Self-Control, Thrill Seeking, and Crime Motivation Matters

Abstract

Self-control theory (SCT), as a control theory, assumes that the pleasures gained from crime are equally obvious and attractive to all. This study brings a consideration of crime as a process into SCT, recognizing that the sensations inherent in offending may not be equally attractive to everyone. In doing so, we test the theory’s equal motivation assumption, bringing a consideration of individual differences in thrill seeking to the fore. Drawing on theory and research on the personality characteristic thrill seeking, we hypothesize that thrill seeking and self-control have independent influences on offending: that motivation to the process of crime matters. In addition, we investigate whether the effects of self-control are contingent on levels of thrill seeking, in part because high thrill seekers are less averse to the process of risk. These hypotheses are tested using data from the Family and Community Health Study, a sample of roughly 700 African American youth and their families. A new measure of self-control is employed in tandem with an existing attitudinal measure of self-control and thrill seeking. Consistent with hypotheses, the results suggest that self-control and thrill seeking have largely independent influences on offending and that the effects of self-control are contingent on levels of thrill seeking. These results provide further evidence that SCT’s assumption of equal motivation to crime is untenable, as individual differences in the personality characteristic thrill seeking influence the likelihood of offending.

Full text

1
CRIMINAL JUSTICE AND BEHAVIOR, Vol. XX, No. X, Month XXXX XX –XX.
DOI: 10.1177/0093854813485575
© 2013 International Association for Correctional and Forensic Psychology
SELF-CONTROL, THRILL
SEEKING, AND CRIME
Motivation Matters
CALLIE H. BURT
Arizona State University
RONALD L. SIMONS
Arizona State University
Self-control theory (SCT), as a control theory, assumes that the pleasures gained from crime are equally obvious and attrac-
tive to all. This study brings a consideration of crime as a process into SCT, recognizing that the sensations inherent in
offending may not be equally attractive to everyone. In doing so, we test the theory’s equal motivation assumption, bringing
a consideration of individual differences in thrill seeking to the fore. Drawing on theory and research on the personality
characteristic thrill seeking, we hypothesize that thrill seeking and self-control have independent influences on offending: that
motivation to the process of crime matters. In addition, we investigate whether the effects of self-control are contingent on
levels of thrill seeking, in part because high thrill seekers are less averse to the process of risk. These hypotheses are tested
using data from the Family and Community Health Study, a sample of roughly 700 African American youth and their fami-
lies. A new measure of self-control is employed in tandem with an existing attitudinal measure of self-control and thrill
seeking. Consistent with hypotheses, the results suggest that self-control and thrill seeking have largely independent influ-
ences on offending and that the effects of self-control are contingent on levels of thrill seeking. These results provide further
evidence that SCT’s assumption of equal motivation to crime is untenable, as individual differences in the personality char-
acteristic thrill seeking influence the likelihood of offending.
Keywords: self-control; thrill seeking; risk taking; motivation; crime; delinquency
S
elf-control theory (SCT; Gottfredson & Hirschi, 1990), as a control theory, views the
motivation to crime as invariant, based on the assumption that the pleasures gained
from crime are equally obvious and attractive to all. This assumption is grounded in the
classical view of humans as hedonistic, and crime as an easy, efficient means to pleasure.
AUTHORS’ NOTE: This research uses data from the Family and Community Health Study (FACHS), a proj-
ect designed by Ron Simons, Frederick Gibbons, and Carolyn Cutrona, and funded by grants from the National
Institute of Mental Health (MH48165, MH62669), the Center for Disease Control (029136-02), the National
Institute on Drug Abuse (DA021898), and the National Institute on Alcohol Abuse and Alcoholism. An earlier
version of this article was presented at the 2011 Annual Meeting of the American Society of Criminology. The
authors would like to thank Kate Fox, Travis Pratt, Carter Rees, and two anonymous reviewers for their help-
ful comments on earlier drafts of the article. Direct all correspondence to Callie H. Burt, School of Criminology
and Criminal Justice, Arizona State University, 411 N. Central Ave., Ste. 600, Phoenix, AZ 85004; e-mail:
chburt@asu.edu.
485575
CJBXXX10.1177/0093854813485575Criminal Justice and BehaviorBurt and Simons / Self-Control and Thrill Seeking
research-article2013

2 CRIMINAL JUSTICE AND BEHAVIOR
This equal motivation assumption, however, focuses on the ends of crime—we all want
“money without work, sex without courtship, revenge without court delays” (Gottfredson
& Hirschi, 1990, p. 89)—and generally ignores another potential source of variability: dif-
ferential attractions to the means.
Crime is a process. An act of law violation involves sensual dynamics (Katz, 1988),
emotional energy (Collins, 2004, 2008), and always entails the risk of punishment, whether
legal, moral, social, and/or physical (Bentham, 1823/2007). When deciding whether to
commit a crime, a potential offender is deciding to seek the end reward of crime and to
engage in a process imbued with risk. Individual differences in attitudes toward risk, then,
could reasonably be expected to influence the likelihood of offending by shaping the per-
ceived (and actual) rewards of the process of crime (e.g., Felson & Osgood, 2008; McCarthy
& Hagan, 2005).
An extensive body of work in psychology demonstrates that individuals differ in their
attitudes toward risk in large part as a function of a personality characteristic known vari-
ously as thrill seeking (Gullone & Moore, 2000), sensation seeking (Zuckerman, 1994,
2007), venturesomeness (Eysenck & Eysenck, 1978), or risk seeking (e.g., Arnett, 1994;
Hagan, Gillis, & Simpson, 1985; Knowles, Cutter, Walsh, & Casey, 1973). In the present
article, we refer to this personality characteristic as thrill seeking.
1
This body of work
evinces that individuals differ in their attitudes toward risky acts in part as a function of
thrill seeking, which shapes the amount of pleasure they receive (or anticipate receiving)
from the psychological and physiological sensations inherent in risky activities (e.g.,
Zuckerman, 2007). Given that crime is inherently risky, the sensations generated by involve-
ment in the process of crime should be more likely to be evaluated as more pleasurable by
those high in thrill seeking and as more painful by those low in thrill seeking. Given the
rational, hedonistic actor assumed by control theories, this implies differential motivation
to crime in part as a function of thrill seeking.
This idea of differential motivation clearly contradicts the equal motivation assumption
of SCT. As Felson and Osgood (2008) explained,
A preference for risk is a motivating factor, not a disinhibition. Thrill seekers are not necessarily
impulsive—that is, they do not necessarily have low self-control. They can make plans to go
bungee jumping or commit robbery. Their behavior is unrelated to their ability to consider
costs when contemplating a crime. (p. 164)
Theoretically, then, the concept of thrill seeking and its role in decision making is distinct
from self-control—defined as variation in the extent to which individuals consider the long-
term consequences of their actions
2
(see also, Jones & Lynam, 2009; Marcus, 2004).
Consistent with this theoretical conceptualization, research provides some evidence that
self-control and attitudes toward risk are empirically distinguishable (e.g., Horvath &
Zuckerman, 1993; Jones & Lynam, 2009; Steinberg et al., 2008; Whiteside & Lynam,
2001). In addition, the few studies that have examined their behavioral sequelae find that
thrill seeking and self-control have independent effects on antisocial behavior (Lynam &
Miller, 2004) and substance use (Jones & Lynam, 2009) as well as different developmental
timetables (e.g., Quinn & Harden, 2013; Steinberg et al., 2008). Based on these ideas, the
present study tests the equal motivation assumption of SCT, examining whether individual
differences in thrill seeking influence the likelihood of offending net of self-control.

Burt, Simons / SELF-CONTROL AND THRILL SEEKING 3
In addition, the current study considers the possibility that the inhibitory effect of self-
control on offending is contingent upon individuals’ levels of thrill seeking. In other words,
we test whether thrill seeking moderates the effects of self-control. Considering and cogniz-
ing the potential negative consequences of possible action alternatives is proposed to inhibit
crime because individuals realize that the potential negative consequences of offending far
outweigh any possible (material) benefits (Gottfredson & Hirschi, 1990). We test the pos-
sibility that for individuals high in thrill seeking, considering potential consequences and
recognizing that an action alternative risks severe long-term consequences may not inhibit
crime because this process of taking risks is itself highly pleasurable. Phrased differently,
the intense reward effects of risky acts, such as crime, may outweigh the potential negative
consequences for the high thrill seekers. However, we would expect that for low thrill seek-
ers, who have a strong aversion to risky behaviors, self-control would be more strongly
related to crime. Rather than being an added potential pleasure, the process of risk taking is
an added pain for individuals low in thrill seeking.
In sum, this study brings a consideration of the process of crime into SCT, recognizing
that the sensations inherent in offending may not be equally attractive to everyone. In the
following pages, we review relevant theoretical and empirical work, namely, theory and
research concerned with SCT as well as that related to individual differences in thrill seek-
ing. Drawing on this work, we hypothesize that self-control and thrill seeking have inde-
pendent influences on offending. In addition, we explore whether the effect of self-control
on offending is contingent upon levels of thrill seeking, such that considering the long-
term costs of actions will be less delinquency-inhibiting among thrill seekers. In doing so,
we utilize a novel measure of self-control that attempts to more directly tap into time per-
spective at the point of decision making without capturing thrill seeking. We test these
hypotheses using data from the Family and Community Health Study (FACHS), a sample
of African American children and their primary caregivers in Georgia and Iowa. Finally,
we discuss what these findings suggest about the assumptions of SCT and directions for
future work.
SELF-CONTROL THEORY
Roughly 20 years ago, Gottfredson and Hirschi (1990) boldly presented their theory of
self-control. Defining crimes as “acts of force or fraud undertaken in the pursuit of self-
interest,” (p. 15) they maintain that the vast majority of offenses are trivial, unspecialized,
involve little or no planning, and require little skill or learning. The theory’s conception of
crime is grounded in the classical tradition’s view of human nature and action (e.g., Beccaria,
1764/1963; Bentham, 1823/2007). SCT assumes that humans are hedonistic and rational.
Therefore, human action is determined by hedonic calculus. From this perspective, people
choose acts, such as crime, when the perceived benefits outweigh the perceived costs.
SCT assumes that criminal acts provide immediate gratification to all but risk severe
long-term punishment (“In all cases, the behavior [crime] produces immediate short-term
pleasure to the actor”; Gottfredson & Hirschi, 1990, p. 83). The theory further assumes that
the perception and appreciation of the short-term consequences of action alternatives is
invariant across individuals, and, therefore, positive motivation to crime is no longer a nec-
essary element of the theory. According to SCT, we are all equally attracted to the immedi-
ate pleasures to be gained from theft, physical assault, rape (“sex without courtship”), and

