Normalizing Memory Recall in Fibromyalgia
With Rehearsal: A Distraction-Counteracting
FRANK LEAVITT AND ROBERT S. KATZ
Objective. To examine the impact of distraction on the retention of rehearsed information in patients with fibromyalgia
Methods. Data refer to the neurocognitive examination of 134 patients (91 with FMS and 43 control subjects) presenting
with memory loss. Four neurocognitive measures free of distraction, along with 2 measures with added distraction, were
completed. Differences in the retention of rehearsed and unrehearsed information with a source of distraction present
Results. Patients with FMS showed normal cognitive functioning on verbal memory tests free of distraction. Adding a
source of distraction caused unrefreshed information to be lost at a disproportionate rate in patients with FMS. Over 87%
of patients with FMS scored in the impaired range on a task of unrehearsed verbal memory. Adding a source of
distraction to well-rehearsed information produced a normal rate of recall in FMS.
Conclusion. Rehearsal mechanisms are intact in patients with FMS and play beneficial roles in managing interference
from a source of distraction. In the absence of rehearsal, a source of distraction added to unrefreshed information signals
a remarkable level of cognitive deficit in FMS that goes undetected by conventionally relied-upon neurocognitive
measures. We present a theory to promote understanding of the cognitive deficit of people with FMS based on reduced
speed of lexical activation and poor recall after distraction.
Cognitive distraction plays a critical role in short-term
memory loss in people with fibromyalgia syndrome (FMS)
(1). In recent research, exposure to a source of distraction
stood out as the central cause of rapid forgetting in people
with FMS. Adding a source of distraction to a standard
memory task eradicated verbal information memory at a
disproportionately high rate in people with FMS. In fact,
people with FMS lost new verbal information at a rate that
was 44% greater than an age-matched control group pre-
senting with memory problems, and was almost 3 times
greater than the normative sample (1). In the absence of
distraction, short-term verbal memory is normal in people
with FMS. Very little is known about why the costs of
distraction are so high in people with FMS.
People with FMS show a prominent disturbance in cog-
nitive performance on the Auditory Consonant Trigram
(ACT) test (1). Properties of the ACT permit an evaluation
of severe forgetfulness in FMS that is rooted in the preven-
tion of rehearsal of information by competition from a
source of distraction (2). The ACT paradigm entails re-
membering a 3-item list (a trigram) of consonant letters
after attending to a distracting, unrelated task. Between
hearing the trigram and recall, subjects count backwards
by threes (the distracter task) from a predetermined num-
ber for intervals of 9, 18, or 36 seconds. Diverting attention
to an unrelated cognitive operation involving serial sub-
traction prevents rehearsal of the 3 consonant letters.
When an unrelated cognitive operation (a distraction) dis-
rupts the encoding of auditory information, the unre-
hearsed information is forgotten almost twice as fast by
individuals with FMS (1).
Why a source of distraction disproportionately hinders
later recall of unrehearsed information in people with
FMS is unclear. A natural question is whether interference
created by a source of distraction impacts differently when
the information to be remembered has been rehearsed. The
objectives of this study were to replicate previous cogni-
tive findings relating cognitive impairment to distraction
while recalling unrehearsed information in people with
FMS, and to determine whether distraction impairs the
retention of rehearsed information in a similar fashion.
Frank Leavitt, PhD, Robert S. Katz, MD: Rush Medical
College, Chicago, Illinois.
Address correspondence to Frank Leavitt, PhD, 513 Illi-
nois Road, Wilmette, IL 60091. E-mail: frankleavitt@
Submitted for publication December 25, 2008; accepted in
revised form March 3, 2009.
Arthritis & Rheumatism (Arthritis Care & Research)
Vol. 61, No. 6, June 15, 2009, pp 740–744
© 2009, American College of Rheumatology
PATIENTS AND METHODS
Patients. Archival data were drawn from the records of
91 women with FMS who consecutively presented for the
evaluation of memory problems by neurocognitive exam-
ination on the basis of clinical referral. The women in the
sample fulfilled the American College of Rheumatology
criteria for the classification of fibromyalgia (3). The diag-
nosis was established by a rheumatologist and was based
on widespread pain in combination with tenderness of
?11 of 18 specific tender point sites (3). The control sam-
ple consisted of 43 women without FMS presenting with
memory problems, who were examined during the same
time period. The control sample was similar to the FMS
group in terms of age and education (mean ? SD age was
46.1 ? 9.9 years and 45.2 ? 11.1 years, respectively, and
the mean ? SD education was 14.8 ? 2.1 years and 14.8 ?
