153

.
ASM20410.1177/10731

Article
Assessment

WAIS-IV Profile of Cognition in

20(4) 462­–473
© The Author(s) 2013
Reprints and permissions:

Schizophrenia sagepub.com/journalsPermissions.nav
DOI: 10.1177/1073191113478153
asm.sagepub.com

Natalie M. Michel1, Joel O. Goldberg1,2, R. Walter Heinrichs1, Ashley A. Miles1,

Narmeen Ammari1, and Stephanie McDermid Vaz2,3

Abstract
The Wechsler Adult Intelligence Scale (WAIS) has been used extensively to study impairment across a range of cognitive
domains in schizophrenia. However, cognitive performance among those with the illness has yet to be examined using
the newest edition of this measure. Hence, the current study aims first, to provide WAIS-IV normative data for Canadian
individuals with schizophrenia of low average intelligence; second, to examine schizophrenia performance on all WAIS-IV
subtest, index and general intelligence scores relative to healthy comparison subjects; and third, to revalidate the pattern
of impairment identified in this clinical group using the WAIS-III, where processing speed (PS) was most affected, followed
by working memory (WM), perceptual reasoning (PR) and verbal comprehension (VC). The WAIS-IV was administered to
outpatients with schizophrenia and their performance compared with age, gender, and education matched controls. WAIS-
IV schizophrenia performance data are provided. Analyses revealed significant impairment on several tasks, including the
new Cancellation subtest and the VC supplemental subtest, Comprehension. At the index score level, group differences in
PS were significantly larger than those observed in all other cognitive domains. Impairments were also observed in WM
amid relatively preserved performance in VC, thereby confirming the pattern of impairment identified using the WAIS-III.

Keywords
schizophrenia, cognition, Wechsler Adult Intelligence Scale, processing speed, working memory, verbal comprehension, can-
cellation

