Distinguishing BAS risk for university students’ drinking, smoking,
and gambling behaviors
Roisin M. O’Connor a,*, Sherry H. Stewart b
, Margo C. Watt a,c
a
Department of Psychology, Dalhousie University, Life Sciences Centre, Halifax, NS, Canada B3H 4J1
b
Departments of Psychiatry & Psychology, Dalhousiec University, Life Sciences Centre, Halifax, NS, Canada B3H 4J1
c
Department of Psychology, Saint Francis Xavier University, P.O. Box 5000, Antigonish, NS, Canada
a r t i c l e i n f o
Article history:
Received 14 May 2008
Received in revised form 6 November 2008
Accepted 3 December 2008
Available online 8 January 2009
Keywords:
BIS
BAS
Alcohol use
Smoking
Gambling
Undergraduates
a b s t r a c t
Individual differences in Gray’s Behavioral Approach System (BAS) and Behavioral Inhibition System (BIS)
have been associated with young adults’ substance misuse and gambling. To clarify the distinct and common
etiology of these behaviors, the current study examined the unique inﬂuence of BAS sub-components
(Reward Responsiveness, Drive, Fun Seeking) and BIS on university students’ drinking, smoking,
and gambling behaviors. Participants included 533 (361 females) undergraduate students who completed
self reports of BIS/BAS (Carver & White, 1994) and retrospective reports of frequency and quantity/money
spent of alcohol use, cigarette use, and gambling. Drinking, smoking, and gambling status
were identiﬁed based on past month behavior. Logistic regressions revealed that Fun Seeking was associated
with increased risk for being a drinker or smoker. BAS and BIS were not supported as unique predictors
of gambling status. Linear regressions revealed that Fun Seeking was positively associated with
drinking, and Drive and Fun Seeking were positively and negatively associated with gambling, respectively.
Findings suggest that different components of BAS sensitivity contribute to drinking and smoking
compared to gambling, such that the tendency to seek out new potentially rewarding experiences puts
individuals at risk for drinking/smoking, while the tendency to pursue appetitive goals increases risk
for gambling.
Crown Copyright Ó 2008 Published by Elsevier Ltd. All rights reserved.
1. Introduction
Rates of drinking among university students continue to be a
concern, with 30–50% of those who drink engaging in heavy episodic
use (e.g., Adlaf, Demers, & Gliksman, 2005). In addition, while
the overall rates of cigarette smoking are declining (CDC, 2005),
about 30% of university students smoke. Of these, almost 13%
smoke at least a pack a day (Rigotti, Lee, & Wechsler, 2000). Gambling
among university students also is a growing concern, with
upwards of 42% of students gambling (LaBrie, Shaffer, LaPlante, &
Wechsler, 2003), and more than 7% gambling at problematic levels
(Shaffer, Hall, & Vander Bilt, 1999). Moreover, there is a high cooccurrence
of these risk behaviors in undergraduates (e.g., Stuhldreher,
Stuhldreher, & Forrest, 2007).
Personality has been identiﬁed as an important inﬂuence on
drinking, smoking, and gambling during early adulthood (Gupta,
Derevensky, & Ellenbogen, 2006; McChargue, Cohen, & Cook,
2004; Sher, Trull, Bartholow, & Vieth, 1999). Addiction models
emphasize reinforcement as a critical factor, such that drinking,
smoking, and gambling are motivated by either the desire to experience
positive (e.g., euphoria) or decrease negative (e.g., offer escape)
outcomes (Cooper, Frone, Russell, & Mudar, 1995; Doran,
McChargue, & Cohen, 2007; Raylu & Oei, 2002). Accordingly, models
of motivationally-relevant individual difference variables provide
a useful framework for examining personality risk for
addictive behaviors.
Gray’s (1975) original reinforcement sensitivity theory attributes
behavior to the relative strength of two motivational
systems – the behavioral approach and behavioral inhibition systems
(BAS and BIS). These are thought to be causal systems that
underlie more secondary personality constructs (e.g., extraversion,
neuroticism) and as such are important for understanding etiology
of addictive behaviors (see Corr, 2008 for review). The BAS
is an approach system that manifests itself in the active pursuit
of reward. Those with a strong BAS are characterized by a sensitivity
to cues signaling potential reward, thus may be more attentive
to the potential rewarding effects of addictive behaviors, such
as the euphoric effects of alcohol (Marlatt, 1987), the pleasurable
effects of smoking (Doran et al., 2007), or the excitement about a
potential win when gambling (Raylu & Oei, 2002). Accordingly,
those with a strong BAS are at risk for engaging in addictive
behaviors for positive reinforcement. Gray and McNaughton
(2000); see Corr, 2008 for review) revised the BIS/BAS theory,
0191-8869/$ - see front matter Crown Copyright Ó 2008 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.paid.2008.12.002
* Corresponding author. Tel.: +1 902 494 3793; fax: +1 902 494 6585.
