Sie sind auf Seite 1von 9

Drug and Alcohol Dependence 63 (2001) 87 95

www.elsevier.com/locate/drugalcdep

Antisocial personality disorder and cocaine dependence: their


effects on behavioral and electroencephalographic measures of
time estimation
Lance O. Bauer *
Department of Psychiatry, Uni6ersity of Connecticut School of Medicine, Farmington, CT 06030 -2103, USA
Received 2 May 2000; received in revised form 31 July 2000; accepted 8 August 2000

Abstract
The present study examined time estimation performance and concurrently-recorded electroencephalographic activity among 57
residential treatment program patients previously dependent on either cocaine or cocaine and alcohol. The patients were assigned
to one of two subgroups based upon the presence (n= 20) versus absence (n= 37) of a comorbid diagnosis of Antisocial
Personality Disorder (APD). Twenty-six subjects, who had no history of substance abuse and no diagnosis of APD, were also
examined. All subjects performed a psychomotor task in which they were asked to press a response key exactly 2 s after the onset
of a visual cue. Analyses revealed that cocaine-dependent patients with APD were often premature in their behavioral estimates
of time passage. The analysis of a slow EEG potential, viz. the Contingent Negative Variation, recorded over the 2 s time
estimation interval, also suggested premature response preparation in the cocaine-dependent, APD-positive group. Correlational
analyses revealed that the number of conduct problems reported prior to age 15 was a better predictor of both premature
responding and CNV amplitude than either severity of cocaine dependence, alcohol use, or anxious or depressed mood. The
potential relevance of these findings for studies of future time orientation and delay discounting behavior are discussed. 2001
Elsevier Science Ireland Ltd. All rights reserved.
Keywords: Antisocial personality disorder; Time perception; Cocaine dependence; Contingent negative variation; Evoked potentials; Conduct
disorder; EEG

1. Introduction
A capacity for accurately estimating the passage of
time is an important element of everyday life. Accurate
time estimation is, for example, important for ensuring
that activities, such as conversations, formal meetings,
or solitary work, do not interfere with activities or
events that follow. Accurate judgments about the duration of an activity are also important in planning and
scheduling its reoccurrence. Obviously, external
prompts, in the form of time-keeping devices and
prompts by other persons, can serve to regulate behavior and inform us about time passage. An impairment
in time sense can thereby be overcome, but only if the
individual utilizes and responds to such prompts.
* Tel.: +1-860-6794154; fax: + 1-860-6794077.
E-mail address: bauer@psychiatry.uchc.edu (L.O. Bauer).

An impaired ability to estimate time has been anecdotally, and sometimes formally, linked to various
forms of psychopathology. Both depression and anxiety
have been related to distortions in time sense, in opposite directions (Hawkins et al., 1988; Meluzzi et al.,
1991; von Kirchenheim and Persinger, 1991). Among
individuals with psychoactive substance use disorders,
an altered time sense has been formally demonstrated.
For example, Smart (1968) found that alcoholics performed less well than social drinkers on Wallaces
(1956) Future Time Perspective task, which measures
the ability to conceptualize the future, in terms of the
timing and ordering of future events. Manganiello
(1978) described a similar finding from his comparison
of 45 adult heroin addicts versus 50 high school students. In addition, Petry and colleagues (1998) compared 34 active heroin addicts to 59 non-drug-using
controls. Heroin addicts scored lower than the control

0376-8716/01/$ - see front matter 2001 Elsevier Science Ireland Ltd. All rights reserved.
PII: S 0 3 7 6 - 8 7 1 6 ( 0 0 ) 0 0 1 9 5 - 2

88

L.O. Bauer / Drug and Alcohol Dependence 63 (2001) 8795

group on both the Future Time Perspective task and


the future orientation subscale of the Stanford Time
Perception Inventory (Zimbardo, 1992). In all three
studies, substance-abusing patients reported shorter
time horizons than the control group. Petry and colleagues (1998) have theorized that shortened time horizons contribute to the disadvantageous preference by
substance abusers for smaller immediate rewards versus
larger delayed rewards, as well as an underappreciation
of the serious, and often delayed, negative consequences of substance abuse (Kirby and Herrnstein,
1995; Madden et al., 1997; Vuchinich and Simpson,
1998; Petry and Casarella, 1999).
To date, there are no studies which have examined
the accuracy of time sense among substance-abusing
patients who have verifiably discontinued active drug
and medication use. Accordingly, it has been difficult to
determine whether it originates from the acute effects of
psychoactive drugs (Lapp et al., 1994). A disordered
time sense may also be related to those personality
disorders, including childhood Conduct Disorder and
its continuance into adulthood as Antisocial Personality
Disorder (APD), which are characterized by impulsivity
and/or inattention. Interestingly, childhood Conduct
Disorder and adult APD are significant risk factors for
substance dependence (Hesselbrock et al., 1985; Malow
et al., 1989; Yates et al., 1989; Kleinman et al., 1990;
Rounsaville et al., 1991; Weiss et al., 1993) as well as
many other forms of adult psychopathology (Robins,
1966; Robins and Price, 1991). Indeed, these premorbid
personality factors might mediate the altered time sense
that has been attributed to both substance abuse and
mood disorders.
The present study is unique in examining the relative
contributions of antisocial personality, depression, and
anxiety level to altered time sense in a group of cocainedependent patients who had demonstrably remained
abstinent from cocaine and other drugs for a 15
month period. The study also examined the relation
between various measures of substance abuse severity
and time estimation. None of the subjects enrolled in
the present study were under the influence of psychoactive medications at the time of testing.
A second unique feature of the present study was the
measurement of the ability to estimate the passage of a
fixed interval of time via key press latency. Because of
its relative simplicity, a manual key press response may
be less dependent than are the aforementioned questionnaires upon group differences in reading skill, verbal comprehension, and motivation. It can also be
hypothesized that response latency is a more valid
estimate of time estimation ability for actual, discrete
events than are perceptions regarding the duration and
sequence of events embedded within fictional scenarios.
Yet another unique feature of the present study was
the use of electroencephalographic activity recorded

