bog.
Neuro-Psychophmmacoi.
& Bml. Psychat.
Copyright
2001, Vol
0 2001
25, pp.
1523-1536
Elsev~er Saence
Inc.
Prmted m the USA. All nghts reserved
0278-5846/O
l/$-see
front matter
ELSEVIER
PII:50278-5846(01)00210-X
RELATIONSHIP OF COGNITIVE ABILITY TO THE DEVELOPMENTAL
COURSE OF ANTISOCIAL BEHAVIOR IN SUBSTANCE-DEPENDENT
PATIENTS
MICHAEL
C. STEVENS, RICHARD F. KAPLAN, AND LANCE 0. BAUER
Department
of Psychiatry, University of Connecticut
Farmington, CT, USA
Health Center,
(Final form, July 200 1)
Abstract
Stevens, Michael C. , Richard F. Kaplan and Lance 0. Bauer: Relationship
Developmental
Course of Antisocial
Behavior in Substance-Dependent
Psychopharmacol.
& Biol. Psychiat. 2001,=,
pp. 1523-1536. 02001 Ek.wkr
of Cognitive Ability to the
Patients.
Prog. NeuroScience
Inc.
1. The present study examined cognitive differences among three groups of abstinent substancedependent patients and a control group of non-drug users. The patient groups were defined according
to their DSM III-R substance dependence diagnosis(es):
heroin, cocaine, or dual alcohol and cocaine
dependence.
2. In the initial analysis, which compared the four subject groups on scores from the Shipley Institute of
Living Scale, no significant differences were found. However, the groups did vary on the number of
Antisocial Personality Disorder (ASPD) behaviors.
3. Another set of analyses was conducted to examine the relationship between ASPD and SILS scores.
Analyses of the effects of ASPD (+/-) across all of the patients revealed lower SILS scores in the
ASPD-positive
group. Additional analyses examined the developmental course of the ASPD effect
by contrasting 1) patient groups characterized by childhood Conduct Disorder (CD) combined with
adult ASPD vs. 2) childhood CD which did not continue into adulthood as ASPD vs. 3) adults who
did not report childhood CD but who met other ASPD behavioral criteria as adults, vs. 4) subjects
who had neither childhood CD nor adult ASPD.
4. In this analysis, it was found that patients who met diagnostic criteria for childhood Conduct
Disorder, but whose antisocial behaviors resolved after age 15, had equivalent SILS scores to those
patients with no childhood CD or adult ASPD. A decrement in SILS scores was only found in those
patients whose antisocial behaviors persisted into adulthood.
5. ASPD adults who did not report childhood CD behaviors had normal SILS scores compared to
Controls.
Key Words:
alcoholism,
antisocial personality,
cognitive ability, conduct disorder, drug dependency
Abbreviations:
Addiction Severity Index (ASI), antisocial personality disorder (ASPD), conceptual
quotient (CQ), conduct disorder (CD), Semi-Structured
Assessment for the Genetics of Alcoholism
(SSAGA), Shipley Institute of Living Scale (SILS).
1523
1524
M.C. Stevens et al.
Introduction
Research that provides better understanding
of the etiology of alcoholism
has the potential to enhance
both primary and secondary prevention
efforts directed at this important individual and societal problem,
Much of the recent research
on the etiology
alcohol information
fathers
dysfunction
1991).
cognitive
of healthy
(1994) has proposed
differences
fathers,
abuse has used family history of
conceptualization
(Knop,
between
samples.
a substantial
the children
of substance-
premise
that cognitive
abuse is fostered (Hesselbrock
as an explanation
Others (Giancola
between
1985), attention
to be “at-risk”
et al., 1989), planning
factors such as co-morbid
Conduct Disorder
and its continuation
that antisocial
subserves behavioral
behavior
regulation
dependence
factors
psychiatric
diagnoses.
One highly relevant
factor in this literature is the co-morbid
into adulthood
as Antisocial
may mark a developmental
Personality
deficit
(Boyle et al.,
diagnosis
of childhood
Disorder.
in information
(Moftitt,
suggests otherwise.
It has been
processing
tests.
Such a mechanism
that
is thought to operate in conjunction
1993b) and would not be expected
to covary
with substance
unless the specific cognitive deficit jointly predicted both ASPD and addiction.
possible that a single biological
account
(Hogan, 1999). It is possible that this deficit may manifest in estimates
of cognitive ability on neuropsychological
with environmental
et
skills (Whipple et al., 1987).