Cognitive impairment is now widely recognized as a central
feature of schizophrenia (Heinrichs, 2005). Difficulties
with attention, working memory and higher order executive
functioning are prominent in this population (Forbes, Car-
rick, McIntosh, & Lawrie, 2008; Johnson-Selfridge &
Zalewski, 2001; Lee & Park, 2005; Reichenberg & Harvey,
2007) and deficits in processing speed are especially pro-
nounced (Knowles, David, & Reichenberg, 2010). In fact,
patients perform 1.5 to 2.0 standard deviations below
healthy controls on many tests of cognitive functioning
(Keefe & Fenton, 2007).
Cognitive impairments in schizophrenia are posited to
reflect an underlying neurological vulnerability to the ill-
ness since they precede illness onset, are stable over time,
persist when symptoms begin to remit, and are associated
with frontal and medial temporal lobe dysfunction (Albus
et al., 2006; Erlenmeyer-Kimling et al., 2000; Hughes et al.,
2002; Joyce & Huddy, 2004; Rund, 1998). Research has
also shown that cognitive functioning is more accurate than
neurobiology in distinguishing schizophrenia patients from
healthy comparison subjects (Davidson & Heinrichs, 2003;
Heinrichs, 2005). Furthermore, difficulties in neurocogni-
tion have important clinical implications. Cognitive deficits
have been linked to poor quality of life, social functioning,
and community outcome (Green, Kern, & Heaton, 2004;
Williams et al., 2008). Given the pervasive and integral
nature of cognitive impairments in schizophrenia, the con-
tinued refinement of methods used to quantify and qualify
these deficits is necessary to advance the field.
The Wechsler Adult Intelligence Scale (WAIS) is a com-
prehensive test of general intellectual ability comprising
subtests spanning four domains of cognitive functioning,
namely, verbal comprehension, perceptual reasoning, work-
ing memory, and processing speed (Wechsler, 2008). For
more than four decades, the WAIS has been used exten-
sively to examine cognitive ability in schizophrenia, with
patients impaired relative to comparison subjects on most
tasks (Allen et al., 1998; Dickinson & Coursey, 2002;
1
York University, Toronto, Ontario, Canada
2
McMaster University, Hamilton, Ontario, Canada
3
St. Joseph’s Healthcare, Hamilton, Ontario, Canada
Corresponding Author:
Joel O. Goldberg, Department of Psychology,York University, 4700 Keele
Street, Toronto, Ontario, M3J 1P3, Canada.
Email: jgoldber@yorku.ca
Michel et al. 463
Dickinson, Iannone, & Gold, 2002; Gold et al., 1995;
Jortner, 1970; Revheim et al., 2006). However, to our
knowledge, the cognitive impairments affecting individuals
with the illness have yet to be assessed using the latest edi-
tion of the Wechsler Adult Intelligence Scale, the WAIS-IV
(Wechsler, 2008).
A number of revisions were made to enhance the
WAIS-IV relative to its predecessors (Frazier, 2011). First,
whereas the WAIS-III (Wechsler, 1997) collapsed tasks
involving verbal and working memory into the Verbal IQ
score, and tasks involving processing speed and perceptual
organization into the Performance IQ score, the WAIS-IV
omits the summary Verbal IQ and Performance IQ scores
and places greater emphasis on the four distinct index-
based measures previously confirmed by factor analysis
(Bowden, Saklofske, & Weiss, 2011a; Dickinson et al.,
2002; Dickinson, Iannone, Wilk, & Gold, 2004). Second,
three new subtests, namely, Visual Puzzles, Figure Weights,
and Cancellation, were added; and two others, Picture
Arrangement and Object Assembly were eliminated
(Hartman, 2009). Subtests used to derive the Perceptual
Reasoning Index (PRI) were also changed. Whereas the
WAIS-III Perceptual Organization Index was derived from
Block Design, Matrix Reasoning, and Picture Completion,
the core subtests of the analogous index measure in the
WAIS-IV include the new Visual Puzzles task and omit
Picture Completion, now deemed a supplemental subtest. In
addition, adjustments to individual items and subtests were
made in an attempt to improve clarity, decrease fine motor
demands, reduce cultural biases, and decrease administra-
tion time (Frazier, 2011; Hartman, 2009). Discontinue rules,
for example, were shortened for 4 of the 15 subtests, item
illustrations were simplified and subtests which made use
of similar sounding letters or numbers (e.g., P and B) were
revised. Subtest instructions were also simplified and dem-
onstration and sample items added. Finally, the WAIS-IV
standardization sample was updated to reflect our aging and
increasingly diverse population (Frazier, 2011; Hartman,
2009; Wechsler, Coalson, & Raiford, 2008). This has the
added advantage of correcting for inflation to intelligence
scores which occurs when dated norms are used (Flynn,
2009). As a result of these changes, the WAIS-IV boasts
substantial improvement over its predecessors.
At present, schizophrenia norms for the WAIS-IV sub-
test, index, and general intelligence scores have not been
reported in the literature, nor have they been examined
alongside a comparable group of healthy individuals.
Likewise, group data for the three new subtests of the
WAIS-IV are lacking. Moreover, studies using earlier ver-
sions of the WAIS have demonstrated a particular pattern of
cognitive impairment, with tasks involving processing
speed yielding the largest group differences, followed by
working memory and perceptual organization, amid rela-
tively well-preserved performance on subtests involving
verbal knowledge (Allen et al., 1998; Chen & Yao, 2009;
Dickinson & Coursey, 2002; Dickinson et al., 2002;
Dickinson et al., 2004; Gold, Queern, Iannone, & Buchanan,
1999; Goldberg & Gold, 1995; G. Goldstein, Beers, &
Shemansky, 1996; Lezak, 1995; Nestor, Kubicki, et al.,
2010; Nestor, Niznikiewicz, & McCarley, 2010; Nuechterlein
et al., 2004; Psychological Corporation, 1997). It is not yet
known whether this pattern of cognitive impairment will
be equally robust when performance is assessed using the
WAIS-IV.
In addition, seldom have neurocognitive studies in
schizophrenia simultaneously matched control and clinical
groups on a number of key demographic variables known
to affect cognition. Gender is one such factor. Sexual
dimorphisms have been noted across cognitive domains in
disordered and nondisordered populations alike, with
women generally outperforming their male counterparts on
tasks of language and memory, and men demonstrating
an advantage on tasks that draw on spatial abilities
(Beatty, Mold, & Gontkovsky, 2003; Feingold, 1993; J. M.
Goldstein et al., 1998; Halari, Mehrotra, Sharma, Ng, &
Kumari, 2006; Hoff, Riordan, O’Donnell, & DeLisi, 1991;
Maylor et al., 2007; Sota & Heinrichs, 2003). Furthermore,
some research suggests that relative to healthy comparison
subjects, cognitive impairments are more pronounced
among men with schizophrenia than among women, indi-
cating that gender may also interact with group member-
ship to affect cognition (Heinrichs & Zakzanis, 1998;
Sota & Heinrichs, 2003). Educational achievement is yet
another variable to consider (Stratta, Prosperini, Daneluzzo,
Bustini, & Rossi, 2001), with years of education account-
ing for a significant proportion of the variance in neuropsy-
chological test performance (G. Goldstein, Zubin, &
Pogue-Geile, 1991; Gontkovsky, Mold, & Beatty, 2002).
This relationship is especially important to be mindful of in
schizophrenia, since most affected individuals achieve no
more than a high school education (Heinrichs, 2005;
Heinrichs & Zakzanis, 1998; Jones, Guth, Lewis, &
Murray, 1994). Finally, age has also been associated with
cognition (Jeste, Wolkowitz, & Palmer, 2011). These find-
ings underscore the importance of closely matching, or
otherwise controlling, for gender, education, and age when
examining WAIS-IV cognitive performance in schizophre-
nia relative to controls. If not, observed between-group dif-
ferences risk being an artifact of variations in group
demographics rather than a reflection of true cognitive
impairment in the clinical group.
Hence, in the current study, WAIS-IV cognitive perfor-
mance was examined in a sample of outpatients with schizo-
phrenia or schizoaffective disorder relative to healthy
control participants individually matched for age, gender,
and education. Study goals were threefold: first, to provide
mean group performance data for a Canadian reference
sample of individuals with schizophrenia on each of the 15
464 Assessment 20(4)
subtests, four index scores and two general intelligence
measures of the WAIS-IV; second, to assess schizophrenia
performance on these scales relative to healthy comparison
subjects; and third, to obtain data on the pattern of perfor-
mance across the four cognitive domains identified by the
WAIS-IV in the clinical group relative to controls.
Consistent with previous research using the WAIS (Allen et
al., 1998; Dickinson & Coursey, 2002; Dickinson et al.,
2004; Gold et al., 1999; Nestor, Kubicki, et al., 2010;
Nestor, Niznikiewicz, & McCarley, 2010; Psychological
Corporation, 1997), we hypothesized that individuals with
schizophrenia will demonstrate a differential pattern of per-
formance wherein the PRI is least impaired, followed by the
Working Memory Index (WMI) and finally, the Processing
Speed Index (PSI) where the greatest impairments are
expected (i.e., PRI > WMI > PSI). Since verbal comprehen-
sion is generally stable over time and relatively resistant to
clinical illness (G. Goldstein et al., 1996; Lezak, 1995;
Nuechterlein et al., 2004), we expected patients to evince
preserved performance on verbal tasks relative to controls.
Method
Participants
Data for 37 outpatients with schizophrenia or schizoaffec-
tive disorder and 185 healthy controls were collected for
this study. Patients were recruited from the Hamilton
Program for Schizophrenia, a community-based, case
management and psychiatric rehabilitation program in
Ontario, Canada. Flyers describing the study were posted
at the location and individuals interested in participating
were asked to communicate their interest via their case-
workers. All clients were assessed using the Structured
Clinical Interview for the Diagnostic and Statistical
Manual of Mental Disorders, fourth edition, text revision
(DSM-IV-TR SCID-I; First, Spitzer, Gibbon, & Williams,
2002) and met DSM-IV-TR criteria (American Psychiatric
Association, 2000) for schizophrenia or schizoaffective
disorder. Participants who presented with Cushing’s dis-
ease, mental retardation, or a comorbid substance use
disorder, learning disorder, developmental disability, neu-
rological disorder, or a thyroid or endocrine problem were
excluded. Participants were also required to have normal
or corrected-to-normal vision and hearing, as well as
English as a first language.
Data for healthy controls were extracted from Pearson
Assessment’s American standardization sample, which
includes 800 examinees ranging from 16 to 90 years of age.
Given limitations in collecting additional data for our clini-
cal group, we chose to consider a larger subset of controls
so as to improve statistical power. Guided by Grimes and
Schulz (2005), who note that increasing the number of con-
trols to a ratio of 4:1 improves a study’s ability to detect
existing differences, whereas increases beyond that ratio do
little to improve power, we considered a control sample five
times the size of our schizophrenia group. Each participant
with schizophrenia was individually matched to five indi-
viduals from the standardization sample who were of the
same gender, had attained the same level of education and
who most closely approached the patient in age. This
yielded a control sample of 185 participants. Information
regarding exclusion criteria for the control sample can be
found in the WAIS-IV Technical and Interpretive Manual
(Wechsler et al., 2008). Table 1 provides the demographic
characteristics of the study participants.
Measures
Structured Clinical Interview for DSM-IV-TR Axis I disor-
ders–patient edition (SCID-I/P; First et al., 2002). The SCID is
a semistructured interview that informs diagnostic deci-
sions based on DSM-IV-TR criteria. The current study used
the SCID-I/P, an adaptation of this measure designed spe-
cifically for use with individuals who have been identified
as psychiatric patients.
Positive and Negative Syndrome Scale (Kay, Fiszbein, &
Opler, 1987). The Positive and Negative Syndrome Scale is
a widely used measure of symptom severity for patients
with schizophrenia and other psychotic disorders. Follow-
ing a semistructured interview, patients are rated on 30
symptom items which range in severity from 1 (absent) to 7
(extreme). Items can be categorized into three dimensions,
namely positive symptoms, negative symptoms, and
general symptoms.
Wechsler Adult Intelligence Scale–fourth edition (Wechsler,
2008). The WAIS-IV is a comprehensive test of intellectual
functioning. It consists of 10 core and 5 supplemental sub-
tests. Raw subtest scores are converted into scaled scores
corrected for age group. Subtest scaled scores are standard-
ized to a mean of 10, with one standard deviation reflected
in three point increments.
Subtests measuring similar aspects of cognitive ability
are combined to derive one of four index scores: Verbal
Comprehension (VCI; Similarities, Vocabulary, and
Information), Perceptual Reasoning (PRI; Block Design,
Matrix Reasoning, and Visual Puzzles), Working Memory
(WMI; Digit Span and Arithmetic), and Processing Speed
(PSI; Symbol Search and Coding). Supplemental subtests
associated with these indices are Comprehension (VCI),
Figure Weights and Picture Completion (PRI), Letter–
Number Sequencing (WMI), and Cancellation (PSI). These
are meant to replace a core subtest when performance on the
latter has been compromised because of factors unrelated to
the participant or task in question. For example, a partici-
pant’s Arithmetic score may not be reflective of his or her
true ability if during the administration of the test, he or she
was distracted by loud noises (Lichtenberger & Kaufman,
Michel et al. 465