continuously throughout the time estimation interval.


By constructing a time-point average of the electroencephalogram over trials, one can detect the emergence
of a slowly developing negative voltage known as the
Contingent Negative Variation (CNV; Rockstroh et al.,
1982). The CNV is recordable during periods of time in
which subjects are anticipating future events of personal or instructed significance (Ruchkin et al., 1977;
Macar and Besson, 1985; Elbert et al., 1991; Hiraku
and Sakuma, 1996). Normal subjects with accurate time
estimation ability have been shown to exhibit CNVs of
a smaller amplitude and slower rise time than normal
subjects with poor time estimation ability (Ladanyi and
Dubrovsky, 1985). The present study utilized the CNV
as an objective marker of time estimation and response
preparation.

2. Method

2.1. Subjects
Fifty-seven cocaine-dependent subjects were recruited
from long-term residential treatment programs in Hartford, Connecticut. In addition, 26 non-drug-dependent
subjects were recruited through newspaper advertisements. All of the subjects signed informed consent
agreements approved by the Universitys Institutional
Review Board. They were paid for their time.
Eligibility for the study was initially assessed through
a brief interview. To be eligible, subjects were required
to demonstrate adequate reading and comprehension
skills for completing the informed consent agreement
and demographic, drug use, and mood state questionnaires. Potential subjects were excluded if they possessed a lifetime history of a major medical disorder
(neurological, hepatic, or cardiovascular); HIV infection; a head injury resulting in a loss of consciousness;
seizures (including drug-related seizures); DSM-III-R
defined opiate, sedative, or barbiturate abuse or dependence; a major psychiatric illness such as schizophrenia,
or bipolar or depressive disorders; or a maternal history
of substance abuse or dependence. Uncorrected visual
or auditory deficits, and current psychoactive medication use, were also exclusionary.
After the screening interview, potential subjects were
transported to the Health Center for a more detailed
clinical evaluation and the laboratory study. The clinical evaluation involved two semi-structured psychiatric
interviews: the Semi-Structured Assessment for the Genetics of Alcoholism (Bucholz et al., 1994; Hesselbrock
et al., 1999) and the Addiction Severity Index (McLellan et al., 1980). The SSAGA permitted the assignment
of Axis I and Axis II diagnoses according to DSM-IIIR criteria. The ASI provided quantitative measures of
alcohol and drug use. In addition, all subjects were

L.O. Bauer / Drug and Alcohol Dependence 63 (2001) 8795

required to complete a number of questionnaires. These


included the Michigan Alcoholism Screening Test
(MAST; Selzer, 1971), Drug Abuse Screening Test
(DAST-10; Skinner, 1982), Beck Depression Inventory
(BDI; Beck et al., 1961), Spielberger State-Trait Anxiety Inventory (STAI; Spielberger, 1983), and the Hartford Shipley/Institute of Living Scale (Shipley, 1940).
Based on the results of the interview, the 83 eligible
subjects were assigned to one of three groups. Group A
was comprised of 20 subjects who met DSM-IIIR diagnostic criteria for both cocaine dependence and APD.
Group C consisted of 37 cocaine-dependent subjects
who did not meet the DSM-III-R diagnostic threshold
for APD. The subjects assigned to Groups A and C
were additionally required to have used at least six
grams of cocaine per month during the year preceding
treatment. They were evaluated after 1 5 months of
verified abstinence. Abstinence was verified by urine
toxicology and breath alcohol screens performed at
frequent (12/week) and irregular intervals by research staff.
The control group Group N was comprised of
26 individuals with no history of substance abuse or
dependence and no diagnosis of Antisocial Personality
Disorder.