1992; Robins, 1966) but often neglected
proposed
and
(Schaeffer
Perhaps the most important limitation of this literature is the failure to control or adequately
for complicating
for
and low-risk
(Drejer et al., 1985; Knop, 1985) abstracting/problem-solving
al., 1984), and visuospatial
on
et al, 1996b; Harden and
at-risk subjects
(Tarter
et al.,
of impairments
the same for samples considered
Many studies have found differences
on tests of vocabulary
More specifically,
based on the conjoint
dysfunction
et al., 1984) have proposed
dependence.
groups
cognitive
tests in substance-dependent
Pihl, 1995; Schaeffer
nonverbal
examining
is heritable and that it is a factor by which substance
neuropsychological
control
towards
and the children
Parsons
substance
of substance
as the key for defining high and low risk groups.
effort has been directed
dependent
focusing
factor underlies both ASPD and substance
Recent findings supports the hypothesis
behavior mark an increased risk to substance dependence
culture, rather than there being a direct relationship
dependence,
that cognitive
some evidence
deficits related to antisocial
through involvement
between cognitive
While it is
in antisocial activities and
factors and addiction (Cadoret et
al., 1995).
Early work (Gorenstein,
impairment
represent
prison
and antisocial
diagnoses
inmates
adolescents
1982; Pontius
behavior
and Ruttiger,
can been criticized
1976) examining
in that they used samples
of CD or ASPD (Kandel and Freed, 1989; Moffitt,
or juvenile
delinquents.
However,
and young adults have found cognitive
a link between
more recent
that may not best
1993a), such as psychopathic
studies
deficits associated
cognitive
of well-defined
samples
with both CD (Moffitt,
of
1993b;
1525
Cognitive ability and ASPD
Moftit and Henry, 1989) and ASPD (Gillen and Hesselbrock,
1992) that discriminate
normal controls.
Many of these studies have focused on decrements
of frontal
functioning,
lobe
Electrophysiological
and Hesselbrock,
ASPD.
consistent
studies have further implicated
and reliable
impairment
1993; Bauer et al., 1994; O’Connor
The convergence
developmental
as the most
in executive
these groups from
functioning,
impairments
of the frontal-lobe
in these
an index
groups.
(Bauer, 1997; Bauer
et al., 1994) in association
with both CD and
of evidence across studies suggests the prefrontal cortex as a possible locus of
impairment
for youth and adults exhibiting antisocial behavior.
Previous work has linked impairments
without substance-related
disorders
in verbal abstraction
ability to diagnosis
(Gillen and Hesselbrock,
of ASPD in subjects
1992; Stevens et al., submitted).
In the
current study, it was desired to see whether a similar effect would be found in a sample of substancedependent
inpatients,
secondary
to substance
was hypothesized
and also whether
dependence
that antisocial
this relationship
would exist regardless
(i.e., type of substance,
behavior
of possible
effects
duration or use, or period of abstinence).
would have a stronger
predictive
relationship
It
to cognitive
ability than family history or substance use characteristics.
Methods
Subiects
Participants
in the current study included 77 male and 45 female substance-dependent
were part of a larger study investigating
participants
were
Connecticut
region,
advertisements
enrolled
electroencephalographic
in long-term
residential
treatment
changes
programs
A control group of 17 male and 8 female participants
at inner city churches and newspapers.
abuse or dependence.
All participants
informed consent agreements
patients who
during abstinence.
located
was recruited
from through
Control subjects had no diagnosis
reported English as their primary language.
approved by the medical school’s Institutional
Patient
in the Hartford,
of substance
All patients signed
Review Board.
Patient Population
Potential participants
were excluded if they were more than 6 months or less than 1 month abstinent
from alcohol or other drugs of abuse.
excluded
infection,
currently
if they possessed
significant
a lifetime
Abstinence
history of neurologic,
head injury, bipolar disorder
be taking psychoactive
was confirmed
medications
Patients were also excluded if they possessed
through urinalysis.
hepatic or cardiovascular
or schizophrenia.
such as methadone,
uncorrected
Eligible
disorders,
participants
anticonvulsants,
visual impairments.
They were also
HIV
could not
or antidepressants.
M.C. Stevens
1526
Assessment
Instruments
Evaluations
took place
psychiatric
interviews
Alcoholism
(Bucholz
SSAGA
et al.
diagnostic
Association,
at the University
were administered
of Connecticut
- the Semi-Structured
et al., 1994) and the Addiction
interview
instrument
1987) diagnoses
provided
of substance
Health
Severity
Center.
Assessment
Cognitive
ability was estimated
SILS has recognized
that details substance
knowledge
ability through assessment
age-corrected
t-scores
of associative
for Vocabulary,
Abstract
possible confounds
Test (Skinner, 1982), Beck Depression
(Spielberger,
the
persons,
of vocabulary
thinking and problem-solving.
and Total.
ability. During this evaluation, subjects also completed the Michigan Alcoholism
Trait Anxiety Inventory
severity.
principally because of its brevity and ease of
It also provides
Quotient, which is a measure of abstraction ability that controls for premorbid
1971), Drug Abuse Screening
The
1940). Although
The SILS provides an estimate of general verbal ability though assessment
The test provides
Conceptual
use history and quantifies
(Harnish et al., 1994) in its norms for older and non-Caucasian
and of abstraction
The
et al., 1999). The Addiction
by the Shipley Institute of Living Scale (Shipley,
limitations
of
Psychiatric
and Axis I and Axis II psychopathology.
it has been widely used in studies of substance dependence,
administration.
et al., 1980).