4 CRIMINAL JUSTICE AND BEHAVIOR
tax evasion, and if long-term consequences of behavior were removed, then presumably we
would all engage in these immediately gratifying acts.
Behavior, however, has both short- and long-term consequences. Due to various sanction
systems—the least of which are not only penal/political sanctions but also moral (social),
religious, and physical sanctions—the criminal act risks objectively more costly conse-
quences in the long run (“in all cases the behavior [crime] tends to entail long-term costs”;
Gottfredson & Hirschi, 1990, p. 83). Thus, a rational being that perceives the full risks of an
act of crime should, according to the theory, choose not to engage in crime.
How then does crime result if individuals are assumed to be both hedonistic and rational?
According to SCT, the individual difference responsible for different cost-benefit calcula-
tions and, thus, the choice of crime or noncrime relates to time perspective in considering the
consequences of action alternatives. Specifically, the theory proposes that individual varia-
tion in the tendency to consider the long-term potential costs of actions accounts for indi-
vidual variations in perceptions of pleasures and pains. At one end of the spectrum are
individuals whose cost-benefit calculations are based on relatively immediate consequences;
at the other end of the spectrum are those that consider the consequences of behavior across
long time frames. Therefore, according to SCT, the former individuals would choose crime
because it is immediately gratifying and the concomitant painful consequences are largely
distant and, therefore, out of the realm of their considerations, whereas the latter avoid crime
because they are able to realize that crime is not utility-maximizing in the long term. This
individual variable (time perspective in considering the costs of actions) allows pleasure-
maximizing individuals to engage in objectively more costly acts because their perceptions
of pleasures and pains are limited to immediate consequences. Thus, offenders remain hedo-
nistic and rational, but are myopic. “Like the family dog, typical offenders think only of the
stimuli they face and the immediate payoff. Dogs and criminals live in the present and there-
fore have difficulty resisting temptation and waiting” (Felson & Osgood, 2008, p. 162).
This myopic time perspective is called low self-control. Specifically, Gottfredson and
Hirschi (1990) defined low self-control as the “tendency of individuals to pursue short-term
gratification without consideration of the long-term consequences of their acts” (p. 177) or
“the tendency to consider or ignore the long-term consequences of one’s acts” (Hirschi &
Gottfredson, 1993, p. 49). Because low self-control is the sole individual variable respon-
sible for offending in SCT, it is tantamount to criminality (the propensity to commit crime).
Importantly, the theory assumes that other individual variables are largely irrelevant to
criminal propensity. Most individual differences theorized to be causes of crime (e.g., devi-
ant peers, school failure, poor relationship/bonds) are said to be manifestations of self-
control and, thus, only spuriously related to crime. Self-control, the tendency to consider the
long-term consequences of action alternatives at the point of decision making, is the pri-
mary source of individual differences in criminal propensity according to the theory.
Empirical assessments of SCT are plentiful. Briefly, tests of the central proposition—
that low self-control is a primary cause of crime—have been largely supportive, even if the
evidence is not as overwhelming as SCT theorists claim (e.g., Gottfredson, 2008), and the
measurement of self-control is problematic (e.g., Felson & Osgood, 2008; Marcus, 2004).
Scholars have concluded that low self-control is “one of the strongest known correlates of
crime,” but that other individual differences identified by rival theories continue to influ-
ence criminal propensity net of self-control (e.g., Pratt & Cullen, 2000). Moreover, although
research on this underlying process is sparse, some research suggests that the cognitive

Burt, Simons / SELF-CONTROL AND THRILL SEEKING 5
evaluation of potential punishments mediates the effects of self-control (e.g., Intravia,
Jones, & Piquero, 2012).
To reiterate, as a control theory, SCT is grounded upon the assumption that crime is a
universal way of maximizing pleasure in the short term; we are all equally motivated to the
immediate pleasures from the act. What distinguishes offenders and nonoffenders are dif-
ferences in their perceptions of long-term costs. Recently, Gottfredson (2011) reiterated,
“Self-control theory is a restraint theory. Motivation is not a variable.”
THRILL SEEKING AND MOTIVATION TO CRIME
Traditionally, theories have focused on the material benefits of crime, such as the mon-
etary or material value of the goods obtained from the illegal act. Other recognized plea-
sures or motivations to crime are noneconomic ends, such as status, respect, and esteem
gained from the successful completion of acts that are prohibited by law (e.g., Anderson,
1999; Cohen, 1955). This focus on ends, however, does not recognize differential attraction
to the means—the process of crime. Sensory stimulation as a form of reward has been
largely neglected as a motivation to crime by dominant criminological theories (Baldwin,
1990; but see McCarthy & Hagan, 2005; Katz, 1988), even as studies have underscored its
importance in criminal behavior (e.g., Baldwin, 1985; Ellis, 1987; Farley & Farley, 1972).
Although it is widely acknowledged that crime is inherently risky, criminological work
has been less attentive to the well-established body of psychological work demonstrating
that individuals differ in the extent to which they enjoy the sensations inherent in risk taking
(e.g., Arnett, 1996; Zuckerman, 2007). The personality trait that corresponds to this indi-
vidual difference is known variously as thrill seeking, preference for risks, venturesome-
ness, or more broadly as sensation seeking (e.g., Zuckerman, 2007). This trait, which we
label thrill seeking, is a continuum.
3
At one end are the high thrill seekers who seek out
intense sensations and experiences because such experiences feel quite pleasurable to them,
and at the other end are the low thrill seekers, who experience such intense sensations as
painful and aversive. Given these differences, and net of other factors, risky behaviors are
more rewarding for high thrill seekers than for lows (e.g., Felson & Osgood, 2008; Lykken,
1995). Consistent with this idea, research suggests that some individuals experience the
inherent risk in committing illegal behaviors as a reward in and of itself—independent of
the rewards from the ends of the behavior—while others are potentially deterred from an act
not only by a recognition of long-term consequences but also by the immediate pains of fear
and anxiety from the process of crime (e.g., Felson & Osgood, 2008; Grasmick & Bursik,
1990; Wood, Pfefferbaum, & Arneklev, 1993). Therefore, all else equal, high thrill seekers
have greater motivation to crime than low thrill seekers, an idea which contrasts with the
SCT assumption of equal short-term motivation to offending.
To be sure, this idea that people differentially engage in risk-taking behaviors such as
crime, in part due to individual differences in the extent to which such behaviors generate
pleasurable arousal, is not new even if it has been largely neglected for a number of years
(e.g., Andrews & Wormith, 1989; Baldwin, 1985; D. C. Gibbons, 1989; Gove & Wilmoth
1990; Katz, 1988; Matza & Sykes, 1961; White, Labouvie, & Bates, 1985; Wood et al.,
1993). Scholars have amassed a considerable body of work suggesting that individuals high
in the personality trait thrill seeking are prone to engaging in criminal and delinquent acts
(Eysenck & Eysenck, 1978; Farley & Farley, 1972; Hansen & Breivik, 2001; Horvath &
Zuckerman, 1993; Pfefferbaum & Wood, 1994; Wood et al., 1993) as well as noncriminal

6 CRIMINAL JUSTICE AND BEHAVIOR
risky behaviors such as skydiving, bungee jumping, roller coaster riding, and driving,
cycling, and skiing at high speeds (e.g., Hansen & Breivik, 2001, see Zuckerman, 2007, for
a review).
4
Matza and Sykes (1961) noted half a century ago, for example, that delinquent
youth are immersed in a constant pursuit of thrills and excitement: “The excitement, then,
that flows from gang rumbles, games of ‘chicken’ played with cars, or the use of drugs is
not merely an incidental byproduct but may instead serve as a major motivating force”
(p. 714). Zuckerman (2007) argued that delinquent acts “can be sources of pleasurable exci-
tation for otherwise idle and bored adolescents, and the risks may add to the pleasure rather
than reducing it . . . sensation seeking may be a motive for some individuals” (p. 169).
Qualitative studies of criminal motives are also consistent with this body of quantitative
work. In her interviews with male criminals, Robertson (1994) found that by engaging in
delinquent behaviors “participants were seeking high levels of sensory arousal that they
described as a rush. Seeking a rush was the primary underlying factor for [their] activity
choices” [italics in original] (p. 38). Similarly, Jacobs and Wright (1999; see also R. T.
Wright & Decker, 1997) highlighted the “open-ended quest for excitement and sensory
stimulation that shaped much of the offenders’ daily lives” (p. 155). Respondents in
Pfefferbaum and Wood’s (1994; see also Agnew, 1990) study who admitted delinquent
behavior were asked to provide one main reason for it. “Fun/thrills” was the reason cited far
more often than any other for delinquent behavior, followed by “feels good.” These stated
offending rationales combined with quantitative work provide strong support for the idea
that for some individuals in some situations, the intrinsic physiological and psychological
rewards of risky delinquent behaviors serve as a major motivating force.
In addition, several studies in the rational choice tradition have successfully incorporated
perceived excitement or thrills as endogenous rewards into their expected utility equations
(e.g., Matsueda, Kreager, & Huizinga, 2006; McCarthy & Hagan, 2005; Piquero, Paternoster,
Pogarsky, & Loughran, 2011). Collectively, then, these results suggest that the sensory
stimulation or thrill inherent in risky behaviors such as crime seduces some individuals to
the act, in part as a function of their thrill-seeking propensity. The idea inherent in the con-
cept of thrill seeking—that people are differentially attracted to the process of risky behav-
iors—is that of differential motivation. To reiterate, thrill seekers may or may not have low
self-control; their preference for risky behaviors is distinct from their time perspective in
considering the consequences of their actions (Felson & Osgood, 2008; Jones & Lynam,
2009). In fact, recent findings suggest that a substantial amount of adolescent risk behavior
is planned instead of impulsive (e.g., Fischer & Smith, 2004; Maslowsky, Keating, Monk,
& Schulenberg, 2011). Moreover, involvement in a number of acts perceived as risky by
many people, such as skydiving, require considerable forethought and planning, even
training.
Although research provides evidence that thrill seeking should influence the likelihood
of offending, research has not thoroughly explored whether thrill seeking matters net of
self-control. Notably, the most popular measures of self-control, namely, the Grasmick,
Tittle, Bursik, and Arneklev (1993) measure and similar ones (e.g., Burt, Simons, & Simons,
2006; Burton, Cullen, Evans, Alarid, & Dunaway, 1998; Gibbs & Giever, 1995) incorpo-
rate risk-taking preferences, which obscure the effects of (thrill seeking) motivation and
(self-control) constraint. Studies that disaggregate the Grasmick et al. type scales into the
individual subscales (e.g., Risk-Seeking, Simplicity, Physicality, Immediate Gratification,
Self-Centeredness, and sometimes Anger/Temper) find that the effects of the Risk-Seeking