E-mail address: Roisin.OConnor@Dal.ca (R.M. O’Connor).
Personality and Individual Differences 46 (2009) 514–519
Contents lists available at ScienceDirect
Personality and Individual Differences
journal homepage: www.elsevier.com/locate/paid
and while the function of the BAS remained largely unchanged, a
new conceptualization of the BIS was offered. Originally the BIS
was thought to inhibit behavior in response to cues of potential
punishment. However, the revised theory views the BIS as a conﬂict
resolution system; one that moves an individual towards a
decision of behavioral approach or avoidance by drawing attention
to the potential dangers of a behavior. Thus, a high BIS individual
will overattend to the warning signs of a behavior, leading
to avoidance. A strong BIS gives rise to high levels of anxiety (Gray
& McNaughton, 2000), which may lead to the inclination to selfmedicate
(Pihl & Peterson, 1995). Accordingly, those with a strong
BIS may be at risk for engaging in addictive behaviors for negative
reinforcement.
The BIS and BAS have been linked to many of the other personality
constructs in the extant literature. For example, Eysenck’s
neuroticism and extraversion (Eysenck & Eysenck, 1991) are
thought to map onto (although not directly) Gray’s model (Corr,
2002). Extraversion is thought to reﬂect the balance of the BIS–
BAS sensitivities, where an imbalance favoring BAS leads to elevated
extraversion. Conversely, neuroticism is thought to reﬂect
the strengths of the BIS–BAS, such that a strong BIS that disproportionably
suppresses the BAS leads to elevated neuroticism.1
While there is theoretical support for the BAS and BIS risk pathways
to addictive behaviors, the BAS pathway has received much
more attention and consistent empirical support. A strong BAS
has been associated with elevated risk for alcohol and illicit drug
abuse (e.g., Franken, Muris, & Georgieva, 2006), as well as the
sub-clinical levels of alcohol misuse that are prevalent on university
campuses (e.g., O’Connor & Colder, 2005). Research examining BAS
as a risk factor for smoking and gambling is more limited. One study
identiﬁed BAS as a predictor of substance use including cigarette
smoking, but the direct link to smoking could not be disaggregated
(Knyazev, 2004). BAS has also been associated with gambling indirectly.
For example, a strong BAS was associated with poor performance
on the Iowa Gambling Task (Suhr & Tsanadis, 2007),
however this is not a direct measure of gambling propensity. While
there is a paucity of research examining the direct association between
BAS and smoking and gambling, there is evidence to suggest
that disinhibited behavior and low constraint – indicators of a
strong BAS – are associated with risk for these behaviors (e.g., Slutske,
Caspi, Mofﬁtt, & Poulton, 2005; White, Pandina, & Chen, 2002).
Together these data suggest that BAS may be a common personality
risk for heavy drinking, smoking, and problem gambling, however,
research incorporating direct measures of BAS is needed.
Empirical research examining BIS as a risk factor for substance
misuse and problem gambling is similarly limited. While there is
an emerging literature supporting use of cigarettes and gambling
for negative reinforcement purposes (e.g., Doran et al., 2007; Raylu
& Oei, 2002) there appears to be no direct examination of the relations
of BIS to smoking and gambling. Furthermore, examination of
BIS-related constructs as risk factors for substance use and gambling
have yielded inconsistent ﬁndings. Some evidence supports
an anxiety-related pathway to heavy drinking, smoking, and problem
gambling (e.g., Audrain, Lerman, Gomez-Caminero, Boyd, & Orleans,
1998; see Carrigan & Randall, 2003; el-Guebaly et al., 2006),
while other ﬁndings do not support BIS and BIS-related constructs
as signiﬁcant predictors (e.g., O’Connor & Colder, 2005). Moreover,
there is some theoretical support for a negative relation between
BIS and these problem behaviors, such that those with a weak
BIS are at greater risk for engaging in impulsive behaviors (Corr,
2002).
Research examining both BAS and BIS as predictors of problematic
alcohol use suggests that a strong BAS may be particularly
important among undergraduates (O’Connor & Colder, 2005).
Accordingly, the primary goal of the current study was to directly
examine BAS as a risk factor for heavy drinking, smoking, and high
levels of gambling in university students. Research suggests that
global BAS is comprised of distinct components (e.g., Carver &
White, 1994), such that there are three lower-order BAS components:
Reward Responsiveness (positive response to reward or
anticipation of reward), Drive (pursuit of rewarding goals), and
Fun Seeking (seeking out new rewards). Fun Seeking and (to a lesser
degree) Drive, but not Reward Responsiveness, are associated
with risk-taking and novelty seeking behavior (Carver & White,
1994; Smillie, Jackson, & Dalgleish, 2006). Moreover, only Fun
Seeking and Drive have been linked to alcohol misuse and illicit
drug use (Franken & Muris, 2006; Franken et al., 2006). These data
suggest that there may be utility in considering the distinct lowerorder
BAS constructs when modeling risk for addictive behaviors.