Axis I and II DSM III-R (American
dependence
Index is a widely used instrument
for the Genetics
Index (McLellan
SSAGA has been shown to have good reliability and validity (Hesselbrock
Severity
Two semistructured
the
general verbal
Screening Test (Selzer,
Inventory (Beck, 1961), and State-
1983). These latter two measures
on cognitive ability related to differential psychiatric
were employed
to control for
status.
Data Analvsis
Participants
substance
were divided
dependence
DSM III-R (American
These included
into four groups.
diagnoses
Psychiatric
@25)
a control
group of subjects
and three other groups of subjects classified
Association,
alcohol and cocaine @=47), or heroin @=38).
according
1987) subtype of substance dependence:
without
to their
cocaine (n=37),
Of note, 8 of the 38 heroin patients (21%) were dually-
dependent on heroin and cocaine.
Demographic
information
1). Other variables,
diagnoses,
alcoholism,
including
months of abstinence,
(Table 2). Chi-square
measures:
(age, gender and years of education) was compared across all groups (Table
gender,
analyses
Beck Depression
and duration of substance
duration of substance use.
score, ASP and CD symptom
use, were compared
were utilized to test for differences
race, and diagnosis.
age, education,
Inventory
BDI depression
ANOVA
across patient groups
in categorical
measures:
was used to test for differences
score, ASP symptom
counts and
counts, months
paternal
in continuous
of abstinence,
and
Cognitive
ability
1527
and ASPD
Next, the Verbal, Abstract and Total raw scores from the SILS were converted
scores
in order to control
dependent
measures
Conceptual Quotient.
for the influence
were calculated:
Zachary
ability is questionably
of Vocabulary
function.
and Abstraction
Two other
t-scores,
(1986) recommends
caution
computing
and the
CQ scores when
low; therefore CQ scores for subjects with Vocabulary
than 23 were not calculated.
corresponding
the difference
of cognitive
t-
Both scores served as estimates of abstraction ability that controlled for premorbid
verbal ability (i.e., Vocabulary).
Vocabulary
of age on estimates
into age-corrected
As a result,
there were a total of 29 missing
raw scores less
values
for CQ; the
analyses for CQ utilized ~=118 subjects.
Scores for the substance-dependent
analysis of covariance
the contribution
(ANCOVA)
of specific
participants
to the control group by univariate
using years of education as the covariate.
substance
analyses included correlations,
were compared
use and personality
variables
Planned analyses examined
to the group differences.
These
additional ANOVAs, and logistic regression.
Results
Samnle Characteristics
There was a significant
16.997, p < .OOl.
correlated
scores.
difference
Additional
among the patient groups and controls for education,
correlation
(p < ,001) with SILS Vocabulary
Given the known linear association
analysis
shows that years of education
(1 = .623), Abstraction
of education
substance-dependent
patients’
SILS scores
was significantly
(r = .532), and Total (1 = ,606) t-
and scores on language-based
like the SILS, these findings indicated a need to control for education differences
comparing
F(3, 143) =
with the normal
cognitive
in subsequent
control
group.
analyses
Means
standard deviations for age, education and gender are displayed in Table 1.
Table 1
Characteristics
of Study Groups=
Normal
(n=25)
Heroin
(n=38)
Cocaine
(n=37)
Alcohol/
Cocaine
(tJ=47)
P
Subject age - Mean (SD)
34.4 (6.53)
33.5 (6.30)
32.9 (5.77)
33.7 (5.47)
1Is
Years education - Mean (SD)
14.4 (1.96)
12.2 (1.55)
11.8 (1.82)
11.8 (1.20)
c.001
68.0
57.9
65.8
62.5
ns
% Male
tests
*Significant differences were tested using one-way ANOVA for continuous measurer and X2 for categorical measures.
and
M.C. Stevens et aI.
1528
Comparison
of patient groups revealed
no significant
number of years of abuse, Beck Depression
for number of months
abstinent,
Inventory score, numbers of antisocial behavior symptoms
childhood
or adulthood,
significant
at a trend level, IJ2, 118) = 2.711, p = .071. Alcohol/Cocaine-dependent
more ASPD symptoms
or diagnoses
differences
of CD or ASPD (Table 2). The number of ASPD symptoms
compared to other substance-dependent
in
was
patients had slightly
groups.
Table 2
Comparison
of Patient Group Characteristics,
Heroin
(n=38)
Cocaine
(n=37)
Alcohol/
Cocaine
(lJ=47)
Mean (SD)
Mean (SD)
Mean (SD)
E
Number months abstinent
3.3 (1.12)
3.2 (1.28)
2.9 (1.30)
ns
Years using drug
11.4 (7.00)
10.7 (5.12)
12.5 (5.15)
ns
BDI score
9.7 (5.72)
8.9 (6.23)
10.0 (6.81)
ns
Number CD symptoms
3.1 (3.04)
2.3 (2.12)
2.7 (2.30)
I1s
Number ASPD symptoms
4.0 (3.71)
3.3 (3.62)
5.1 (3.46)
p = .071
% ASPD diagnosis
31.6
34.3
42.6
ns
% CD diagnosis in childhood
44.7
43.2
44.7
ns
% Paternal substance history
64.7
65.6
68.4
ns:
.Significant differences were tested using one-way ANOVA for continuous measures and X2 for categorical measures.