Table 1. Demographic Characteristics. scores alone. All four index scores, in addition to the FSIQ
and GAI, are standardized to a mean of 100, with one stan-
Schizophrenia Controls
dard deviation reflected in 15-point increments.
N 37 185
Sex male, n (%) 25 (67.6) 125 (67.6)
Age in years (mean ± SD)a 45.08 ± 7.83 44.49 ± 9.63 Procedure
Education in years, n (%)   This research protocol was approved by the Ethics Review
  0-11 (Did not graduate from 9 (24.32) 45 (24.32) Board and the Human Participants Review Sub-Committee
high school) at York University. It also conforms to the standards of the
  12 (High school graduate or 16 (43.24) 80 (43.24) Canadian Tri-Council research ethics guidelines.
equivalent)
Prior to beginning the study, participants were thor-
  13-15 (Some postsecondary) 6 (16.21) 30 (16.21)
oughly briefed on the research protocol and intent, and writ-
  16 or More (completed 6 (16.21) 30 (16.21)
postsecondary degree or ten informed consent was obtained. Participants in our
more) clinical group were assessed on two occasions and were
Ethnicity, n (%)   provided remuneration for their time. During the initial
 Caucasian 34 (91.89)   assessment, chart reviews and brief semistructured inter-
 Non-Caucasian 2 (5.41)   views were carried out to gather relevant demographic and
 Undisclosed 1 (2.70)   clinical information. This included the participant’s age,
Clinical diagnosis, n (%)   date of birth, education level, ethnicity, formal diagnosis,
 Schizophrenia 33 (89.19)   age at onset of illness, and a list of his or her prescribed
  Schizoaffective disorder 4 (10.81)   medications. Comorbid psychiatric condition(s) were also
Age at onset of illness in years 20.31 ± 3.98   noted. The SCID-I/P was then administered to confirm a
(mean ± SD) diagnosis of schizophrenia or schizoaffective disorder and
Duration of illness in years 24.71 ± 8.27   participants’ symptoms were assessed using the Positive
(mean ± SD)
and Negative Syndrome Scale rating measure. During a
PANSS scoreb (mean ± SD)  
second meeting, participants completed the core and sup-
 Positive 18.14 ± 5.11  
plemental subtests of the WAIS-IV.
 Negative 15.92 ± 5.19  
 General 31.95 ± 7.35  
Antipsychotics, n (%)   Data Analysis
  Taking typical or first- 6 (16.21)  
generation antipsychotics Scaled scores. Data for this study were collected as part of
  Taking atypical or second- 28 (75.68)   a larger examination of a new measure of social perception
generation antipsychotics developed by Pearson Assessment, for which only American
  Taking a combination 2 (5.41)   standardization information is presently available. This has
of typical and atypical two noteworthy and related implications for the current
antipsychotics study. First, since funding was awarded for the express pur-
  Not taking any antipsychotic 1 (2.70)   pose of providing remuneration to patients, control data
medication
were extracted from the WAIS-IV American standardization
  Dosage in mg/dayc (mean 14.16 ± 10.07  
sample and as such, our Canadian clinical group was com-
± SD)
pared with an American control population. Second, WAIS-
Note. PANSS = Positive and Negative Syndrome Scale. IV cognitive raw scores for our schizophrenia sample were
a. Independent samples t test revealed the difference in age across groups converted into subtest and index scaled scores using Ameri-
was not statistically significant.
b. Mean PANSS scores are reflective of individuals who are relatively can norms so as to maintain consistency in the derivation of
stable but receiving outpatient treatment through a rehabilitation pro- clinical and control group data used in our analyses. Note,
gram (Whelton, Pawlick, & Cook, 1999). scaled scores for the schizophrenia group were also derived
c. In Haldol equivalents (Woods, 2003). using Canadian norms, and are provided as a reference in
Table 2 since Canadian population norms for this clinical
group have not yet been published in the literature.
2009). In this instance, the test administrator could use the Although using American norms to derive scaled scores
Letter–Number Sequencing score instead of the Arithmetic for the schizophrenia group introduces a potential bias,
score to calculate the WMI for this participant. findings generated in this manner are argued to be conser-
All core subtest scores contribute to the determination of vative. Research suggests that on average, Canadians score
the Full Scale IQ (FSIQ), whereas the General Ability Index slightly higher on the WAIS-IV than Americans do (Bowden
(GAI) is composed of the six VCI and PRI core subtest et al., 2011a, 2011b). For example, mean raw score
466 Assessment 20(4)

Table 2. Group Differences on WAIS-IV Index and Subtest Scaled Scores.

Schizophrenia (n = 37) Controls (n = 185)

Cohen’s
Tests Meana SD Meanb SD Meanb SD Fc pc dc
Verbal Comprehension Index 92.05 17.35 95.76 14.86 96.92 14.15 0.21 .650 0.08
 Similarities 8.16 3.44 8.81 2.92 9.42 2.70 1.53 .217 0.21
 Vocabulary 8.49 3.81 9.38 3.25 9.33 2.91 0.01 .928 −0.03
 Information 9.24 2.72 9.54 2.64 9.72 2.93 0.12 .732 0.07
 Comprehension 6.49 3.68 7.76 3.26 9.56 3.08 10.37 .001 0.56
Perceptual Reasoning Index 86.41 17.35 91.38 16.20 97.01 15.26 4.11 .044 0.36
  Block Design 7.27 3.07 8.24 2.86 9.61 2.92 6.74 .010 0.48
  Matrix Reasoning 8.27 3.18 8.92 3.10 9.29 3.10 0.43 .511 0.13
  Visual Puzzles 7.76 3.40 8.46 3.41 9.69 3.19 4.50 .035 0.36
  Figure Weights 7.89 2.98 8.32 3.07 9.59 3.21 4.85 .029 0.41
  Picture Completion 7.41 2.64 7.32 2.59 9.83 3.49 17.17 <.001 0.81
Working Memory Index 84.19 19.50 89.35 18.77 96.57 15.27 6.37 .012 0.42
  Digit Span 7.24 3.44 7.95 3.54 9.49 3.08 7.32 .007 0.45
 Arithmetic 7.51 3.74 8.32 3.70 9.35 3.04 3.22 .074 0.33
  Letter–Number Sequencing 7.76 3.32 8.51 2.98 9.30 2.73 2.47 .118 0.28
Processing Speed Index 78.97 15.43 82.95 14.60 95.71 14.69 23.32 <.001 0.87
  Symbol Search 6.49 2.98 7.11 2.91 9.28 2.98 16.46 <.001 0.75
 Coding 6.16 2.76 6.57 2.78 9.16 3.10 22.21 <.001 0.88
 Cancellation 5.84 2.58 5.92 2.52 9.47 3.04 44.45 <.001 1.30
Full Scale IQ 82.84 18.85 88.51 17.02 95.85 14.53 7.45 .007 0.47
General Ability 88.16 18.03 92.86 16.39 96.66 14.66 1.99 .160 0.24
Note. WAIS-IV = Wechsler Adult Intelligence Scale–fourth edition. Mean differences in scaled scores derived using American versus Canadian norms are
0.61 (SD = 0.34) and 4.46 (SD = 0.72) for subtests and index scaled scores, respectively.
a. Derived using Canadian norms.
b. Derived using American norms.
c. Analysis of variance results comparing control and patient groups’ WAIS-IV performance derived using American norms.