2.0 years, respectively). Among the controls, 5 reported a
history of seizures/epilepsy, 4 reported a history of head
trauma or transient ischemic attack, and 3 reported hypo-
thyroidism, substance abuse, or obstructive sleep apnea.
The inclusion criteria were that all participants should be
age 18–65 years, be female, and be fluent in English. The
exclusion criteria were a history of drug or alcohol abuse,
psychiatric treatment in the past 3 years, auditory impair-
ment that might interfere with cognitive testing, or a lack
of fluency in English.
Patients with FMS reported a median length of memory
problems of 20 months (range 1 month to 18 years); 41%
percent reported memory problems ?1 year in duration.
By comparison, controls reported a median length of mem-
ory problems of 18 months (range 1 month to 22 years);
36% reported memory problems for ?1 year in duration.
Differences were not statistically significant (P ? 0.05).
Measures. The Logical Memory and Paired Associates
subtests of the Wechsler Memory Scale, Third Edition
assess memory performance free of stimulus competition
at encoding (4). The Logical Memory subtest entails re-
membering 2 story paragraphs, with recall tested immedi-
ately and then after a 30-minute delay. Each paragraph
contains 25 segments, and the score is the number of
The Paired Associates subtest assesses formation of as-
sociations between 8 word pairs, some unrelated (e.g.,
truck/arrow), over 4 learning trials with cued recall tested
after each trial (4). The scores are the sum of word pairs
recalled over the 4 trials in addition to a 30-minute de-
layed cued recall trial.
The ACT entails remembering 3-item lists (trigrams) of
consonant letters after performing a distracting task that
interferes with the input and encoding of new information
(the trigram) (2). Immediately after hearing the trigram,
subjects count backwards by threes (the distracter task)
from a number for intervals of 9, 18, or 36 seconds, fol-
lowed by free recall of the trigram. Five trials at each
distracter interval are administered, along with a no-dis-
traction condition (0 seconds). The maximum score for
each interval is 15.
The Rey Auditory Verbal Learning Test (RAVLT) (5) is a
word list–learning task consisting of 5 verbal presentations
of a 15-word list of nouns (list 1) with free recall following
each presentation. Thus, in this part, subjects rehearse the
list information 5 times. Next, a distracter list of 15 nouns
(list 2) is read, followed by free recall. On the postdistrac-
tion trial, the subject is asked to recall the first list (list 1).
Four scores are reported: a summary learning measure (the
total words recalled on the first 5 presentations), a predis-
traction measure (the number of words recalled on trial 5),
a postdistraction measure (the number of words recalled
from list 1 after distraction), and an interference measure
(the difference between scores on the predistraction and
The vocabulary subtest of the Wechsler Adult Intelli-
gence Scale, Revised served as an estimate of general in-
telligence (4). Participants provided definitions of words
presented in the order of increasing difficulty.
The Beck Depression Inventory II, a 21-item, 4-point
self-report scale, was used to measure symptoms of de-
Statistical analysis. Differences in memory perfor-
mance between the 2 groups were analyzed. Mean ? SD
was used to define the demographic and clinical features
of the 2 groups. To assess group differences on the neuro-
cognitive measures, we used a multivariate analysis of
covariance with the 5 neurocognitive measures as the de-
pendent variables; the clinical groups as the between-
subject variable; and age, education, vocabulary knowl-
edge, and depression as covariates. With a significant
overall multivariate effect, univariate effects were exam-
ined to assess differences between the groups on individ-
ual cognitive tests. Categorical variables were compared
with Pearson’s chi-square test. For RAVLT continuous
variables, comparisons were made using the Student’s
2-tailed t-test. P values less than 0.05 were considered
The cognitive performance of the 2 groups along with the
age-appropriate normative values are shown in Table 1.
Test scores are expressed in scale scores rather than raw
scores to enable direct comparison of the level of perfor-
mance across different tests. The mean ? SD standardized
normative mean score for each test is 10 ? 3. On cognitive
measures free of distraction, patients with FMS performed
essentially at the normative mean. By contrast, they per-
formed significantly below the established normative
value on the measure with a source of distraction (ACT
score ?2.5 SDs below the normative mean).