No research to date has elucidated the unique and shared associations
between these three BAS constructs and smoking and gambling
behavior. In the current study, Reward Responsiveness,
Drive, and Fun Seeking were examined as unique predictors of
undergraduates’ drinking, smoking, and gambling behavior. This
allowed us to consider the common and possibly distinct etiology
of these behaviors. We hypothesized that like alcohol use, Fun
Seeking and/or Drive would be important predictors of smoking
and gambling. In addition, a secondary goal was to examine BIS
as a predictor of these three behaviors.
2. Method
2.1. Participants and procedure
Participants included 533 (361 females) undergraduate students
enrolled at an Eastern Canadian university. On average, participants
were 18.9 years old (SD = 3.0), the majority (84%) were in
their ﬁrst year, and 58% were Euro-Canadian and 30% identiﬁed as
‘‘other”. Minority groups included Aboriginal, Asian-Canadian, and
African-Canadian. The measures for this study were completed as
part of the Psychology Department’s mass screening questionnaire
battery. Participation was compensated with course credit.
2.2. Measures
BIS/BAS. Carver and White’s (1994) 20-item measure, which is
based on Gray’s (1975) original BIS/BAS theory, includes one BIS
scale (7 items, e.g., I worry about making mistakes) and three BAS
scales: Reward Responsiveness (5 items, e.g., When I get something
I want, I feel excited and energized), Drive (4 items, e.g., If I see a
chance to get something I want, I move on it right away), and Fun
Seeking (4 items, e.g., I crave excitement and new sensations). Responses
were made on 4-point scales (1 = strong disagreement to
4 = strong agreement), and were totaled to form scale scores. The
psychometric properties of these scales are well supported (e.g.,
Heubeck, Wilkinson, & Cologon, 1998; Ross, Millis, Bonebright, &
Bailley, 2002). In the present study, the Cronbach’s alpha for the
BIS scale was .75, and for the BAS scales the alphas were .81 for Reward
Responsiveness, .88 for Drive, and .82 for Fun Seeking.
Drinking, Smoking, and Gambling. Behaviors over the past 30
days were assessed. Four-point scales were used to assess frequency
of alcohol use (Did Not Drink or 1 = once to 4 = 6 or more
times), cigarette use (Did Not Smoke or 1 = 1 or 2 days to 4 = 15–
30 days), and gambling (Did Not Gamble or 1 = once to 4 = more than
6 times) in the past 30 days. Five-point scales were used to assess
quantity of alcohol use (Did Not Drink or 1 = one [drink] to 5 = 10 or
1
Providing a full review of the theoretical links between Gray’s model and the
many other existing personality constructs is beyond the scope of this paper. The
reader is referred to the very recently published book by Corr (2008) which provides
an excellent review.
R.M. O’Connor et al. / Personality and Individual Differences 46 (2009) 514–519 515
more [drinks]) and cigarette use (Did Not Smoke or 1 = <1 (couple of
drags) to 5 = 1 pack or more), and money spent gambling (Did Not
Gamble or 1 = $1–$10 to 5 = more than $200) on a typical day in
the past month. Dichotomous variables were also formed indicating
whether or not the person was a drinkers, smokers, or gamblers.
Individuals were identiﬁed as such if they engaged in the
respective behavior at least once in the past month.
3. Results
3.1. Descriptives and correlation analysis
Ninety one percent of the sample were identiﬁed as drinkers,
21% as smokers, and 20% as gamblers, with 7% endorsing all three
behaviors. Positive zero-order correlations were supported between
the three BAS scales (rs = .36–.45, ps < .001). Zero-order correlations
(Table 1) supported signiﬁcant positive correlations
between drinking behavior and the Drive and Fun Seeking scales
and between being a smoker and the Fun Seeking scale. The BAS
scales were not signiﬁcant correlates of gambling behavior. The
correlations between total BAS and the risk behaviors are also presented.
Total BAS was signiﬁcantly and positively correlated with
drinking behaviors as expected and with being a smoker. BIS was
a signiﬁcant negative correlate of heavy drinking and being a
gambler.
3.2. Regression analyses: BIS/BAS predicting risk behaviors
Age and sex were signiﬁcant correlates of some of the criterion
variables, and thus were controlled for in the models. Using hierarchical
models, age and sex were entered on the ﬁrst step. Final
models and change in model ﬁt from the ﬁrst to ﬁnal model are
presented in Tables 2 and 3.
First, the utility of the BIS/BAS scales in predicting being a drinker,
smoker, and gambler, respectively, were examined using a set
of hierarchical logistic regressions (Table 2). The BIS/BAS scales
predicted signiﬁcant incremental variance over and above demographics
for drinking and smoking, but not gambling. Examination
of ﬁrst order effects revealed that the odds of being a drinker and
smoker signiﬁcantly increased by 1.23 and 1.17, respectively, with
each unit increase on Fun Seeking scores.