Burt, Simons / SELF-CONTROL AND THRILL SEEKING 7
subscale equals or exceeds the global self-control construct and its components (e.g.,
Arneklev, Grasmick, Tittle, & Bursik, 1993; Longshore, Turner, & Stein, 1996; Romero,
Gómez-Fraguela, Luengo, & Sobral, 2003; Wood et al., 1993). Such findings led Iovanni
and Miller (2008) to remark, “Given the primacy of the risk seeking component [measures
of self-control] may simply be a less precise measure of risk seeking . . . perhaps a prefer-
ence for risk seeking in disguise” (p. 130).
In addition to studies disaggregating the Grasmick et al. (1993) scale, at least three stud-
ies provide direct evidence for an independent effect of thrill seeking on offending net of
self-control.
5
Nagin and Paternoster (1993) found that respondents’ perceived pleasure
from crime was related to offending intentions in vignette scenarios. Net of self-control,
sanctions, and shame, the “perceived utility” of offending, measured as “How much fun or
how much of a ‘kick’ would it be for you if you did what [the scenario character] did under
the circumstances,” was strongly related to offending intentions. Replicating and extending
this work, Piquero and Tibbets (1996) found that the perceived pleasure of crime was
strongly related to offending intentions net of self-control, shame, sanctions, and moral
beliefs. They found that low self-control was strongly related to perceived pleasures from
offending but not related to perceived sanctions, a finding providing support for the notion
that the Grasmick et al. (1993) scale captures positive motivations to crime such as thrill
seeking and not merely self-control. More recently, Jones and Lynam (2009) found that
both thrill seeking and lack of premeditation, a construct analogous to self-control,
6
had
independent effects on crime, a finding that lead them to “call for a more nuanced measure
of self-control” (p. 317; see also Jones, Lynam, & Piquero, 2011).
Based on this body of work and contrary to SCT’s equal motivation assumption, we test
two hypotheses. First, we predict that net of self-control, thrill seeking will be positively
related to increases in offending. Indeed, as we have noted, there is some reason to believe
that thrill seeking will be equal to or stronger than self-control in predicting offending.
Furthermore, we depart from most prior work in examining the effects of thrill seeking net
of attitudinal measures of self-control, and we introduce a new measure of self-control that
attempts to capture individuals’ time perspectives in decision making without capturing
thrill-seeking motivation.
In addition to the direct effects of thrill seeking, we are interested in understanding how
these factors may interact to influence offending. Given the idea that thrill seekers are
attracted to risky behaviors because of pleasurable sensations accompanying such behavior,
it is possible that exercising self-control (i.e., recognizing that behaviors have potential
long-term consequences) may not be as strong of a constraint on offending for these indi-
viduals. Consonant with this idea, Rosenbloom (2003) concluded from her study of thrill
seeking, risk evaluation, and risk taking that “the more high sensation seekers evaluate an
activity as dangerous the more they will be attracted to it” (p. 384), because danger is a
source of pleasure to these individuals (see also, Maslowsky, Buvinger, Keating, Steinberg,
& Cauffman, 2011). For low sensation seekers, however, the opposite was true—they were
more likely to avoid the act as a function of the extent to which they viewed it as dangerous
(see also Gullone, Moore, Moss, & Boyd, 2000; Horvath & Zuckerman, 1993). In addition,
Jones et al. (2011) showed that the effect of thrill seeking on substance use was mediated
by perceived costs of engaging in substance use behaviors.
Further justification for the interaction between thrill seeking and self-control is found in
Lykken’s (1995) and Gray’s (1987) work on the BIS (behavioral inhibition system) and

8 CRIMINAL JUSTICE AND BEHAVIOR
BAS (behavioral activation system).
7
According to their view, individuals differ in their
BAS, which deals with responses to rewards or approach, and BIS, which deals with their
responsivity to punishments or loss (avoidance). Individuals with a high BAS respond with
more intensity to potentially rewarding stimuli (such as that promised by thrilling and/or
risky behavior), whereas those with lower BIS are less responsive to potentially punishing
stimuli and experience less anxiety or frustration with the potential for punishment.
According to Lykken, then, for individuals with a high BAS, the constraining effects of the
BIS (responsivity to potential punishments) may be weaker. Thus, conceiving of thrill seek-
ing and self-control as part of the BAS and BIS systems provides further support for expect-
ing an interaction between the two. Based on these ideas, we hypothesize that self-control
and thrill seeking will interact, such that self-control will be less strongly related to decreased
offending at higher levels of thrill seeking.
Notably, we are not the first to predict an interaction between self-control and other
criminogenic factors. For example, Jones and Lynam (2009) and Ousey and Wilcox (2007)
examined the link between self-control/criminal propensity and various risk factors.
8
These
studies found support for the argument that self-control matters more in the absence of
inhibiting factors, such as low neighborhood control (e.g., Jones & Lynam, 2009), opportu-
nity (Ousey & Wilcox, 2007), as well as strain and social learning variables (Ousey &
Wilcox, 2007; B. R. E. Wright, Caspi, Moffitt, & Silva, 2001). Rather than viewing our
hypothesis as competing, we see our argument as complimenting these works, as all attempt
to further specify the individual and situational factors that condition the influence of low
self-control on offending. It is also worth noting that the motivating factor we consider is
distinct from that of general motivation. We do not imply that the effects of motivation to
crime generally should matter less at high levels of self-control, but rather that this particu-
lar motivating factor creates a situation such that the mechanism through which low self-
control is said to operate—the recognition of the pain inherent in potential negative
consequences of crime—is mitigated by the pleasurable sensations risky acts provide for
people with high thrill seeking.
CURRENT STUDY
In the present study, we bring a consideration of motivation to the process of crime into
a test of SCT. In doing so, we use a novel measure of (low) self-control, which focuses on
consequence considerations in decision making. There is an ongoing measurement debate
about behavioral versus attitudinal measures, with Gottfredson and Hirschi (1990) prefer-
ring the latter given their certitude that their theory is correct, and other scholars preferring
the former given that the link between self-control and behavior is not yet established and
thereby produces a tautology (Akers, 1991; see Marcus, 2004, and Piquero, 2008, for more
information on self-control measurement). This article takes a different approach and intro-
duces a new operationalization of self-control: considering future consequences of actions
and acting rationally based on these considerations. This measure of self-control is not
flawless but improves upon most existing measures in its isolation of cost considerations at
the point of decision making. Most extant scales of self-control identify a number of person-
ality characteristics, namely, the six “elements of self-control”: impulsivity, preference for
simple tasks, preference for physical (vs. mental) activity, temper, risk taking, and self-
centeredness (e.g., Grasmick et al., 1993). These facets, which are assumed to be conse-
quences of myopia in cost considerations, have been criticized as measures of self-control

Burt, Simons / SELF-CONTROL AND THRILL SEEKING 9
for their multidimensionality (e.g., Longshore et al., 1996; Wood et al., 1993) and validity
(e.g., Marcus, 2004). For present purposes, such measures are particularly inappropriate
because they incorporate risk-taking preferences, which may result not only from inade-
quate forethought but also from thrill-seeking motivation.
9
Thus, prior to testing the focal hypotheses, we also compare the predictive effects of this
novel measure of self-control with an attitudinal measure of self-control as well as thrill-
seeking preferences. Insofar as the new measure is strongly related to the attitudinal mea-
sures and/or mediates their influence on offending, this provides support for the use of
personality measures of self-control as it suggests that myopia underlies these differences.
Insofar as there is not a strong relation between the different measures and they indepen-
dently predict crime, this suggests that existing personality/attitudinal measures of self-
control might be capturing something other than lack of foresight. These other things may
include motivational factors.
These hypotheses will be tested using data from the first four waves of the FACHS, a
longitudinal, multisite investigation of health and development among African American
families (Gibbons, Gerrard, Cleveland, Wills, & Brody, 2004; Simons et al., 2002). The
FACHS sample consists of several hundred African American families residing in Iowa and
Georgia at the first wave of data collection. The study was designed to analyze the particu-
lar risks and resources that disrupt or promote African American family functioning and
youth development in various contexts. The sites sampled included rural, suburban, and
metropolitan communities.
METHOD
SAMPLING
The FACHS data were collected in Georgia and Iowa using identical research proce-
dures; the samples were combined after data analyses indicated that they were comparable
on demographic, community, and family process variables (Cutrona, Russell, Hessling,
Brown, & Murry, 2000). The families lived in neighborhoods that varied considerably on
demographic characteristics, such as racial composition and economic level. In selecting
neighborhoods from which to draw the sample, researchers examined neighborhood char-
acteristics at the level of block groups (BGs).
10
Using 1990 Census data, BGs were identi-
fied in both Iowa and Georgia in which the percentage of African American families was
high enough to make recruitment economically practical (10% or higher) and in which the
percentage of families with children living below the poverty line varied considerably. In
general, the sample was representative of the African American populations of the com-
munities from which participants were recruited (Cutrona et al., 2000). Caregivers received
US$100 and youth received US$70 for their participation in Waves 1 and 2, and US$125
each in Waves 3 and 4.
SAMPLE
A total of 897 African American families (475 in Iowa and 422 in Georgia) participated
in the first wave of the FACHS. Each family included a fifth-grade target youth who was 10
(52%), 11 (45%), or 12 (3%) years old at Wave 1. Slightly more than half (54%) were
female. The targets and their primary caregivers were interviewed simultaneously but sepa-
rately. Most (84%) of the primary caregivers were the target’s biological mothers, of whom