Patients with FMS recalled information on distraction-
free measures better than controls on both Logical Memory
paragraphs and the Paired Associates. On both measures,
patients with FMS recalled more information immediately
after exposure, as well as following a 30-minute delay. By
contrast, on a measure involving unrehearsed information
and a source of distraction, the patients with FMS recalled
significantly less information on the ACT than the controls
Effects of Distraction and Rehearsal on Memory in FMS 741
The proportion of scores falling in the impaired range
for each group on each neurocognitive test is shown in the
right panel of Table 1. Impairment was defined as scores
?1.67 SDs below the age-appropriate normative mean (be-
low the fifth percentile). These data show that the cogni-
tive performance of patients with FMS was disproportion-
ately impaired by distraction. Collectively, 87.9% of
patients with FMS scored in the impaired range on the
ACT. This proportion was significantly higher than the
proportion found in the control sample. By contrast, the
great majority of patients with FMS were functioning at
normal levels on the 4 measures that were free of distrac-
tion. Impairment on these measures ranged from a low of
7.7% on Logical Memory delayed recall to a high of 12.1%
on Paired Associates.
Differences in the groups’ recall of the RAVLT 15-item
word list are presented in Table 2. On average, total recall
across 5 presentations was 5 words greater in the FMS
group. Recall at the fifth presentation trial (predistraction
trial) was also significantly higher in the FMS group. The
difference between trial V and trial VI represents the ef-
fects of having been distracted by the introduction of a
second word list (list 2). Both groups recalled fewer words
after the presentation of a distracter word list, but the
mean postdistraction loss was equivalent in the 2 groups.
The postdistraction information loss was small, ?2 words.
No significant differences from reference norms (mean ?
SD score 1.7 ? 2.0) were noted for the postdistraction trial,
indicating that a source of distraction did not greatly in-
terfere with the recall of information rehearsed over 5
Rehearsal mechanisms are intact in patients with FMS,
and play a beneficial role in managing a source of distrac-
tion. The retention of repeated information interrupted by
a source of distraction is normal in patients with FMS,
indicating that distraction does not disrupt memory for
well-rehearsed information in patients with FMS. Patients
with FMS forget well-rehearsed information at a rate that
is very similar to a healthy population following a distrac-
tion. Data of this nature imply that rehearsal counteracts
interference from a source of distraction. One way to sup-
port memory in patients with FMS may be added re-
Our results confirm previous findings and provide com-
pelling evidence that deficits in the management of dis-
traction are an important feature of cognitive dysfunction
in FMS (1). As measured by the ACT, which provides a
distraction task that prevents rehearsal between auditory
presentation of information and recall (7), short-term
memory for verbal information that is unrefreshed by re-
hearsal is significantly impaired in patients with FMS. In
the aggregate, psychometric-based evidence of cognitive
abnormality was found in 87.9% of patients with FMS. As
a group, people with FMS performed ?2 SDs below the
normative score for their age on the ACT, highlighting
their incapacity to retain unrefreshed information when
their attention is split between relevant and distracting
information. Adding a source of distraction to neurocog-
nitive testing is critical to the objective demonstration of
significant memory problems in people with FMS.
Table 1. Mean summary data and percentage of impaired scores in patients with fibromyalgia and age-matched controls
presenting with memory problems*
Mean summary data Impaired scores, no. (%)
normative meanFMS groupControl group FMS group Control group
No distraction (WMS-III)
Logical memory, delayed
Paired associates, delayed
With distraction (ACT test)
10.0 ? 3.0
10.0 ? 3.0
10.0 ? 3.0
10.0 ? 3.0
10.0 ? 3.0
9.7 ? 2.5†
10.8 ? 2.6§
9.5 ? 2.7§
9.9 ? 2.9§
2.3 ? 2.6
7.8 ? 3.1
8.2 ? 3.3
7.4 ? 3.0
7.5 ? 3.5
5.1 ? 4.2†
* Values are the mean ? SD unless otherwise indicated. FMS ? fibromyalgia syndrome; WMS-III ? Wechsler Memory Scale, Third Edition; ACT ?