performance on the Block Design task is approximately 41
for Canadians and 34 for Americans. However, when devel-
oping standardization data for each population, mean sub-
test raw scores are independently set to a scaled score of 10
(i.e., in the example above, both raw scores would convert
to a scaled score of 10 using the appropriate norms).
Conversely, raw scores typically convert into higher scaled
scores when Americans norms are used resulting in a slight
overestimate in performance relative to a Canadian refer-
ence sample (see Table 2). Therefore, the chances of finding
mean differences between the patient and control groups in
this study are reduced, and any significant group differ-
ences which are observed are likely reflective of legitimate
cognitive impairment in the schizophrenia group.
WAIS-IV performance and group differences. Mean group
scaled scores were calculated for each of the 15 subtests,
four index scores, and two general intelligence measures of
the WAIS-IV. Scaled scores derived using American norms
were used in the analyses that follow.
The effect of group on the WAIS-IV subtest scores was
examined using a multivariate analysis of variance
(MANOVA). Post hoc univariate analyses of variance
(ANOVAs) were then carried out to explicate between
group differences on each of these scales. Group differences
on the index and global cognitive scores were also assessed
using univariate ANOVAs. Alpha levels were adjusted at
each stage of the analysis using a Bonferroni correction. To
evaluate the magnitude of the effect of group on each of the
dependent variables, effect size calculations (Cohen’s d;
Cohen, 1992) were undertaken by dividing the mean group
difference by the pooled standard deviation for each of
the scales.
To investigate the pattern of performance across groups
on the VCI, PRI, WMI, and PSI, a multivariate approach to
repeated measures analysis was used, followed by post hoc
contrasts.
Results
WAIS-IV Cognitive Profile in Schizophrenia
WAIS-IV mean subtest, index, and general intelligence
scaled scores derived using American and Canadian norms
are presented in Table 2.
Michel et al. 467
10.50
10.00
9.50
Subtest Scaled Scores
9.00
8.50
8.00
7.50
7.00
6.50
6.00
5.50
5.00
Schizophrenia
Figure 1. Group performance on WAIS-IV cognitive subtest scaled scores.
Note. Error bars indicate 95% confidence intervals. WAIS-IV = Wechsler Adult Intelligence Scale–fourth edition.
*p < .003.
MANOVA was first used to assess whether schizophre-
nia group membership conferred an overall disadvantage on
the 15 WAIS-IV subtests relative to mentally healthy con-
trols. All assumptions were met and analyses revealed a
main effect of group, Λ = 0.72, F(15, 206) = 5.25, p < .001.
Results for the post hoc univariate ANOVAs are presented
in Table 2 and illustrated in Figure 1. Mean differences for
5 of the 15 subtests were significant after adjusting the
alpha level using a Bonferroni correction. The magnitudes
of group differences on the Comprehension and Symbol
Search subtests were in the medium range (Cohen’s d =
0.56 and 0.75, respectively). Large effect sizes were noted
for the Picture Completion subtest (Cohen’s d = 0.81), as
well as two of the three processing speed subtests:
Cancellation and Coding (Cohen’s d = 1.30 and 0.88,
respectively). Mean differences on the two new PRI sub-
tests, Figure Weights, and Visual Puzzles, were not signifi-
cant following a Bonferroni correction (p = .029 and p =
.035, respectively). All scaled scores were normally distrib-
uted. However, the assumption of equality of variance was
met for all but the Picture Completion variable. Hence,
group differences on this subtest were further examined
using post hoc independent samples t tests (equal variances
not assumed), as well as a Mann–Whitney nonparametric
test. Results converged on both post hoc tests, demonstrat-
ing a significant between-group difference, t(65.17) = 5.04,
p < .001; U = 1998.50, z = −4.01, p < .001.
Univariate ANOVAs were then carried out for each of
the four index scores. All assumptions were met, and results
were in keeping with our hypothesis. The schizophrenia
group evinced a pattern of increasing impairment on the
Healthy Controls
PRI, WMI, and PSI, with the PSI yielding large effect size
differences across groups (p < .001, Cohen’s d = 0.87).
Differences on the WMI were also significant (p = .012,
Cohen’s d = 0.42), but the PRI only showed a trend toward
impairment after adjusting for multiple comparisons (p =
.044, Cohen’s d = 0.36). In contrast, patient performance on
the VCI was comparable with controls (p = .65). Group dif-
ferences on the WAIS-IV index scores are presented in
Table 2 and depicted in Figure 2.
Between-group differences on the GAI and FSIQ were
also assessed using univariate ANOVAs. Analyses revealed
that schizophrenia group performance differed significantly
from healthy comparisons on the FSIQ (F = 7.42, p = .007,
Cohen’s d = 0.47) but not the GAI (F = 1.99, p = .160,
Cohen’s d = 0.24).
Pattern of Performance on the WAIS-IV
Index Scores
A multivariate approach to repeated measures analysis was
used to investigate the pattern of performance across cogni-
tive domains in the schizophrenia group relative to con-
trols. All assumptions were met, and analyses revealed a
significant group-by-domain interaction, Λ = 0.91, F(3,
218) = 6.84, p < .001. Follow-up post hoc contrasts
revealed that relative to controls, the degree of impairment
on the PSI in the schizophrenia group was significantly
larger than that observed on the VCI, PRI, and WMI, F(1,
220) = 19.31, p < .001; F(1, 220) = 7.17, p = .008; and F(1,
220) = 3.90, p = .050, respectively. Individuals with schizo-
phrenia also performed more poorly on the WMI than they
468 Assessment 20(4)
102.00
100.00
98.00
96.00
Index Scaled Scores 94.00
92.00
90.00
88.00
86.00
84.00
82.00
80.00
78.00
76.00
Verbal Perceptual
Comprehension Reasoning
Schizophrenia
Figure 2. Group performance on WAIS-IV cognitive index scores.
Note. Error bars reflect 95% confidence intervals. WAIS-IV = Wechsler Adult Intelligence Scale–fourth edition.
*p < .0125.
did on the VCI, F(1, 220) = 7.06, p = .008. In contrast,
group differences on the PRI did not differ significantly
from those observed on the VCI and WMI, F(1, 220) =
3.74, p = .054 and F(1, 220) = 0.35, p = .557, respectively.
Discussion
The present study is the first to examine cognitive perfor-
mance in schizophrenia using the latest edition of the
WAIS. Results demonstrate that the WAIS-IV is sensitive to
schizophrenia, with patients performing below demograph-
ically matched control participants on most subtests and
index scores. Statistically significant impairments were
evident on the processing speed and working memory sum-
mary indices, as well as 5 of the 15 subtests after applying
a Bonferroni correction. In addition, schizophrenia patients
showed the expected pattern of differential impairment
across the four cognitive domains, as previously identified
using the WAIS-III.
The introduction of new WAIS subtests has the potential
to uncover subtle but important differential sensitivities, to
which the clinician needs to be alert in case work. Three
new subtests were added to the WAIS-IV. Although results
are suggestive of a trend toward impairment on Visual
Puzzles and Figure Weights, patients demonstrated greatest
difficulty on the PSI supplemental subtest, Cancellation, a
search task where subjects are asked to identify target
objects of a particular shape and color from a visual display
(McCrae & Robinson, 2011). In fact, effect size analyses
revealed that the mean group difference on Cancellation
surpassed that observed on all other WAIS-IV subtests,
