Subvocal articulatory rehearsal during verbal working memory in multiple sclerosis

Abstract

This study was designed to examine verbal working memory (VWM) components among multiple sclerosis (MS) patients and determine the influence of information processing speed. Of two frequently studied VWM sub-components, subvocal rehearsal was expected to be more affected by MS than short-term memory buffering. Furthermore, worse subvocal rehearsal was predicted to be specifically related to slower cognitive processing. Fifteen MS patients were administered a neuropsychological battery assessing VWM, processing speed, mood, fatigue, and disability. Participants performed a 2-Back VWM task with modified nested conditions designed to increase subvocal rehearsal (via inter-stimulus interval) and short-term memory buffering demands (via phonological similarity). Performance during these 2-Back conditions did not significantly differ and both exhibited strong positive correlations with disability. However, only scores on the subvocal rehearsal 2-Back were significantly related to performance on the remaining test battery, including processing speed and depressive symptoms. Findings suggest that performance during increased subvocal rehearsal demands is specifically influenced by cognitive processing speed and depressive symptoms.

Full text

PLEASE SCROLL DOWN FOR ARTICLE
This article was downloaded by:
[Canadian Research Knowledge Network]
On:
20 April 2010
Access details:
Access Details: [subscription number 783016864]
Publisher
Psychology Press
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-
41 Mortimer Street, London W1T 3JH, UK
Neurocase
Publication details, including instructions for authors and subscription information:
http://www.informaworld.com/smpp/title~content=t713658146
Subvocal articulatory rehearsal during verbal working memory in multiple
sclerosis
Lawrence H. Sweet a; Susan D. Vanderhill a; Beth A. Jerskey a; Norman M. Gordon a; Robert H. Paul
a;Ronald A. Cohen a
a The Warren Alpert Medical School of Brown University, Butler Hospital, Providence, RI, USA
First published on: 15 April 2010
To cite this Article Sweet, Lawrence H. , Vanderhill, Susan D. , Jerskey, Beth A. , Gordon, Norman M. , Paul, Robert H.
andCohen, Ronald A.(2010) 'Subvocal articulatory rehearsal during verbal working memory in multiple sclerosis',
Neurocase,, First published on: 15 April 2010 (iFirst)
To link to this Article: DOI: 10.1080/13554791003620314
URL: http://dx.doi.org/10.1080/13554791003620314
Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf
This article may be used for research, teaching and private study purposes. Any substantial or
systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or
distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses
should be independently verified with primary sources. The publisher shall not be liable for any loss,
actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly
or indirectly in connection with or arising out of the use of this material.

NEUROCASE
2010, iFirst, 1–8
© 2010 Psychology Press, an imprint of the Taylor & Francis Group, an Informa business
http://www.psypress.com/neurocase DOI: 10.1080/13554791003620314
NNCS Subvocal articulatory rehearsal during verbal working
memory in multiple sclerosis
Rehearsal & Working Memory in MS Lawrence H. Sweet, Susan D. Vanderhill, Beth A. Jerskey,
Norman M. Gordon, Robert H. Paul, and Ronald A. Cohen
The Warren Alpert Medical School of Brown University, Butler Hospital, Providence, RI, USA
This study was designed to examine verbal working memory (VWM) components among multiple sclerosis (MS)
patients and determine the influence of information processing speed. Of two frequently studied VWM sub-
components, subvocal rehearsal was expected to be more affected by MS than short-term memory buffering. Fur-
thermore, worse subvocal rehearsal was predicted to be specifically related to slower cognitive processing. Fifteen
MS patients were administered a neuropsychological battery assessing VWM, processing speed, mood, fatigue,
and disability. Participants performed a 2-Back VWM task with modified nested conditions designed to increase
subvocal rehearsal (via inter-stimulus interval) and short-term memory buffering demands (via phonological sim-
ilarity). Performance during these 2-Back conditions did not significantly differ and both exhibited strong positive
correlations with disability. However, only scores on the subvocal rehearsal 2-Back were significantly related to
performance on the remaining test battery, including processing speed and depressive symptoms. Findings suggest
that performance during increased subvocal rehearsal demands is specifically influenced by cognitive processing
speed and depressive symptoms.
Keywords: Information processing speed; Depression; n-Back paradigm; Short-term memory buffering; Verbal
working memory; Subvocal articulatory rehearsal; Multiple sclerosis.