2.2. Procedures
The laboratory study was always scheduled to begin
in the morning; the clinical evaluation and interviews
followed. On arrival, the subject was asked to provide
breath and urine samples for drug screening. If the
samples tested positive for the presence of alcohol,
cocaine, heroin, amphetamine, or marijuana, then the
subject was immediately dismissed from the study.
The subject was escorted into the laboratory where
electrodes for recording electroencephalographic and
electro-oculographic activity were applied. Tin electrodes were applied to 15 scalp sites (FZ, CZ, PZ, F3,
F4, C3, C4, P3, P4, T3, T4, T7, T8, O1, O2). Electrodes
of the same type were applied to the tip of the subjects
nose (non-cephalic reference), mid-forehead (ground),
and in a diagonal orientation above and below the left
eye (eye movement). Inter-electrode impedances were
maintained below 5 KV.
Following the application of electrodes, the subject
was escorted into an adjacent room and seated inside a
soundproofed chamber. The EEG electrode cap (ECI,
Eaton, Ohio) was connected to a bank of amplifiers.
The quality of the recording was then verified.
The subject performed a number of cognitive and
neurophysiological tests within this laboratory setting.
The results of many of these tests have already been
described in the literature (Bauer, 1996; Bauer and
Easton, 1996; Bauer and Mott, 1996; Bauer, 1997a,b;
Costa and Bauer 1997; Easton and Bauer, 1997; Bauer,

89

1998; Costa and Bauer, 1998). For the presently described test, the subject was instructed that he/she
should press a response key to designate the passage of
a 2 s interval following the onset of each of 50 cue
stimuli. The cue stimulus was the letter X presented in
the middle of a computer display for 50 ms. A small
fixation spot was presented in the center of the monitor
at all other times.
The computer was programmed to present either a
500 or 2000 Hz tone 3 s after response execution. The
500 Hz tone was presented if response latency was
within a 500 ms (9 250 ms) window of the designated
2 s target. A 2000 Hz tone was presented if the response
latency was outside of this range. The next trial commenced 510 s later. Subjects were allowed to practice
the task for five to ten trials to verify their comprehension of the instructions.

2.3. Data reduction


Across 50 trials of the task, the number of key press
responses were counted within each each half of the 2 s
time estimation interval. Responses that occurred more
than 2 s after the cue were not included in the analysis
because such responses were rare. Expanding the frequency distribution so as to accomodate these rare, late
responses would have been problematic for the assumptions of the analysis.
EEG and EOG signals were appropriately amplified
(EEG gain= 20 K; EOG gain= 2 K) and filtered
(bandpass= 0.01 12.5 Hz). The amplified signals were
digitized at a rate of 200 Hz per channel. EEG activity
recorded from each electrode site was tested offline for
A-D converter saturation and eye movement (peak-topeak EOG deviation\ 50 microvolts) artifacts. A positive test for either condition resulted in the exclusion of
both EEG and task performance data for that trial.
A minimum of 20 artifact-free epochs, commencing
at stimulus onset and spanning the 2000 ms time estimation interval, were required for the data to be retained for further analysis. The epochs were averaged
separately for each electrode site. The averages were
then baseline-corrected relative to the average voltage
during the initial 20 ms of the post-stimulus period.
CNV amplitude was measured in the averaged waveform within three time periods: 600 610 ms, 1200
1210 ms, and 1700 1710 ms post-stimulus onset. These
three periods respectively estimated the amplitude of
the early, middle, and late components of the CNV.
As an additional data reduction tool, a principal
components analysis was applied to the amplitude of
each CNV component. The PCA was used to reveal
electrode sites that produced highly intercorrelated activity. As a result, CNV component amplitudes could
be averaged across these intercorrelated sites yielding a
more reliable estimate of the true amplitude as well as

L.O. Bauer / Drug and Alcohol Dependence 63 (2001) 8795

90

a reduced likelihood of Type I error. A varimax-rotation of the principal components yielded two factors.
The factor structure was the same for the early, middle,
and late CNV components. The first factor was comprised of amplitudes measured at posterior electrode
sites, viz. PZ, P3, P4, O1, O2, T3, T4, T7, and T8. The
other factor was comprised of amplitudes measured at
anterior or central sites, viz. FZ, CZ, F3, F4, C3, and
C4. Amplitudes were averaged within each factor yielding a single amplitude measure for each combination of
the two scalp regions (anterior/posterior) and three
CNV components (early/middle/late).
3. Results

3.1. Demographic, psychological, and drug use


characteristics
Table 1 summarizes the background characteristics
of the three subject groups. One-way ANOVAs were
used to evaluate group equivalence on continuous measures. Pearsons Chi-Square test was applied to categorical measures.

On average, the subjects were 33.8 years of age.


Approximately 64% of the subjects were male. An
identical percentage were members of a racial/ethnic
minority. The subject groups did not differ in these
demographic features.
The groups did differ significantly with respect to the
number of self-rated alcohol [F(2, 80)= 25.4, P B0.001]
and drug abuse [F(2, 80)= 229, PB 0.001] problems,
and the number of depression [F(2, 80)= 8.3, P B
0.002] and anxiety [trait: F(2, 80)= 9.5, PB0.001;
state: F(2, 80)= 8.5, PB 0.002] symptoms. Tukey post
hoc tests revealed that the source of these significant
main effects could be traced to a difference between the
control group and the two cocaine-dependent groups.
The cocaine-dependent groups collectively also differed
from the control group with respect to the prevalence
of paternal substance dependence [ 2(2df) = 10.6, PB
0.005]. Cocaine-dependent groups with versus without
APD did not differ on MAST, DAST-10, BDI, or
STAI scores.
The cocaine-dependent patient groups did not differ
in the percentage of their membership that was also
alcohol-dependent or that possessed a paternal history