Working
* *
Processing
Memory Speed
Healthy Controls
including Coding. This finding is interesting in light of
recent meta-analytic work by Dickinson, Ramsey, and Gold
(2007) that shows that impairments on Digit Symbol Coding
are more substantial than those observed on all other neuro-
psychological measures in the review, and may therefore
reflect the single largest cognitive deficit in schizophrenia.
Future research should examine WAIS-IV performance on
both the Coding and Cancellation subtests, so as to deter-
mine whether the latter task is in fact a more sensitive mea-
sure of processing speed impairment in this population.
This is especially important given that the mean scatter of
scores derived using Canadian and American norms dif-
fered in this study, and as a result, individual subtest dis-
crepancies may be more or less evident when norms from
one country are used over another.
Also noteworthy is the variability in performance across
subtests of a given domain. For instance, although perfor-
mance on the core subtests of the VCI was relatively intact
in the patient group, individuals with schizophrenia evinced
significant impairment on the supplemental subtest of this
index, namely Comprehension. A similar pattern of results
has been noted using the WAIS-IV in other clinical groups
as well (Holdnack, Goldstein, & Drozdick, 2011). Several
items in this subtest rely on abstract reasoning. Participants
may be asked, for example, to explain metaphors or the
meaning of less well-known proverbs (Lichtenberger &
Kaufman, 2009). This may be especially challenging for
individuals who engage in ‘unusual and idiosyncratic’
thinking, which is often the case in schizophrenia.
These findings call attention to the potential implications
of substituting core subtest scores with supplemental subtest
Michel et al. 469
scores when working with individuals affected by schizophre-
nia. The supplemental VCI subtest, Comprehension, appears
to be more challenging than the core VCI subtests for this
population. Therefore, performance on this subtest may not
be comparable to performance on Vocabulary, Similarities,
and Information. Should clinicians use Comprehension to
replace a spoiled core subtest in calculating the VCI, they
should be aware that this may result in a lower index level
score than expected. A similar argument may be made for
the Cancellation subtest of the PSI. Further research directly
assessing the impact of supplemental subtest substitution in
the WAIS-IV across clinical and nonclinical groups is war-
ranted, particularly since substantial variability in perfor-
mance across core and supplemental subtests has not been
observed in nonschizophrenia normative samples (see con-
trol group data Table 2; see also Bowden et al., 2011a,
2011b; Ryan & Glass, 2011).
Results from analyses examining the differential pattern
of impairment across the four cognitive domains were con-
sistent with previous literature (Allen et al., 1998; Chen &
Yao, 2009; Dickinson & Coursey, 2002; Dickinson et al.,
2002; Dickinson et al., 2004; Gold et al., 1999; Goldberg &
Gold, 1995; G. Goldstein et al., 1996; Lezak, 1995; Nestor,
Kubicki, et al., 2010; Nestor, Niznikiewicz, & McCarley,
2010; Nuechterlein et al., 2004; Psychological Corporation,
1997) and in line with our predictions. Individuals with
schizophrenia evinced most impairment on the PSI relative
to age, education, and gender-matched controls. Impairments
were also noted on the WMI and a trend toward impairment
was noted on the PRI. In contrast, preserved performance
was observed on the VCI.
In keeping with meta-analytic work by Knowles et al.
(2010), group differences on the PSI were especially nota-
ble. Effect sizes were in the large range, and between-group
differences were significantly larger than those observed on
all other index measures. Likewise, schizophrenia perfor-
mance on the PSI subtests (Symbol Search, Coding, and
Cancellation) was also indicative of substantial impairment,
with effect sizes in the medium to large range.
Impaired functioning in the schizophrenia group was
also apparent on the WAIS-IV WMI, and group differences
here were significantly larger than those observed on the
VCI. This is consistent with research in the field, which has
repeatedly found that the short-term maintenance and
manipulation of information in memory is an area of diffi-
culty for this population (Forbes et al., 2008; Gold,
Carpenter, Randolph, Goldberg, & Weinberg, 1997; Lee &
Park, 2005). Although, mean differences on tasks associ-
ated with the WMI were not significant on their own, core
WMI subtests certainly showed a trend toward impairment.
In fact, group differences on Digit Span would have reached
significance (p = .007) were it not for the strict application
of a Bonferroni correction to control for the family-wise
error rate (alpha required for statistical significance = .003).
Performance differences on Arithmetic were more mild (p =
.074), however, demonstrating variability in task perfor-
mance across WMI subtests.
Still, effect size differences on the WMI and its subtests
were smaller than anticipated. Whereas previous research
using the WAIS-III working memory subtests and similar
measures has yielded effects sizes in the medium to large
range (Dickinson et al., 2007; Forbes et al., 2008; Lee &
Park, 2005), small to medium effect sizes were noted in the
current study. We offer a few potential explanations for this
discrepancy. To begin, the clinical sample in the present
study consisted of a group of outpatients with schizophre-
nia. Conversely, meta-analytic studies identifying medium
to large effect size differences in working memory have
collapsed data across inpatient and outpatient samples (see
Forbes et al., 2008; Lee & Park, 2005). In addition, as part
of the changes made to the WAIS-IV, a number of WMI
subtest items were revised. For instance, WAIS-III Letter–
Number Sequencing and Digit Span items that were
deemed confusing (e.g., items that included similar sound-
ing digits or letters), were adjusted or removed to improve
task clarity in the WAIS-IV (Hartman, 2009). The
Arithmetic subtest was also altered to decrease the empha-
sis on mathematical skills and the English measurement
system. Together, these changes may have resulted in a
slight decrease in the measure’s sensitivity to impairments
in schizophrenia than was previously the case. Future con-
current validity research comparing WAIS-III and WAIS-IV
WMI scores with other measures of working memory
should clarify this discrepancy.
WAIS-IV PRI clinical group performance was compa-
rable to previous literature using the WAIS-III in this popu-
lation (Dickinson et al., 2002; Nestor, Kubicki, et al., 2010;
Nestor, Niznikiewicz, & McCarley, 2010). Relative to con-
trols, individuals with schizophrenia expressed greater dif-
ficulty in perceptual reasoning with group differences
showing a trend toward significance after adjusting for the
family-wise error rate.
However, mean performance differences on the PRI
were not significantly larger than those observed on the
VCI, where preserved performance was found. Findings
here are not surprising. Previous literature supports the
notion that tasks relying on perceptual organization skills
are generally more difficult than tasks assessing verbal
abilities in schizophrenia (Allen et al., 1998; Dickinson et
al., 2004; Nuechterlein et al., 2004), as was noted in the
present study. However, findings regarding the degree of
impairment on Perceptual Reasoning tasks are mixed.
Whereas some studies find medium to large group differ-
ences on PRI subtests (Dickinson et al., 2007), other
