Content uploaded by Jeremy M Silverman
Author content
All content in this area was uploaded by Jeremy M Silverman on Mar 12, 2014
Content may be subject to copyright.
ROITMAN, CORNBLATT, BERGMAN, ET AL.ATTENTIONAL FUNCTIONI NGAm J Psychiatry 154:5, May 1997
Attentional Functioning in Schizotypal Personality Disorder
Sonia E. Lees Roitman, B.S., Barbara A. Cornblatt, Ph.D., Andrea Bergman, Ph.D.,
Michael Obuchowski, Ph.D., Vivian Mitropoulou, M.A., Richard S.E. Keefe, Ph.D.,
Jeremy M. Silverman, Ph.D., and Larry J. Siever, M.D.
Objective: Previous research has shown biological, phenomenological, and cognitive similari-
ties between schizophrenic patients and individuals with schizophrenia-related personality dis-
orders and features. Evidence further suggests that of these common dysfunctions, abnormal
attention is one of the most promising indicators of a biological susceptibility to schizophrenia-
related disorders. Although attentional dysfunctions have been reliably detected in schizophrenic
patients as well as in a variety of populations at risk for schizophrenia, few studies have investi-
gated attention in clinical patients with schizotypal personality disorder. In this study, the extent
of attentional impairment was assessed in subjects with schizotypal personality disorder, normal
comparison subjects, patients with other personality disorders, and schizophrenic patients.
Method: Thirty subjects with schizotypal personality disorder, 35 subjects with other personality
disorders (i.e., clinic patients with non-odd cluster personality disorders), 36 subjects with schizo-
phrenia, and 20 comparison subjects who did not meet criteria for any axis I or axis II disorder
participated in this study. All subjects were diagnosed according to DSM-III criteria. Attention
was assessed by using the Continuous Performance Test, Identical Pairs Version. Results: Analy-
ses indicated that subjects with schizotypal personality disorder, like schizophrenic subjects, per-
formed significantly worse than comparison subjects on both the verbal and spatial tasks of the
Continuous Performance Test, Identical Pairs Version. In contrast, patients with other person-
ality disorders performed similarly to comparison subjects across conditions. Conclusions: These
results suggest that patients with schizotypal personality disorder are impaired in their atten-
tional functioning relative to normal comparison subjects and that they display deficits that are
similar to the pattern characterizing schizophrenic patients.
(Am J Psychiatry 1997; 154:655–660)
It has been suggested that the boundaries of schizo-
phrenia extend beyond the core psychotic illness to
include milder schizophrenia-related personality disor-
ders and that the genetic and neurocognitive factors im-
plicated in schizophrenia are expressed across the full
range of schizophrenia-related illnesses. Furthermore, re-
sults from family studies suggest that schizotypal person-
ality disorder, the prototype for the milder schizophre-
nia-spectrum personality disorders, may be a more
common phenotypic expression of the schizophrenia-re-
lated genotype than chronic schizophrenia (1–4). Because
of the genetic relationship between schizophrenia and
schizotypal personality disorder, the study of the phe-
nomenology, biology, and neurocognition of patients
with schizotypal personality disorder provides an op-
portunity to investigate the pathophysiology of schizo-
phrenia without the confounding factors (such as medi-
cation effects, hospitalization, and psychosis) often
found in research with chronic schizophrenic patients.
Although attention, as measured by the Continuous
Performance Test, has been identified as a potential
phenotypic indicator of schizophrenia (5, 6), atten-
tional functioning in subjects with schizotypal person-
ality disorder has not been extensively studied. Atten-
tional impairments have been found in both actively
psychotic schizophrenic patients and those in remission
(7–9), as well as in first-degree relatives of schizo-
phrenic individuals (10–12) and psychometrically de-
fined schizotypic volunteers (13). These studies suggest
that abnormal attention is centrally related to the biol-
ogy of schizophrenia spectrum disorders and raise the
possibility that attentional difficulties may provide a
biobehavioral indicator of liability to schizophrenia (5).