In an ANCOVA controlling
dependent
groups and normal controls.
142) = 2.207, ns, Abstract
difference
for education effects, SILS mean t-scores did not differ among substance-
between
F(3, 113) = .907, s.
t-score,
Vocabulary
Although
There were no main effects of group on Vocabulary
E(3, 142) = 1.464, m, Total t-score,
and Abstract t-scores, F(3, 142) = .897, s,
not presented
t-score, F(3,
F(3, 142) = 1.974, B, the
or the Conceptual
in Table 2, there were no differences
Quotient,
in SES among the
patient groups.
Analyses of Factors Influencing
Cognitive Ability
The authors then examined factors that might affect scores on cognitive tests for substance-dependent
patients using bivariate Pearson correlations
substance-dependent
SILS t-scores
patients (excluding
and paternal
history
and logistic regression.
The authors examined the groups of
the normal control group).
of substance-dependence,
There was no relationship
months
of abstinence,
between
total years of
1529
Cognitive ability and ASPD
substance use or Beck Depression
scores and number of childhood
and adult ASPD symptoms
regression
analyses
However,
showed
significant
However, relationships
significant
relationships
between
or through examination
Individual
ASPD diagnosis
either through
analysis
of the Conceptual
The B-coefficients,
between
SILS t-
t-score, 1 = -.215)
t-score, r = -.203, p < .05) were significant.
with antisocial diagnosis.
interval) and significance
score.
t-score, 1 = -.206, and Abstraction
level was controlled
and Abstraction,
relationships
depression
CD (Vocabulary
(Vocabulary
when Abstraction
Vocabulary
Inventory
logistic
and SILS t-scores.
of the mean difference
of
Quotient score, neither showed
odds-ratios
(and 95% confidence
level results of each of the logistic regression analyses are displayed in Table 3.
Table 3
Association
of ASPD Diagnosis to SILS Scores:
Individual Logistic Regression
Results
B
OR
95% CI
Iz
SILS Vocabulary
-0.0829
0.9204
0.8812 - 0.9614
< .Ol
SILS Abstract
-0.0912
0.9129
0.8699 - 0.9580
< ,001
SILS Total
-0.0910
0.9130
0.8724 - 0.9555
< .OOOl
SILS V-A Difference
-0.0281
0.9723
0.9236 - 1.0236
ns
SILS Conceptual
-0.0086
0.9915
0.9588 - 1.0253
ns
Quotient,
,Conceptual Quotient scores were calculatedusing n=94 subjects.
These results were consistent
diagnosis
of ASPD.
with a two-way
Three SILS T-scores
patients;VocabularyF(l,
ANOVA
examining
were significantly
116)= 16.841,~<.001,AbstractF(l,
116) = 20.246, p < .OOl. There were no differences
different
SILS scores by gender and by
for ASPD versus non-ASPD
116)=17.655,~<.OOl,andTotal~(l,
for ASPD diagnosis for V-A Difference,
0.607, a or CQ, F(l, 90) = 0.935, m. There were no SILS scores differences
F( 1, 116) =
as a function of gender or
the gender x ASPD interaction.
Further Analysis of the Relationship
Because
of the significant
examined
this relationship.
examined
the relationship
rules violations,
groups.
Between Conduct Disorder and Cognitive Ability
association
Using
between
and destructiveness
of number of CD symptoms
bivariate
correlations
with SILS scores,
with two-tailed
SILS scores and severity of the aggressive,
we further
tests of significance,
deceitfulness
we
and theft,
subtypes of childhood conduct disorder for the subjects in the patient
It should be noted that the relationships
the higher the number of CD symptoms,
found were in the negative direction (i.e., indicating that
the lower the adult SILS score).
the three SILS t-scores were found to be correlated
Consistent
with earlier results,
(E < .Ol) with number of aggression
CD symptoms.
1530
M.C. Stevens et al.
Abstract
and Total scores,
violation
symptoms.
There was no significant
Difference or Conceptual
Relationshin
Finally,
but not Verbal scores, were significantly
relationship
correlated
between
(p < .05) with rules
CD symptom
type and V-A
Quotient.
of Cognitive Abilitv and Continuity of Antisocial Behavior
we were interested
in whether
adulthood would affect SILS t-scores.
diagnosis yielding 4 groups:
were diagnosed
the continuity
Therefore,
of antisocial
the authors classified
behavior
from childhood
to
patients with regard to the CD
1) n=36 patients with no diagnosis of CD or ASPD, 2) a=10 patients who
with CD, but whose reported
antisocial
behaviors
resolved
so that ASPD diagnostic
criteria were not met, 3) n=44 patients with both CD and later diagnosis of ASPD, and 4) _n=31 patients
who had no diagnosis
of CD as youth, but whose adult behavior met criteria for ASPD (Table 4). These
latter patients do not technically
group.
qualify for a diagnosis
of ASPD and should be considered
Note, one subject did not have data for numbers
of child and adult antisocial
a distinct
behaviors
and
diagnoses of CD and ASPD could not be determined.