Table 1
Demographic, drug use, and psychological characteristics of study groups mean 9 SD or percent
Non-dependent
APD-negative (Group N)
N
26
Age
35.1 96.2
Yrs. of
14.1 9 1.8
education
% Male
61.5
% Caucasian
46.2
IQ Shipley
102.8 9 12.6
scale
% Paternal
31.8
addiction
% Alcohol
NA
dependent
Mos. of cocaine
NA
abstinence
Yrs. of cocaine
NA
use
c Alcohol
2.0 9 1.7
problems
MAST
c Drug
0.2 9 0.5
problems
DAST-10
Beck depression
4.4 9 4.4
scale
State anxiety 31.3 9 8.7
STAI
Trait anxiety 35.9 9 9.9
STAI
a

PB0.05.

Coc-dependent
APD-negative (Group C)

Coc-dependent
APD-positive (Group A)

Test result

Significant pairwise
comparisons

37
33.8 9 6.3
12.1 91.6

20
32.0 9 5.1
11.1 9 1.4

F= 1.4
F= 19.6a

N\C, N\A

64.9
37.8
92.4 9 13.3

65
25.0
81.7 9 10.9

2 =0.1
2 =7.7
F =15.0a

N\C, N\A, C\A


NBC, NBA

71.9

75.0

2 =10.6a

51.4

60

2 =0.1

3.19 1.5

2.8 91.4

F =0.4

11.9 9 5.9

12.5 95.2

F =0.1

12.3 9 7.2

11.8 9 7.0

F=25.4a

NBC, NBA

7.4 9 1.8

8.2 9 1.4

F= 229.1a

NBC, NBA

10.1 9 6.2

9.8 9 6.3

F= 8.3a

NBC, NBA

41.1 9 10.4

38.5 97.7

F= 8.5a

NBC, NBA

45.0 9 8.3

45.3 98.0

F= 9.5a

NBC, NBA

L.O. Bauer / Drug and Alcohol Dependence 63 (2001) 8795

Fig. 1. Average ( +1 SE) number of key press responses within each


half of the 2 s time estimation interval as a function of Group.
*P B0.05.

of substance dependence. The patient groups were likewise similar in the number of years of cocaine use
(minus years abstinent) and the number of months of
cocaine abstinence. The only variable which differentiated among all three groups was the total IQ score
from the Shipley Institute of Living Scale [F(2, 80)=
15, PB 0.001].

3.2. Task performance


A multivariate ANOVA, with age as a covariate,
revealed a significant main effect of Group on reaction
time and the number of responses executed within the 2
s time estimation interval [MANCOVA F(2, 80)= 3.4,
P B 0.05]. Univariate ANCOVAs, accompanied by
Tukey post hoc tests, revealed significantly [F(2, 80)=
6.5, P B0.005] shorter reaction times among APD-positive cocaine-dependent subjects (1.099 0.18 s) versus
subjects enrolled in the APD-negative, cocaine-dependent (1.2790.24) and APD-negative, non-dependent
(1.27 90.20) groups.
These results were validated by the results of analyzing the number of responses emitted during each half of
the time estimation interval. On average, members of
the APD-positive, cocaine-dependent group executed
significantly more responses during the first half of the
time estimation interval in comparison to the members
of the two other groups [F(2, 80)=2.98, P B0.05].
During the second half of the interval, the differences
among the groups in response frequency were statistically nonsignificant. Fig. 1 illustrates these results.

3.3. CNV amplitude


CNV amplitudes were analyzed separately for the
anterior and posterior regions. The analytic method
was a 3 (Group) by 3 (CNV component) repeated-measures ANOVA with age as a covariate. The degrees of
freedom were reduced by the Geisser-Greenhouse
method so as to compensate for violations of the
sphericity assumption that are inherent in time-series

91

data. Corrected degrees of freedom are reported. Tukey


post hoc tests were used to identify the source of
significant main or interaction effects.
Analyses of anterior region CNV data revealed a
significant interaction between Group and CNV component [F(3, 30)=2.75, PB0.05]. Post hoc analyses
revealed that the early component of the CNV was
significant larger in the cocaine-dependent, APD-positive group ( 3.499 6.74 mV) than in the cocaine-dependent, APD-negative (0.769 3.82 mV) or control
( 0.589 1.59 mV) groups. The main effects of Group
and CNV component were not significant. Analyses of
posterior region CNV data revealed no significant main
or interaction effects.
Group-averaged CNV waveforms are presented for
all 15 electrode sites in Fig. 2.