INTRODUCTION
Multiple sclerosis (MS) is the most common non-
traumatic disabling neurological disorder in 20–59-
year-olds (Rao, 1990). Although it is associated
with sensory, motor, and neuropsychiatric symp-
toms, it has been well documented that cognitive
deficits are present in 43–70% of cases (Amato,
Zipoli, & Portaccio, 2006; Chiaravalloti & DeLuca,
2008; Rao, Leo, Bernardin, & Unverzagt, 1991). In
contrast to deficits usually seen in areas of informa-
tion processing speed, working memory, executive
functioning, and long-term memory encoding and
retrieval, other domains such as procedural mem-
ory, forgetting rates, recognition, and immediate
recall are usually not impaired (Calabrese, 2006;
Chiaravalloti & DeLuca, 2008). This pattern sug-
gests that memory encoding and retrieval are often
impaired, while storage may be spared. It is unclear
whether similar patterns of impairment exist within
the component processes of the verbal working
memory (VWM) system, or if they are more uni-
formly affected.
Models of VWM suggest a coordination of
encoding, storage, retrieval, and management of
online information, with executive components
managing these cognitive resources (e.g., Baddeley
& Hitch, 1974; Cowan, 1988; Ericsson & Kintsch,
1995). Baddeley and colleagues proposed the model
with the greatest empirical support in MS
This study was supported by a NINDS grant (F32NS042404) to Dr Sweet.
Address correspondence to Lawrence H. Sweet, Ph.D., The Warren Alpert Medical School of Brown University, Butler Hospital,
345 Blackstone Boulevard, Providence, RI 02906, USA. (E-mail: Lawrence_Sweet@Brown.edu).
Downloaded By: [Canadian Research Knowledge Network] At: 18:25 20 April 2010

2 SWEET ET AL.
(D’Esposito et al., 1996; Diamond, DeLuca, Kim,
& Kelley, 1997; Hillary et al., 2003; Lengenfelder
et al., 2006), including functional neuroimaging
(Hillary et al., 2003; Sweet, Rao, Primeau, Durgerian,
& Cohen, 2006; Sweet, Rao, Primeau, Mayer, &
Cohen, 2004). According to Baddeley’s model, the
executive component of working memory coordi-
nates the short-term memory (STM) buffering sub-
systems of at least two different modalities (e.g.,
visual-spatial, verbal). The phonological loop,
depicted schematically in Figure 1, is the buffering
subsystem that is used to hold verbal information
online, whether the stimulus is presented in phone-
mic form or is automatically translated (e.g., when
reading). Phonemes in the STM buffer of the pho-
nological loop decay in about 2 s unless they are
maintained by subvocal articulatory rehearsal
(Baddeley, Thompson, & Buchanan, 1975). This
two-part mechanism of STM buffering and subvo-
cal rehearsal has been examined via paradigms that
have yielded reliable effects upon VWM compo-
nents. For instance, the phonological similarity
effect denotes that dissimilar sounding phonemes
are recalled better than similar phonemes due to the
lack of unique features that aid in storage (Badde-
ley, 1966). A word length effect has been demon-
strated in which shorter words are recalled better
than longer words due to greater demands placed
upon rehearsal (Baddeley et al., 1975). Other inves-
tigators have manipulated retention intervals to
study the effects of increased rehearsal demands
(Barch et al., 1997).
The VWM system has strong construct validity,
including support for subsystems such as the
phonological loop (Baddeley, 1966; Baddeley &
Lieberman, 1980; Barch et al., 1997; Logie, 1986;
Logie, Zucco, & Baddeley, 1990) and its compo-
nents (i.e., STM buffer; Baddeley, 1966) and subvo-
cal articulatory rehearsal (Baddeley et al., 1975).
Functional neuroimaging literature suggests that
these VWM components can be dissociated and
mapped to unique brain regions using experimental
manipulations in healthy volunteers (Barch et al.,
1997; Paulesu, Frith, & Frackowiak, 1993; Salmon,
Van der Linden, Collette, & Delfiore, 1996; Smith
& Jonides, 1997; Sweet et al., 2008). For example,
increased demands on subvocal articulatory
rehearsal by increasing interstimulus intervals (ISI)
has been associated with increased brain activation
in Broca’s area, while increased storage demands
resulted in increased posterior parietal cortex activ-
ity (Barch et al., 1997; Smith & Jonides, 1997).
Sweet et al. (2008) examined phonological similar-
ity effects using functional magnetic resonance
imaging during a 2-Back VWM paradigm. In con-
trast to the standard 2-Back condition, the phono-
logical similarity condition consisting of rhyming
stimuli yielded a significant positive relationship
between performance accuracy and brain response
in inferior parietal regions.
Several lines of evidence suggest that subvocal
articulatory rehearsal may be specifically affected
in MS. Firstly, experimental protocols using articu-
latory suppression during verbal serial recall tasks
have discovered exaggerated word length effects
(Litvan, Grafman, Vendrell, & Martinez, 1988a;
Litvan et al., 1988b; Rao et al., 1993). This suggests
that poor VWM performance demonstrated by MS
patients could be attributed, at least in part, to a
rehearsal dysfunction within the phonological loop.