Next, the BIS/BAS scales were examined as predictors of drinking,
smoking, and gambling behaviors using hierarchical linear
regressions (Table 3). Cohen’s (1988) f2
was used as an index of effect
size. BIS/BAS scales predicted signiﬁcant incremental variance
over and above demographics for drinking and gambling (marginal
DR2
for gambling frequency), but not smoking behaviors. First order
effects revealed elevated Fun Seeking scores were signiﬁcantly
associated with risk for heavy and frequent drinking (f2
s = .03),
while low scores were associated with risk for increased spending
when gambling (f2
= .04). These effect sizes were in the small
range. Elevated Drive scores were signiﬁcantly associated with risk
for frequent gambling (f2
= .07) and increased spending when gambling
(f2
= .08). These effect sizes were in the moderate range.
4. Discussion
Consistent with motivational models of BIS and BAS (Fowles,
2001; Gray, 1975), and extant literature supporting distinct components
of the BAS (Carver & White, 1994), we examined BIS,
BAS-Reward Responsiveness, BAS-Drive, and BAS-Fun Seeking as
unique predictors of drinking, smoking, and gambling. While the
effect sizes of our ﬁndings were in the small to moderate range,
the theoretical implications of clarifying the BIS/BAS risk speciﬁc
to these potentially addictive behaviors supports the interpretation
of these small and moderate effects (Cohen, 1988).
Our simultaneous examination of the BAS components revealed
that Fun Seeking – the inclination to seek out new rewards – increased
the risk for being a drinker, engaging in heavy and frequent
drinking, and being a smoker. This positive relation between Fun
Seeking and alcohol misuse has been previously documented in
the literature (Franken & Muris, 2006). While not as consistently
supported as Fun Seeking, some of this former work has also found
a strong Drive to be predictive of alcohol misuse. In the current
study a strong Drive was correlated with all of the alcohol use measures
in bivariate correlations; however, Drive was not uniquely
associated when examined in the context of the other predictors.
The risk for smoking associated with a strong motivation to seek
out fun is supported in the literature, as Fun Seeking is thought
to capture impulsivity (Smillie et al., 2006) which has been implicated
in smoking risk (e.g., Doran et al., 2007). Notably, the present
ﬁndings suggest that the motivation to seek out fun and new reward
put individuals at risk for being a smoker, but did not further
inﬂuence the extent of cigarette use. Unlike alcohol use, which is
more socially accepted and thus used by a broader range of individuals,
smokers may be a distinct group from non-smokers.
Accordingly, Fun Seeking may systematically increase risk for heavy
and frequent drinking, while it may distinguish smokers from
Table 1
Descriptive statistics for all variables and correlations between predictor and criterion variables.
Predictor variables Criterion variables
Drinker (N = 533) Drinkers only
(N = 485)
Smoker (N = 533) Smokers only
(N = 112)
Gambler (N = 533) Gamblers only (N = 106)
Alc Frq Alc Qty Cig Frq Cig Qty Gmbl Frq Gmbl $Spent M SD
BAS .14**
.15**
.16**
.11**
.04 À.02 .02 À.06 À.10 40.82 5.97
Reward Responsiveness .05 .03 .02 .02 .01 À.01 À.04 À.15 À.19 17.72 2.40
Drive .12**
.11*
.12**
.08 .05 .02 .06 .13 .11 10.93 2.65
Fun Seeking .16**
.19**
.23**
.16**
.03 À.08 .03 À.12 À.18 12.18 2.60
BIS À.02 À.05 À.17**
À.08 À.03 À.06 À.20**
À.16 À.08 20.74 3.48
Sex .10*
À.10*
À.40**
À.08 À.04 À.10 À.32*
À.19 À.11 0.68
Age À.02 À.19**
À.21**
.03 .36**
.39**
À.03 .01 .29**
18.87 3.04
M .91 3.10 3.27 .21 2.10 1.86 .20 1.76 1.30
SD .97 1.07 1.22 .95 .87 .69
Note: Drinker, Smoker, Gambler: 1 = Yes, 0 = No; Sex: 1 = female, 0 = male; BAS = Behavioral Approach System; BIS = Behavioral Inhibition System; Alc Frq = Frequency of
alcohol use in past month; Alc Qty = Quantity of alcohol use on a typical day in past month; Cig Frq = Frequency of cigarette use in past month; Cig Qty = Quantity of cigarette
use on a typical day in past month; Gmbl Frq = Frequency of gambling in past month; Gmbl $Spent = Money spent gambling on a typical day in past month.
*
p <.05.
**
p <.01.
516 R.M. O’Connor et al. / Personality and Individual Differences 46 (2009) 514–519
non-smokers, but have no additional impact on speciﬁc smoking
behaviors.
