Professional Documents
Culture Documents
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
Sage Publications, Inc., Association for Psychological Science are collaborating with
JSTOR to digitize, preserve and extend access to Perspectives on Psychological Science
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
aps mĒĒĒmmĒm
mĒĒĒmmĒm I association for I association for
PSYCHOLOGICAL SCIENCE
$SAGE
Hiroki P. Kotabe1 and Wilhelm Hofmann2
department of Psychology and Center for Decision Research, University of Chicago, and
2Social Cognition Center, University of Cologne
Abstract
As the science of self-control matures, the organization and integration of its key concepts becomes increasing
important. In response, we identified seven major components or "nodes" in current theories and research bearing
on self-control: desire , higher order goal, desire-goal conflict , control motivation, control capacity, control effort , and
enactment constraints. To unify these diverse and interdisciplinary areas of research, we formulated the interplay
these components in an integrative model of self-control. In this model, desire and an at least partly incompatibl
higher order goal generate desire-goal conflict, which activates control motivation. Control motivation and contr
capacity interactively determine potential control effort. The actual control effort invested is determined by seve
moderators, including desire strength, perceived skill, and competing goals. Actual control effort and desire stren
compete to determine a prevailing force, which ultimately determines behavior, provided that enactment constrain
do not impede it. The proposed theoretical framework is useful for highlighting several new directions for research on
self-control and for classifying self-control failures and self-control interventions.
Keywords
self-control, self-regulation, willpower, desire, temptation
Science is built up with facts, as a house is with studied and discussed in isolation, making it difficult to
stones. But a collection of facts is no more a science grasp how they interconnect. For example, the construct of
than a heap of stones is a house. desire itself has recently received theoretical treatments on
- Jules Henri Poincaré, Science and Hypothesis how it emerges and operates (Hofmann & Van Dillen,
2012; Kavanagh, Andrade, & May, 2005; see also Hofmann
& Nordgren, 2015). Goals and goal pursuit have a long his-
Self-control has fascinated and perplexed many of the
tory in psychology (Carver & Scheier, 1982; Kruglanski
great thinkers of our past, dating back to Socrates (470-
et al., 2002). Intrapsychic conflict, pertinent to understand-
399 BC), Plato (437-347 BC), and Aristotle (384-322 BC).
ing incompatibilities between desires and higher order
For millennia, such giants have asked the same quintes-
goals, has received considerable attention by cognitive
sential question of self-control that we ask today: Whyneuroscientists (Botvinick, Braver, Barch, Carter, & Cohen,
do we act on passion when reason knows better? 2001; Carter & Van Veen, 2007). Self-control motivation is a
Contemporary psychologists have been investigating the
major topic in self-control research (Fujita, 2011; Muraven
question now for almost half a century, beginning with & Slessareva, 2003), with some explicitly pitting it against
Walter Mischel and his colleagues studying the ability of the idea that self-control relies on some depletable (physi-
children to delay gratification (Mischel & Ebbesen, 1970;
ological) resource or capacity (Beedie & Lane, 2012;
Mischel, Zeiss, & Ebbesen, 1972). Since then, many Inzlicht, Schmeichel, & Macrae, 2014; Molden et al., 2012).
insights have been made, shedding light, bit by bit, on
Others assume depletable self-control resources explain
this unique part of the human condition.
Yet, researchers find themselves overwhelmed by a
Corresponding Author:
large collection of facts that do not easily cohere. One Hiroki P. Kotabe, Department of Psychology, University of Chicago,
major challenge, a consequence of "zooming in," is that5848 South University Ave., Chicago, IL 60637
self-control involves several components that are often E-mail: hkotabe@uchicago.edu
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
Integrative Self-Control Theory 619
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
620 Kotabe, Hofmann
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
Integrative Self-Control Theory 621
Pre-Behavior Behavior
I #
i EXERTION
j CLUSTER
! Prevailing Force Is I 1 kJ - ñ - : - ř - I - I - I
i : ■ñ Degjre
i Facilitation/ ;
! Inhibition ACTIVATION i
j CLUSTER
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
622 Kotabe, Hofmann
Kavanagh et toward
al., immediate,
2005), rewarding stimuli.
we A person
as c
emerges in a relatively
experience desire even when he or she does aut
not know
resource-independent) man
why he or she is experiencing desire (e.g., imagine
centers in the
gamblerbrain
wants to take another evaluate
chance at roulette eve
background of
though he internal
thinks he is done for the night). need Desire orig
(for a more natesdetailed
as a state of wanting (Berridge explan et al., 2009) whe
Dillen, 2012).subcortical
In reward contrast,
processing regions (e.g., the we nucle
nisms involved accumbens)
inevaluate external stimuli (instigating
effortful selfac-
tively controlled
tors) against the (i.e.,
backdrop of internal rule-bneed states and
manner. one's learning history
However, a (impelling
strict factors; Hofmann on &
and control Kotabe,
to 2013).