research finds that impairment on these tasks is more mod-
erate (Dickinson et al., 2004).
Moreover, given recent changes to the tasks that com-
prise the core subtests of the PRI in the WAIS-IV, it was
470 Assessment 20(4)
difficult to predict whether the index score would be more or
less sensitive to perceptual reasoning impairments in our
clinical group, relative to studies examining similar deficits
using previous versions of the WAIS. Results of the present
study indicate that omitting Picture Completion in favor of
Visual Puzzles as a core PRI subtest may result in an index
score that is less sensitive to impairment in schizophrenia.
Of all the PRI subtests, Picture Completion appears to be the
most challenging for individuals with the illness, with effect
sizes in the large range (p < .001, Cohen’s d = 0.81). In con-
trast, our clinical group was less impaired on Visual Puzzles,
a subtest which is new to the WAIS. Future research may
attempt to assess the relative sensitivity of these measures to
impairments in perceptual organization in schizophrenia.
In contrast to all other index measures, performance on
the VCI was not different between groups, reflecting pre-
served verbal ability in our outpatient sample relative to
demographically matched controls. This finding was
expected since a majority of VCI subtest items draw on
crystallized verbal knowledge (Dawes, Jeste, & Palmer,
2011), which is known to be relatively resilient to clinical
illness and stable over time (Dickinson et al., 2002; Gold et
al., 1999; Goldberg & Gold, 1995; G. Goldstein et al., 1996;
Groth-Marnat, 2009; Nuechterlein et al., 2004). Mean VCI
performance in our group was also comparable to prior
research using the WAIS-III to assess cognition in schizo-
phrenia (Dickinson et al., 2002; Nestor, Niznikiewicz, &
McCarley, 2010).
Nevertheless, the finding of preserved verbal task perfor-
mance in schizophrenia is a controversial one, with other
research finding impairment on the Vocabulary, Similarities,
and Information subtests of the WAIS (Dickinson et al.,
2007). Variations in the way control and clinical groups are
matched may partially explain this difference. It is not
always the case, for instance that groups are matched on
demographic variables known to affect cognitive perfor-
mance. This is problematic when between-group differences
along these variables are apparent. As an example, in many
studies control participants are on average slightly more
educated than individuals affected by schizophrenia
(Dickinson et al., 2007), thereby conferring a disadvantage
to patient groups as compared with controls on cognitive
measures. In other research, participants are matched on
parental level of education. Parents of individuals with
schizophrenia typically achieve higher levels of education
than their children whose academic pursuits are often inter-
rupted by illness onset. Hence, matching on parental, and not
participant, level of education may similarly result in
increased performance discrepancies between schizophrenia
and control groups. Continued research examining perfor-
mance on VCI subtests between patients and demographi-
cally matched controls is needed to determine if, in fact,
these measures are intact in schizophrenia.
In summary, the overall pattern of impairment observed
across WAIS-IV index scores was in keeping with our
hypothesis, with patients expressing most impairment on
tasks of processing speed, and least impairment on tasks rely-
ing on verbal knowledge. It is likely that this pattern broadly
characterizes schizophrenia as a diagnostic group. Findings
of the present study provide a reference against which clini-
cians and researchers may compare WAIS-IV impairment
patterns among people with schizophrenia, and may serve to
complement more comprehensive clinical and cognitive
assessments. Schizophrenia group scaled score performance
derived using Canadian norms may also prove useful in dis-
criminating meaningful impairment when individual perfor-
mance on a particular subtest or index score is of interest.
However, clinicians are cautioned against applying the
pattern of performance observed herein as some kind of
definitive schizophrenia rubric. Significant variability in
performance across cognitive domains exists between par-
ticipants. Not all individuals with schizophrenia show
greatest impairment on the PSI, for instance, and some
express difficulty on verbal tasks. Similarly, some individu-
als with the illness appear cognitively unimpaired relative
to comparable healthy subjects, scoring in the high average
range across subtests. Furthermore, our sample consisted of
a relatively high-functioning group of Canadian outpatients,
and differences in the degree of impairment across scores
are also likely affected by other factors, such as the severity
of illness and geographic location. Therefore, clinicians are
encouraged instead to use the current study’s findings as
general guidelines for the idiographic interpretation of
neurocognitive assessments.
Study limitations include a relatively small schizophre-
nia sample size. Future research should endeavor to collect
larger samples, as this would allow for a more detailed
examination of WAIS-IV cognitive performance than was
possible here. The effect of ethnicity and antipsychotic dos-
age, for instance, which have previously been found to
influence cognitive performance in this population
(Knowles et al., 2010; Walker, Batchelor, & Shores, 2009)
may be assessed if larger samples are available. Also, ran-
dom selection of study participants was not possible.
Samples were instead selected based on availability, affect-
ing the generalizability of these findings. Additionally,
future studies should endeavor to examine patient and con-
trol samples of the same nationality. This will likely result
in greater between-group performance discrepancies for
some subtests and index scores. Studies examining clinical
and control groups matched on other variables of interest,
such as socioeconomic status and parental level of educa-
tion, may also provide new insights on WAIS-IV cognitive
performance in schizophrenia.
Conclusions
The present study was the first to examine cognitive per-
formance in a group of individuals with schizophrenia
using the WAIS-IV (Wechsler, 2008). Results indicate that
Michel et al. 471
relative to age-, gender-, and education level–matched
controls, patients express difficulty on several WAIS-IV
subtests and domains. They further suggest that the pattern
of cognitive impairment identified in schizophrenia using
the WAIS-III is robust and remains when cognitive ability
is assessed using the WAIS-IV. Patient and control group
performance was comparable on tasks relying on crystal-
lized verbal knowledge. In contrast, the schizophrenia
group performed below controls on measures of perceptual
reasoning, working memory, and processing speed, with
impairments on the latter being significantly larger than
those observed on all other index measures. Similar cogni-
tive profiles have been observed in other clinical groups
including individuals with traumatic brain injury and
Alzheimer’s disease, and may reflect a common underly-
ing neuropathology (Wechsler et al., 2008). Findings may
have implications for predicting functional competence
and community independence in schizophrenia (Heinrichs,
Ammari, Miles, & McDermid Vaz, 2010).
Acknowledgments
The authors would like to thank Ryan Barnhart, Hugh McCague,
Mirka Ondrack, and Xiaobin Zhou for their support in this
research; the clients and staff at the Hamilton Program for
Schizophrenia for their participation and assistance; and the