Received March 25, 1996; revision received Dec. 3, 1996; accepted
Dec. 18, 1996. From the Department of Psychiatry, Mount Sinai
School of Medicine, New York; Bronx VA Medical Center, Bronx,
N.Y.; Psychiatry Research, Hillside Hospital; and Department of Psy-
chology, St. John’s University, Flushing, N.Y. Address reprint re-
quests to Dr. Cornblatt, Psychiatry Research, Hillside Hospital, 75-59
263rd St., Glen Oaks, NY 11004; cornblat@lij.edu (e-mail).
Supported in part by NIH grant RR-00071 from the National Cen-
ter for Research Resources to the Mount Sinai Medical Center, NIMH
grant MH-41131 and Department of Veterans Affairs merit award
7609004 to Dr. Siever, and a research grant award from the Scottish
Rite Schizophrenia Research program, Northern Masonic Jurisdic-
tion, to Dr. Cornblatt.
Am J Psychiatry 154:5, May 1997 655
Two studies have looked at Continuous Performance
Test scores in subjects with schizotypal personality disor-
der who were diagnosed through use of DSM criteria. In
one study, schizotypal individuals with and without a
family history of schizophrenia were compared on a sim-
ple single-digit version of the Continuous Performance
Test (14). Individuals with schizotypal personality disor-
der and a family history of schizophrenia were identified
from a sample of male siblings of male schizophrenic pro-
bands. Those without a family history of schizophrenia
were recruited through newspaper advertisements. Both
groups of patients with schizotypal personality disorder
showed attentional deficits compared to normal subjects.
However, neither group contained patients recruited
from clinical settings, and studies have shown that
schizotypal volunteers are often less impaired (15) and
etiologically more heterogeneous than clinic populations
(16). Furthermore, because that study (14) contained no
comparison group whose members manifested psycho-
pathology outside of the schizophrenia spectrum, it is dif-
ficult to differentiate the influences of general psychopa-
thology from those specifically associated with
schizophrenia-related disorders.
In a second study that has looked at attentional im-
pairments in a DSM-III-diagnosed schizotypal popula-
tion, a simple, low processing load version of the Con-
tinuous Performance Test was administered to patients
with personality disorders (schizotypal personality dis-
order and other personality disorders) and comparison
subjects recruited at our center (17). On the nonde-
graded version of the Continuous Performance Test,
subjects with schizotypal personality disorder and
other personality disorders performed similarly, both
making few errors of omission or commission. But
when the target stimuli were degraded, patients with
schizotypal personality disorder made significantly
more errors of omission than patients with other per-
sonality disorders (17). The results of this study suggest
that clinically identified patients with schizotypal per-
sonality disorder exhibit attentional impairments that
are less severe than those of schizophrenic patients and
that these impairments may only be apparent when
there is an increased perceptual processing load.
On the basis of these preliminary data with an easier
version of the Continuous Performance Test, we wanted
to more thoroughly explore the attentional deficits exhib-
ited by schizotypal patients by using a more difficult test
that taps into higher-level attentional processes. In con-
trast to the variant of the Continuous Performance Test
used in the study by Harvey et al. (17), the present study
used the Continuous Performance Test, Identical Pairs
Version (18, 19), which is a more complex attentional
measure that focuses on cognitive, as opposed to percep-
tual, processing and on working memory. This version of
the test was specifically developed to cognitively chal-
lenge clinically unaffected subjects who show subtle in-
formation processing disturbances associated with the bi-
ology of schizophrenia (20). The Continuous
Performance Test, Identical Pairs Version, has been used
extensively to demonstrate the potential of impaired at-
tention as a phenotypic marker of schizophrenia. Chil-
dren at genetic risk for developing schizophrenia (10), as
well as students with high perceptual aberration scores
and therefore considered at psychometric risk for schizo-
phrenia (13), have been found to perform deficiently on the
Continuous Performance Test, Identical Pairs Version.
On the basis of previous work using the Continuous
Performance Test, Identical Pairs Version, our primary
hypotheses were that 1) schizotypal subjects would
show a global impairment (i.e., in both verbal and spa-
tial processing) in their attentional capacities when con-
trasted with normal comparison subjects, 2) subjects
with other personality disorders would not be impaired
relative to a normal comparison group, and 3) subjects
with schizotypal personality disorder would be im-
paired compared to subjects with other personality dis-
orders. As shown in previous studies, our secondary hy-
pothesis was that schizophrenic patients would show a
global attentional impairment when contrasted with
normal comparison subjects.