3.4. Correlational analyses


In an attempt to identify factors that might amplify
or moderate the relationship between APD and impaired time estimation ability in cocaine-dependent patients, a small number of correlations were performed.
The factors chosen for inclusion in the correlation
matrix were justified on the basis of published speculation or clinical lore. More specifically, the following
variables were chosen: depression, anxiety, severity of
cocaine use, duration of cocaine abstinence, and the
number of comorbid alcohol or drug use problems. In
addition, the number of diagnostic criterion items for
APD, reflecting both childhood and adult conduct
problems, were examined.
Table 2 presents the results of these correlational
analyses. As can be seen, only the number of conduct
problems reported prior to age 15 (i.e. the number of
Conduct Disorder criteria) correlated significantly with
reaction time (rpartial = 0.36, PB 0.05) and early component CNV amplitude at anterior electrode sites
(rpartial = 0.41, PB 0.05).

4. Discussion
The present study demonstrated that cocaine-dependent patients with APD executed more premature responses during a time estimation task than
cocaine-dependent patients and non-drug abusing controls who did not meet APD diagnostic criteria. The
electroencephalogram of the APD-positive, cocaine-dependent group was also characterized by the emergence
of an early, frontally-dominant negativity (see Fig. 2)
whose latency coincided with the time period (Fig. 1)
containing this excessive number of responses. Cocainedependent patients and healthy controls who failed to
meet APD diagnostic criteria exhibited a smaller electroencephalographic change of this type. The identical

92

L.O. Bauer / Drug and Alcohol Dependence 63 (2001) 8795

pattern of differences among the groups (i.e. Group


A "Groups C and N) found for reaction time, response frequency, and CNV amplitude convergently
validates the hypothesis that APD is associated with
premature response preparation and execution.
The frontally-dominant focus of the group difference
in early CNV amplitude agrees with the frontal focus of
neuropsychological and P300 abnormalities associated
with Antisocial Personality Disorder. Cocaine- (Bauer,
1997a,b) and alcohol-dependent (Costa et al., 2000)
patients with APD have been found to exhibit small
frontal P300 amplitudes in comparison to peers without
APD. Similar P300 and neuropsychological (Gillen and
Hesselbrock, 1992) decrements, related to APD, have
been reported from studies of children (Bauer and
Hesselbrock, 1999) or young adults (Bauer et al.,
1994a,b; OConnor et al., 1994) who were neither drug
nor alcohol dependent. It is important to recognize that
the present demonstration of enhanced frontal CNVs
among APD-positive subjects does not contradict
demonstrations of reduced frontal P300s found in studies of other APD-positive subject groups. Indeed, the
early onset of the enhanced CNV can be interpreted as
reflecting an inappropriately-timed activation of neurons involved in the preparation and execution of mo-

tor responses (Rockstroh et al., 1982) and a deficit in


the regulation of this process. Similar selective enhancements of the early component of the CNV have been
shown in studies of healthy adults characterized by
poor time estimation ability (Ladanyi and Dubrovsky,
1985) and incarcerated criminals (Forth and Hare,
1989). Individuals who exhibit elevated scores on psychopathy checklists have likewise been shown to exhibit
enhanced CNV amplitude (Fenton et al., 1978; Howard
et al., 1984; Raine et al., 1990). Many of these studies
used the conventional, paired-stimulus paradigm for
eliciting the CNV rather than the time estimation
paradigm used presently.
The significant correlations found between the number of childhood Conduct Disorder criterion behaviors
and both reaction time and early CNV amplitude are
interesting, because they suggest that poor time estimation ability is related abnormal personality and neural
processes which predate the onset of substance dependence. An association between APD and poor time
estimation ability was originally suggested by Petry and
colleagues (1998). Unfortunately, the association could
not be formally tested because Petry and colleagues did
not include a structured assessment of APD in their
protocol.

Fig. 2. Group-averaged event related electroencephalographic potentials recorded at each of the 15 electrode sites. Each epoch commences with
the onset of the visual cue and spans the entire 2.0 s time estimation interval. The arrow presented in the top, center panel points to the early
component of the Contingent Negative Variation. Note that the amplitude of this component, and the magnitude of the group difference, is
greatest at frontal electrode sites.

L.O. Bauer / Drug and Alcohol Dependence 63 (2001) 8795

93

Table 2
Partial correlations, adjusted for age, between time estimation performance, frontal region CNV amplitude, and selected psychological and drug
use measures. Cocaine-Dependent Groups Only (N =57)

c Childhood conduct disorder criterion behaviors


c Adult APD behaviors
Mos. of cocaine abstinence
Yrs. of cocaine use
Trait anxiety STAI
State anxiety STAI
Beck depression score
c Alcohol problems MAST
c Drug problems DAST-10
a

Reaction time

Anterior region CNV amplitude early component


(microvolt change from baseline)

0.36a
0.04
0.19
0.11
0.09
0.11
0.09
0.11
0.11

0.41a
0.05
0.05
0.11
0.01
0.06
0.06
0.18
0.04

PB0.05.