Secondly, a study of VWM components in MS
patients found evoked potential phase shifts in left
anterior brain regions that coincided with pronun-
ciation rate and pronunciation errors (Ruchkin et al.,
1994). The authors interpreted this abnormality as
consistent with articulatory rehearsal difficulty.
Thirdly, the typical pattern of verbal memory
impairment in MS includes decreased explicit mem-
ory encoding and retrieval, but unimpaired storage
(see Rao, 1986). This pattern raises the possibility
that subvocal articulatory rehearsal components
might disproportionately influence VWM perform-
ance in MS. Active subvocal rehearsal and STM
buffering components of VWM have not yet been
studied simultaneously or in relationship to ubiqui-
tous processing speed impairments, fatigue, and
depression that may underlie MS effects on VWM.
Figure 1. Schematic of the phonological loop.
Downloaded By: [Canadian Research Knowledge Network] At: 18:25 20 April 2010

REHEARSAL & WORKING MEMORY IN MS 3
Information processing speed, a measure of the
efficiency of cognitive function, is assessed using
timed tests that typically challenge relatively simple
cognitive operations (e.g., Symbol Digit Modalities
Test; SDMT). However, the efficiency of cognitive
function has also been assessed using timed tests
that require more complex operations (e.g., Paced
Auditory Serial Addition Test; PASAT). Slowing
on such measures is among the most prominent
cognitive symptoms of MS. Many studies have
reported the co-occurrence of VWM and process-
ing speed deficits (e.g., Archibald & Fisk, 2000;
Beatty, Goodkin, Monson, & Beatty, 1989; Bernardin
et al., 1993; Grigsby, Ayarbe, Kravcisin, &
Busenbark, 1994; Paul, Beatty, Schneider, Blanco,
& Hames, 1998; Rao et al., 1991, 1993). There is
evidence that processing speed may differentially
affect components of VWM, whereby, active sub-
vocal rehearsal may be attenuated, while STM
buffering might not (D’Esposito et al., 1996; Litvan
et al., 1988a, 1988b; Rao et al., 1993; Ruchkin et al.,
1994).
This study was designed to examine component
processes of VWM, a cognitive function frequently
impaired among MS patients. Our specific aims
were to determine if rehearsal and buffering com-
ponents are differentially affected in MS, and to
examine the relationship between information pro-
cessing speed and these VWM components. We
used a modified n-Back paradigm to test the effects
of subvocal articulatory rehearsal (see Barch et al.,
1997) and phonological similarity (see Sweet et al.,
2008) challenges on VWM performance. We
hypothesized that VWM rehearsal performance
would be worse than VWM STM buffering per-
formance, and that lower performance during the
VWM rehearsal challenge would be specifically
related to information processing speed. Such find-
ings would have potential clinical significance, as
evidence that processing speed underlies other
cognitive impairments associated with MS would
suggest that primary focus should be directed at
this slowing and its relationship to other key MS
symptoms (e.g., depression).
METHOD
Participants
Participants were fifteen MS patients who were
recruited from a university affiliated neurology
clinic. Patients were recruited based on medical
records that indicated a diagnosis of MS. Diagnosis
was confirmed by a licensed and board certified neu-
rologist using the McDonald criteria (McDonald
et al., 2001). The sample consisted of three men and
12 women with a mean age of 41.93 and mean edu-
cation of 14.29. Three patients exhibited a second-
ary progressive course and 12 exhibited a relapsing-
remitting course. Mean disability assessed on the
Multiple Sclerosis Functional Composite (MSFC)
was in the moderate range (mean z= –2.21, median
z = –1.85, SD = 2.78; Fox, Lee, & Rudick, 2007).
Exclusion criteria included a diagnosis of a current
psychiatric disorder (e.g., depressive episode, active
substance dependence) or neurological disorder
(excluding MS), a history of psychiatric hospitaliza-
tion, severe visual impairment (corrected vision
worse than 20/70), or failure to perform above
chance on n-Back practice items. An interview was
conducted to take a brief psychiatric and neurologi-
cal history, and screen for grossly impaired visual
acuity. The study was conducted in compliance
with the Helsinki Declaration. All participants
signed an Institutional Review Board approved
informed consent form before any study procedures
were administered.
Neuropsychological battery
A neuropsychological battery assessing VWM,
processing speed, mood, fatigue, and disability was
administered in the same order to each participant.
Cognitive measures included the 2-Back task, the
Paced Auditory Serial Addition Test (PASAT;
Gronwall, 1977) and the oral version of the Sym-
bol Digit Modalities Test (SDMT; Smith, 1973).