METHOD
Subjects
All subjects in this study were part of an ongoing program investi-
gating mood and personality disorders. They were identified from the
outpatient and inpatient clinical units of the Bronx Veterans Affairs
Medical Center and Mount Sinai Hospital or were referred by local
practitioners.
The group was diagnosed with DSM-III criteria and consisted of
30 subjects with schizotypal personality disorder, 35 subjects with
another non-odd cluster personality disorder (excludes those patients
who also met criteria for schizotypal, schizoid, and paranoid person-
ality disorders), 36 subjects with schizophrenia, and a normal com-
parison group of 20 subjects. DSM-III criteria were used for subjects
with personality disorders because the study was initiated before
structured interviews for axis II diagnoses of DSM-III were estab-
lished. Comparison subjects were free of a personal or family history
of axis II disorder (see section on diagnostic assessment). Exclusion
criteria for all subjects consisted of systemic medical illness, deter-
mined on the basis of history and results of laboratory tests and physi-
cal examination; a history of substance abuse in the past 6 months;
past history of substance dependence ; any history of intravenous drug
abuse; or a history of long-term serious alcohol abuse, positive results
on an initial urine toxicology screen, or positive results on a test to
determine the alcohol content of a breath sample. Written informed
consent was obtained from all subjects before testing.
Demographic information is presented in table 1. The four groups
did not differ significantly on age. Subjects with schizotypal personality
disorder, subjects with other personality disorders, and comparison
subjects did not differ on education level, whereas schizophrenic pa-
tients had significantly fewer years of education than the other groups.
All subjects, except those in the schizophrenic group, were medi-
cation free for at least 2 weeks before testing (6 weeks for those pa-
tients who had previously taken fluoxetine). Schizophrenic patients
were all stabilized on a regimen of a standard neuroleptic drug for at
least 2 weeks. Subjects were asked to abstain from alcohol for at least
24 hours before testing and were informed that on the day of testing
they would be given a test to determine the alcohol content of a breath
sample. All subjects had negative results on this test.
Diagnostic Assessment
Patients were evaluated for axis I pathology according to the Re-
search Diagnostic Criteria (RDC) by two experienced raters, who
ATTENTIONAL FUNCTIONING
656 Am J Psychiatry 154:5, May 1997
were blind to the subject’s performance on the Continuous Perform-
ance Test and who used the Schedule for Affective Disorders and
Schizophrenia (21). Patients who met the RDC or DSM-III axis I cri-
teria for major psychiatric syndromes other than schizophrenia, met
criteria for current substance abuse, or had a current or past history
of substance dependence were excluded from the study. Patients in-
cluded in the schizophrenic group were inpatients who met DSM-III
criteria for schizophrenia.
All patients with personality disorders were studied as outpatients
and were interviewed by one rater with the Structured Interview for
DSM-III Personality Disorders (22). If a patient with personality disor-
der met criteria for major depressive disorder, he or she was included
only if it was determined that the depression was not responsible for the
impaired interpersonal functioning and schizophrenia-related symp-
toms. This was assessed in several ways. First, the Structured Interview
for DSM-III Personality Disorders is an instrument that assesses lifetime
personality traits independent of affective symptoms, and second, we
include supplemental questions for the schizotypal criteria (formulated
by L.J.S. and J.M.S.) in order to differentiate between social withdrawal
due to paranoid fears or due to problems with self-esteem. A second
rater administered an independent Structured Interview for DSM-III
Personality Disorders to an individual close to the patient. Consensus
diagnoses were reached in a meeting of all raters with an expert diag-
nostician (J.M.S.) (kappa=0.73 for schizotypal personality disorder).
The group of patients with other personality disorders consisted of
patients in the following diagnostic subgroups: histrionic personality
disorder (N=14, 40%), borderline personality disorder (N=13, 37%),
and compulsive personality disorder (N=9, 26%) (Only these person-
ality disorders representing 25% or more of the other personality
disorders group are reported; the percentages total more than 100%,
since some patients met criteria for more than one personality disor-
der. Patients with schizotypal personality disorder had an average of
2.30 personality disorders [SD=1.34]; patients with other personality
disorders had an average of 1.54 personality disorders [SD=0.78]).
The Beck Depression Inventory (23), an assessment of severity of cur-
rent depression, was also given to all subjects the day of testing.