Table 4
Cross-Tablulation
of Subjects Who Meet Behavioral Criteria for Childhood Conduct Disorder and
Adult Antisocial Personality Diagnoses
Conduct Disorder -
Conduct Disorder +
Antisocial Personality
Disorder -
36
10
Antisocial
Disorder +
31
44
Personality
In an ANCOVA
Vocabulary
controlling
for years of education,
there was a significant
of group on
t-score F(3, 116) = 4.156, p < .Ol, Abstract T-score F(3, 116) = 8.782, p < .Ol, and Total t-
score, F(3, 116) = 4.793, p < .Ol), but not for V-A Difference,
Quotient,
effect
F(3, 90) = 0.140, m.
Vocabulary,
Abstract,
Bonferroni
post-hoc
F(3, 116) = 0.187, s,
tests reveal that the significant
and Total scores is due to the difference
Antisocial Only patients.
or Conceptual
@ < .Ol) between
difference
CD/ASPD+
in
and
These results are displayed in Table 5.
Results show that patients with antisocial
behaviors
across their lifetimes
(i.e., patients
with CD as
children who currently meet criteria for ASPD) have the lowest t-scores on the SILS, even statistically
controlling
for differences
in education
the CD-Only
level.
CD/ASPD
diagnosis,
premorbid
levels of verbal ability by holding
The scores are equivalent
group, and ASPD-Only
Vocabulary
group.
for those patients
When one attempts
ability constant,
with no
to control for
there are no significant
Cognitive
differences
in abstraction
A Difference
ability
1531
and ASPD
for any group. It should be noted that the observed power of the analyses of V-
and CQ were .09 and .l 1 respectively.
Table 5
SILS Scores by Presence or Absence of Conduct Disorder or Antisocial Personality
Disorder,
No Diagnosis
@=36)
CD Only
(n=lO)
CD/ASPD
(n=44)
Antisocial
Adult
(n=3 1)
Mean (SD)
Mean (SD)
Mean (SD)
Mean (SD)
p
44.7 (10.05)
47.0 (10.54)
37.5 (10.87)
46.0 (6.81)
< .Ol
SILS Abstract
50.5 (8.23)
53.4 (7.04)
45.3 (8.73)
52.4 (7.17)
< .Ol
SILS Total
50.1 (9.32)
5 1.2 (7.98)
41.1 (9.87)
50.4 (6.86)
< .Ol
SILS V-A Difference
47.6 (7.38)
50.5 (9.45)
41.1 (7.57)
49.6 (6.83)
ns
SILS Conceptual
91.4 (12.67)
93.2 (15.50)
91.0 (14.50)
92.8 (13.24)
ns
SILS Vocabulary
Quotient,
,Conceptual Quotient scores were calculated using a=95 subjects.
Discussion
Relationshin
to Previous Studies
The present results are consistent
submitted).
with previous studies (Gillen and Hesselbrock,l992.
Stevens et al.,
These studies examined the effects of ASPD in a large number of young adults, who had not
yet attained diagnosable
is capable of influencing
From the present
abuse/dependence,
levels of substance dependence.
study, we might similarly
that the presence
than possible deleterious
It would appear from these studies that ASPD
cognitive function in the absence of comorbid substance
conclude
of co-morbid
that for inpatients
abuse or dependence.
with diagnoses
of substance
ASPD is a more potent predictor of cognitive
effects on the nervous system due to substance use. These conclusions
ability
may or
may not generalize to patients studied after briefer periods of abstinence or greater duration or severity of
lifetime abuse.
Also, it is important to acknowledge
that while there was no statistical association
found
between duration and severity of substance use and cognitive test scores, this does not necessarily
imply
there is no relationship.
Previous reviews (Parsons and Leber, 1981) have commented
findings often reflect the heterogeneity
of findings and large individual differences.
that such negative
M.C. Stevens
1532
The diagnosis
consistent
with the literature
diagnosable
patient
of ASPD was associated
antisocial
groups
individual
attainment
ability.
(Frost et al., 1989).
that educational
However, it is acknowledged
of educational
of educational
does not contribute
the effects
These results are also
lower verbal ability estimates
The equivalence
attainment
in distinguishing
verbal ability.
between
that a measure of educational
Yet, achievement
individual differences
with poorer premorbid
that has noted a relationship
behavior
suggests
et al.
achievement
and
levels across the
to the group
difference.
would be superior to a measure
of education
from those of personality
measures would probably also be correlated
and
with personality
and
in learning ability.
Patients reporting antisocial behaviors persisting throughout their lifetimes had the lowest t-scores on
the SILS. This finding that conduct problems which began early in development
scores in adulthood
suggests
which is in accord
with Moffitt’s
behaviors
that cognitive
only during childhood
scores were not significantly
adult ASPD.