10 CRIMINAL JUSTICE AND BEHAVIOR
37% were married at Wave 1. The remaining primary caregivers were grandmothers (6%),
biological fathers (5%), or other adults (5%). The caregivers’ mean age at Wave 1 was
37 years, ranging from 23 to 80 years; 92% identified themselves as African American. The
mean family income across the four waves of data collection was US$32,259.
Of the 897 families that originally participated in the study, 779 (87%) remained in the
sample at Wave 2, 767 (86%) at Wave 3, and 714 (80%) at Wave 4.
11
Data collection was
completed for the waves in 1998, 2001, 2004, and 2007. The first wave was collected when
the youth were in late childhood, ages 10 to 12; subsequent waves were collected approxi-
mately 2, 5, and 7 years thereafter. Thus, the youth were ages 12 to 14, 15 to 17, and 17 to 20
in Waves 2 through 4, respectively. This data set allows for the examination of hypotheses
with data spanning adolescence, a time when both offending (e.g., Hirschi & Gottfredson,
1983) and thrill seeking peak (e.g., Baldwin, 1985; Steinberg et al., 2008). Given the sam-
pling design, this sample represents Black youth from the two research sites who come from
extremely poor to middle-class families and who reside in neighborhoods that exhibit sig-
nificant variability in economic status, racial composition, and other factors. There is no
reason to expect that the processes under consideration would be unique to this sample.
Thus, we assume that we are examining processes relevant to adolescents in general.
MEASURES
Delinquency
The primary dependent variable was generated using youth self-reports at Wave 4 and
measures the number of different delinquent acts (out of 17) the respondents committed in
the past year, such as binge drinking (24%), shoplifting (19%), aggravated assault (9%),
marijuana use (33%), vandalism (8%), theft of personal property (17%), physical abuse of
an animal (2%), breaking and entering (3%), assault with a weapon (6%), and completed or
attempted robbery (1%).
12
Previous research has shown that self-reported survey items are
reasonably reliable indicators of delinquent behavior and preferable to police reports (e.g.,
Huizinga & Elliott, 1986). The items were culled from the Diagnostic Interview Schedule
for Children–Version 4 (DISC-IV; American Psychiatric Association, 1994). Although
these items vary in seriousness, Gottfredson and Hirschi (1990) argued that the effect of
self-control is general across offenses. The variety scale gives one point for every type of
delinquency committed by the respondents in the past year, regardless of how often they
committed the crime.
13
Offending variety scales are preferable to frequency and dichoto-
mous scales, which largely reflect variation in the least serious offenses (Sweeten, 2012, see
also Hirschi & Gottfredson, 1995). Considerable variation exists in delinquency among the
study respondents. The mean number of delinquent acts committed was 2.52 at Wave 4, and
scores ranged from 0 (30%) to 15 (0.3%); 283 respondents (40%) reported committing at
least 3 different offenses in the past year. The Kuder–Richardson coefficient of reliability
(KR
20
; Kuder & Richardson, 1937), designed to assess the reliability of dichotomously
scored scales, was .87.
14
The control for previous delinquency was created by averaging
Waves 1, 2, and 3 scores for the same instrument.
Cognitive Self-Control
Six items developed by the FACHS researchers were used to create a new measure
of self-control that focuses more explicitly on individual variation in consideration of

Burt, Simons / SELF-CONTROL AND THRILL SEEKING 11
consequences.
15
The first three items assess the extent to which respondents considered (or
would consider if they had not done the behavior) the possible negative consequences (1 =
not at all to 4 = a lot) the first time they considered drinking alcohol, using drugs like crack
or cocaine, or driving after drinking. These items were combined with three items that
assess the extent to which respondents deliberate at the point of decision making. For these
items, respondents were asked, “When deciding about [drinking alcohol, using drugs like
crack or cocaine, or having sex without a condom] for the first time, did you or would you,”
(1) “Just do what feels right”; (2) “Think about the pros and cons and then just do what feels
right”; or (3) “Carefully weigh the pros and cons before deciding what to do.” These six
items loaded on a single factor (all loadings >.60) and were standardized and averaged to
create the measure self-control. This scale assesses the extent to which respondents con-
sider potential negative consequences and make a nonimpulsive (deliberative) choice based
on cost-benefit calculations at the point of decision making. Cronbach’s alpha for the scale
is .81.
Attitudinal Measures of Self-Control and Thrill Seeking
As noted above, self-reported attitudinal/personality measures of self-control are most
frequently used in research on SCT. In the present study, we utilize instruments that have
previously been used as measures of self-control in tests of SCT (Burt et al., 2006). Three
subscales without thrill seeking comprise the attitudinal self-control measure. Notably,
exploratory and confirmatory factor analyses support the separation of thrill seeking from
the other three subscales, as the thrill seeking measure loads poorly on the higher order fac-
tor with the other three attitudinal self-control scales.
Thrill seeking is measured with four items from Eysenck and Eysenck’s (1978) personal-
ity inventory that assess whether respondents enjoy taking risks and are bored with a life
without danger (e.g., “You enjoy taking risks”). These items capture the extent to which
respondents experience the stimulation inherent in risky acts as pleasurable or rewarding.
Response categories range from 1 (not at all true) to 3 (very true). High scores indicate
higher thrill-seeking preferences. Cronbach’s alpha for the four-item scale is .77.
Three subscales make up the attitudinal self-control measure. The items were gleaned
from the good and poor self-control scales from Kendall and Wilcox’s (1979) self-control
inventory, revised for self-reporting. Notably, the conception of self-control that underlies
these two scales is distinct from that of Gottfredson and Hirschi (1990). In the broader and
more widely accepted conceptualization adopted by Kendall and Wilcox, self-control refers
to an individual’s ability to restrain from behavior that is tempting because one appreciates
that the behavior is potentially costly (e.g., Baumeister, Vohs, & Tice, 2007; Tangney,
Baumeister, & Boon, 2004; Wikström & Treiber, 2007). Ability to see long-term conse-
quences is not the focal aspect of this widely accepted understanding of self-control.
Nonetheless, current measures of self-control used to test SCT resemble this attitudinal
measure (e.g., Grasmick et al., 1993; Reisig, Wolfe, & Pratt, 2012).
Exploratory and confirmatory factor analyses directed the formation of the attitudinal
self-control measure. The three subscales can be described as Carelessness (α = .57), Lack
of Perseverance (α = .56), and Inability to Delay Gratification or Narrow Impulsivity (α =
.53). The three subscales were standardized and averaged to create the attitudinal self-con-
trol measure (α = .66).
16

12 CRIMINAL JUSTICE AND BEHAVIOR
Control Variables
Consistent with previous research on SCT, we control for age in years at Wave 4 (stan-
dardized in the analyses) and sex (1 = male, 0 = female) of the respondent. Additional con-
trols were considered, including primary caregiver race, age, and sex; the presence of a
second caregiver in the home; and household income. None of these variables significantly
influenced the processes under consideration and, thus, were not included in the model.
ANALYTIC STRATEGY
Our analyses consist of a series of negative binomial models, given that our dependent
variable (delinquency) is a variety count.
17
Notably, because we are interested in the effects
of the variables on changes in delinquency, we estimated the change in the outcome by
controlling for the averaged Waves 1, 2, and 3 scores to predict the Wave 4 outcome (the
regressor variable method; Allison 1990).
18
In the first few models, we compare the effects
of the various measures of self-control in predicting the change in offending. Next, we test
the hypothesis that motivation to the process of crime matters by examining the influence
of thrill seeking on offending net of self-control. Finally, we evaluate the extent to which
thrill seeking moderates the effects of self-control by incorporating product terms (standard
protocol delineated by Aiken & West, 1991) into the models. The standard errors in these
models were adjusted with the Huber/White sandwich estimator using the BGs as the clus-
tering units and were estimated in Stata 11 (StataCorp, 2011). Given the paucity of missing
data (2 cases), we utilize listwise deletion, which resulted in a sample size of 712.
RESULTS
Table 1 presents the means, standard deviations, and correlation matrix for the study
variables. At the zero order, there are significant correlations between most of the variables.
The low magnitude of the correlations between the new cognitive self-control measure and
the attitudinal self-control measures presented in Table 1 suggests that they have some com-
mon variance, yet are independent constructs. The new cognitive self-control has a correla-
tion of .13 (p < .01) with the attitudinal measure of self-control and −.19 (p < .001) with
thrill seeking. All of the self-control measures are related to delinquency in the expected
direction. Notably, and consistent with some past works (e.g., Arneklev et al., 1993), at the
zero order, the thrill seeking measure is more strongly related to delinquency
W4
(r = .33)
TABLE 1: Descriptive Statistics and Correlation Matrix for Study Variables (n = 712).
M SD 1 2 3 4 5 6
1. Delinquency
W4
2.52 2.64 
2. Prior delinquency
W3 + W2 + W1
−0.02 0.78 .34**** 
3. Thrill seeking 0.00 1.00 .33**** .17**** 
4. Attitudinal self-control 0.00 1.00 −.19**** −.22**** −.25**** 
5. Cognitive self-control 0.00 1.00 −.16**** −.12**** −.19**** .13**** 
6. Sex (1 = male) 0.00 1.00 .09*** .17**** .11**** .06 −.13*** 
7. Age (standardized) 0.44 0.50 .14**** .21**** .03 −.03 −.08 −.02
***p < .01. ****p < .001.