The MSFC was used to assess disability (Cutter et
al., 1999; Kalkers et al., 2000) and the Chicago
Multiscale Depression Inventory (CMDI) was
administered to assess depressive symptoms
(Nyenhuis et al., 1998). All tests were administered
following standard clinical procedures and super-
vised by a licensed and board certified clinical neu-
ropsychologist. All participants were assessed in
the morning to avoid the possibility of higher levels
of fatigue later in the day. Measures are described
in more detail below.
The oral version of the SDMT was utilized as a
measure of information processing speed. This task
requires rapid comparisons of simple geometric
designs. The number of correct SDMT matches at
90 s was used in the analyses. The oral version of the
SDMT has been used as a measure of information
Downloaded By: [Canadian Research Knowledge Network] At: 18:25 20 April 2010

4 SWEET ET AL.
processing speed in several studies of MS (Beatty
et al., 1989; Rao et al., 1991) to minimize the influ-
ence of motor impairments common to MS.
The n-Back paradigm, a cognitive task widely
used in functional neuroimaging research (Braver
et al., 1997; Smith & Jonides, 1997; Sweet et al.,
2008), provides a method for assessing overall
VWM and offers design flexibility that allows
experimental manipulation of component proc-
esses. Participants performed a 2-Back VWM task
with modified nested conditions designed to
increase subvocal articulatory rehearsal (via inter-
stimulus interval) and STM buffering demands
(via phonological similarity). During the standard
2-Back a series of 15 consonants is presented visu-
ally. The participant makes a yes/no response fol-
lowing each consonant – whether it is the same as,
or different from, the consonant presented two
earlier. Thus, a target consonant is separated from
the reference consonant by one other consonant.
The 2-Back task requires encoding, maintaining,
buffering, matching, and updating phonemes.
The 2-Back task designed for this study had
three conditions. The three conditions included the
standard presentation as well as the manipulations
of subvocal articulatory rehearsal (Barch et al.,
1997) and phoneme similarity. The nested condi-
tions were as follows: the rehearsal condition
included increased interstimulus intervals (ISI)
from 2500 ms (used in the other conditions) to
5000 ms in order to increase rehearsal demands
(see Barch et al., 1997); and the STM buffering con-
dition included rhyming consonants (e.g., C, V, Z,
G . . .; see Sweet et al., 2008) as compared to the
different-sounding stimuli in the standard and
rehearsal conditions (e.g., T, K, X, Q . . .).
Each condition was presented three times in
three randomized cycles that appeared as a single
blocked task with one set of instructions. The 2-
Back task was presented using E-prime software
installed on a laptop computer with a 15” screen. A
response box was utilized to collect yes/no
responses. In total, the 2-Back paradigm was
composed of nine 15-consonant lists requiring 9
min to complete. Each 15-consonant list con-
tained five target consonants (i.e., ‘yes’ is the cor-
rect response).
The PASAT was administered to assess both
information processing speed and VWM. The
PASAT requires addition of two single-digit num-
bers, reporting the sum, and retaining the later of
the two numbers for addition to the next number
presented. Numbers are presented at rates of 2 and
3 s during two subtests. The PASAT has been used
as a measure of VWM and information processing
speed in several studies of MS (Beatty et al., 1989;
Rao et al., 1991), and it is a primary component
used to calculate MSFC disability scale scores
(Cutter et al., 1999). The other components of the
MSFC, nine-hole pegs and 25’ walk, were also
administered to derive the disability index (Cutter
et al., 1999; Fox et al., 2007; Mathiowetz, Weber,
Kashman, & Volland, 1985).
The remaining battery consisted of the CMDI, a
depression inventory designed to assess mood,
evaluative, and vegetative components of depres-
sion; and a novel measure of fatigue. The fatigue
score was calculated upon the completion of the
neuropsychological battery by summing the partic-
ipants’ subjective rating (on a scale of 1–5) on the
following items based on their current state (poor,
fair, average, good, or excellent): general energy
level, muscle strength, concentration/ memory
level, ability to finish task, ability to solve prob-
lems, overall well-being, feeling of sleepiness, qual-
ity of last night’s sleep, and feeling of fatigue (in
order of presentation).
RESULTS
Means and normed z-scores are presented in Table 1.
The mean MSFC disability score (z = –2.21) indi-
cates an average level of moderate disability
among this patient sample. Consistent with MS-
related disability, mean test scores in the impaired
range (z < –1.00) were observed on the nine-hole
pegs, 25’ walk, and the vegetative scale of the
CMDI. In addition, mean scores on the remaining
tests were all below average (i.e., 25th percentile or
lower).
Paired contrasts of 2-Back conditions revealed
no significant differences in accuracy (two-tailed
t(14) > 1.669, p > .117). Table 2 reports 2-Back
relationships to other variables in the battery.