The failure of time estimation ability to correlate with


severity of cocaine use and the duration of cocaine
abstinence lends further credence to the hypothesis that
impairments in time estimation ability are related to
premorbid personality. Yet, one should always be cautious in interpreting a nonsignificant finding because of
the potential for Type II error. Furthermore, the lack of
a significant correlation between time estimation and
cocaine use among abstinent, residential treatment program patients does not exclude the possibility that drugs
of abuse can acutely alter time sense. In fact, several
studies have found that acute doses of alcohol, amphetamine, heroin, and other drugs of abuse alter the
perception of elapsed time (Lapp et al., 1994; Mohs et
al., 1978, 1980). The null correlations between cocaine
use and time estimation do suggest, however, that
previous cocaine abuse has no significant residual effects
on time estimation.
The absence of a significant correlation between time
perception and both depression and anxiety should
likewise be interpreted cautiously. Patients who were
receiving psychoactive medications or who met DSMIIIR diagnostic criteria for schizophrenia, or bipolar or
major depressive disorders, were excluded from participating in our study. Accordingly, the range of depression
and anxiety symptoms was restricted, and the power for
detecting a significant correlation was thereby muted.
The null correlations can be interpreted as suggesting
that subclinical levels of depression and anxiety do not
bear a strong relationship to time estimation.
The present demonstration of an association between
APD and time estimation may be relevant for studies of
the future time perspective of substance abusers. For
example, one previous study (Alvos et al., 1993), using
Wallaces (1956) Future Time Perspective task, found
that opioid-dependent patients who were untreated and
active users of heroin reported a shorter time horizon
than patients enrolled in methadone maintenance therapy. Importantly, however, the absence of active heroin

use by the methadone-maintained group was not verified


by urine toxicology screens or collateral informants.
More importantly, it has repeatedly been shown that
relapse and noncompliance with treatment are strongly
associated with the presence of comorbid APD (Leal et
al., 1994; Rounsaville et al., 1986; Woody et al., 1985).
Therefore, comparisons of the future time perspective of
active versus treated and abstinent drug abusers are very
likely to be confounded by a group difference in the
prevalence of APD. Unfortunately, we did not measure
future time perspective in the present study and cannot
formally verify that it measures the same underlying
construct as is measured by our time estimation task. Yet,
it seems probable that APD would influence performance
on the Future Time Perspective task as it did in our time
estimation task.
The present findings may likewise have implications
for studies of delay-discounting. Delay-discounting
refers to the unwise choice of smaller, immediate rewards
over larger, delayed rewards. Substance-dependent patients have repeatedly been shown to engage in delay
discounting (Bretteville-Jensen, 1999). They also tend to
minimize the adverse medical, legal, and psychological
consequences of drug abuse, which are typically experienced after a delay of many years. It has not been
determined whether impaired time estimation ability
correlates with delay discounting. Yet, to the extent that
time estimation and delay discounting are correlated, the
present findings suggest that APD may be an important
factor to consider in the dysfunctional cost-benefit analysis which prompts some individuals to develop drug
dependence.

Acknowledgements
Support for this research was provided by Grants
R01-DA05826, R01-MH61346, and P50AA03510. The

94

L.O. Bauer / Drug and Alcohol Dependence 63 (2001) 8795

author thanks Dr. Nancy Petry for her comments on an


earlier draft of this manuscript.
References
Alvos, L., Gregson, R.A.M., Ross, M.W., 1993. Future time perspective in current and previous injecting drug users. Drug Alcohol
Depend. 31, 193 197.
Bauer, L.O., 1996. Resting hand tremor in abstinent cocaine dependent, alcohol dependent and polydrug dependent patients. Alcohol Clin. Exp. Res. 20, 11961201.
Bauer, L.O., 1997a. Frontal P300 decrements, childhood conduct
disorder, family history, and the prediction of relapse among
abstinent cocaine abusers. Drug Alcohol Depend. 44, 110.
Bauer, L.O., 1997b. Smooth pursuit eye movement dysfunction in
abstinent cocaine abusers: Effects of a family history of alcoholism. Alcohol Clin. Exp. Res. 21, 910915.
Bauer, L.O., 1998. Effects of chronic opioid dependence and HIV-1
infection on the pattern shift visual evoked potential. Drug Alcohol Depend. 50, 147156.
Bauer, L.O., Easton, C.E., 1996. Pattern shift VEPs in abstinent
cocaine-dependent, alcohol-dependent, and co-dependent patients.
Drug Alcohol Depend. 40, 203209.
Bauer, L.O., Mott, A.E., 1996. Differential effects of cocaine, alcohol,
and nicotine dependence on olfactory evoked potentials. Drug
Alcohol Depend. 42, 2126.
Bauer, L.O., Hesselbrock, V.M., 1999. P300 decrements in teenagers
with conduct problems: Implications for substance abuse risk and
brain development. Biol. Psychiatry 46, 263272.
Bauer, L.O., Hesselbrock, V., OConnor, S., Roberts, L., 1994a. P300
differences between nonalcoholic young men at average and
above average risk for alcoholism: effects of distraction and task
modality. Prog. Neuropsychopharmacol. Biol. Psychiatry 18,
263 277.
Bauer, L.O., OConnor, S., Hesselbrock, V.M., 1994b. Frontal P300
decrements in Antisocial Personality Disorder. Alcohol Clin. Exp.
Res. 6, 1300 1305.
Beck, A.T., Ward, C.H., Mendelson, M., Mock, J., Erbaugh, J.,
1961. An inventory for measuring depression. Arch. Gen. Psychiatry 4, 561 568.
Bretteville-Jensen, A.L., 1999. Addiction and discounting. J. Health
Econ. 18, 393 407.
Bucholz, K.K., Cadoret, R., Cloninger, C.R., Dinwiddie, S.H., Hesselbrock, V.M., Nurnberger, J., Jr., Reich, T., Schmidt, I.,
Schuckit, M.A., 1994. A new semi-structured psychiatric interview
for use in genetic linkage studies: A report on the reliability of the
SSAGA. J. Stud. Alcohol 55, 149158.
Costa, L., Bauer, L.O., 1997. Quantitative electroencephalographic
differences associated with alcohol, cocaine, heroin, or dual substance dependence. Drug Alcohol Depend. 46, 8794.
Costa, L., Bauer, L.O., 1998. Smooth pursuit eye movement dysfunction in substance dependent patients: Mediating effects of Antisocial Personality Disorder. Neuropsychobiology 37, 117123.
Costa, L., Bauer, L.O., Kuperman, S., Porjesz, B., OConnor, S.,
Hesselbrock, V.M., Rohrbaugh, J., Begleiter, H., 2000. Frontal
P300 decrements, alcohol dependence and Antisocial Personality
Disorder. Biol. Psychiatry 47, 10641071.
Easton, C.J., Bauer, L.O., 1997. Neuropsychological differences between alcohol dependent and cocaine dependent patients with or
without problematic drinking. Psychiatry Res. 71, 97103.
Elbert, T., Ulrich, R., Rockstroh, B., Lutzenberger, W., 1991. The
processing of temporal intervals reflected by CNV-like brain
potentials. Psychophysiology 28, 648655.
Fenton, G.W., Fenwick, P.B., Ferguson, W., Lam, C.T., 1978. The
contingent negative variation in antisocial behaviour: a pilot study
of Broadmoor patients. Br. J. Psychiatry 132, 368377.