Continuous Performance Testing
The Continuous Performance Test, Identical Pairs Version, was
administered to all subjects. This version of the test is a computerized
measure of sustained attention that consists of two sets of complex
stimuli: nonsense shapes and four-digit numbers. The numbers and
shapes conditions are psychometrically matched in adults for diffi-
culty level and error variance and are considered to assess verbal
(numbers) versus spatial (shapes) attentional processing (5, 19).
Numbers are considered verbal stimuli because they are read from left
to right and have no arithmetic properties. Nonsense shapes are con-
sidered to be spatial stimuli resistant to verbal labeling. In this study,
the conditions administered included the basic 150 trial numbers con-
dition and the basic 150 trial shapes condition.
The stimuli are flashed on a video monitor at a constant rate of 1 per
second, with each stimulus exposed for 50 msec. The subject’s task is to
respond by lifting his or her finger from a reaction time key as quickly
as possible whenever two identical stimuli are presented in a row. There
are 30 target pairs requiring a response in each condition. Responses to
the target trials (the second stimulus of the target pair) are considered
correct detections or hits (20% of the total number of trials). Each con-
dition also contains 30 “catch” pairs, in which the two stimuli in a row
are almost the same but not completely identical (i.e., they generally
differ by one digit in the numbers condition and one feature in the
shapes condition). A response to the second stimulus of a catch pair is
labeled a false alarm, a type of commission error. The remaining trials
in each condition are randomly organized fillers. Responses on any of
these trials are considered to be random commission errors.
The primary Continuous Performance Test, Identical Pairs Version,
performance index used in this report is d′. This is one of two indices
derived from signal detection theory (24) and measures an individual’s
ability to discriminate a signal from background noise. The second in-
dex is beta (typically expressed as natural logB) and is a measure of a
subject’s response bias or tendency to overrespond versus underrespond
on a task. In general, d′ is typically considered a measure of attentional
capacity, while logB appears more directly related to motivational state
(5). Previous studies of attention suggest that d′ is more reliable than
logB (19). The higher the d′, the better the attentional performance; the
higher the logB, the more conservative the response style.
Statistical Analysis
Chi-square tests were performed to compare the two groups with
personality disorders (schizotypal personality disorder and other per-
sonality disorders) on demographic and clinical characteristics. All hy-
potheses were tested through use of a group- (schizotypal personality
disorder, other personality disorders, comparison subjects, schizo-
phrenic patients) by-stimuli (four-digit numbers, shapes) repeated meas-
ures analysis of variance (ANOVA) with planned contrasts (25). In this
report, only d′ will be presented in detail, since this index is considered
to provide the most direct assessment of attentional processing capacity.
TABLE 1. Demographic and Clinical Characteristics of Normal Comparison Subjects and of Patients With Schizotypal Personality Disorder,
Other Personality Disorders, and Schizophrenia
Patients
Characteristic
Schizotypal
Personality
Disorder
(N=30)
Other
Personality
Disorders
(N=35) Schizophrenia
(N=36)
Normal
Comparison
Subjects
(N=20)
Mean SD Mean SD Mean SD Mean SD
Age (years) 37.9 9.9 39.3 11.6 39.9 10.6 37.7 10.2
Education (years)a13.8 2.7 14.8 2.5 12.8 1.1 15.4 2.1
Number of previous hospitalizations 0.53 0.96 0.34 1.00 — —
Beck Depression Inventory score 15.5 8.9 18.7 7.8 — — 2.9 4.3
N% N % N% N%
Genderb
Male 22 17 32 9
Female 8 18 4 11
Never took neuroleptic medication 18 60 30 86 0 0
Never took antidepressant medication 16 53 17 49 — —
Never took other psychiatric medication 20 67 22 63 — —
Ever depressedc14 47 29 83 — —
aF=3.5, df=117, p<0.05. bχ2=14.7, df=3, p<0.01. cχ2=7.9, df=1, p<0.01.
ROITMAN, CORNBLATT, BERGMAN, ET AL.