Auditory Consonant Trigram.
† P ? 0.001. Adjusted for age, education, vocabulary, and depression.
‡ P ? 0.01.
§ P ? 0.001.
¶ P ? 0.05.
Table 2. Group differences on recall and distraction trials of the Rey Auditory Verbal Learning Test
Total words recalled,
trial VI–trial V*
Fibromyalgia syndrome, mean ? SD
Control, mean ? SD
43.8 ? 8.3†
38.4 ? 10.7
11.2 ? 2.3‡
10.1 ? 2.9
9.5 ? 2.9‡
7.8 ? 2.8
?1.7 ? 1.9
?2.3 ? 1.8
* Normative mean ? SD difference 1.7 ? 2.0.
† P ? 0.01.
‡ P ? 0.05.
742Leavitt and Katz
Abnormalities in memory among patients with FMS
may be missed because measures traditionally relied upon
for neurocognitive examination fail to include a source of
distraction (8). On 4 such measures of verbal memory,
patients with FMS demonstrated auditory memory skills
equal to those of the average, healthy adult. On distraction-
free measures, patients with FMS resembled the controls
on both immediate and delayed recall of both narrative
and unstructured verbal information. The fact that patients
with FMS can lay down new memories at an age-appro-
priate rate in conditions free of distraction suggests that
some parts of memory work well when dealing with one
source of information at a time. Results in the normal
range on measures conventionally relied upon to assess
auditory memory functions imply that primary encoding
and retrieval mechanisms are intact.
We envisage the culprit of poor recall in people with
FMS as a weak memory trace brought on by reduced pro-
cessing time, and maintained by distraction that prevents
rehearsal of relevant information. In support of this hy-
pothesis, we turn to a timing deficit uncovered among
people with FMS, specifically slow naming speed (9).
Naming comes so naturally that we forget that naming
speed is determined by the time course of lexical activa-
tion, which is why naming speed has become a common
tool in estimating lexical speed (10,11). Faster naming
speed is associated with faster lexical activation, and gen-
erally reflects the amount of time consumed in probing the
lexical system for matching mental representations (12),
which is a crucial stage in transforming letter strings into
a recognizable word form. Knowledge stored in the lexical
system derives from a buildup of mental traces linked to
previous exposure to stimuli (13). The more time needed
to find the best-matching lexical trace for word identifica-
tion, the less time available for processing information at
the next stage (14).
People with FMS typically take more time to name
words, implying reduced speed of lexical activation. In
fact, the average deficiency in word-naming speed in peo-
ple with FMS is ?203 msec (Table 3) (9). The extra con-
sumption of time delays the flow of information from the
lexical stage to the next stage. It is as though a 203-msec
time delay is put on information entering the short-term
memory bank of individuals with FMS, thus shortening
the time available to lay down a memory trace by 203
msec. The implication is that 3 seconds of incoming infor-
mation processes as a 2.380-second stimulus (Table 3).
Without this built-in delay of 203 msec, processing takes
place over 2.583 seconds in short-term memory. The
longer that information is processed in short-term mem-
ory, the stronger the memory trace will be (15). The pre-
sumption is that built-in delays operate to shorten process-
ing time, causing individuals with FMS to lay down
weaker memory traces (15). Adding a source of distraction
following a 3-second presentation of the stimulus adds to
the difficulty by preventing rehearsal of the primary infor-
mation, sending weaker memory traces into long-term
memory without any chance of solidifying memory traces
by rehearsal (16).
Presumably, distraction is more disruptive to individu-
als with fibromyalgia because their memory traces have
not built sufficient velcro to stick in memory on their own
(17). Adding rehearsal practice seemingly makes up for
losses in processing time and creates a durable memory
trace that is available for later recall and less affected by
distraction. Presumably, another way to support recall in
people with FMS is to alter lexical processing to work
Our theory is also capable of describing the phenome-
non known as fibrofog (18). Sensory information enters
and leaves the lexical storehouse millions of times a day.