Forth, A.E., Hare, R.D., 1989. The contingent negative variation in


psychopaths. Psychophysiology 26, 676 682.
Gillen, R., Hesselbrock, V.M., 1992. Cognitive functioning, ASP, and
a family history of alcoholism in young men at risk for alcoholism. Alcohol Clin. Exp. Res. 16, 206 214.
Hawkins, W.L., French, L.C., Crawford, B.D., Enzle, M.E., 1988.
Depressed affect and time perception. J. Abnorm. Psychol. 97,
275 280.
Hesselbrock, M., Meyer, R., Keener, J., 1985. Psychopathology in
hospitalized alcoholics. Arch. Gen. Psychiatry 42, 10501055.
Hesselbrock, M.N., Easton, C., Bucholz, K.K., Schuckit, M.A.,
Hesselbrock, V.M., 1999. A validity study of the SSAGA a
comparison with the SCAN. Addiction 94, 1361 1370.
Hiraku, S., Sakuma, H., 1996. Effects on contingent negative variation of set created by anticipating variable foreperiods. Percept.
Mot. Skills 83, 1163 1169.
Howard, R.C., Fenton, G.W., Fenwick, P.B., 1984. The contingent
negative variation, personality, and antisocial behaviour. Br. J.
Pyschiatry 144, 463 474.
Kirby, K.N., Herrnstein, R.J., 1995. Preference reversals due to
myopic discounting of delayed rewards. Psychol. Sci. 6, 8389.
Kleinman, P.H., Miller, A.B., Millman, R.B., Woody, G.E., Todd,
T., Kemp, T., 1990. Psychopathology among cocaine abusers
entering treatment. J. Nerv. Ment. Dis. 178, 442 447.
Ladanyi, M., Dubrovsky, B., 1985. CNV and time estimation. Int. J.
Neurosci. 26, 253 257.
Lapp, W.M., Collins, R.L., Zywaik, W.H., Izzo, C.V., 1994. Psychopharmacological effects of alcohol on time perception: The
extended balanced placebo design. J. Stud. Alcohol 55, 96112.
Leal, J., Ziedonis, D., Kosten, T., 1994. Antisocial personality disorder as a prognostic factor for pharmacotherapy of cocaine dependence. Drug Alcohol Depend. 35, 31 35.
Macar, F., Besson, M., 1985. Contingent negative variation in processes of expectancy, motor preparation, and time estimation.
Biol. Psychology 21, 293 307.
Madden, G., Petry, N.M., Badger, G.J., Bickel, W.K., 1997. Impulsive and self-control choices in opioid-dependent and non-drug
using controls: drug and monetary rewards. Exp. Clin. Psychopharmacol. 5, 256 272.
Malow, R.M., West, J.A., Williams, J.L., Sutker, P.B., 1989. Personality disorders classification and symptoms in cocaine and opioid
addicts. J. Consult Clin. Psychol. 57, 765 767.
Manganiello, J.A., 1978. Opiate addiction: A study identifying three
systematically related psychological correlates. Int. J. Addictions
13, 839 847.
McLellan, A.T., Luborsky, L., Woody, G.E., OBrien, C.P., 1980. An
improved diagnostic evaluation instrument for substance abuse
patients the Addiction Severity Index. J. Nerv. Ment. Dis. 168,
26 33.
Meluzzi, A., Gasco, A., Francetti, G., Gecele, M., 1991. Time in
depression. Minerva Psichiatrica 32, 179 188.
Mohs, R.C., Tinklenberg, J.R., Roth, W.T., Kopell, B.S., 1978.
Methamphetamine and diphenhydramine effects on the rate of
cognitive processing. Psychopharmacol. 59, 13 19.
Mohs, R.C., Tinklenberg, J.R., Roth, W.T., Kopell, B.S., 1980.
Sensitivity of some human cognitive functions to effects of
methamphetamine and secobarbital. Drug Alcohol Depend. 5,
145 150.
OConnor, S., Bauer, L.O., Tasman, A., Hesselbrock, V.M., 1994.
Reduced P3 amplitudes of ERPs are associated with both a family
history of alcoholism and anti-social personality disorder. Prog.
Neuropsychopharmacol. Biol. Pyschiatry 18, 1307 1321.
Petry, N.M., Bickel, W.K., Arnett, M., 1998. Shortened time horizons and insensitivity to future consequences in heroin addicts.
Addiction 93, 729 738.