(1993b) proposal.
exhibited
is a factor in these behaviors,
In contrast,
comparatively
different from patients who never possessed
or vice-versa,
who reported
normal adult SILS t-scores
childhood
antisocial
- that is, their
conduct disorder or
lag which normalized
with
This finding is relatively unique and important because it suggests that personality
ability on the SILS may have a similar, trait-like
whether treatment
or prevention
that focus on antisocial
and/or a reduced risk for substance dependence.
empirically
test whether
the lower SILS estimates
disorder or attention deficit hyperactivity
feature.
SILS test
Also, the present analyses were not able to
may be due to co-morbid
disorder, as this information
with this is the difference
It remains to be determined
behavior would result in improved
performance
Consistent
patients
This suggests that these subjects may have had a developmental
continued maturation.
and cognitive
development
were related to SILS t-
of mean cognitive
diagnoses
of learning
was not collected in the protocol.
ability estimates
between
patients
who
showed antisocial behavior in both childhood and adulthood, and those patients who only met criteria for
ASPD as adults.
Controls.
The SILS t-scores
This finding
dependence
supports
may in fact represent
for these latter ASPD-only
the idea that adults diagnosed
two distinct groups, possibly
those with a primary ASPD diagnosis
dependence
who acquire antisocial
who have problem
behaviors
as a result.
patients were indistinguishable
with co-morbid
with different
drinking,
consistent with several previously proposed typology systems (Cloninger,
Finally, our data showed
disorder and cognitive ability.
ability.
a link between
dependence
-
substance
based on cognitive
ability is
198 1; Jellinek, 1960).
of retrospectively-diagnosed
using aggression
boys (Seguin et al., 1995), conduct-disordered
risk for substance
substrates
childhood
conduct
Higher childhood aggression was associated with lower general cognitive
This finding is similar to findings
preadolescent
subtypes
ASPD and substance
biological
and those with primary
This distinction
from
scales and neuropsychological
tests with
females (Giancola et al., 1998), and boys at
through a positive family history of alcoholism
(Giancola et al., 1996a).
1533
Cognitive ability and ASPD
Interestingly,
however,
a negative
Violation subtypes of childhood
specific
subtest
specific,
deficit
localized
was also detected
CD and Abstraction
raises the intriguing
deficit
in cognitive
between
aggression
association
found
neurological
substrates of aggressive
Limitations
relationship
that rule-breaking
subserved
by frontal
and all three SILS subtests
limitations.
and to various
complications,
brain areas.
suggests,
measure
a
The broader
by contrast,
that the
forms
of substance
dependence,
the present
ability relates to
study has several
of cognitive ability to factors like
This limits the generalizability
or learning disabilities.
of the current
Second, the power of some analyses were limited by small sample size (e.g., comparing
scores by continuity
of CD to ASPD).
of cognitive
ability.
neuropsychological
tests.
antisocial
disorders
behavior
Perhaps the most important
The SILS lacks the specificity
of dependence,
and cognition.
limitation
may interact with the effects of various substances
as this suggests
The collection
greater
support
selected
battery of
of how neurobehavioral
the lack of differences
SILS
is the use of only one
of a more rigorously
Such a battery would permit consideration
tests would make more meaningful
substances
behavior
might reflect
of how cognitive
First, data were not available to examine the relationship
ADHD, perinatal
specific
behavior
This
of the Current Study
behavior
findings.
of the Rules
behavior may be more diffusely localized.
While the current findings suggest avenues for future exploration
antisocial
the severity
subtest scores, but not for the other subtests.
possibility
processing
between
of abuse.
among persons
for the relationship
aspects of
Also, more
with difference
between
of such data in the future would be an interesting
antisocial
direction
to
pursue.
Conclusion
The present study demonstrated
abstinent,
applied
substance
abusing patients was not substance
to SILS scores
correlational
that the primary determinant
failed
to reveal
significant
analyses failed to reveal an association
use (i.e. dose-response)
or months
of abstinence
demonstrated
that ASPD, a personality
(Hesselbrock
et al., 1985), was reliably
antisocial personality
and ASPD-negative
via correlations
groups.
and logistic regressions
differences
in l-5 month
The initial overall analysis of variance
differences
among
the groups.
Furthermore,
between SILS scores and either years of substance
(i.e., time-response).
Instead,
the present
disorder which frequently occurs in substance dependent
associated
with cognitive
disorder and cognitive impairment
diagnosed
abuse.
of cognitive
ability.
The association
was shown via a comparison
But more convincingly,
the association
study
patients
between
of ASPD-positive
was also demonstrable
that included measures of severity of ASP, CD/ASPD subtypes,
and analyses of the effects of ASP progression
versus remission.
This finding for ASPD appears to be
M.C. Stevens et al.
1534
related to premorbid
ability, as analyses
that controlled
showed no differences
for abstraction for ASPD+ subjects.
for general verbal (i.e., Vocabulary)
ability
Acknowledgement
This research was supported in part by Public Health Service Grants T32AA07290
AMERICAN PSYCHIATRIC
ASSOCIATION
(1987)
Disorders, 3’d ed. rev., Washington, DC: Author.