Burt, Simons / SELF-CONTROL AND THRILL SEEKING 13
than the other measures of self-control. Not surprisingly, Table 1 reveals that males have
higher levels of thrill seeking and lower cognitive self-control, but no sex/gender differ-
ences are observed for the attitudinal self-control measure.
Table 2 displays the results of the negative binomial models of delinquency on the vari-
ous measures of self-control and thrill seeking. The table presents the percentage change in
the expected count for a standard deviation increase in the predictor (% ), holding all other
variables constant, calculated as (100 × [e

– 1]). In the first six models, we sought to deter-
mine whether and to what extent the various measures of self-control overlap or indepen-
dently influence offending. Model 1 contains only the control variables and shows that a
standard deviation increase in the best predictor of future delinquency (past delinquency)
increases the expected count of delinquency at Wave 4 by roughly 35%, while a standard
deviation increase in age augments the expected count by 10%.
In Models 2 and 3, we add thrill seeking and the attitudinal measures of self-control to
the models separately. Thrill seeking is added to Model 2 and is related as expected with
increased offending; a standard deviation increase in thrill seeking decreases the expected
count of delinquency by roughly 31%, net of other variables, and slightly attenuates the
effects of prior delinquency. As shown in Model 3, the attitudinal measure of self-control is
negatively related to the expected count of offending (%  = −13.4) but is less than half the
size of the thrill seeking measure from the prior model. Model 4 includes both the thrill
seeking and the attitudinal measure. Here we can see that both thrill seeking and attitudinal
self-control have an independent effect on offending, but the influence of thrill seeking (%
 = 28.5) is more than 3 times the size of that for attitudinal self-control (%  = −8.0). Wald
tests of the equality of the coefficients indicate that this difference in coefficient size is
statistically significant, χ
2
= 43.1(1); p < .001.
19
Consistent with previous research, thrill
seeking appears to be the driving factor behind the attitudinal self-control measures.
20
The results thus far suggest that thrill-seeking preferences do matter and drive the
observed effects of this attitudinal measure of self-control on offending. We have argued,
however, that the attitudinal measures do not isolate the central construct at the heart of
SCT. Model 5 adds the new measure of self-control to the baseline model. As expected, the
coefficient is significant and negative, such that a standard deviation increase in cognitive
TABLE 2: Models Examining the Effects of Self-Control Measures and Thrill Seeking on Delinquency
(n = 712).
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7
Independent variables %  %  %  %  %  %  % 
Prior delinquency
W1 + W2 + W3
35.0**** 29.3**** 30.6**** 27.3**** 33.5*** 26.8**** 27.2****
Sex (1 = male) 3.6 1.2 5.3 2.6 1.4 0.9 0.6
Age 10.2** 8.9** 9.2**** 9.5** 8.2* 8.9** 9.5**
Thrill seeking 31.5**** 28.5**** 27.2*** 30.5****
Attitudinal self-control −13.4**** −8.0** −7.0 −6.5
Cognitive self-control −13.3**** −9.7*** −10.9***
Thrill Seeking × Cognitive Self-Control 12.9**
Maximum likelihood (Cox–Snell) R
2
.09 .15 .11 .16 .11 .16 .17
Note. Standardized estimates shown. The percentage change in the expected count of crime for a standard devia-
tion increase in the predictor, net of other variables, is indicated by % .
*p < .07. **p < .05. ***p < .01. ****p < .001 (two-tailed tests).

14 CRIMINAL JUSTICE AND BEHAVIOR
self-control decreases the expected count of delinquency by approximately 13%. Notably,
the effect of this self-control measure is less than half the size of the thrill seeking measure
and roughly the same size as the attitudinal measure. To test the hypothesis that thrill-
seeking motivation matters net of self-control, Model 6 includes both the cognitive and
attitudinal measure of self-control as well as thrill seeking.
21
Here one can see that thrill
seeking and cognitive self-control are only slightly reduced when incorporated simultane-
ously, indicating that individual differences in self-control do not account for the bulk of the
effects of differences in thrill seeking on delinquency. Instead, they have independent influ-
ences on offending. The attitudinal measure is reduced slightly and is no longer statistically
significant. These results suggest that both thrill seeking as a motivational factor and cogni-
tive self-control, an inhibiting factor, matter for delinquency, but the effects of thrill seeking
are significantly stronger than that of (cognitive) self-control. Consistent with past research
(e.g., Tittle & Botchkovar, 2005), motivation appears more important than self-control
(constraint) in explaining offending.
The second goal of the study was to assess the hypotheses that the effects of self-control
on offending are contingent upon individuals’ levels of thrill seeking. We hypothesized that
among those who enjoy taking risks, self-control would be less of a constraint on offending.
The results displayed in Model 7 provide tentative evidence in support of the hypothesis;
the interaction term is significant and positive (%  = 12.9; p < .05). To facilitate the inter-
pretation of the effect, we graphed the interaction. Figure 1 displays the marginal effects of
self-control across levels of thrill seeking, and, consistent with our expectations, indicates
that the inhibiting effect of self-control on delinquency declines as thrill seeking increases,
such that at slightly above the mean of thrill seeking, the effect of self-control is no longer
significant. An alternative depiction of this relationship is displayed in Figure 2. This figure
shows the effect of self-control on delinquency at high thrill seeking (1 standard deviation
W4
Figure 1. Graph depicting the marginal effects of self-control on delinquency across levels of thrill seeking.

Burt, Simons / SELF-CONTROL AND THRILL SEEKING 15
Figure 2. Graph depicting the effects of self-control on delinquency at high and low levels of thrill seeking.
above the mean) and low thrill seeking (1 standard deviation below the mean). This graph
reveals that self-control decreases delinquency at low levels of thrill seeking, but has no
effect on delinquency at high levels of thrill seeking.
22
Collectively, then, these results sug-
gest not only that motivation to the process of crime matters in predicting delinquency but
also that motivation influences the effects of self-control.
DISCUSSION AND CONCLUSION
Gottfredson and Hirschi’s (1990) SCT challenged many beliefs in the field, instigated a
wealth of research on its clearly stated propositions, and has earned commendation for its
parsimony, logical consistency, and scope (e.g., Akers & Sellers, 2004; Tittle, 1995). The
theory is grounded in a number of empirical assumptions, some of which, as scholars have
noted, may be invalid (e.g., Matsueda, 2008; Tittle & Botchkovar, 2005). In this study, we
examined the equal motivation assumption of SCT, bringing a consideration of motivation
to the risky process of crime to the fore. In doing so, we introduced a new measure of self-
control and explored two questions. First, we examined the veracity of SCT’s equal motiva-
tion assumption, investigating whether motivation to the process of crime in the form of
thrill seeking influences offending net of self-control. In addition, we investigated whether
self-control and thrill seeking interact to influence delinquency, such that the inhibiting
effects of self-control on offending are contingent upon levels of thrill seeking. Several
important findings emerged from this analysis and they are discussed below. This is fol-
lowed by a consideration of the implications of these findings for SCT, including the modi-
fications it may require to continue to be a dominant explanation of individual differences
in offending.

16 CRIMINAL JUSTICE AND BEHAVIOR
SCT assumes that motivation to crime is ubiquitous. We all want free money, quick
revenge, and uncomplicated sex, and therefore, as a constant, motivation can be removed
from a theory of crime (Gottfredson & Hirschi, 1990). This SCT assumption of universally
desired ends of crime, however, largely ignores and thereby rejects the possibility that indi-
vidual differences in thrill seeking may influence attraction to the process of crime. In the
present study, we examined individual differences in the process of crime, recognizing as
others have before, that crimes are inherently risky and individuals vary in the extent to
which they enjoy taking risks (e.g., Baldwin, 1990; Farley & Farley, 1972; Felson &
Osgood, 2008; Wood et al., 1993). In particular, we investigated whether individual differ-
ences in attitudes toward thrill seeking, conceptualized as a personality trait, influenced
offending net of self-control. Contrary to SCT’s equal motivation assumption, we found
that motivation to the process of crime matters. Thrill seeking was strongly related to
offending net of self-control measures. Moreover, consistent with prior work (Arneklev et
al., 1993; Longshore et al., 1996; Wood et al., 1993), thrill seeking actually had a stronger
effect on offending than both the measures of self-control we employed.
In addition, we explored whether thrill seeking moderated the effects of self-control on
offending. Based on the idea that the intense sensations that are generated by engaging in
risky acts is pleasurable for individuals high in thrill seeking and painful for those low in
thrill seeking, we predicted that considering future consequences (self-control) would be
less crime inhibiting among high thrill seekers. The rationale being that the mechanism
through which self-control operates to reduce crime—the realization that crime risks much
greater pains than pleasures—will operate differently as a function of individual differences
in attitudes toward the sensations accompanying risk-taking behaviors. The results sup-
ported our hypothesis. Self-control was most strongly related to less offending among those
low in thrill seeking, whereas among individuals more than 1 standard deviation above the
mean in thrill seeking, self-control did not significantly reduce offending. These results
suggest that motivation to crime in the form of thrill seeking matters directly as well as
indirectly by influencing the effects of self-control.
These results combined with others (e.g., Antonaccio, Botchkovar, & Tittle, 2011;
Brezina & Piquero, 2003; Nagin & Paternoster, 1993; Piquero & Tibbetts, 1996; Tittle &
Botchkovar, 2005) provide evidence contradicting the equal motivation assumption of
SCT. Research shows that individuals are not equally motivated to the ends of crime (e.g.,
Tittle & Botchkovar, 2005) or, as the present study shows, the process of crime (see also
Farley & Farley, 1972, Pfefferbaum & Wood, 1994). Insofar as these results are replicated,
either SCT needs to be revised to take account of differential motivation, a revision that
some argue can be made without destroying the theory (e.g., Tittle, 2011), or scholars
should recognize that the theory is based on a faulty assumption, and is therefore not an
adequate explanation of crime. Humans may be hedonistic, but growing evidence suggests
that individuals differ in their needs, wants, and desires, and that these differences influence
whether individuals find pleasure or pain in reading, knitting, skydiving, running, using
psychedelic drugs, getting drunk, boxing, gardening, or riding roller coasters. In this study,
we focused on individual differences in thrill seeking as affecting the process of crime, but
there are certainly other individual variables that may play a role in influencing motivation,
such as individuals’ needs and abilities to meet wants and desires, including desires for
thrills, legitimately (see, for example, Farley & Farley, 1972; McCarthy, 1995).
In addition, this study contributes to the sparse body of work identifying factors that
condition the effects of self-control (Hay & Forrest 2008; Jones & Lynam, 2009; Ousey &