different Fast associative processes give
leve rise t
ble for at least
apparentlythree
spontaneous, intrusive reasons.thoughts about the
evidence suggesting
appetitive target. When those thatintrusive thoughts des sign
processes the possibility
(e.g., planning of pleasure or relief, cognitivewher elaborati
motivated usually ensues (Kavanagh et al.,
reasoning) as 2005). itThroughen cogni-
becomes more elaborated
tive elaboration, desires can "crowd out" concurrent (De co
Ridder, 2013; nitive activity associated with higher
Hofmann, order goals
Fries
2011; Hofmann (Hofmann et& al., 2011;
VanHofmann & Van Dillen,
Dillen, 2012;
Nordgren & Kavanagh et al., 2005). Such
Chou, elaborative processes main-
2013; Van
2013). tain the desired
Second, some target in working memory
contro over an
effortful extended period.
self-control may tur
processes Desire and time
over related concepts (e.g., impulse,
as craving)
they
(Fishbach, Friedman,are studied from various levels of psychological
& analysis.
Krug
Drolet, 2013; Approaches Ouellette that help bridge the social-cognitive
& and
Woo
hierarchical nature
neural levels of analyses have of action
been particularly fruitful
1982, 1990; Powers,
here. From such a perspective, 1973),
the current consensus is con
a higher order that the wanting
goal aspect of desire
may is mediated by largely
indu
more automatic subcortical neural systems that include mesolimbic dopa-
subsystems i
tion, response selection, and response execution mine projections. Research has shown that specific parts
(Dreisbach & Haider, 2009; Folk, Remington, & Johnston, of the nucleus accumbens in the ventral striatum medi-
1992; Miller & Cohen, 2001). From such a hierarchical ates opioid-stimulated increases in wanting (Berridge
perspective of automaticity and control, effortful self- et al., 2009; Pecina & Berridge, 2005). Reward signals
control must work with an auxiliary of automatic pro- from midbrain regions are then forwarded to regions in
cesses. For these reasons, we refrain from subsuming thethe brain involved in reward representation and integra-
components of self-control into a strict automatic versus tion, with the orbitofrontal cortex being among the
controlled dichotomy. Rather, we assume that each com- regions most consistently implicated in the conscious
ponent has automatic and controlled aspects, and their representation of desire (Van der Laan, De Ridder,
interplay within and across components is an important Viergevera, & Smeetsa, 2011). Broader behavioral neuro-
topic for future research. scientific models, such as the balance model by
Heatherton and Wagner (2011), suggest that humans fail
to control desire when lateral prefrontal cortical (PFC)
The activation cluster
regions do not exert enough top-down control over sub-
cortical
The activation cluster comprises the first three reward-processing due to either strong reward
compo-
signals
nents of SCT: desire, higher order goal, and D-G (desire) or impaired prefrontal function due to
conflict
(see Figure 1). contextual factors (e.g., depletion of control capacity).
Desire is necessary but not sufficient for the activation
of effortful
Desire . Although the term desire is colloquially used to self-control processes. To activate effortful
refer to all kinds of wishes and wants (e.g., self-control,
"he desires one needs to activate both a lower order
desire and
to build his own car"; "the student desires straight a conflicting higher order goal. It is important
A's"),
to note at this
we advance a technical definition to help distinguish it point that our analysis is at the subjective
level
as a unique construct: Desire is a psychological throughout. The absence of self-control activation
driving
force that varies in strength and is rooted indoes not or
innate imply that a given desire is nonproblematic
when judged
learned need states (e.g., for food, alcohol, drugs, sex, by an outsider (e.g., parent, spouse, policy
rest, social connection, gambling). It directs maker).