Am J Psychiatry 154:5, May 1997 657
Correct response and false alarms are not directly discussed, since pre-
liminary analyses indicated that they follow the same pattern as d′ and
are therefore redundant. Means and standard deviations are presented
for logB but not discussed, since, as indicated earlier, this index is con-
sidered to reflect unstable motivational factors that are not the focus of
this study. On the basis of our hypotheses, individual F tests were used
(25) to evaluate the following planned comparisons: 1) subjects with
schizotypal personality disorder and normal volunteers, 2) subjects with
other personality disorders and normal volunteers, and 3) subjects with
schizotypal personality disorder and subjects with other personality dis-
orders. An additional planned comparison (schizophrenic patients and
comparison subjects) was included to assess the consistency of the cur-
rent data with previous findings in the literature. All analyses performed
use an alpha level of 0.05 and are two-tailed even though we have spe-
cific hypotheses.
RESULTS
Demographic and Clinical Variables
Demographic and clinical variables are reported in table
1. Age did not differ among the groups; however, since
schizophrenic patients had fewer years of education than
any of the other groups, we covaried for years of educa-
tion. The gender distribution was significantly different
among the groups, but gender did not have a significant
effect on performance (F=0.90, df=1, 4). Since patients
with personality disorders (schizotypal personality disor-
der and other personality disorders) had Beck Depression
Inventory scores that were significantly higher than those
of normal comparison subjects, we examined the rela-
tionship between depression and test performance. Cur-
rent depressive symptoms (total score on the Beck Depres-
sion Inventory) were not significantly related to task
performance (r=–0.08). Furthermore, among patients
with personality disorders, there was no statistical differ-
ence in scores on the Continuous Performance Test be-
tween those with a comorbid
history of depression (N=22,
mean=1.7, SD=0.9) and those
without such a history (N=43,
mean= 1.7, SD=0.9) (t=0.39,
df=63). There were also no
significant differences be-
tween patients with other
personality disorders and
those with schizotypal per-
sonality disorder in number
of psychiatric hospitaliza-
tions or history of anti-
depressant medication (table
1). In addition, scores on the
Continuous Performance Test
and days without neurolep-
tic medication were not sig-
nificantly related to one an-
other (r=0.17), and neither
were scores on the Continu-
ous Performance Test and
total number of schizotypal
traits (r=0.13).
Group Differences in Continuous Performance Test
Scores
The mean performance of each group is reported in
table 2. Confidence intervals are reported in figures 1 and
2. Subjects with schizotypal personality disorder per-
formed significantly worse overall than both subjects
with other personality disorders and comparison subjects
(p<0.05 and p<0.001, respectively). Subjects with other
personality disorders and comparison subjects did not
differ significantly in their attentional performance (F=
3.60, df=1, 117). As has been found in previous studies,
schizophrenic patients were significantly impaired on
the Continuous Performance Test when contrasted with
comparison subjects (p<0.001). There was no significant
difference in the performance between the two stimulus
conditions (four digits, shapes; F=0.27, df=1, 117) and
no interaction between group membership and type of
stimulus. The findings remained significant when we co-
varied for years of education (F=8.48, df=3, 104, p<0.001)
and depression (F=4.85, df=2, 73, p=0.01).
Figures 1 and 2 present 95% confidence intervals for
the numbers and shapes conditions, respectively, by
group. These figures indicate that although there were
clear group differences in the hypothesized directions,
there was some degree of overlap between contiguous
groups. This is consistent with the expected heterogene-
ity within the schizotypal personality disorder and
other personality disorders groups and with the overlap
between groups in personality disorder features and co-
morbid diagnoses.
Because a number of axis II disorders that character-
ized the other personality disorders group were also sec-
ondary disorders among patients with schizotypal per-
sonality disorder, a series of t tests were conducted
TABLE 2. Performance of Normal Comparison Subjects and of Patients with Schizophrenia, Schizo-
typal Personality Disorder, and Other Personality Disorders on the Four-Digit and Shape Conditions of
the Continuous Performance Test, Identical Pairs Version
Patients
Schizotypal
Personality
Disordera
(N=30)
Other
Personality
Disorders
(N=35) Schizophreniab
(N=36)
Normal
Comparison
Subjects
(N=20)
Test Condition Mean SD Mean SD Mean SD Mean SD
d′Four digits 1.52 0.81 1.90 0.84 1.13 0.81 2.30 1.00
Shapes 1.55 0.74 1.86 0.85 1.07 0.57 2.19 0.93
Hits
Four digits 0.65 0.22 0.71 0.20 0.57 0.23 0.84 0.17
Shapes 0.69 0.21 0.77 0.16 0.61 0.20 0.84 0.18
False alarms
Four digits 0.16 0.11 0.13 0.10 0.21 0.12 0.14 0.14
Shapes 0.21 0.15 0.18 0.10 0.25 0.12 0.17 0.10
LogB
Four digits 0.28 0.79 0.34 0.80 0.38 0.71 0.07 0.68
Shapes 0.18 0.90 0.09 0.79 0.13 0.60 –0.45 0.79
aPatients with schizotypal personality disorder performed significantly worse than patients with other
personality disorders and normal subjects (F=4.33, df=1, 117, p<0.05, and F=13.39, df=1, 117, p<
0.001, respectively).