Burt, Simons / SELF-CONTROL AND THRILL SEEKING 17
Wilcox, 2007; Wright et al., 2001). To these studies that specify the social factors that shape
the influence of self-control, we add a consideration of a specific individual motivating fac-
tor in thrill seeking. Future research should seek to extend our more refined understanding
of what situational and individual factors moderate the influence of self-control and thereby
increase or decrease the risk of offending.
This study also has implications for the measurement of self-control. We introduced a
new measure of self-control that attempts to more directly assess the concept of self-control
and its focus on time perspective in decision making. Our results provide provisional evi-
dence that considering future consequences at the point of decision making and acting based
on these considerations does not account for much of the variation in attitudinal measures
of self-control and is only weakly-to-moderately related to offending. These findings are
consistent with scholarship suggesting that the findings using the Grasmick et al. (1993)
and similar measures may be driven by thrill seeking (aka risk-seeking preferences) and
underscore the need to revise measures of self-control so that ideas contrary to the logic of
the theory are not included in the measure (e.g., Marcus, 2004). In short, this work high-
lights the need for greater attention to the measure of self-control and comparisons with
other similar constructs and measures, such as impulsivity, lack of premeditation, and con-
sideration of future consequences (e.g., Jones & Lynam, 2009).
The import of thrill seeking as well as incorporation of thrill-seeking/risk-taking prefer-
ences in the most commonly used measures of self-control might have implications for
interpreting existing tests of several aspects of SCT. For example, thrill seeking may also
play a role in explaining offender versatility and stability. While SCT argues that offenders
are versatile, research clearly shows that offenders vary in the extent to which they special-
ize (e.g., Deane, Armstrong, & Felson, 2005; Osgood & Schreck, 2007; Sullivan, McGloin,
Pratt, & Piquero, 2006). A focus on thrill seeking would predict greater versatility among
high thrill seekers, as repetition breeds habituation and skill, making repeated offenses less
risky and therefore less stimulating, whereas we would expect more stability among low
thrill seekers. Net of other factors, the familiarity and skill bred by repetition would make
the act less risky and therefore more pleasing to the systems of moderate to low thrill seek-
ers, which prefer lower stimulation and arousal.
A consideration of individual differences in thrill seeking and offending may also shed
light on socioeconomic differences in offending. Individuals differ not only in their thrill-
seeking preferences, but also, as Farley and Farley (1972) postulated some years ago, in
their opportunities for obtaining and channeling thrill-seeking needs and desires. Individuals
who have more wealth are able to satisfy their desires for risky activities through skydiving,
downhill skiing, cycling at high speeds, driving ATVs (all-terrain vehicles) or motorcycles,
and other potentially dangerous acts that require monetary resources. Even acts such as
risky or fast driving requires the use of a car; poor individuals have to steal a car to engage
in such acts (potentially heightening the experience) while more wealthy individuals can
drive their own automobiles at high rates of speed. Years ago, Farley and Farley (1972)
argued that individuals who need or prefer higher levels of stimulation but have limited
legitimate opportunities to satisfy these needs will be more likely to turn to delinquent
behavior to do so. Consistent with this idea, Robertson’s (1994) interviews with criminals
indicated that crime is often an alternative when there is a lack of opportunities to engage in
thrilling and exciting legal activities (see also Farley & Farley, 1972; Hansen & Breivik,
2001; Iso-Ahola & Crowley, 1988). These findings highlight the importance of examining

18 CRIMINAL JUSTICE AND BEHAVIOR
sociostructural influences on thrill seeking and suggest ways that crime as a response to a
need or desire for thrill seeking can be reduced.
Although we believe that the results of this study contribute to knowledge on individual
differences in offending propensities and SCT, it is not without limitations. First, the sam-
ple is comprised of African American youth who resided in Iowa and Georgia at the first
wave of data collection. While there is no reason to believe that the processes under consid-
eration would be limited to an African American sample or these geographic contexts, these
issues should be explored in other contexts with racially and ethnically diverse samples
comprising different age groups. An additional caveat should be noted about the measures.
While we have argued that the measure of self-control that we introduce is consistent with
SCT’s conceptualization of self-control, more research is needed on our measure and simi-
lar ones and how these measures relate to the “personality measures” of self-control (e.g.,
Intravia et al., 2012; Jones & Lynam, 2009). Finally, we have examined these relations in
isolation from the central causal variables of competing theories. While we believe that a
relatively narrow focus on motivation to the process of crime and self-control was a neces-
sary first step, future work should examine these processes controlling for other variables
such as deviant norms, definitions, or personal morality (Akers, 1973; Tittle, 1995;
Wikström, 2006); social bonding (Hirschi, 1969); negative emotionality (Agnew, 1992);
and motivation to the ends of crime (Tittle & Botchkovar, 2005).
Notwithstanding these limitations, we believe that these findings represent a useful con-
tribution to the body of work testing SCT and motivation to offending. The present results
suggest that motivation to the process of crime does matter. Although research depicts the
sensations and emotions inherent in the process of crime (e.g., Collins, 2008; Katz, 1988),
the influence of sensory rewards (or punishments) as a motivation or impediment to crime
has not been incorporated in dominant explanations of individual differences in offending.
Insofar as these results are replicated, this state of affairs may need to change. As Hirschi
(1969) quipped some years ago about why we do not commit crime, “There is much evi-
dence that we would if we dared” (p. 34). But, daring does not repel us equally. Indeed,
some revel in risky acts. These are the high thrill seekers who, rather than being deterred by
the potential negative consequences of crimes, can be seduced by its sensory thrills.
NOTES
1. Our selection of the name thrill seeking over the primary alternative sensation seeking was guided in large part by the
much broader meaning of the term sensation seeking in Zuckerman’s (2007) popular conceptualization. Sensation seek-
ing, according to Zuckerman, comprised four underlying dimensions, one of which is thrill and adventure seeking. Thus,
we adopt the more narrow term thrill seeking.
2. Although Gottfredson and Hirschi (1990) have lumped a number of personality characteristics together, including pref-
erences for risk taking, in their discussion of self-control, and many researchers have followed its implications, in this
study, we follow the lead of Felson and Osgood (2008) in “ignor[ing] those other differences that [Gottfredson and
Hirschi] sometimes lump together” (p. 161) and focus instead on the theory’s narrow conceptualization of self-control
as a “faulty time perspective” (see, for example, Jones & Lynam, 2009, for a similar usage). Given that this article is a
test of self-control theory (SCT), when we use the term self-control in this article, we refer to Gottfredson and Hirschi’s
conceptualization of this construct (for a discussion of the different usages of the term self-control, see, for example,
Wikström & Treiber, 2007).
3. Although noting that thrill seeking is a continuum, for clarity we often speak in categorical terms (i.e., high thrill seekers
and low thrill seekers).
4. This does not include studies utilizing the Grasmick, Tittle, Bursik, and Arneklev (1993) scale.
5. Given the first two of these studies measure self-control with the Grasmick et al. (1993) scale, it is likely the case that
their findings were conservative.

Burt, Simons / SELF-CONTROL AND THRILL SEEKING 19
6. Jones and Lynam’s (2009) construct lack of premeditation “refers to a consideration of the consequences before acting”
(p. 312).
7. We thank the anonymous reviewer who pointed us to Lykken’s (1995) work on the behavioral activation system/behav-
ioral inhibition system (BAS/BIS) to provide further support for our argument.
8. Although Wikström’s (Wikström & Treiber, 2007) situational action theory (SAT) proposes an interaction between self-
control and morality and has sparked research testing such an interaction (e.g., Antonaccio & Tittle, 2008; Wikström &
Svensson, 2010), SAT relies on a different conceptualization of self-control than SCT’s version, and, as such, does not
directly speak to the contingency effects proposed here.
9. This new measure is also consistent with Hirschi’s (2004) reconceptualization of self-control from the tendency to con-
sider long-term costs to “the tendency to consider the full range of potential costs of a particular act” (p. 543, italics in
original).
10. A block group (BG) is a cluster of blocks within a census tract. The Census Bureau strives to use naturally occurring
neighborhood boundaries when constructing BGs (Cutrona et al., 2000).
11. The youth who dropped out of the study between Waves 1 and 4 did not differ significantly (p < .05) in measures of
self-control or delinquency. The sex composition of the study changed significantly, as more males dropped out than
females. The percentage of females in the study sample increased by 4% from Wave 1 to Wave 4 (to 56%).
12. Most of these delinquent acts are not status offenses and thus are also illegal for adults as well. Models were reestimated
using a measure of crime (i.e., only violations of the adult criminal code) and the pattern of results was the same.
13. The results of this study are robust to alternative scoring of crime. We also estimated models using a frequency count of
crime (censored at 10 and 20) that replicated the pattern of results observed here.
14. The formula for the KR
20
is KR
20
= N / (N – 1)[1 − Σp
i
q
i
/ σ
2
x
], where N is the number of dichotomous items, p
i
is the
proportion responding “positively” to the ith item, q
i
is the proportion responding negatively (i.e., zero, and is equal to
1 − p
i
), and σ
2
x
is the variance of the total composite. The KR
20
is a special case of Cronbach’s alpha and is interpreted
in the same manner.
15. A list of these questionnaire items and their factor loadings are available from the first author upon request.
16. These Cronbach’s alphas are lower than desirable. Nonetheless, the average interitem correlation is greater than .20 for
all the scales, which is above the .15 threshold recommended by Clark and Watson (1995). Notably, the model is robust
to alternative measures of the attitudinal measure, including an average scale of all items.
17. For all of the negative binomial models, we reject the null hypothesis (at p < .001) that the residual variance parameter
is 0, thus establishing that a negative binomial model fits the data better than a Poisson model would (Long, 1997).
18. We also estimated the models without the control for prior delinquency. The pattern of findings for these models are
equivalent to that presented here (and are available upon request).
19. The equality of the coefficients was calculated in Stata with a Wald test with a chi-square distribution and 1 degree of
freedom. The null hypothesis that the coefficients for thrill seeking and attitudinal self-control are equal was clearly
rejected.
20. We also estimated a model where the thrill seeking scale is combined into a composite attitudinal measure consistent
with prior works. The coefficient for the combined measure was smaller than that for thrill seeking alone.
21. Even though the zero-order correlations supported the distinction between these three subscales, we tested for multicol-
linearity, and in no case was this a problem. Ordinary least squares (OLS) models were estimated to examine variance
inflation factors (VIFs), and in no case did the VIFs exceed 2.0.
22. At the suggestion of an anonymous reviewer, we estimated an alternative (left-censored) Tobit model to test whether
the observed significant interaction was an artifact of our negative binomial estimator (Osgood, Finken, & McMorris,
2002). The results from the Tobit models were analogous to those presented here, giving us greater confidence that the
significant interaction is not a method artifact (results available upon request).
REFERENCES
Agnew, R. (1990). The origin of delinquent events: An examination of offender accounts. Journal of Research in Crime &
Delinquency, 27, 267-294.
Agnew, R. (1992). Foundation for a general strain theory of crime and delinquency. Criminology, 30, 47-88.
Aiken, L., & West, S. (1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: SAGE.
Akers, R. L. (1973). Deviant behavior: A social learning approach. Belmont, CA: Wadsworth.
Akers, R. L. (1991). Self-control as a general theory of crime. Journal of Quantitative Criminology, 7, 201-211.
Akers, R. L., & Sellers, C. S. (2004). Criminological theories: Introduction, evaluation, and application. Los Angeles, CA:
Roxbury.
Allison, P. (1990). Change scores as dependent variables in regression analysis. Sociological Methodology, 20, 93-114.
American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC:
Author.
Anderson, E. (1999). Code of the street: Decency, violence, and the moral life of the inner city. New York, NY: W. W. Norton.