Recurrent time lags, further disrupted by outside distrac-
tions, may distort the synchrony of communication in
neural transmission of information, thus forming the basis
for the blurring of mental clarity commonly referred to as
There were several limitations to this study. First, 3
subjects in the control sample did not meet full exclusion-
ary criteria. However, data analysis without these 3 control
subjects did not change the statistical results in any sig-
nificant way. Second, cognitive test performance was not
subjected to effort testing; however, suboptimal effort is
unlikely to account for inferior test performance on the
ACT. Deficits among patients with FMS are selective and
occur in the context of normal test performance on the
majority of the cognitive measures administered. Finally,
variables such as sleep disruption, fatigue, and pain can
affect cognitive performance but were not measured in this
study. Future research should be directed toward the con-
tribution of these factors in explaining differences between
FMS and control groups.
All authors were involved in drafting the article or revising it
critically for important intellectual content, and all authors ap-
Table 3. Differences in the time course of information transmission from presentation of sensory information of a 3-second
duration to short-term memory in the neural networks of the participant groups, and their impact on the strength of the
memory trace sent into long-term memory
Time consumed in
Time available for
processing information in
2 minutes, 380 msec
2 minutes, 583 msec
Weak memory trace
Strong memory trace
* 203 msec represents the extra consumption of time in the lexical storehouse (620 msec ? 417 msec ? 203 msec) in fibromyalgia.
Effects of Distraction and Rehearsal on Memory in FMS743
proved the final version to be submitted for publication. Dr.
Leavitt had full access to all of the data in the study and takes
responsibility for the integrity of the data and the accuracy of the
Study conception and design. Leavitt, Katz.
Acquisition of data. Leavitt, Katz.
Analysis and interpretation of data. Leavitt, Katz.
1. Leavitt F, Katz RS. Distraction as a key determinant of im-
paired memory in fibromyalgia. J Rheumatol 2006;33:127–32.
2. Stuss DT, Stethem LL, Pelchat G. Three tests of attention and
rapid information processing: an extension. Clin Neuropsy-
3. Wolfe F, Smythe HA, Yunus MB, Bennett RM, Bombardier C,
Goldenberg DL, et al. The American College of Rheumatology
1990 criteria for the classification of fibromyalgia: report of
the Multicenter Criteria Committee. Arthritis Rheum 1990;33:
4. Wechsler D. Wechsler Adult Intelligence Scale. 3rd ed. San
Antonio (TX): Psychological Corporation; 1997.
5. Rey A. L’examen clinique en psychologie. Paris: Presses Uni-
versitaires de France; 1964.
6. Beck AT, Steer RA, Brown GK. Beck Depression Inventory:
BDI-II manual. 2nd ed. San Antonio (TX): Harcourt Brace;
7. Zeineh MM, Engel SA, Thompson PM, Bookheimer SY. Dy-
namics of the hippocampus during encoding and retrieval of
face-name pairs. Science 2003;299:577–80.
8. Kawas C, Resnick S, Morrison A, Brookmeyer R, Corrada M,
Zonderman A, et al. A prospective study of estrogen replace-
ment therapy and the risk of developing Alzheimer’s disease.
9. LeavittF, KatzRS.Speed
fibromyalgia: a selective naming speed deficit. J Clin Rheuma-
10. Brouwer WH. Limitations of attention after closed head
injury: detection and 0.10 Hz heart rate variability in a low
event rate vigilance task. Cortex 1985;21:111–9.
11. Kello CT. Control over time course of cognition in the tempo-
naming task. J Exp Psychol Hum Percept Perform 2004;30:
12. Wagner RK, Torgerson JK. The nature of phonological pro-
cessing and its causal role in the acquisition of reading skills.
Psychol Bull 1987;101:192–212.
13. Allor JH, Fuchs D, Mathes PG. Do students with and without
lexical retrieval weaknesses respond differently to instruc-
tion? J Learn Disabil 2001;34:264–75.
14. Ericsson KA. Memory skill. Can J Psychol 1985;39:188–23.
15. Craik FI, Lockhart RS. Levels of processing: a framework for
memory research. J Verb Learn Verb Be 1972;11:671–84
16. Lezak MD. Neuropsychological assessment. New York: Ox-
ford University; 1995.
17. Dalenoort GJ. The representation of tasks in active cognitive
networks. Cogn Syst 1985;1:253–72.
18. Katz RS, Heard AR, Mills M, Leavitt F. The prevalence and
clinical impact of reported cognitive difficulties (fibrofog) in
rheumatic patients with and without fibromyalgia. J Clin
744Leavitt and Katz