bSchizophrenic patients performed significantly worse than normal subjects (F=37.32, df=1, 117, p<0.001).
ATTENTIONAL FUNCTIONING
658 Am J Psychiatry 154:5, May 1997
among the entire personality disorder cohort in order to
determine whether attentional impairment was found
among any other (non-odd cluster) personality disorder
diagnosis significantly represented in our group. The per-
formance of patients who met criteria for borderline per-
sonality disorder did not differ from that of patients who
did not meet criteria for the disorder (t=0.33, df=63). The
performance of patients who met criteria for histrionic
personality disorder did not differ significantly from that
of patients who did not meet criteria for the disorder
(t=1.8, df=63). In addition, Continuous Performance
Test scores of patients with compulsive personality dis-
order did not differ significantly from scores of patients
who did not meet criteria for the disorder (t=1.3, df=63).
DISCUSSION
To our knowledge, this study is the first to assess,
through use of a high cognitive processing load version
of the Continuous Performance Test, sustained attention
in clinical patients diagnosed with DSM-III schizotypal
personality disorder. Consistent with previous findings
from our laboratory that used a simpler version of the
Continuous Performance Test (17), we found that DSM-
III-defined schizotypal subjects showed significant im-
pairment in attention, as measured by d′, in contrast to
normal comparison subjects and patients with non-
schizophrenia-related personality disorders. Patients
with schizotypal personality disorder showed atten-
tional deficits similar to but less severe than those seen
in medicated schizophrenic patients.
In contrast to patients with schizotypal personality dis-
order, subjects with non-odd cluster DSM-III personality
disorders performed similarly to comparison subjects,
providing support for the hypothesis that sustained at-
tention deficits are specifically associated with member-
ship in the schizophrenia spectrum and are not related to
general psychopathology. Findings from studies of at-risk
populations, such as relatives of schizophrenic individu-
als (11, 12, 14) and psychometrically defined schizotypic
individuals (13), are consistent with this hypothesis. Taken
together, these findings also indicate that attentional defi-
cits may serve as a marker of biological susceptibility to
schizophrenia, even in nonpsychotic populations.
Attentional impairment on the Continuous Perform-
ance Test can be viewed as a continuum (figures 1 and
2), with impairment monotonously worsening from the
comparison group through the patients with other per-
sonality disorders and from patients with schizotypal
personality disorder to schizophrenic patients. While
there may be overlap between contiguous groups, this
overlap is consistent with the heterogeneity within these
groups, e.g., the other personality disorders group may
include some patients with schizotypal personality dis-
order traits, as well as some patients indistinguishable
from comparison subjects, and the schizotypal person-
ality disorder group may include patients with a severe
spectrum of attentional impairment similar to that seen
in schizophrenia, as well as some patients with a milder
impairment similar to the one seen in the other person-
ality disorders group.
This impairment may play an increasingly important
role in the development of the symptoms that character-
ize schizophrenia-related disorders as the environment
makes greater demands on these vulnerable individuals.
It is then that they are more likely to fail. On the basis of
the theories proposed by Cornblatt et al. (5, 11) and
Siever and Davis (26), an individual who has difficulties
processing incoming information may have particular
difficulties in social situations, where interpersonal cues
and communications are often quite complicated and
subtle. This social deficit may, in turn, lead the individual
to withdraw from most social interactions, thus leading
FIGURE 1. 95% Confidence Intervals for Four-Digit Condition of the
Continuous Performance Test for Patients With Schizophrenia,
Schizotypal Personality Disorder, and Other Personality Disorders
and for Normal Comparison Subjects
FIGURE 2. 95% Confidence Intervals for Shapes Condition of the
Continuous Performance Test for Patients With Schizophrenia,
Schizotypal Personality Disorder, and Other Personality Disorders
and for Normal Comparison Subjects
ROITMAN, CORNBLATT, BERGMAN, ET AL.