20 CRIMINAL JUSTICE AND BEHAVIOR
Andrews, D. A., & Wormith, J. S. (1989). Personality and crime: Knowledge destruction and construction in criminology.
Justice Quarterly, 6, 289-309.
Antonaccio, O., Botchkovar, E. V., & Tittle, C. R. (2011). Attracted to crime: Exploration of criminal motivation among
respondents in three European cities. Criminal Justice and Behavior, 38, 1200-1221.
Antonaccio, O., & Tittle, C. R. (2008). Morality, self-control, and crime. Criminology, 46, 479-509.
Arneklev, B., Grasmick, H. G., Tittle, C. R., & Bursik, R. J. (1993). Low self-control and imprudent behavior. Journal of
Quantitative Criminology, 9, 225-247.
Arnett, J. J. (1994). Sensation seeking: A new conceptualization and a new scale. Personality and Individual Differences, 16,
289-296.
Arnett, J. J. (1996). Sensation seeking, aggressiveness, and adolescent reckless behavior. Personality and Individual
Differences, 20, 693-702.
Baldwin, J. D. (1985). Thrill and adventure seeking and the age distribution of crime: Comment on Gottfredson and Hirschi.
American Journal of Sociology, 90, 1326-1330.
Baldwin, J. D. (1990). The role of sensory stimulation in criminal behavior, with special attention to the age peak in crime. In
L. Ellis & H. Hoffman (Eds.), Crime in biological, social, and moral contexts (pp. 204-217). New York, NY: Praeger.
Baumeister, R. F., Vohs, K. D., & Tice, D. M. (2007). The strength model of self-control. Current Directions in Psychological
Science, 16, 351-355.
Beccaria, C. (1963). On crimes and punishments. Indianapolis, IN: Bobbs-Merrill. (Original work published 1764)
Bentham, J. (2007). An introduction to the principles and morals of legislation. Mineola, NY: Dover Publications. (Original
work published 1823)
Brezina, T., & Piquero, A. R. (2003). Exploring the relationship between social and non-social reinforcement in the con-
text of social learning theory. In L. R. Akers & G. Jensen (Eds.), Social learning theory and the explanation of crime
(pp. 265-288). New Brunswick, NJ: Transaction Publishers.
Burt, C. H., Simons, R. L., & Simons, L. G. (2006). A longitudinal test of the effects of parenting and the stability of self-
control: Negative evidence for the general theory of crime. Criminology, 44, 353-396.
Burton, V., Cullen, F., Evans, T. D., Alarid, L., & Dunaway, R. G. (1998). Gender, self-control, and crime. Journal of
Research in Crime & Delinquency, 35, 123-147.
Clark, L. A., & Watson, D. (1995). Constructing validity: Basic issues in objective scale development. Psychological
Assessment, 7, 309-317.
Cohen, A. K. (1955). Delinquent boys: The culture of the gang. New York, NY: Free Press.
Collins, R. (2004). Interaction ritual chains. Princeton, NJ: Princeton University Press.
Collins, R. (2008). Violence: A micro-sociological theory. Princeton, NJ: Princeton University Press.
Cutrona, C. E., Russell, D. W., Hessling, R. M., Brown, P. A., & Murry, V. (2000). Direct and moderating effects of commu-
nity context on the psychological well-being of African American women. Journal of Personality and Social Psychology,
79, 1088-1101.
Deane, G., Armstrong, D. P., & Felson, R. B. (2005). An examination of offense specialization using marginal logit models.
Criminology, 43, 955-988.
Ellis, L. (1987). Religiosity and criminality from the perspective of arousal theory. Journal of Research in Crime &
Delinquency, 24, 215-232.
Eysenck, S., & Eysenck, H. J. (1978). Impulsiveness and venturesomeness: Their position in a dimensional system of person-
ality description. Psychological Reports, 43, 1247-1255.
Farley, F. H., & Farley, S. V. (1972). Stimulus-seeking motivation and delinquent behavior among institutionalized delin-
quent girls. Journal of Consulting and Clinical Psychology, 39, 94-97.
Felson, R. B., & Osgood, D. W. (2008). Violent crime. In E. Goode (Ed.), Out of control: Assessing the general theory of
crime (pp. 160-172). Stanford, CA: Stanford University Press.
Fischer, S., & Smith, G. T. (2004). Deliberation affects risk taking beyond sensation seeking. Personality and Individual
Differences, 36, 527-537.
Gibbons, D. C. (1989). Comment-personality and crime: Non-issues, real issues, and a theory and research agenda. Justice
Quarterly, 6, 311-323.
Gibbons, F. X., Gerrard, M., Cleveland, M. J., Wills, T. A., & Brody, G. (2004). Perceived discrimination and substance use in
African American parents and their children: A panel study. Journal of Personality and Social Psychology, 86, 517-529.
Gibbs, J., & Giever, D. (1995). Self-control and its manifestations among university students: An empirical test of Gottfredson
and Hirschi’s general theory. Justice Quarterly, 12, 231-255.
Gottfredson, M. R. (2008). The empirical status of control theory in criminology. In F. T. Cullen, J. P. Wright, & K. R. Blevins
(Eds.), Taking stock: The status of criminological theory: Advances in criminological theory (Vol. 15, pp. 77-100). New
Brunswick, NJ: Transaction Publishers.
Gottfredson, M. R. (2011, November). The role of choice in crime causation: Self-control theory. Paper presented at the
annual meeting of the American Society of Criminology, Washington, DC.
Gottfredson, M. R., & Hirschi, T. (1990). A general theory of crime. Stanford, CA: Stanford University Press.

Burt, Simons / SELF-CONTROL AND THRILL SEEKING 21
Gove, W. R., & Wilmoth, C. (1990). Risk, crime, and neurophysiologic highs: A consideration of brain processes that may
reinforce criminal and delinquent behavior. In L. Ellis & H. Hoffman (Eds.), Crime in biological, social, and moral
contexts (pp. 261-293). New York, NY: Praeger.
Grasmick, H. G., & Bursik, R. J. (1990). Conscience, significant others, and rational choice: Extending the deterrence model.
Law & Society Review, 24, 837-861.
Grasmick, H. G., Tittle, C. R., Bursik, R. J., & Arneklev, B. (1993). Testing the core empirical implications of Gottfredson
and Hirschi’s general theory of crime. Journal of Research in Crime & Delinquency, 30, 5-29.
Gray, J. A. (1987). The psychology of fear and stress (2nd ed.). Cambridge, UK: Cambridge University Press.
Gullone, E., & Moore, S. (2000). Adolescent risk-taking and the five-factor model of personality. Journal of Adolescence,
23, 393-407.
Gullone, E., Moore, S., Moss, S., & Boyd, C. (2000). The adolescent risk-taking questionnaire: Development and psychomet-
ric evaluation. Journal of Adolescent Research, 15, 231-250.
Hagan, J., Gillis, A. R., & Simpson, J. (1985). The class structure of gender and delinquency: Toward a power-control theory
of common delinquent behavior. American Journal of Sociology, 90, 1151-1178.
Hansen, E., & Breivik, G. (2001). Sensation seeking as a predictor of positive and negative risk behaviour among adolescents.
Personality and Individual Differences, 30, 627-640.
Hay, C., & Forrest, W. (2008). Self-control theory and the concept of opportunity: The case for a more systematic union.
Criminology, 46, 1039-1072.
Hirschi, T. (1969). Causes of delinquency. Berkeley: University of California Press.
Hirschi, T. (2004). Self-control and crime. In R. F. Baumeister & K. D. Vohs (Eds.), Handbook of self-regulation (pp. 537-
552). New Brunswick, NJ: Transaction Publishers.
Hirschi, T., & Gottfredson, M. R. (1983). Age and the explanation of crime. American Journal of Sociology, 89, 552-611.
Hirschi, T., & Gottfredson, M. R. (1993). Commentary: Testing the general theory of crime. Journal of Research in Crime
and Delinquency, 30, 47-54.
Hirschi, T., & Gottfredson, M. R. (1995). Control theory and the life course perspective. Studies on Crime & Crime Prevention,
4, 131-142.
Horvath, P., & Zuckerman, M. (1993). Sensation seeking, risk appraisal, and risky behavior. Personality and Individual
Differences, 14, 41-52.
Huizinga, D., & Elliott, D. (1986). Reassessing the reliability and validity of self-report delinquency measures. Journal of
Quantitative Criminology, 2, 293-326.
Intravia, J., Jones, S., & Piquero, A. R. (2012). The roles of social bonds, personality, and perceived costs: An empirical inves-
tigation into Hirschi’s “new” control theory. International Journal of Offender Therapy and Comparative Criminology,
56, 1182-1200.
Iovanni, L., & Miller, S. L. (2008). A feminist consideration of gender and crime. In E. Goode (Ed.), Out of control: Assessing
the general theory of crime (pp. 127-141). Stanford, CA: Stanford University Press.
Iso-Ahola, S. E., & Crowley, E. D. (1988). Adolescent substance abuse and leisure boredom. Journal of Leisure Research,
23, 260-271.
Jacobs, B. A., & Wright, R. (1999). Stick-up, street culture, and offender motivation. Criminology, 37, 149-173.
Jones, S., & Lynam, D. R. (2009). In the eye of the impulsive beholder: The interaction between impulsivity and perceived
informal social control on offending. Criminal Justice and Behavior, 36, 307-321.
Jones, S., Lynam, D. R., & Piquero, A. R. (2011). Substance use, personality, and inhibitors: Testing Hirschi’s predictions
about the reconceptualization of self-control. Crime & Delinquency. Advance online publication. Retrieved from http://
cad.sagepub.com/content/early/2011/09/10/0011128711420109
Katz, J. (1988). Seductions of crime: Moral and sensual attractions in doing evil. New York, NY: Basic Books.
Kendall, P. C., & Wilcox, L. E. (1979). Self-control in children: Development of a rating scale. Journal of Consulting and
Clinical Psychology, 47, 1020-1029.
Knowles, E. S., Cutter, H. S. G., Walsh, D. H., & Casey, N. A. (1973). Risk-taking as a personality trait. Social Behavior and
Personality, 1, 123-126.
Kuder, G., & Richardson, M. (1937). The theory of the estimation of test reliability. Psychometrika, 2, 151-160.
Long, J. S. (1997). Regression models for categorical and limited dependent variables. Thousand Oaks, CA: SAGE.
Longshore, D., Turner, S., & Stein, J. (1996). Self-control in a criminal sample: An examination of construct validity.
Criminology, 34, 209-228.
Lykken, D. T. (1995). The antisocial personalities. Hillsdale, NJ: Lawrence Erlbaum.
Lynam, D. R., & Miller, J. D. (2004). Personality pathways to impulsive behavior and their relations to deviance: Results from
three samples. Journal of Quantitative Criminology, 20, 319-341.
Marcus, B. (2004). Self-control in the general theory of crime: Theoretical implications of a measurement problem. Theoretical
Criminology, 8, 33-55.
Maslowsky, J., Buvinger, E., Keating, D. P., Steinberg, L., & Cauffman, E. (2011). Cost-benefit analysis mediation of the
relationship between sensation seeking and risk behavior among adolescents. Personality and Individual Differences,
51, 802-806.