reviewers of this article for their thoughtful comments and sug-
gestions. We would also like to thank NCS Pearson for granting
us permission to use the American standardization data from the
Wechsler Adult Intelligence Scale–Fourth Edition (WAIS-IV).
Copyright © 2008 NCS Pearson, Inc. All rights reserved.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect
to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support
for the research, authorship, and/or publication of this article:
Natalie M. Michel was supported by a Frederick Banting and
Charles Best Canada Graduate Scholarship from the Canadian
Institutes of Health Research. Minor grant funding for client
honoraria was provided by the Community Social and
Vocational Rehabilitation Foundation (CSVR) in Hamilton,
Ontario.
References
Albus, M., Hubmann, W., Mohr, F., Hecht, S., Hinterberger-Weber,
P., Seitz, N. N., & Küchenhoff, H. (2006). Neurocognitive func-
tioning in patients with first-episode schizophrenia: Results
of a prospective 5-year follow-up study. European Archives
of Psychiatry and Clinical Neuroscience, 256, 442-451.
doi:10.1007/s00406-006-0667-1
Allen, D. N., Huegel, S. G., Seaton, B. E., Goldstein, G., Gurklis, J.
A., & Kammen, D. P. V. (1998). Confirmatory factor analysis
of the WAIS-R in patients with schizophrenia. Schizophrenia
Research, 34, 87-94. doi:10.1016/S0920-9964(98)00090-5
American Psychiatric Association. (2000). Diagnostic and statisti-
cal manual of mental disorders (4th ed., text revision). Wash-
ington, DC: Author.
Beatty, W. W., Mold, J. W., & Gontkovsky, S. T. (2003). RBANS
performance: Influences of sex and education. Journal of
Clinical and Experimental Neuropsychology, 25, 1065-1069.
doi:10.1076/jcen.25.8.1065.16732
Bowden, S. C., Saklofske, D. H., & Weiss, L. G. (2011a). Aug-
menting the core battery with supplementary subtests:
Wechsler Adult Intelligence Scale–IV measurement invari-
ance across the United States and Canada. Assessment, 18,
133-140. doi:10.1177/1073191110381717
Bowden, S. C., Saklofske, D. H., & Weiss, L. G. (2011b). Invari-
ance of the measurement model underlying the Wechsler
Adult Intelligence Scale-IV in the United States and Canada.
Educational and Psychological Measurement, 71, 186-199.
doi:10.1177/0013164410387382
Chen, H., & Yao, S. Q. (2009). A comparison between nonclini-
cal and schizophrenia sample on the results measured by the
WAIS-III Chinese version. Chinese Journal of Clinical Psy-
chology, 17, 157-160.
Cohen, J. (1992). Statistical power analysis. Current Directions
in Psychological Science, 1, 98-101. doi:10.1111/1467-8721.
ep10768783
Davidson, L., & Heinrichs, R. W. (2003). Quantification of brain
imaging findings on the frontal and temporal lobes in schizo-
phrenia: A meta-analysis. Psychiatry Research: Neuroimag-
ing, 122, 69-87. doi:10.1016/S0925-4927(02)00118
Dawes, S. E., Jeste, D. V., & Palmer, B. W. (2011). Cognitive pro-
files in persons with chronic schizophrenia. Journal of Clinical
and Experimental Neuropsychology, 33, 929-936. doi:10.1080/
13803395.2011.578569
Dickinson, D., & Coursey, R. D. (2002). Independence and over-
lap among neurocognitive correlates of community function-
ing in schizophrenia. Schizophrenia Research, 56, 161-170.
doi:10.1016/S0920-9964(01)00229-8
Dickinson, D., Iannone, V. N., & Gold, J. M. (2002). Factor struc-
ture of the Wechsler Adult Intelligence Scale–III in schizophre-
nia. Assessment, 9, 171-180. doi:10.1177/10791102009002008
Dickinson, D., Iannone, V. N., Wilk, C. M., & Gold, J. M.
(2004). General and specific cognitive deficits in schizophre-
nia. Biological Psychiatry, 55, 826-833. doi:10.1016/j.bio-
psych.2003.12.010
Dickinson, D., Ramsey, M. E., & Gold, J. M. (2007). Overlook-
ing the obvious: A meta-analytic comparison of digit symbol
coding tasks and other cognitive measures in schizophrenia.
Archives of General Psychiatry, 64, 532-542. doi:10.1001/
archpsyc.64.5.532
Erlenmeyer-Kimling, L., Rock, D., Roberts, S. A., Janal, M.,
Kestenbaum, C., Cornblatt, B., & Gottesman, I. I. (2000).
Attention, memory, and motor skills as childhood predictors
of schizophrenia-related psychoses: The New York High Risk
472 Assessment 20(4)
Project. American Journal of Psychiatry, 157, 1416-1422.
doi:10.1176/appi.ajp.157.9.1416
Feingold, A. (1993). Cognitive gender differences: A devel-
opmental perspective. Sex Roles, 29, 91-112. doi:10.1007/
BF00289998
First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (2002).
Structured clinical interview for DSM-IV-TR Axis I disorders,
research version, patient edition (SCID-I/P). New York: Bio-
metrics Research, New York State Psychiatric Institute.
Flynn, J. R. (2009). The WAIS-III and WAIS-IV: Daubert
motions favor the certainly false over the approximately
true. Applied Neuropsychology, 16, 98-104. doi:10.1080/
09084280902864360
Forbes, N. F., Carrick, L. A., McIntosh, A. M., & Lawrie, S. M. (2008).
Working memory in schizophrenia: A meta-analysis. Psycholog-
ical Medicine, 39, 889-905. doi:10.1017/S0033291708004558
Frazier, T. W. (2011). Introduction to the special section on
advancing WAIS-IV and WMS-IV clinical interpretation.
Assessment, 18, 131-132. doi:10.1177/1073191111408581
Gold, J. M., Blaxton, T. A., Hermann, B. P., Randolph, C.,
Fedio, P., Goldberg, T. E., & Weinberger, D. R. (1995).
Memory and intelligence in lateralized temporal lobe epi-
lepsy and schizophrenia. Schizophrenia Research, 17, 59-65.
doi:10.1016/0920-9964(95)00030-P
Gold, J. M., Carpenter, C., Randolph, C., Goldberg, T. E., & Wein-
berg, D. R. (1997). Auditory working memory and Wisconsin
Card Sort Test performance in schizophrenia. Archives of Gen-
eral Psychiatry, 54, 159-165.
Gold, J. M., Queern, C., Iannone, V. N., & Buchanan, R. W.
(1999). Repeatable battery for the assessment of neuropsycho-
logical status as a screening test in schizophrenia, I: Sensitiv-
ity, reliability, and validity. American Journal of Psychiatry,
156, 1944-1950.
Goldberg, T., & Gold, J. (1995). Neurocognitive functioning in
patients with schizophrenia: An overview. In D. Kupfer & F.
Bloom (Eds.), Psychopharmacology: A fourth generation of
progress (pp. 1245-1257). New York, NY: Raven Press.
Goldstein, G., Beers, S. R., & Shemansky, W. J. (1996). Neuro-
psychological differences between schizophrenic patients
with heterogeneous Wisconsin Card Sorting Test perfor-
mance. Schizophrenia Research, 21, 13-18. doi:10.1016/0920-
9964(96)00019-9
Goldstein, G., Zubin, J., & Pogue-Geile, M. F. (1991). Hospitaliza-
tion and the cognitive deficits of schizophrenia: The influences
of age and education. Journal of Nervous and Mental Disease,
179, 202-206. doi:10.1097/00005053-199104000-00004
Goldstein, J. M., Seidman, L. J., Goodman, J. M., Koren, D., Lee, H.,
Weintraub, S., & Tsuang, M. T. (1998). Are there sex differences
in neuropsychological functions among patients with schizo-
phrenia? American Journal of Psychiatry, 155, 1358-1364.
Gontkovsky, S. T., Mold, J. W., & Beatty, W. W. (2002). Age and edu-
cational influences on RBANS index scores in a nondemented
geriatric sample. Clinical Neuropsychologist, 16, 258-263.
doi:10.1076/clin.16.3.258.13844
Green, M. F., Kern, R. S., & Heaton, R. K. (2004). Longitudinal
studies of cognition and functional outcome in schizophrenia:
Implications for MATRICS. Schizophrenia Research, 72, 41-51.
doi:10.1016/j.schres.2004.09.009
Grimes, D. A., & Schulz, K. F. (2005). Compared to what? Find-
ing controls for case-control studies. Lancet, 365, 1429-1433.
doi:10.1016/S0140-6736(05)66379-9
Groth-Marnat, G. (2009). Comparing scores on the WAIS-IV and
the WMS-IV. In G. Groth-Marnet (Ed.), Handbook of psy-
chological assessment (5th ed., pp. 183-206). New York, NY:
Wiley.