In contrast to the BAS risk for drinking and smoking, BAS did not
appear to be a risk factor for being a gambler. However, a strong
Drive – the motivation to pursue rewarding goals -was found to increase
the risk for frequent gambling and for heavy spending when
gambling. The former ﬁnding may reﬂect our overly broad deﬁnition
of gambling, which allowed us to capture behaviors that are
not typical of gambling (e.g., buying lottery tickets). It is possible
that BAS would predict gambling status if a more narrow deﬁnition
of gambling was used or if speciﬁc examples of types of gambling
were given (e.g., slots, sports betting). In contrast, the latter ﬁnding
Table 2
Final model of hierarchical logistic regression: BIS/BAS scales predicting drinking, smoking, and gambling status.
N = 533 Drinker (1 = Yes, 0 = No) Smoker (1 = Yes, 0 = No) Gambler (1 = Yes, 0 = No)
Predictor variables B SE Odds ratio B SE Odds ratio B SE Odds ratio
Sex (1 = female, 0 = male) 1.03**
.34 2.81 À.23 .24 .79 À1.43**
.24 .24
Age À.01 .05 .99 .04 .03 1.04 À.04 .05 .96
Reward Responsiveness À.09 .08 .91 À.04 .06 .97 .02 .06 1.02
Drive .13 .07 1.14 .03 .05 1.03 .05 .05 1.05
Fun Seeking .21**
.07 1.23 .16**
.05 1.17 À.04 .05 .96
BIS À.03 .05 .97 À.04 .03 .96 À.10**
.04 .91
Cox and Snell (1989) Index of goodness of ﬁt .05 .04 .11
LR v2
Diff Test (df = 4) 19.94**
16.22**
8.75
Note: BIS = Behavioral Inhibition System; LR v2
Diff Test = Logistic Regression v2
Difference Test (Cohen, Cohen, West, & Aiken, 2003), used to assess added predictability of
ﬁnal model over and above model including only sex and age; Wald v2
statistic used to assess signiﬁcance of Bs.
**
p < .01.
Table 3
Final model of hierarchical linear regression: BIS/BAS scales predicting drinking, smoking, and gambling behaviors.
Predictor variables Drinkers only (N = 485)
Alcohol use frequency Alcohol use quantity
B SE b B SE b
Sex (1 = female, 0 = male) À.15 .10 À.07 À.88**
.10 À.38
Age À.06**
.02 À.17 À.07**
.02 À.19
Reward Responsiveness À.02 .02 À.04 .01 .02 .02
Drive .02 .02 .06 .02 .02 .04
Fun Seeking .06**
.02 .15 .07**
.02 .15
BIS À.00 .01 À.01 À.01 .01 À.04
R2
.08 .24
DR2
.03**
.03**
Smokers only (N = 112)
Cigarette use frequency Cigarette use quantity
B SE b B SE b
Sex (1 = female, 0 = male) À.02 .24 À.01 À.11 .18 À.05
Age .14**
.04 .38 .12**
.03 .42
Reward Responsiveness .02 .06 .04 .04 .04 .12
Drive À.00 .05 À.00 À.00 .04 À.00
Fun Seeking À.00 .05 À.00 À.05 .04 À.15
BIS .04 .04 À.11 À.04 .03 À.15
R2
.14 .19
DR2
.01 .03
Gamblers only (N = 106)
Gambling frequency Gambling money spent
B SE b B SE b
Sex (1 = female, 0 = male) À.21 .18 À.12 À.08 .14 À.06
Age .00 .05 .00 .12**
.04 .28
Reward Responsiveness À.05 .04 À.15 À.05 .03 À.20
Drive .10*
.04 .30 .08**
.03 .31
Fun Seeking À.07 .04 À.20 À.07*
.03 À.26
BIS À.02 .03 À.07 À.00 .02 À.01
R2
.12 .20
DR2
.08 
.11*
Note: BIS = Behavioral Inhibition System; DR2
: added predictability of ﬁnal model over and above model including only sex and age.
 
p = .07.
*
p < .05.
**
p < .01.
R.M. O’Connor et al. / Personality and Individual Differences 46 (2009) 514–519 517
makes intuitive sense, as a strong drive to pursue rewarding goals
seems essential for being a frequent, heavy spending gambler,
where losses are frequent and wins are quite infrequent. Unexpectedly,
Fun Seeking was found to be negatively associated with
money spent when gambling, suggesting that low motivation to
seek out new rewards increases gambling persistence. On the one
hand, this makes conceptual sense as those who do not need
immediate and new rewards may be able to endure the intermittent
positive outcomes involved with gambling and thus continue
to spend money. This would ﬁt with new speculation (see Corr,
2008) that the BAS is a complex and multidimensional system,
which may consist of separate but overlapping processes. At the
early or anticipatory stage there is the drive process, and this concerns
active pursuit of goals. At the later stage, the BAS is involved
in the emotional excitement that is in response to the biological
reinforcer. Corr (2008) suggests that Carver and White’s (1994)
Fun Seeking assesses the behaviors that are closer to the reinforcer.