Am J Psychiatry 154:5, May 1997 659
to the asociality that is considered a core characteristic of
the schizophrenia spectrum disorders.
In this group, subjects with schizotypal personality
disorder and subjects with other personality disorders
had similar mean levels of depression but were more de-
pressed than comparison subjects. In both groups, self-
reported depression was not significantly related to atten-
tional deficit, suggesting that subjective state factors do
not seem to have a global impact on attentional abilities
in either group of patients. Other factors, such as history
of medication (antipsychotic and antidepressant) and
number of previous hospitalizations, also did not seem to
have a significant effect on the Continuous Performance
Test scores of patients with schizotypal personality dis-
order or other personality disorders. Schizophrenic pa-
tients were included as an additional secondary compari-
son group, although they were studied while they were
taking medication and were inpatients, while the person-
ality disorder patients were studied while they were
not taking medication and were outpatients. Studies
have shown that neuroleptic medication does not affect
scores on high processing load Continuous Performance
Test, Identical Pairs Version (10, 27), and acute hospital-
ized schizophrenic patients show impairment similar to
that seen in remitted schizophrenic patients (8, 9). Thus,
there is little evidence that clinical (e.g., acute psychosis
and neuroleptic use) and demographic characteristics
account for the schizophrenic patients’ poor perform-
ance. In fact, there is evidence suggesting that impaired
attention is a stable trait that is relatively invariant across
development and clinical states (5).
In conclusion, the results of this study suggest that
schizotypal personality disorder and schizophrenia are
characterized by a common global attentional impair-
ment that varies in severity. The deficits found in this
study, in combination with studies suggesting that cogni-
tive impairment in schizophrenia may be a better predic-
tor of social functioning than symptom measures (28,
29), provide further evidence that attentional impairment
may be a core deficit of the schizophrenia spectrum.
REFERENCES
1. Kety SS, Rosenthal D, Wender PH: Mental illness in the biologi-
cal and adoptive families of adopted individuals who have be-
come schizophrenic: a preliminary report based on psychiatric
interviews, in Genetic Research in Psychiatry. Edited by Fieve R,
Rosenthal D, Brill H. Baltimore, Johns Hopkins Press, 1975, pp
147–165
2. Baron M, Gruen R, Asnis L, Kane J: Familial relatedness of
schizophrenia and schizotypal states. Am J Psychiatry 1983; 140:
1437–1442
3. Siever LJ, Silverman JM, Horvath TB, Klar H, Coccaro EF, Keefe
RSE, Pinkham L, Rinaldi P, Mohs RC, Davis KL: Increased mor-
bid risk for schizophrenia-related disorders in relatives of schizo-
typal personality disordered patients. Arch Gen Psychiatry 1990;
47:634–640
4. Kendler KS, Gruenberg AM, Strauss JS: An independent analysis
of the Copenhagen sample of the Danish adoption study of
schizophrenia. Arch Gen Psychiatry 1981; 38:982–987
5. Cornblatt BA, Keilp JG: Impaired attention, genetics, and patho-
physiology of schizophrenia. Schizophr Bull 1994; 20:31–46
6. Keefe RSE, Silverman JM, Cornblatt BA: Refining phenotype
characterization in genetic linkage studies of schizophrenia. Soc
Biol 1992; 38:197–218
7. Nuechterlein KH, Dawson ME: Information processing and at-
tentional functioning in the developmental course of schizo-
phrenic disorders. Schizophr Bull 1984; 10:160–202
8. Wohlberg GW, Kornetsky C: Sustained attention in remitted
schizophrenics. Arch Gen Psychiatry 1973; 28:533–537