Although it was predicted that information pro-
cessing speed assessed using the SDMT would be
related to 2-Back performance, this was only
observed during the rehearsal 2-Back, MSFC, and
all three CMDI subscales (Mood, r = .694, p =
.006; Evaluative, r = .772, p = .001; Vegetative, r =
.669, p = .009). Higher scores on the PASAT and
SDMT were significantly related to better
rehearsal 2-Back accuracy and less disability (i.e.,
higher MSFC z-scores). Likewise, more endorsed
depressive symptoms on the mood and vegetative
Downloaded By: [Canadian Research Knowledge Network] At: 18:25 20 April 2010

REHEARSAL & WORKING MEMORY IN MS 5
TABLE 1
Mean of raw scores and individually normed z-scores
Mean SD
Age 41.93 9.97
Education 14.29 1.98
Years Since Diagnosis 5.86 4.05
Standard 2-Back (percent correct) 71.32 19.99
Similarity 2-Back (percent correct) 74.18 13.21
Rehearsal 2-Back (percent correct) 79.37 17.54
Fatigue (raw) 28.54 9.11
Mean z SD Percentile
Multiple Sclerosis Functional Composite −2.21 2.78 ** 1.74
Paced Auditory Serial Addition Test 3” −0.75 1.46 * 22.66
Paced Auditory Serial Addition Test 2” −0.78 1.57 * 21.77
9 Hole Pegs (dominant) −1.24 1.72 ** 10.75
9 Hole Pegs (nondominant) −0.67 1.61 25.14
25 Foot Walk −4.91 6.30 *** <1.00
Oral Symbol Digit Modalities Test −0.74 1.19 * 22.97
CMDI Mood −0.89 0.14 *** 18.67
CMDI Evaluative −0.87 0.23 *** 19.22
CMDI Vegetative −1.44 0.15 *** 7.49
Notes: *p < .05, **p < .01, ***p < .001 (significantly lower than the population
mean reported in test norms). CMDI, Chicago Multiscale Depression
Inventory.
TABLE 2
Relationship of 2-back performance to other measures
MSFC Standard Similarity Rehearsal
MSFC r−.424 .652 .729
p−.131 .012 .003
PASAT 2” r.826 .492 .613 .680
p<.001 .062 .015 .005
PASAT 3” r.661 .275 .240 .618
p.010 .324 .389 .015
9 Hole Pegs (d) r.834 .096 .538 .504
p<.001 .745 .047 .066
9 Hole Pegs (nd) r−.688 −.261 −.450 −.205
p.007 .368 .106 .483
25 Foot Walk r.891 .458 .553 .620
p<.001 .116 .050 .024
Oral SDMT r.579 .131 .132 .611
p.030 .655 .652 .020
CMDI Mood r.754 .116 .160 .632
p.002 .694 .584 .015
CMDI Vegetative r.556 .161 .253 .599
p.039 .582 .382 .024
CMDI Evaluative r.479 .057 .177 .363
p.083 .846 .546 .202
Notes: PASAT, Paced Auditory Serial Addition Test; CMDI, Chicago Multiscale Depression Inventory; MSFC,
Multiple Sclerosis Functional Composite; SDMT, Symbol Digit Modalities Test; d, dominant hand; nd, nondomi-
nant hand.
Downloaded By: [Canadian Research Knowledge Network] At: 18:25 20 April 2010

6 SWEET ET AL.
scales of the CMDI (i.e., lower z-scores) were
significantly related to lower rehearsal 2-Back
accuracy and greater MSFC disability (i.e., lower
z-scores), but not to performance during the other
2-Back conditions. Neither MSFC nor any of the
2-Back measures were significantly related to the
evaluative scale of the CMDI, level of education,
or post-assessment fatigue ratings. Overall, the pat-
tern of significant relationships between rehearsal
2-Back and the remaining tests resembled the pat-
terns of the MSFC disability index.
DISCUSSION
The aim of this study was to determine if compo-
nents of VWM (i.e., subvocal articulatory
rehearsal and STM buffering) were differentially
affected in MS and to examine the role of informa-
tion processing speed using a 2-Back task with
three nested conditions (rehearsal, buffering, and
standard). We did not find the expected worse per-
formance during subvocal articulatory rehearsal
processes compared to STM buffering challenge;
however, we identified differential relationships
between these components and information
processing speed and depressive symptoms. Lower
performance during the subvocal articulatory
rehearsal challenge was significantly related to
slowed processing and greater symptoms of
depression. Overall, performance on the rehearsal
component of the VWM task exhibited a very sim-
ilar pattern of significant relationships to the
remaining battery as the MSFC disability index,
while the other 2-Back conditions did not.
These findings are consistent with previous liter-
ature indicating that slowed information process-
ing and depressive symptoms are associated with
the performance of other cognitive functions,
including working memory (e.g., Arnett et al.,
2001; Chiaravalloti & DeLuca, 2008; Diamond
et al., 2008; Landro et al., 2004). Our results extend
these findings to link information processing speed
and depressive symptoms specifically to the subvo-
cal articulatory rehearsal component of VWM.