Taken together, variance in the strength of these different facets of
the BAS may lead to risk for different behaviors.
The ﬁnding of a negative association between Fun Seeking and
gambling is surprising, however, when considered in the context of
drinking and smoking, which are positively associated with Fun
Seeking, and which have been shown to co-occur with gambling
(Stuhldreher et al., 2007). While replication and further investigation
of this ﬁnding is needed, we can offer some speculations at
this point. One possible explanation is that Fun Seeking is not
the common risk factor for all three risk behaviors. Possibly elevated
Fun Seeking is not even the broad risk factor for alcohol
use, but rather it is linked to seeking out certain enjoyable social
contexts, such as university house parties, where drinking (but
not gambling) opportunities are typically present. Another explanation
is that having both elevated Fun Seeking and Drive may
put individuals at risk for all three risk behaviors as the strong
Drive mitigates the negative effects of Fun Seeking on gambling
behavior. Alternatively, those with a weak motivation for Fun
Seeking may tend not to drink or smoke and, thus, the negative
association with money spent when gambling may be driven by
gamblers only. Additional analyses of the current data offer some
preliminary support for this, in that the negative association between
Fun Seeking and gambling behavior was not supported
when only drinkers and gamblers were considered. An important
direction for further research is to explicate the speciﬁc risk associated
with being only a drinker, smoker, or gambler, compared
with being some combination of these. The current study was limited
for such an examination due to small sample size (e.g., only 7%
identiﬁed as drinkers, smokers, and gamblers).
Contrary to theory (e.g., Pihl & Peterson, 1995), and our hypothesis
– which was based on the emerging literature that supports a
negative reinforcement pathway to smoking and gambling (e.g.,
Doran et al., 2007; Raylu & Oei, 2002) – a strong BIS was not associated
with drinking, smoking, or gambling risk in the current
study. In fact, zero-order correlations indicated that a weak BIS
was associated with heavy drinking and being a gambler. There
is some, albeit mixed, support for the role of a low BIS in externalizing
behaviors (e.g., drug use; Hundt, Kimbrel, Mitchell, & NelsonGray,
2008). However, BIS was not supported as a unique predictor
in the context of demographic and BAS inﬂuences in the present
study. Our ﬁndings, as well as those of extant research (e.g., Hundt
et al., 2008; O’Connor & Colder, 2005) suggest that the BIS inﬂuence
on risk behaviors may be complex. For example, the null ﬁndings
could reﬂect a curvilinear relation between BIS and addictive
behaviors, such that those with either a strong or weak BIS are at
risk. Alternatively, moderators may be involved in the complex
inﬂuence of a strong BIS on these risk behaviors. On the one hand,
those who are high on BIS may experience anxiety and be at risk
for self-medicating with substances or using gambling as an avoidance
strategy. On the other hand, these individuals may avoid
these types of behaviors because of their heightened focus on the
associated potential dangers (e.g., hangovers; societal disapproval;
loss of money). Finally, an alternative explanation for the null ﬁndings
is that an elevated BIS (i.e., anxiety) may only be a risk factor
for very heavy or disordered levels of substance use (e.g., Ball,
2005) and gambling, levels of use that were not assessed in this
study.
While these ﬁndings do further the etiological addictions literature,
and suggest directions for future research, a few limitations
are noteworthy. First, BAS was originally theorized as a single construct;
thus, examinations of sub-components may not capture the
true aspect of BAS that is linked to substance use and gambling.
Notably, however, a modiﬁed version of the widely used Sensitivity
to Punishment and Sensitivity to Reward Questionnaire (Torrubia,
Ávila, Moltó, & Caseras, 2001) which was designed to asses a single
BAS construct, revealed the same three BAS components in factor
analyses (Colder & O’Connor, 2004). Second, our deﬁnition of drinker,
smoker, and gambler (i.e., drank/smoked/gambled P1/past
month) may have excluded those who engage in these behaviors
less frequently. Also, our assessments of past month typical quantity
and frequency of substance use and gambling may have been
inﬂuenced by transient factors such as exam periods, and the lack
of distinction between weekday and weekend behaviors may have
obscured potential relations between Fun Seeking and elevated
weekend risky behaviors, for example. Third, the Carver and White
(1994) measure of BIS that was used is based on the original theory
(Gray, 1975), and thus does not measure its potential role in conﬂict
resolution (Gray & McNaughton, 2000). However, this measure
does assess the heightened attention to warning signs that is paramount
to anxiety, and which may be central to a negative reinforcement
pathway to addictive behaviors. Lastly, caution should
be given to the generalizability of these ﬁndings to students in
their upper years and non-university student samples. An important
direction for future research will be extending these risk models
to other young adult samples.