9. Asarnow RF, MacCrimmon DJ: Residual performance deficit in
clinically remitted schizophrenics: a marker of schizophrenia? J
Abnorm Psychol 1978; 87:597–608
10. Erlenmeyer-Kimling L, Cornblatt BA: A summary of attentional
findings in the New York High-Risk Project. J Psychiatr Res
1992; 26:405–426
11. Cornblatt BA, Lenzenweger MF, Dworkin RH, Erlenmeyer-
Kimling L: Childhood attentional dysfunctions predict social
deficits in unaffected adults at risk for schizophrenia. Br J Psy-
chiatry 1992; 161(suppl 18):59–64
12. Nuechterlein KH: Signal detection in vigilance tasks and behav-
ioral attributes among offspring of schizophrenic mothers and
hyperactive children. J Abnorm Psychol 1983; 92:4–28
13. Lenzenweger MF, Cornblatt BA, Putnick M: Schizotypy and sus-
tained attention. J Abnorm Psychol 1991; 100:84–89
14. Condray R, Steinhauer SR: Schizotypal personality disorder in in-
dividuals with and without schizophrenic relatives: similarities and
contrasts in cognitive functioning. Schizophr Res 1992; 7:33–41
15. Thaker G, Moran M, Lahti A, Adami H, Tamminga C, Schulz
SC: Pilot studies in schizotypal subjects, in Schizophrenia Re-
search, vol 1: Advances in Neuropsychiatry and Psychopharma-
cology. Edited by Tamminga CA, Schultz SC. New York, Raven
Press, 1991, pp 201–208
16. Chapman LJ, Chapman JM, Kwapil TR, Eckblad M, Zinser MC:
Putatively psychosis-prone subjects 10 years later. J Abnorm
Psychol 1994; 103:171–183
17. Harvey PD, Keefe RSE, Mitropoulou V, DuPre R, Lees Roitman
SE, Mohs RC, Siever LJ: Attentional markers of vulnerability to
schizophrenia: performance of patients with schizotypal and
nonschizotypal personality disorders. Psychiatry Res 1996; 60:
49–56
18. Cornblatt BA, Risch NJ, Faris G, Friedman D, Erlenmeyer-Kim-
ling L: The Continuous Performance Test, Identical Pairs Ver-
sion (CPT-IP), I: new findings about sustained attention in nor-
mal families. Psychiatry Res 1988; 26:223–238
19. Cornblatt BA, Lenzenweger MF, Erlenmeyer-Kimling L: The
Continuous Performance Test, Identical Pairs Version, II: con-
trasting attentional profiles in schizophrenic and depressed pa-
tients. Psychiatry Res 1989; 29:65–85
20. Lenzenweger MF, Loranger AW: Psychosis proneness and clini-
cal psychopathology: examination of the correlates of schizo-
typy. J Abnorm Psychol 1989; 98:3–8
21. Spitzer RL, Endicott J: Schedule for Affective Disorders and
Schizophrenia (SADS), 3rd ed. New York, New York State Psy-
chiatric Institute, Biometrics Research, 1978
22. Stangl D, Pfohl B, Zimmerman M: A structured interview for
DSM-III personality disorders: a preliminary report. Arch Gen
Psychiatry 1985; 42:591–596
23. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J: An in-
ventory for measuring depression. Arch Gen Psychiatry 1961;
4:561–571
24. Rutschmann J, Cornblatt BA, Erlenmeyer-Kimling L: Sustained
attention in children at risk for schizophrenia: report on a con-
tinuous performance test. Arch Gen Psychiatry 1977; 34:571–575
25. Winer BJ: Statistical Principles in Experimental Design, 2nd ed.
New York, McGraw-Hill, 1971
26. Siever LJ, Davis KL: A psychobiological perspective on the per-
sonality disorders. Am J Psychiatry 1991; 148:1647–1658
27. Earle-Boyer EA, Serper MR, Davidson M, Harvey PD: Auditory
and visual continuous performance tests in schizophrenic pa-
tients: medication status and motoric correlates. Psychiatry Res
1991; 37:47–56
28. Gold JM, Harvey PD: Cognitive deficits in schizophrenia. Psy-
chiatr Clin North Am 1993; 16:295–312
29. Wykes T, Katz R, Sturt E, Helmsley D: Abnormalities of re-
sponse processing in a chronic psychiatry group: a possible pre-
dictor of failure in rehabilitation programmes? Br J Psychiatry
1992; 160:244–252
ATTENTIONAL FUNCTIONING
660 Am J Psychiatry 154:5, May 1997