L.O. Bauer / Drug and Alcohol Dependence 63 (2001) 8795


Petry, N.M., Casarella, T., 1999. Excessive discounting of delayed
rewards in substance abusers with gambling problems. Drug
Alcohol Depend. 56, 2532.
Raine, A., Venables, P.H., Williams, M., 1990. Relationships between
N1, P300, and contingent negative variation recorded at age 15
and criminal behavior at age 24. Psychophysiology 27, 567 574.
Robins, L.N., 1966. Deviant Children Grown-up. Williams &
Wilkins, Baltimore.
Robins, L.N., Price, R.K., 1991. Adult disorders predicted by childhood conduct problems: Results from the NIMH Epidemiological
Catchment Area Project. Psychiatry 54, 116125.
Rockstroh, B., Elbert, T., Birbaumer, N., Lutzenbereger, W., 1982.
Slow Brain Potentials and Behavior. Urban & Schwarzenberg,
Baltimore.
Rounsaville, B.J., Kosten, T.R., Weissman, M.M., Kleber, H.D.,
1986. Prognostic significance of psychopathology in treated opiate
addicts. Arch. Gen. Pyschiatry 43, 739745.
Rounsaville, B.J., Foley, S., Carroll, K., Budde, D., Prusoff, B.A.,
Gawin, F., 1991. Psychiatric diagnoses of treatment seeking cocaine abusers. Arch. Gen. Pyschiatry 48, 4351.
Ruchkin, D.S., McCalley, M.G., Glaser, E.M., 1977. Event related
potentials and time estimation. Psychophysiology 14, 451 455.
Selzer, M.L., 1971. The Michigan Alcoholism Screening Test: The
quest for a new diagnostic instrument. Am. J. Pyschiatry 127,
1653 1658.
Shipley, W.C., 1940. A self-administering scale for measuring intellectual impairment and deterioration. J. Psychol. 9, 371377.

95

Skinner, H.A., 1982. The Drug Abuse Screening Test. Addict Behav.
7, 363 371.
Smart, R.G., 1968. Future time perspectives in alcoholics and social
drinkers. J. Abnorm. Psychol. 73, 81 83.
Spielberger, D.C., 1983. State Trait Anxiety Inventory. Consulting
Psychologist Press, Palo Alto, CA.
von Kirchenheim, C., Persinger, M.A., 1991. Time distortion a
comparison of hypnotic induction and progressive relaxation procedures: a brief communication. Int. J. Clin. Exp. Hypn. 39,
63 66.
Vuchinich, R.E., Simpson, C.A., 1998. Delayed reward discounting in
alcohol abuse. In: Chaloupka, F., Bickel, W., Grossman, M.,
Saffer, H. (Eds.), The Economic Analysis of Substance Use and
Abuse: An Integration of Econometric and Behavioral Research.
University of Chicago Press, Chicago.
Wallace, M., 1956. Future time perspective in schizophrenia. J.
Abnorm. Soc. Psychol. 52, 240 245.
Weiss, R.D., Mirin, S.M., Griffin, M.L., Gunderson, J.G., Hufford,
C., 1993. Personality disorders in cocaine dependence. Compr.
Psychiatry 34, 145 149.
Woody, G., McLellan, A.T., Luborsky, L., OBrien, C., 1985. Sociopathy and psychotherapy outcome. Arch. Gen. Pyschiatry 42,
1081 1086.
Yates, W.R., Fulton, A.I., Gabel, J.M., Brass, C.T., 1989. Personality
risk factors for cocaine abuse. Am. J. Public Health 79, 891892.
Zimbardo, P.G., 1992. Stanford Time Perception Inventory Manual.
Stanford University Department of Psychology, Stanford, CA.