians. J
Psychiatric Res 1975; 12: 189–98.
Godefroy O, Azouvi P, Robert P, Roussel M, LeGall D, Meulemans T.
On the behalf of the GREFEX* study group Dysexecutive syndrome.
Diagnostic criteria and validation study. Ann Neurol 2010; 68: 855–64.
Godefroy O, Cabaret M, Petit-Chenal V, Pruvo JP, Rousseaux M.
Control functions of the frontal lobes. Modularity of the central super-
visory system? Cortex 1999; 35: 1–20.
Godefroy O, Duhamel A, Leclerc X, Saint Michel T, Hénon H, Leys D.
Brain-behaviour relationships: models for the study of brain-damaged
patients. Brain 1998; 121: 1545–56.
Godefroy O, Lhullier C, Rousseaux M. Non-spatial attention disorders in
patients with frontal or posterior brain damage. Brain 1996; 119:
191–202.
Godefroy O, Lhullier-Lamy C, Rousseaux M. SRT lengthening: role of an
alertness deficit in frontal damaged patients. Neuropsychologia 2002;
40: 2234–41.
Goldman-Rakic PS. Circuitry of primate prefrontal cortex and regulation
of behavior by representational memory. In: Plum F, Mountcastle V,
editors. The nervous system. Higher hunctions of the brain. Handbook
of physiology, Vol. 5. Bethesda, MD: The American Physiological
Society; 1987. p. 373–417.
Huntley JD, Howard RJ. Working memory in early Alzheimer’s disease:
a neuropsychological review. Int J Geriatr Psychiatry 2010; 25:
121–32.
Hillary FG, Medaglia JD, Gates K, Molenaar PC, Slocomb J, Peechatka A,
et al. Examining working memory task acquisition in a disrupted neural
network. Brain 2011; 134: 1555–70.
Leclercq M, Couillet J, Azouvi P, Marlier N, Martin Y, Strypstein E, et al.
Dual-task performance after severe diffuse traumatic brain injury or
vascular prefrontal damage. J Clin Exp Neuropsychol 2000; 22:
339–50.
Dysexecutive syndrome and working memory
Lezak MD. The problem of assessing executive functions. Int J Psychol
1982; 17: 281–97.
Luria AR. Higher cortical functions in man. New York, NY: Basic Books;
1966.
McDowell S, Whyte J, D’Esposito M. Working memory impairments in
traumatic brain injury: evidence from a dual-task paradigm.
Neuropsychologia 1997; 35: 1341–53.
Miyake A, Shah P. Toward unified theories of working memory: emer-
ging general consensus, unresolved theoretical issues, and future re-
search directions. In: Miyake A, Shah P, editors. Models of working
memory: mechanisms of active maintenance and executive control.
New York: Cambridge University Press; 1999. p. 442–81.
Miyake A, Friedman NP, Emerson MJ, Witzki AH, Howerter A,
Wager TD. The unity and diversity of executive functions and their
contributions to complex “Frontal Lobe” tasks: a latent variable ana-
lysis. Cogn Psychol 2000; 41: 49–100.
Morris N, Jones DM. Memory updating in working memory: the role of
the central executive. Br J Psychol 1990; 81: 111–21.
Morris RG, Downes JJ, Sahakian BJ, Evenden JL, Heald A, Robbins TW.
Planning and spatial working memory in Parkinson’s disease. J Neurol
Neurosurg Psychiatry 1988; 51: 757–66.
Müller NG, Knight RT. The functional neuroanatomy of working memory
contributions of human brain lesion studies. Neuroscience 2006; 139:
51–8.
Nelson HE. A modified card sorting test sensitive to frontal lobe defects.
Cortex 1976; 12: 313–24.
Norman DA, Shallice T. Attention to action: willed and automatic control
of behavior. In: Davidson RG, Schwartz GE, Shapiro D, editors.
Consciousness and self regulation: advances in research. New York:
Plenum Press; 1986. p. 1–18.
Owen AM, Downes JJ, Sahakian BJ, Polkey CE, Robbins TW. Planning
and spatial working memory following frontal lobe lesions in man.
Neuropsychologia 1990; 28: 1021–34.
Paulesu E, Frith CD, Frackowiak RS. The neural correlates of the verbal
component of working memory. Nature 1993; 362: 342–5.
Pillon B, Boller F, Levy R, Dubois B. Cognitive deficits in Parkinson’s
disease. In: Boller F, Grafman J, editors. Handbook of neuropsych-
ology. Vol. Vol. 6. Amsterdam: Elsevier; 2001. p. 311–71.
Pollack I, Johnson LB, Knaff PR. Running memory span. J Exp Psychol
1959; 57: 137–46.
Posner MI, Peterson SE. The attention system of the human brain. Annu
Rev Neurosci 1990; 13: 25–42.
Brain 2012: 135; 2192–2201 | 2201
Raven JC. Guide to using the coloured progressive matrices. London, UK:
Lewis (Pub); 1965.
Reitan RM. Validity of the trail making test as an indicator of organic
brain damage. Percept Mot Skills 1958; 8: 271–6.
Sachdev PS, Brodaty H, Valenzuela MJ, Lorentz L, Looi JC, et al. The
neuropsychological profile of vascular cognitive impairment in stroke
and TIA patients. Neurology 2004; 62: 912–9.
Shallice T. Specific impairments of planning. Philos Trans R Soc Lond B
Biol Sci 1982; 298: 199–209.
Shallice T, Vallar G. The impairment of auditory verbal short-term storage.
In: Vallar G, Shallice T, editors. Neuropsychological impairments of short
term memory. New York: Cambridge University Press; 1990. p. 11–54.
Signoret JL. La batterie d’évaluation mnésique - BEM 144. Paris: Elsevier;
1991.
Stablum F, Umilta C, Mogentale C, Carlan M, Guerrini C. Rehabilitation
of executive deficits in closed head injury and anterior communicating
artery aneurysm patients. Psychol Res 2000; 63: 265–78.
Stroop JR. Studies of interference in serial verbal reactions. J Exp Psychol
1935; 18: 643–62.
Tatemichi TK, Desmond DW, Stern Y, Paik M, Sano M, Bagiella E.
Cognitive impairment after stroke: frequency, patterns, and relation-
ship to functional abilities. J Neurol Neurosurg Psychiatry 1994; 57:
202–7.
Vallar G, Di Betta AM, Silveri MC. The phonological short-term store-
rehearsal system: patterns of impairment and neural correlates.
Neuropsychologia 1997; 35: 795–812.
Vilkki J, Levänen S, Servo A. Interference in dual-fluency tasks after
anterior and posterior cerebral lesions. Neuropsychologia 2002; 40
(3): 340–8.
Downloaded from by guest on October 21, 2014
Vilkki J, Virtanen S, Surma-Aho O, Servo A. Dual task performance after
focal cerebral lesions and closed head injuries. Neuropsychologia 1996;
34: 1051–6.
Warrington EK, Shallice T. The selective impairment of auditory verbal
short-term memory. Brain 1969; 92: 885–96.
Waters GS, Caplan D. The measurement of verbal working memory
capacity and its relation to reading comprehension. Q J Exp Psychol
1996; 49: 51–75.
Waters GS, Rochon E, Caplan D. The role of high-level speech planning
in rehearsal: evidence from patients with apraxia of speech. J Mem
Lang 1992; 31: 54–73.
Wechsler D. W.A.I.S.R. - Echelle d’intelligence de Wechsler pour adulte,
révisée. Paris: Editions du centre de psychologie appliquée; 1989.