The positive correlation between increased
VWM rehearsal performance and faster processing
speed likely reflects a greater need for efficient pro-
cessing during a challenge that was expected to be
more difficult for MS patients (relative to the STM
buffering or standard 2-Back challenges). If the
conditions were equally difficult, it might be
assumed that the slower pace of the rehearsal
condition would not require rapid processing
speed, resulting in better performance. However,
performance did not significantly differ, and the
added time in this condition appears to have
resulted in a paradoxical need for more efficient
rapid processing.
Since the addition of more time seems to be par-
adoxical to the apparent need for faster processing,
the explanation may be related to the increased
challenge to the attentional (e.g., sustained atten-
tion and attentional capacity; Cohen, 1993) and
executive systems (e.g., central executive; Baddely
& Hitch, 1974). It is likely that the coordination of
the STM buffer, rehearsal, updating etc., over
longer periods of time before the next stimulus
appears, requires more efficient attentional and
executive processing than the standard or buffering
2-Back challenges. Therefore, more time to
respond may only result in improved performance
if the participant is motivated and efficient in
attentional and executive functioning.
These conclusions are supported by the results
from another more complex measure of informa-
tion processing speed, the PASAT. Often also clas-
sified as a VWM task, the PASAT is a timed test
that also requires rapid cognitive processing and
executive processing. PASAT 3” performance was
significantly positively correlated with perform-
ance on the rehearsal 2-Back and all three sub-
scales of the CMDI (Mood, r = .553, p = .040;
Evaluative, r = .660, p = .010; Vegetative, r = .733,
p = .003). This offers convergent evidence that the
rehearsal condition was associated to a greater
extent with processing speed and attentional/exec-
utive functions in comparison to standard 2-Back
performance.
Although this pattern of results also suggests a
strong association between depressive symptoms
and processing speed, this study was not designed
to determine cause and effect, or to identify mecha-
nisms shared by both that affect VWM. Slow pro-
cessing speed during high VWM and rehearsal
demands may be due to low effort and motivation
secondary to mood. In addition to strong correla-
tions with measures of processing speed, greater
endorsement of depressive symptoms on the Mood
subscale of the CMDI was significantly associated
with slower performance on the 25’ walk time (r =
.638, p = .019) and dominant hand nine-hole pegs
(r = .765, p = .001), suggesting that lower perform-
ance on these tasks might be influenced by emo-
tional state. Alternatively, cognitive deficits
associated with MS, such as slowed processing
Downloaded By: [Canadian Research Knowledge Network] At: 18:25 20 April 2010

REHEARSAL & WORKING MEMORY IN MS 7
speed, may lead to depressive symptoms. An
advantage of the use of the CMDI is that we may
conclude that mood and vegetative, rather than
evaluative, depressive symptoms are involved.
The finding that slowed processing and depres-
sive symptoms differentially influence effortful
cognitive challenges, such as subvocal articulatory
rehearsal during VWM, suggests that a focus of
cognitive assessments should be information pro-
cessing speed and its relationship to other common
MS symptoms (e.g., mood and vegetative symp-
toms of depression). Relative to a standard admin-
istration of this widely used VWM challenge,
which typically use ISI of 2500 ms, simply increas-
ing the ISI increases its sensitivity to key indices of
MS impairment, including depression, processing
speed, and disability. A possible explanation for
this finding is that the increased rehearsal time may
be most useful for those who are better able to
coordinate and maintain effort and processing
speed. Therefore, patients with symptoms of
depression and cognitive slowing may be most sus-
ceptible to inefficient processing during this type of
challenge.
Limitations to this study include a small sample
size and the absence of a healthy control group.
Therefore further research is needed to assess the
generalizability and specificity to larger samples of
MS patients. Our 2-Back conditions allowed exam-
ination of only two components of VWM. Future
studies are needed to determine the role of the other
components, such as executive functions and atten-
tion. Although we conclude that increased
rehearsal time was associated with processing speed
and motivation, several other factors may have
influenced performance, such as increased atten-
tional and executive demands. More sophisticated
methods designed to test mediation models would
facilitate a better understanding of the relationship
between VWM components and processing speed,
and important associated factors such as depres-
sion, effort, attention, and executive functions.
Original manuscript received 4 August 2009
Revised manuscript accepted 14 December 2009
First published online
REFERENCES
Amato, M. P., Zipoli, V., & Portaccio, E. (2006). Multi-
ple sclerosis-related cognitive changes: A review of
cross-sectional and longitudinal studies. Journal of
the Neurological Sciences, 245(1–2), 41–46.