Diagnostic
and Statistical
and ROlDA05826.
Manual
of Mental
BAUER, L. 0. (1997) Frontal P300 decrements, childhood conduct disorder, family history
prediction of relapse among abstinent cocaine abusers. Drug Alcohol Depend. 44: I-10.
BAUER, L. 0. and HESSELBROCK,
V. M. (1993) EEG, autonomic
risk for alcoholism. J. of Stud. Alcohol 54: 577-589.
and subjective
correlates
BAUER, L. O., O’CONNOR, S. and HESSELBROCK, V. M. (1994) Frontal P300 decrements
social personality disorder. Alcohol. Clin. Exp. Res. 6: 1300-1305.
BECK, A. T., WARD, C. H., MENDELSON, M., MOCK, J. and ERBAUGH,
for measuring depression. Arch. Gen. Psychiatry 4: 561-568.
and the
J. (1961)
of the
in anti-
An inventory
BOYLE, M. H., OFFORD, D. R., RACINE, Y. A., SZATMARY, P., FLEMING, J. E. and LINKS, P. S.
(1992) Predicting substance use in late adolescence: Results from the Ontario child health study
follow-up. Am. J. Psychiatry 149: 761-767.
BUCHOLZ, K. K., CADORET, R., CLONINGER, C. R., DINWIDDIE, S. H., HESSELBROCK, V. M.,
NURNBERGER,
J. I., REICH, T., SCHMIDT, I. and SCHUCKIT, M. A. (1994) A new semistructured psychiatric interview for use in genetic linkage studies: a report on the reliability of the
SSAGA. J. Stud. Alcohol 55: 149-158.
CADORET, R. J., YATES, W. R., TROUGHTON, E., WOODWORTH, G. and STEWART, M. A.
(1995) Adoption study demonstrating two genetic pathways to drug abuse. Arch. of Gen. Psychiatry
2: 42-52.
CLONINGER, C. R., BOHMAN, M. and SIVARDSSON,
S. (1981) Inheritance
Cross-fostering analysis of adopted men. Arch. Of Gen. Psychiatry 3: 861-868.
of alcohol
abuse:
DREJER, K., THEILBAARD,
A., TEASDALE, T. W., SCHULSINGER,
F. and GOODWIN, D.W.
(1985)
A prospective
study of young men at high risk for alcoholism:
Neuropsychological
assessment. Alcohol. Clin. Exp. Res. 9: 498-502.
FROST, L. A., MOFFITT,
T. E. and MCGEE, R. (1989)
psychopathology
in an unselected cohort of young adolescents.
Neuropsychological
function
J. Ab. Psycho1 98: 307-3 13.
and
GIANCOLA, P., MOSS, H., MARTIN, C., KIRISCI, L. and TARTER, R. (1996a) Executive cognitive
functioning predicts reactive aggression in boys at high risk for substance dependence: A prospective
study. Alcohol. Clin. Exp. Res. 20: 740-744.
Cognitive ability and ASPD
1535
GIANCOLA, P. R., MARTIN, C., TARTER, R., PELHAM, W. E. and MOSS, H. (1996b) Executive
cognitive functioning
and aggressive behavior in preadolescent
boys at high risk for substance
abuse/dependence.
J. Stud. Alcohol 57: 352-359.
GIANCOLA, P. R, MEZZICH, A. C. and TARTER, R. E. (1998) Executive cognitive functioning,
temperament, and antisocial behavior in conduct-disordered
adolescent females. J. Abnorm. Psychol.
107: 629-641.
GORENSTEIN,
E. E. (1982) Frontal lobe functions in psychopaths.
J. Abnorm. Psychol. 91: 368-379.
HARDEN, P. W. and PIHL, R. 0. (1995) Cognitive function, cardiovascular
boys at high risk for alcoholism. J. Abnorm. Psychol. 104: 94-103.
reactivity,
and behavior in
HESSELBROCK,
M. N., EASTON, C., BUCHOLZ, K. K., SCHUCKIT, M. and HESSELBROCK
V.
(1999) A validity study of the SSAGA--a comparison with the SCAN. Addiction. 94: 136 l-l 370.
HESSELBROCK,
V., BAUER,
L. O., HESSELBROCK,
M. N. and GILLEN,
R. (1991)
Neuropsychological
factors in individuals at high risk for alcoholism. In: Recent Developments
in
Alcoholism, Vol. 9, M. Galanter, (Ed.), pp 21-39, Plenum Press, New York.
HESSELBROCK, V., MEYER, R. and KEENER, J. (1985) Psychopathology
Arch. Gen. Psychiatry 42: 1050-l 055.
in hospitalized
alcoholics.
HOGAN, A. E. (1999) Cognitive functioning in children with ODD and CD. In: Handbook of
Disruptive Behavior Disorders, H. C. Quay and A. E. Hogan (Eds.) pp 3 17-335, Kluwer, New York.