Halari, R., Mehrotra, R., Sharma, T., Ng, V., & Kumai, V. (2006).
Cognitive impairment but preservation of sexual dimorphism
in cognitive abilities in schizophrenia. Psychiatry Research,
141, 129-139. doi:10.1016/j.psychres.2005.07.021
Hartman, D. E. (2009). Wechsler Adult Intelligence Scale IV
(WAIS-IV): Return of the gold standard. Applied Neuropsy-
chology, 16, 85-87. doi:10.1080/09084280802644466
Heinrichs, R. W. (2005). The primacy of cognition in schizophre-
nia. American Psychologist, 60, 229-242. doi:10.1037/0003-
066X.60.3.229
Heinrichs, R. W., Ammari, N., Miles, A. A., & McDermid Vaz,
S. (2010). Cognitive performance and functional competence
as predictors of community independence in schizophrenia.
Schizophrenia Bulletin, 36, 381-387. doi:10.1093/schbul/
sbn095
Heinrichs, R. W., & Zakzanis, K. K. (1998). Neurocognitive deficit
in schizophrenia: A quantitative review of the evidence. Neu-
ropsychology, 12, 426-445. doi:10.1037/0894-4105.12.3.426
Hoff, A. L., Riordan, H., O’Donnell, D. W., & DeLisi, L. E.
(1991). Cross-sectional and longitudinal neuropsycho-
logical test findings in first episode schizophrenic patients.
Schizophrenia Research, 5, 197-198. doi:10.1016/0920-
9964(91)90070-8
Holdnack, J., Goldstein, G., & Drozdick, L. (2011). Social per-
ception and WAIS-IV performance in adolescents and adults
diagnosed with Asperger’s syndrome and autism. Assessment,
18, 192-200. doi:10.1177/1073191110394771
Hughes, C., Kumari, V., Soni, W., Das, M., Binneman, B., Drozd,
S., & Sharma, T. (2002). Longitudinal study of symptoms and
cognitive function in chronic schizophrenia. Schizophrenia
Research, 59, 137-146. doi:10.1016/S0920-9964(01)00393-0
Jeste, D. V., Wolkowitz, O. M., & Palmer, B. W. (2011). Diver-
gent trajectories of physical, cognitive, and psychosocial
aging in schizophrenia. Schizophrenia Bulletin, 37, 451-455.
doi:10.1093/schbul/sbr026
Johnson-Selfridge, M., & Zalewski, C. (2001). Moderator vari-
ables of executive functioning in schizophrenia: Meta-analytic
findings. Schizophrenia Bulletin, 27, 305-316.
Jones, R., Guth, L., Lewis, S., & Murray, R. (1994). Low intel-
ligence and poor educational achievement precede early onset
psychosis. In A. S. David & J. C. Cathy (Eds.), The neuro-
psychology of schizophrenia (pp. 131-144). Hove, England:
Lawrence Erlbaum.
Michel et al. 473
Jortner, S. (1970). Overinclusion responses to WAIS similarities
as suggestive of schizophrenia. Journal of Clinical Psychol-
ogy, 26, 346-348. doi:10.1002/1097-4679(197007)
Joyce, E., & Huddy, V. (2004). Defining the cognitive impairment
in schizophrenia. Psychological Medicine, 34, 1151-1155.
doi:10.1017/S0033291704003472
Kay, S. R., Fiszbein, A., & Opler, L. A. (1987). The positive
and negative syndrome scale (PANSS) for schizophrenia.
Schizophrenia Bulletin, 13(2), 261-276. doi: 10.1093/schbul/
13.2.261
Keefe, R. S. E., & Fenton, W. S. (2007). How should DSM-V cri-
teria for schizophrenia include cognitive impairment? Schizo-
phrenia Bulletin, 33, 912-920. doi:10.1093/schbul/sbm046
Knowles, E. E. M., David, A. S., & Reichenberg, A. (2010). Pro-
cessing speed deficits in schizophrenia: Reexamining the
evidence. American Journal of Psychiatry, 167, 828-835.
doi:10.1176/appi.ajp.2010.09070937
Lee, J., & Park, S. (2005). Working memory impairments in
schizophrenia: A meta-analysis. Journal of Abnormal Psychol-
ogy, 114, 599-611. doi:10.1037/0021-843X.114.4.599
Lezak, M. D. (1995). Neuropsychological assessment (3rd ed.).
New York, NY: Oxford University Press.
Lichtenberger, E. O., & Kaufman, A. S. (2009). Essentials of
WAIS-IV assessment. New York, NY: Wiley.
Maylor, E. A., Reimers, S., Choi, J., Collaer, M. L., Peters, M.,
& Silverman, I. (2007). Gender and sexual orientation differ-
ences in cognition across adulthood: Age is kinder to women
than to men regardless of sexual orientation. Archives of Sex-
ual Behavior, 36, 235-249. doi:10.1007/s10508-006-9155-y
McCrae, S. M., & Robinson, T. P. (2011). Visual puzzles, figure
weights, and cancellation: Some preliminary hypotheses on
the functional and neural substrates of these three new WAIS-
IV subtests. International Scholarly Research Network, 2011,
1-19. doi:10.5402/2011/123173
Nestor, P. G., Kubicki, M., Nakamura, M., Niznikiewicz, M.,
McCarley, R. W., & Shenton, M. E. (2010). Comparing pre-
frontal gray and white matter contributions to intelligence and
decision making in schizophrenia and healthy controls. Neuro-
psychology, 24, 121-129. doi:10.1037/a0016981
Nestor, P. G., Niznikiewicz, M., & McCarley, R. W. (2010).
Distinct contributions of working memory and social com-
prehension failures in neuropsychological impairment in
schizophrenia. Journal of Nervous and Mental Disorders, 198,
206-212. doi:10.1097/NMD.0b013e3181d14143
Nuechterlein, K. H., Barch, D. M., Gold, J. M., Goldberg, T.
E., Green, M. F., & Heaton, R. K. (2004). Identification of
separable cognitive factors in schizophrenia. Schizophrenia
Research, 72, 29-39. doi:10.1016/j.schres.2004.09.007
Psychological Corporation. (1997). WAIS-III WMS-III technical
manual. San Antonio, TX: Author.
Reichenberg, A., & Harvey, P. D. (2007). Neuropsychological
impairment in schizophrenia: Integration of performance-
based and brain imaging findings. Psychological Bulletin,
133, 833-858. doi:10.1037/0033-2909.133.5.833
Revheim, N., Schechter, I., Kim, D., Silipo, G., Allingham, B.,
Butler, P., & Javitt, D. C. (2006). Neurocognitive and symp-
tom correlates of daily problem-solving skills in schizophre-
nia. Schizophrenia Research, 83, 237-245. doi:10.1016/j.
schres.2005.12.849
Rund, B. R. (1998). A review of longitudinal studies of cognitive
function in schizophrenia patients. Schizophrenia Bulletin, 24,
425-435.
Ryan, J. J., & Glass, L. A. (2010). Substitution of supplementary
subtests for core subtests on composite reliability of WAIS-
IV indexes. Psychological Reports, 106, 13-18. doi:10.2466/
PR0.106.1.13-18
Sota, T. L., & Heinrichs, R. W. (2003). Sex differences in verbal
memory in schizophrenia patients treated with “typical” neu-
roleptics. Schizophrenia Research, 62, 175-182. doi:10.1016/
S0920-9964(02)00373-0
Stratta, P., Prosperini, P., Daneluzzo, E., Bustini, M., & Rossi, A.
(2001). Educational level and age influence spatial work-
ing memory and Wisconsin Card Sorting Test perfor-
mance differentially: A controlled study in schizophrenic
patients. Psychiatry Research, 102, 39-48. doi:10.1016/
S0165-1781(01)00230-X
Walker, A. J., Batchelor, J., & Shores, A. (2009). Effects of edu-
cation and cultural background on performance on WAIS-
III, WMS-III, WAIS-R and WMS-R measures: Systematic
review. Australian Psychologist, 44, 216-223. doi:10.1080/
00050060902833469
Wechsler, D. (1997). Wechsler Adult Intelligence Scale—Third
edition. San Antonio, TX: Psychological Corporation.
Wechsler, D. (2008). Wechsler Adult Intelligence Scale—Fourth
edition. San Antonio, TX: Pearson Assessment.
Wechsler, D., Coalson, D. L., & Raiford, S. E. (2008). WAIS-
IV technical and interpretive manual. San Antonio, TX:
Pearson.
Whelton, C., Pawlick, J., & Cook, P. E. (1999). Growing with
people with a psychiatric disability in a psychosocial rehabili-
tation program. Psychiatric Rehabilitation, 22, 290-293.
Williams, L. M., Whitford, T. J., Flynn, G., Wong, W., Liddell, B.
J., Silverstein, S., & Gordon, E. (2008). General and social
cognition in first episode schizophrenia: Identification of sep-
arable factors and prediction of functional outcome using the
IntegNeuro test battery. Schizophrenia Research, 99, 182-191.
doi:10.1016/j.schres.2007.10.019
Woods, S. W. (2003). Chlorpromazine equivalent doses for the
newer atypical antipsychotics. Journal of Clinical Psychiatry,
64, 663-667.