Archibald, C. J., & Fisk, J. D. (2000). Information pro-
cessing efficiency in patients with multiple sclerosis.
Journal of Clinical and Experimental Neuropsychol-
ogy, 22(5), 686–701.
Baddeley, A. (1966). Short-term memory for word
sequences as a function of acoustic, semantic, and
formal similarity. Quarterly Journal of Experimental
Psychology, 18(362–365).
Baddeley, A. D., & Hitch, G. (1974). Working memory.
In G. H. Bower (Ed.), The psychology of learning and
motivation (Vol. 8, pp. 47–90). New York; Academic
Press.
Baddeley, A., & Lieberman, K. (1980). Spatial working
memory. In R. Nickerson (Ed.), Attention and per-
formance (Vol. VIII, pp. 521–539). Hillsdale, NJ:
Lawrence Erlbaum Associates.
Baddeley, A., Thompson, N., & Buchanan, M. (1975).
Word length and the structure of short-term memory.
Journal of Verbal Learning and Verbal Behavior, 14,
575–589.
Barch, D., Braver, T., Nystron, L., Forman, S., Noll, T.,
& Cohen, J. (1997). Dissociating working memory
from effort in human prefrontal cortex. Neuropsycho-
logia, 35, 1373–1380.
Beatty, W. W., Goodkin, D. E., Monson, N., & Beatty,
P. A. (1989). Cognitive disturbances in patients with
relapsing remitting multiple sclerosis. Archives of
Neurology, 46, 1113.
Bernardin, L., Rao, S. M., Lucchetta, T. L.,
Ellington, L., Unverzagt, F., Swanson, S., et al.
(1993). A prospective, long-term, longitudinal
study of cognitive dysfunction in multiple sclero-
sis. Journal of Clinical and Experimental Neuropsy-
chology, 15, 17.
Braver, T. S., Cohen, J. D., Nystrom, L. E., Jonides, J.,
Smith, E. E., & Noll, D. C. (1997). A parametric
study of prefrontal cortex involvement in human
working memory. Neuroimage, 5(1), 49–62.
Calabrese, P. (2006). Neuropsychology of multiple scle-
rosis – an overview. Journal of Neurology, 253, (Suppl
1), i10–i15.
Chiaravalloti, N. D., & DeLuca, J. (2008). Cognitive
impairment in multiple sclerosis. Lancet Neurology,
7(12), 1139–1151.
Cowan, N. (1988). Evolving conceptions of memory
storage, selective attention, and their mutual con-
straints within the human information-processing
system. Psychological Bulletin, 104(2), 163–191.
Cohen R. A. (1993). Neuropsychology of attention. New
York, Plenum.
Cutter, G. R., Baier, M. L., Rudick, R. A., Cookfair, D.
L., Fischer, J. S., Petkau, J., et al. (1999). Develop-
ment of a multiple sclerosis functional composite as a
clinical trial outcome measure. Brain: A Journal of
Neurology, 122(Pt 5), 871–882.
D’Esposito, M., Monish, K., Thompson, A., Robinson,
K., Armstrong, C., Grossman, M., et al. (1996).
Working memory impairment in multiple sclerosis:
Evidence from a dual-task paradigm. Neuropsychol-
ogy, 10, 51–56.
Diamond, B. J., DeLuca, J., Kim, H., & Kelley, S. M.
(1997). The question of disproportionate impair-
ments in visual and auditory information processing
Downloaded By: [Canadian Research Knowledge Network] At: 18:25 20 April 2010

8 SWEET ET AL.
in multiple sclerosis. Journal of Clinical and Experi-
mental Neuropsychology, 19(1), 34–42.
Ericsson, K. A., & Kintsch, W. (1995). Long-term work-
ing memory. Psychological Review, 102(2), 211–245.
Fox, R. J., Lee, J. C., & Rudick, R. A. (2007). Optimal
reference population for the multiple sclerosis func-
tional composite. Multiple Sclerosis, 13(7), 909–914.
Grigsby, J., Ayarbe, S. D., Kravcisin, N., & Busenbark,
D. (1994). Working memory impairment among per-
sons with chronic progressive multiple sclerosis. Jour-
nal of Neurology, 241(3), 125–131.
Gronwall, D. M. (1977). Paced auditory serial-addition
task: A measure of recovery from concussion. Percep-
tual and Motor Skills, 44(2), 367–373.
Hillary, F. G., Chiaravalloti, N. D., Ricker, J. H., Stef-
fener, J., Bly, B. M., Lange, G., et al. (2003). An
investigation of working memory rehearsal in multi-
ple sclerosis using fMRI. Journal of Clinical and
Experimental Neuropsychology, 25(7), 965–978.
Kalkers, N. F., de Groot, V., Lazeron, R. H. C.,
Killestein, J., Adèr, H. J., Barkhof, F., et al. (2000).