GILLEN, R. and HESSELBROCK,
V. (1992) Cognitive functioning, ASPD, and family history
alcoholism in young men at risk for alcoholism. Alcohol. Clin. Exp. Res. 16: 206214.
of
HARNISH, M. J., BEATTY, W. W., NIXON, S. J. and PARSONS, 0. A. (1994) Performance
normal subjects on the Shipley Institute of Living Scale. J. Clin. Psychol. 50: 881-883.
by
JELLINEK,
E. M. (1960) Alcoholism:
KANDEL, E. and FREED, D. (1989)
Clin. Psychol. 45: 404-413.
A genus and some of its species.
Frontal-lobe
dysfunction
CMAJ u:
and antisocial
1341-1345.
behavior:
A review. J.
KNOP, J. (1985) Premorbid assessment of young men at high risk for alcoholism.
In: Recent
Developments in Alcoholism, Vol. 3, M. Galanter (Ed.), pp 53-64, Plenum Press, New York
MCCLELLAN, A. T., LUBORSKY, L., WOODY, G. E. and O’BRIEN, C. P. (1980) An improved
diagnostic evaluation instrument for substance abuse patients (the Addiction Severity Index). J. Nerv.
Men. Dis. 167: 26-33.
MOFFITT, T. E. (1993a)
Adolescence-limited
and life-course-persistent
developmental taxonomy. Psychol. Rev. 100: 674701.
MOFFITT, T. E. (1993b)
The neuropsychology
antisocial
of conduct disorder. Dev. Psychopathol.
MOFFITT, T. E. and HENRY, B. H. (1989) Neuropsychological
self-reported delinquents. Dev. Psychopathol. 1: 105-l 18.
assessment
behavior:
A
5: 135-l 5 1.
of executive
functions
in
O’CONNOR, S, BAUER, L. TASMAN, A. and HESSELBROCK, V. (1994) Reduced P3 amplitudes are
associated with both a family history of alcoholism and antisocial personality
disorder. Prog
Neuropsychopharmacol
Biol Psychiatry Is: 1307-l 321.
1536
M.C. Stevens et al.
PARSONS, 0. (1994)
Neuropsychological
measures and event-related
potentials
in alcoholics:
Interrelationships,
long-term reliabilities, and prediction of resumption of drinking. J. Clin. Psychol.
3: 3746.
PARSONS, 0. and LEBER, W. R. (1981) The relationship between cognitive dysfunction and brain
damage in alcoholics: Causal, interactive, or epiphenomenal Alcohol. Clin. Exp. Res. 5: 326-343.
PONTIUS, A. A. and RUTTIGER, K. F. (1976)
delinquents shown in narratives test. Adolescence
ROBINS, L. N. (1966) Deviant Children Grown-Up,
Frontal-lobe system
1: 509-5 18.
maturational
lag in juvenile
Williams & Wilkins, Baltimore.
SCHAEFFER, K., PARSONS, 0. and YOHMAN, J. (1984) Neuropsychological
differences between
male familial and nonfamilial alcoholics and nonalcoholics. Alcohol. Clin. Exp. Res. 8: 347-35 1.
SEGUIN, J. R, PIHL, R. O., HARDEN, P. W., TREMBLAY,
Cognitive and neuropsychological
characteristics of physically
104: 614-624.
SELZER, M. L. (1971) The Michigan Alcoholism
instrument. Am. J. Psychiatry 127: 1652-1658.
SHIPLEY, W. C. (1940)
A self-administering
deterioration. J. Psychol. 9: 371-377.
Screening
R. E. and BOULERICE,
B. (1988)
aggressive boys. J. Abnorm. Psychol.
Test: The quest for a new diagnostic
scale for measuring
intellectual
impairment
and
SKINNER, H. A. (1982) The Drug Abuse Screening Test. Addict. Behav. 1: 363-371.
SPIELBERGER,
CA.
D. C. (1983) State-Trait Anxiety Inventory.
Consulting Psychologists
STEVENS, M. C., KAPLAN, R. F. and HESSELBROCK,
V. M. (submitted)
Functioning in the Development of Antisocial Personality Disorder.
TARTER, R., JACOB, T. and BREMER,
Alcohol. Clin. Exp. Res. 13: 232-235.
D. A. (1989)
Cognitive
WHIPPLE, S. C., PARKER, E. S. and NOBLE, E. P. (1987)
alcoholic fathers and their sons. J. Stud. Alcohol 49: 240-244.
ZACHARY, R. A. (1986)
Angeles, CA.
Shipley
Institute
of Living Scale.
Inquiries and reprint requests should be addressed to:
Richard F. Kaplan, Ph.D.
Department of Psychiatry, MC-2103
University of Connecticut Health Center
Farmington, CT 06030-2103, USA
Tel: 860/679-3497
Fax: 8601679-4077
Email: [email protected]
Press Palo Alto,
Executive
Cognitive
status of sons of alcoholic
An atypical neurocognitive
Western
Psychological
men.
profile
Services
in
Los