In conclusion, our hypothesis that the Drive and Fun Seeking
components of BAS, which are indicators of risk taking and novelty
seeking (Carver & White, 1994; Smillie et al., 2006), would be positively
associated with university students’ drinking, smoking, and
gambling behaviors was partially supported. Moreover, our ﬁndings
generally supported an elevated BAS compared to BIS as an
important predictor of undergraduates’ risk behaviors. This could
suggest an individual difference (i.e., gene) by environment interaction,
such that during this period when parental control is low
and availability of substances and opportunities to gamble are
high, the predisposition for risk taking/novelty seeking manifests
itself in substance use and gambling. In terms of clinical implications,
these ﬁndings may suggest the need to target different motivations
for drinkers and smokers compared to gamblers. For
example, it may be beneﬁcial for treatments to focus on identifying
alternative novel and rewarding experiences for those at risk for
drinking and smoking, while identifying alternative appetitive
goals for those at risk for gambling. Furthermore, treatments may
beneﬁt from a multi-focus for those engaging in multiple addictive
behaviors.
Acknowledgements
The authors gratefully acknowledge Kerry MacSwain for her
assistance with data management. Dr. O’Connor was supported
by a Canadian Institutes of Health Research (CIHR) Fellowship during
completion of part of this work. Dr. Stewart is supported by an
Investigator Award from CIHR and a Killam Research Professorship
from the Faculty of Science at Dalhousie University. Dr. Watt is
supported by Social Sciences and Humanities Research Council
518 R.M. O’Connor et al. / Personality and Individual Differences 46 (2009) 514–519
(SSHRC) and Sport Canada Research Initiative (SCRI) funding. Portions
of this article were presented at the annual meeting of the Research
Society on Alcoholism, July 2007, Chicago, Illinois.
References
Adlaf, E. M., Demers, A., & Gliksman, L. (Eds.). (2005). Canadian campus survey 2004.
Toronto, ON: Centre for Addiction and Mental Health.
Audrain, J., Lerman, C., Gomez-Caminero, A., Boyd, N. R., & Orleans, C. T. (1998). The
role of trait anxiety in nicotine dependence. Journal of Applied Biobehavioral
Research, 3, 29–42.
Ball, S. A. (2005). Personality traits, problems, and disorders: Clinical applications to
substance use disorders. Journal of Research in Personality, 39, 84–102.
Carrigan, M. H., & Randall, C. L. (2003). Self-medication in social phobia: A review of
the alcohol literature. Addictive Behaviors, 28, 269–284.
Carver, C. S., & White, T. L. (1994). Behavioral inhibition, behavioral activation, and
affective responses to impending reward and punishment: The BIS/BAS scales.
Journal of Personality and Social Psychology, 67, 319–333.
CDC (2005). Cigarette smoking among adults – United States, 2003. Morbidity and
Mortality Weekly Report CDC Surveillance Summary, 54, 509–513.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (20th ed.).
Hillsdale, NJ: Erlbaum.
Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/
correlation analysis for the behavioral sciences (3rd ed.). Hillsdale, NJ: Erlbaum.
Colder, C. R., & O’Connor, R. M. (2004). Gray’s reinforcement sensitivity model and
child psychopathology: Laboratory and questionnaire assessment of the BAS
and BIS. Journal of Abnormal Child Psychology, 32, 435–451.
Cooper, M. L., Frone, M. R., Russell, M., & Mudar, P. (1995). Drinking to regulate
positive and negative emotions: A motivational model of alcohol use. Journal of
Personality and Social Psychology, 69, 990–1005.
Corr, P. (2002). J.A. Gray’s reinforcement sensitivity theory: Tests of the joint
subsystems hypothesis of anxiety and impulsivity. Personality and Individual
Differences, 33, 511–532.
Corr, P. J. (Ed.). (2008). The Reinforcement Sensitivity Theory of personality.
Cambridge, UK: Cambridge University Press.
Cox, D. R., & Snell, E. J. (1989). The analysis of binary data (2nd ed.). London, UK:
Chapman & Hall.
Doran, N., McChargue, D., & Cohen, L. (2007). Impulsivity and the reinforcing value
of cigarette smoking. Addictive Behaviors, 32, 90–98.
el-Guebaly, N., Patten, S. B., Currie, S., Williams, J. V. A., Beck, C. A., & Maxwell, C. J.
(2006). Epidemiological associations between gambling behavior, substance
use and mood and anxiety disorders. Journal of Gambling Studies, 22, 275–287.
Eysenck, H. J., & Eysenck, S. B. G. (1991). Manual of the Eysenck personality scales.
London, UK: Hodder and Stoughton.
Fowles, D. C. (2001). Biological variables in psychopathology: A psychobiological
perspective. In H. E. Adams & P. B. Sutker (Eds.), Comprehensive handbook of
psychopathology (pp. 85–141). New York: Kluwer Academic/Plenum.
Franken, I. H. A., & Muris, P. (2006). BIS/BAS personality characteristics and college
students’ substance use. Personality and Individual Differences, 40, 1497–1503.