MS functional composite: Relation to disease pheno-
type and disability strata. Neurology, 54(6), 1233–1239.
Lengenfelder, J., Bryant, D., Diamond, B. J., Kalmar, J.
H., Moore, N. B., & DeLuca, J. (2006). Processing
speed interacts with working memory efficiency in
multiple sclerosis. Archives of Clinical Neuropsychol-
ogy, 21(3), 229–238.
Litvan, I., Grafman, J., Vendrell, P., & Martinez, J. M.
(1988a). Slowed information processing in multiple
sclerosis. Archives of Neurology, 45(3), 281–285.
Litvan, I., Grafman, J., Vendrell, P., Martinez, J. M.,
Junque, C., Vendrell, J. M., et al. (1988b). Multiple
memory deficits in patients with multiple sclerosis.
Exploring the working memory system. Archives of
Neurology, 45(6), 607–610.
Logie, R. H. (1986). Visuo-spatial processing in working
memory. The Quarterly Journal of Experimental Psy-
chology A: Human Experimental Psychology, 38A(2),
229–247.
Logie, R. H., Zucco, G. M., & Baddeley, A. D. (1990).
Interference with visual short-term memory. Acta
Psychologica, 75(1), 55–74.
Mathiowetz, V., Weber, K., Kashman, N., & Volland,
G. (1985). Adult norms for the Nine Hole Peg Test of
finger dexterity. Occupational Therapy Journal of
Research, 5(1), 24–38.
McDonald, W. I., Compston, A., Edan, G., et al.,
(2001). Recommended diagnostic criteria for multiple
sclerosis: Guidelines from the international panel on
the diagnosis of multiple sclerosis. Annals of Neurol-
ogy, 50(1), 121–127.
Nyenhuis, D. L., Luchetta, T., Yamamoto, C., Terrien,
A., Bernardin, L., Rao, S. M., et al. (1998). The
development, standardization, and initial validation
of the Chicago Multiscale Depression Inventory.
Journal of Personality Assessment, 70(2), 386–401.
Paul, R. H., Beatty, W. W., Schneider, R., Blanco, C., &
Hames, K. (1998). Impairments of attention in indi-
viduals with multiple sclerosis. Multiple Sclerosis,
4(5), 433–439.
Rao, S. M. (1986). Neuropsychology of multiple sclero-
sis: A critical review. Journal of Clinical and Experi-
mental Neuropsychology, 8(5), 503–542.
Rao, S. M. (1990). Neurobehavioral aspects of multiple
sclerosis. New York: Oxford University Press.
Rao, S. M., Leo, G. J., Bernardin, L., & Unverzagt, F.
(1991). Cognitive dysfunction in multiple sclerosis. I.
Frequency, patterns, and prediction. Neurology,
41(5), 685–691.
Rao, S. M., Grafman, J., DiGiulio, D., Mittenberg, W.,
Bernardin, L., Leo, G. J., et al. (1993). Memory dys-
function in multiple sclerosis: Its relation to working
memory, semantic encoding, and implicit learning.
Neuropsychology, 7, 364–374.
Ruchkin, D., Grafman, J., Krauss, G., Johnson, R.,
Canoune, H., & Ritter, W. (1994). Event-related
brain potential evidence for a verbal working mem-
ory deficit in multiple sclerosis. Brain, 117, 289–305.
Schneider, W., Detweiler, M., & Bower, G. H. (1987). A
connectionist/control architecture for working mem-
ory. In: The psychology of learning and motivation:
Advances in research and theory (Vol. 21, pp. 53–119).
San Diego, CA: Academic Press.
Smith, A. (1973). Symbol Digit Modalities Test. Los
Angeles, CA: Western Psychological Services.
Smith, E. E., & Jonides, J. (1997). Working memory: A
view from neuroimaging. Cognitive Psychology,
33(1), 5–42.
Sweet, L. H., Rao, S. M., Primeau, M., Mayer, A. R.,
& Cohen, R. A. (2004). Functional magnetic reso-
nance imaging of working memory among multiple
sclerosis patients. Journal of Neuroimaging, 14(2),
150–157.
Sweet, L. H., Rao, S. M., Primeau, M., Durgerian, S., &
Cohen, R. A. (2006). Functional magnetic resonance
imaging response to increased verbal working mem-
ory demands among patients with multiple sclerosis.
Human Brain Mapping, 27(1), 28–36.
Sweet, L. H., Paskavitz, J. F., Haley, A. P., Gunstad,
J. J., Mulligan, R. C., Nyalakanti, P. K., et al.
(2008). Imaging phonological similarity effects on
verbal working memory. Neuropsychologia, 46(4),
1114–1123.
Downloaded By: [Canadian Research Knowledge Network] At: 18:25 20 April 2010