22 CRIMINAL JUSTICE AND BEHAVIOR
Maslowsky, J., Keating, D. P., Monk, C. S., & Schulenberg, J. (2011). Planned versus unplanned risks: Neurocognitive
predictors of subtypes of adolescents’ risk behavior. International Journal of Behavioral Development, 35, 152-160.
Matsueda, R. L. (2008). On the compatibility of social disorganization and self-control. In E. Goode (Ed.), Out of control:
Assessing the general theory of crime (pp. 102-126). Stanford, CA: Stanford University Press.
Matsueda, R. L., Kreager, D. A., & Huizinga, D. (2006). Deterring delinquents: A rational choice model of theft and violence.
American Sociological Review, 71, 95-122.
Matza, D., & Sykes, G. M. (1961). Juvenile delinquency and subterranean values. American Sociological Review, 26,
712-719.
McCarthy, B. (1995). Not just “for the thrill of it”: An instrumentalist elaboration of Katz’s explanation of sneaky thrill prop-
erty crimes. Criminology, 33, 519-538.
McCarthy, B., & Hagan, J. (2005). Danger and the decision to offend. Social Forces, 83, 1065-1096.
Nagin, D. S., & Paternoster, R. (1993). Enduring individual differences and rational choice theories of crime. Law & Society
Review, 27, 467-496.
Osgood, D. W., Finken, L. L., & McMorris, B. J. (2002). Analyzing multiple-item measures of crime and deviance II: Tobit
regression analysis of transformed scores. Journal of Quantitative Criminology, 18, 319-347.
Osgood, D. W., & Schreck, C. (2007). A new method for studying the extent, stability, and predictors of individual specializa-
tion in violence. Criminology, 45, 273-312.
Ousey, G. C., & Wilcox, P. (2007). The interaction of antisocial propensity and life-course varying predictors of delinquent
behavior: Differences by method of estimation and implications for theory. Criminology, 45, 313-354.
Pfefferbaum, B., & Wood, P. (1994). Self-report study of impulsive and delinquent behavior in college students. Journal of
Adolescent Health, 15, 295-302.
Piquero, A. R. (2008). Measuring self-control. In E. Goode (Ed.), Out of control: Assessing the general theory of crime
(pp. 26-37). Stanford, CA: Stanford University Press.
Piquero, A. R., Paternoster, R., Pogarsky, G., & Loughran, T. (2011). Elaborating the individual difference component in
deterrence theory. Annual Review of Law and Social Science, 7, 335-360.
Piquero, A. R., & Tibbetts, S. (1996). Specifying the direct and indirect effects of low self-control and situational factors
in offenders’ decision making: Toward a more complete model of rational offending. Justice Quarterly, 13, 481-510.
Pratt, T., & Cullen, F. (2000). The empirical status of Gottfredson and Hirschi’s general theory of crime: A meta-analysis.
Criminology, 38, 931-964.
Quinn, P. D., & Harden, K. P. (2013). Differential changes in impulsivity and sensation seeking and the escalation of
substance use from adolescence to early adulthood. Development and Psychopathology, 25, 223-239. doi:10.1017/
S0954579412000284
Reisig, M. D., Wolfe, S. E., & Pratt, T. C. (2012). Low self-control and the religiosity-crime relationship. Criminal Justice
and Behavior, 39, 1172-1191.
Robertson, B. J. (1994). Leisure in the lives of male adolescents who engage in delinquency activity for excitement. Journal
of Park and Recreation Administration, 12, 29-46.
Romero, E. J., Gómez-Fraguela, A., Luengo, M. A., & Sobral, J. (2003). The self-control construct in the general theory of
crime: An investigation in terms of personality psychology. Psychology, Crime and Law, 9, 61-86.
Rosenbloom, T. (2003). Risk evaluation and risky behavior of high and low sensation seekers. Social Behavior and
Personality, 31, 375-386.
Simons, R. L., Murry, V., McLoyd, V., Lin, K., Cutrona, C., & Conger, R. D. (2002). Discrimination, crime, ethnic iden-
tity and parenting as correlates of depressive symptoms among African American children: A multilevel analysis.
Development and Psychopathology, 14, 371-393.
StataCorp. (2011). Stata Statistical Software: Release 12 [Computer software]. College Station, TX: Author.
Steinberg, L., Albert, D., Cauffman, E., Banich, M., Graham, S., & Woolard, J. (2008). Age differences in sensation seeking
and impulsivity as indexed by behavior and self-report: Evidence for a dual systems model. Developmental Psychology,
44, 1764-1778.
Sullivan, C. J., McGloin, J. M., Pratt, T. C., & Piquero, A. R. (2006). Rethinking the “norm” of offender generality:
Investigating specialization in the short-term. Criminology, 44, 199-233.
Sweeten, G. (2012). Scaling criminal offending. Journal of Quantitative Criminology, 28, 533-557.
Tangney, J. P., Baumeister, R. F., & Boone, A. L. (2004). High self-control predicts good adjustment, less pathology, better
grades, and interpersonal success. Journal of Personality, 72, 271-324.
Tittle, C. R. (1995). Control balance: Toward a general theory of deviance. Los Angeles, CA: Roxbury.
Tittle, C. R. (2011, November). The role of choice in crime causation: Discussion. Paper presented at the Annual Meeting of
the American Society of Criminology, Washington, DC.
Tittle, C. R., & Botchkovar, E. V. (2005). Self-control, criminal motivation and deterrence: An investigation using Russian
respondents. Criminology, 43, 307-354.
White, H. R., Labouvie, R. W., & Bates, M. E. (1985). The relationship between sensation seeking and delinquency: A longi-
tudinal analysis. Journal of Research in Crime & Delinquency, 22, 197-211.

Burt, Simons / SELF-CONTROL AND THRILL SEEKING 23
Whiteside, S. P., & Lynam, D. R. (2001). The five factor model and impulsivity: Using a structural model of personality to
understand impulsivity. Personality and Individual Differences, 30, 669-689.
Wikström, P.-O. H. (2006). Individuals, settings, and acts of crime: Situational mechanisms and the explanation of crime.
In P.-O. H. Wikström & R. J. Sampson (Eds.), The explanation of crime: Contexts, mechanisms, and development
(pp. 61-107). New York, NY: Cambridge University Press.
Wikström, P.-O. H., & Svensson, R. (2010). When does self-control matter? The interaction between morality and self-control
in crime causation. European Journal of Criminology, 7, 395-410.
Wikström, P.-O. H., & Treiber, K. (2007). The role of self-control in crime. European Journal of Criminology, 4, 237-264.
Wood, P., Pfefferbaum, B., & Arneklev, B. (1993). Risk-taking and self-control: Social psychological correlates of delin-
quency. Journal of Crime and Justice, 16, 111-130.
Wright, B. R. E., Caspi, A., Moffitt, T. E., & Silva, P. A. (2001). The effects of social ties on crime vary by criminal propen-
sity: A life-course model of interdependence. Criminology, 39, 321-352.
Wright, R. T., & Decker, S. H. (1997). Armed robbers in action. Boston, MA: Northeastern University Press.
Zuckerman, M. (1994). Behavior expressions and biosocial bases of sensation seeking. New York, NY: Cambridge University
Press.
Zuckerman, M. (2007). Sensation seeking and risky behavior. Washington, DC: American Psychological Association.
Callie H. Burt is an assistant professor in the School of Criminology and Criminal Justice at Arizona State University. Her
research focuses on criminological theories, with particular emphasis on elucidating the social-psychological mechanisms
through which social factors, such as racial discrimination, community crime, parenting practices, and life transitions, influ-
ence criminal offending across the life course. Her work has recently appeared in the American Journal of Sociology, the
American Sociological Review, and Criminology.
Ronald L. Simons is a foundation professor in the School of Criminology and Criminal Justice at Arizona State University
and a fellow in the Institute for Behavioral Research at the University of Georgia. His research has focused upon onset, ampli-
fication, and desistance from various externalizing and internalizing problems. Recently, he has expanded his research pro-
gram to include genes and other biological variables.