Franken, I. H. A., Muris, P., & Georgieva, I. (2006). Gray’s model of personality and
addiction. Addictive Behaviors, 31, 399–403.
Gray, J. A. (1975). Elements of a two-process theory of learning. London, UK: Academic
Press.
Gray, J. A., & McNaughton, N. (2000). The neuropsychology of anxiety: An enquiry into
the functions of the septo-hippocampal system (2nd ed.). New York: Oxford
University Press.
Gupta, R., Derevensky, J. L., & Ellenbogen, S. (2006). Personality characteristics and
risk-taking tendencies among adolescent gamblers. Canadian Journal of
Behavioural Science, 38, 201–213.
Heubeck, B. G., Wilkinson, R. B., & Cologon, J. (1998). A second look at Carver and
White’s (1994) BIS/BAS scales. Personality and Individual Differences, 25,
785–800.
Hundt, N. E., Kimbrel, N. A., Mitchell, J. T., & Nelson-Gray, R. O. (2008). High BAS, but
not low BIS, predicts externalizing symptoms in adults. Personality and
Individual Differences, 44, 565–575.
Knyazev, G. G. (2004). Behavioural activation as predictor of substance use:
Mediating and moderating role of attitudes and social relationships. Drug and
Alcohol Dependence, 75, 309–321.
LaBrie, R. A., Shaffer, H. J., LaPlante, D. A., & Wechsler, H. (2003). Correlates of college
student gambling in the United States. Journal of American College Health, 52,
53–62.
Marlatt, G. A. (1987). Alcohol, expectancy, and emotional states: How drinking
patterns may be affected by beliefs about alcohol’s effects. Alcohol, Health, and
Research World, 11, 10–14.
McChargue, D. E., Cohen, L. M., & Cook, J. W. (2004). The inﬂuence of personality and
affect on nicotine dependence among male college students. Nicotine and
Tobacco Research, 6, 287–294.
O’Connor, R. M., & Colder, C. R. (2005). Predicting alcohol patterns in ﬁrst year
college students via motivational systems and reasons for drinking. Psychology
of Addictive Behaviors, 19, 10–20.
Pihl, R. O., & Peterson, J. B. (1995). Alcoholism: The role of different motivational
systems. Journal of Psychiatry and Neuroscience, 20, 372–396.
Raylu, N., & Oei, T. P. S. (2002). Pathological gambling: A comprehensive review.
Clinical Psychology Review, 22, 1009–1061.
Rigotti, N. A., Lee, J. E., & Wechsler, H. (2000). US college students’ use of tobacco
products: Results of a national survey. Journal of the American Medical
Association, 284, 699–705.
Ross, S. R., Millis, S. R., Bonebright, T. L., & Bailley, S. E. (2002). Conﬁrmatory factor
analysis of the behavioral inhibition and activation scales. Personality and
Individual Differences, 33, 861–865.
Shaffer, H. J., Hall, M. N., & Vander Bilt, J. (1999). Estimating the prevalence of
disordered gambling behavior in the United States and Canada: A research
synthesis. American Journal of Public Health, 89, 1369–1378.
Sher, K. J., Trull, T. J., Bartholow, B. D., & Vieth, A. (1999). Personality and alcoholism:
Issues, methods, and etiological processes. In K. E. Leonard & H. T. Blain (Eds.),
Psychological theories of drinking and alcoholism (2nd ed., pp. 54–105). New
York: Guilford.
Slutske, W. S., Caspi, A., Mofﬁtt, T. E., & Poulton, R. (2005). Personality and problem
gambling: A prospective study of a birth cohort of young adults. Archives of
General Psychiatry, 62, 769–775.
Smillie, L. D., Jackson, C. J., & Dalgleish, L. I. (2006). Conceptual distinctions among
Carver and White’s (1994) BAS scales: A reward-reactivity versus trait
impulsivity perspective. Personality and Individual Differences, 40, 1039–1050.
Stuhldreher, W. L., Stuhldreher, T. J., & Forrest, K. Y. Z. (2007). Gambling as an
emerging health problem on campus. Journal of American College Health, 56,
75–83.
Suhr, J. A., & Tsanadis, J. (2007). Affect and personality correlates of the Iowa
Gambling Task. Personality and Individual Differences, 43, 27–36.
Torrubia, R., Ávila, C., Moltó, J., & Caseras, X. (2001). The sensitivity to punishment
and sensitivity to reward questionnaire (SPSRQ) as a measure of Gray’s anxiety
and impulsivity dimensions. Personality and Individual Differences, 31, 837–862.
White, H. R., Pandina, R. J., & Chen, P.-H. (2002). Developmental trajectories of
cigarette use from early adolescence into young adulthood. Drug and Alcohol
Dependence, 65, 167–178.
R.M. O’Connor et al. / Personality and Individual Differences 46 (2009) 514–519 519