a person Some desires are consensually harmful yet are
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
Integrative Self-Control Theory 623
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
624 Kotabe, Hofmann
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
Integrative Self-Control Theory 625
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
626 Kotabe, Hofmann
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
Integrative Self-Control Theory 627
MODERATORS
y* Desire strength
/ z* Perceived skill
// • Competing goals
Potential
Effort
^ ļj|gļj[ļj|^_ Effort
Activation
The multiplicative relation implies that, in terms of using cognitive effort or "taking it easy") resemble labor
determining the level of potential control effort, control leisure decisions predicted by economic models of lab
motivation and control capacity are functionally inter- supply. The point is that multiple streams of researc
changeable (see Kruglanski et al., 2012). Further, the converge on the idea that people tiry to efficiently al
terms imply that how much control effort could be cate control effort to effectively deal with the desire
invested in principle is a joint function of the various hand. Thus, it follows that people usually allocate les
sources that factor into control motivation and of trait dif- effort to control weak desires than strong desire
ferences (e.g., in executive functions) and state influ- (Hofmann & Van Dillen, 2012; Kavanagh et al., 200
ences (e.g., cognitive load, alcohol intoxication, However, this linear relationship holds up only to th
stereotype threat) that factor into control capacity. Thepoint where the desire strength is too high in relation
range of M implies that potential effort is capped by theone's potential control effort - resulting in (temporar
control capacity at a given time. disengagement (see Brehm & Self, 1989; Gendolla
Why do people not always fully exert themselves in aRichter, 2010). For example, take the relationship between
self-control episode? Control motivation and control
desire strength and actual control effort at different lev
capacity determine how much control effort can beof potential control effort because of depletion reduci
exerted, but how much control effort is actually exertedcontrol motivation and control capacity (Baumeister
may depend on additional moderators (see Figure 2). Vohs, 2007; Muraven & Slessareva, 2003). Whereas
Formally stated, actual control effort expenditure CEA) is depleted and nondepleted people would invest similar
limited by potential control effort (Ą>) and is reducedamounts of control effort in controlling weak desires,
further by additional moderators of control effort CEM): depleted people would disengage much sooner as desire
strength exceeds potential control effort CEP), leading to
higher rates of control failure.
El=Ev-Eu.
A second moderator is perceived skill. Some people
may see themselves as more tacitly able to use their avail-
First, according to the effort mobilization ableliterature,
control capacity in the service of self-control than oth-
actual control effort investment should depend on the
ers (Reber, 1989; Wagner & Sternberg, 1985). Additionally,
some people
difficulty of controlling desire. Most centrally, effort allo-may believe that they have more or better
cation is assumed to be guided by a basic concern for strategies at their disposal in a given situation
self-control
energy conservation (Brehm & Self, 1989; Fiskethan others (see Sheppes & Meiran, 2008). In both cases,
& Taylor,
1991; Kruglanski et al., 2012). Further, recent research
the degree of perceived skiļl would inversely vary with
control
suggests that cognitive effort is intrinsically costly effort engagement because of energy conservation
(Kool,
McGuire, Wang, & Botvinick, 2013). These views concerns.
supportThe emphasis on perceived skill suggests that
that, like the use of money or time, the use people
of effort
may downplay the difficulty of combating a given
should be economical and contingent on one'sdesireavailable
or make overly confident judgments about their skill
"budget." In fact, Kool and Botvinick (2014) showed that it, perhaps because of unrealistic perceptions
in controlling
labor/leisure decisions in humans (i.e., choosing between
of self-efficacy. Such "control illusions" may lead them to
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
628 Kotabe, Hofmann
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
Integrative Self-Control Theory 629
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
630 Kotabe, Hofmann
Self-Control
Failures
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
Integrative Self -Control Theory 631
reducing
Desire-based failures. These effectivefailures
self-control (Clarkson, Hirt, Jia, &
resonate w
folk concept of overpowering passion. Accordin
Alexander, 2010). Moreover, recent work suggests that
elaboration-intrusionmerely
theory ofconsumption,
expecting glucose desire (Kavana
compared with
2005) and a recent review of the role of desire in self- actually ingesting it, is sufficient to buffer against typi-
control (Hofmann & Van Dillen, 2012), the dynamicalcal depletion effects (Molden et al., 2012; Sanders, Shirk,
reprocessing of desire stimuli uses up space in work- Burghin, & Martin, 2012).
ing memory that otherwise could be engaged in other Self-control research has also lately been linked closely
competing and concurrent cognitive tasks. In this way, awith motivated reasoning (De Witt Huberts et al., 2013;
potent desire may temporarily consume mental facultiesKunda, 1990). The core idea in this research suggests that
that otherwise could be used to represent higher ordercontrol motivation may be susceptible to reasoning that
goals. In such cases, potent desire may cause the higher promotes desire or demotes higher order goals. Motivated
order goal to be temporarily "forgotten" (Hofmann et al., reasoning is thought to bias people toward particular
2011; Hofmann et al., 2008). preferred conclusions by making cognitions that support
the preferred conclusions more accessible. In our discus-
Higher order ; goal-based failures. This type of failuresion on different sources of control motivation, we noted
results when attention to higher order goals is temporar- that people may hold specific beliefs about how to bal-
ily lacking at the outset of the self-control episode. Peo-ance between desire and higher order goal enactment in
ple may enter a self-control episode neglecting higher their daily lives. When these beliefs offer too much lee-
order goals because they are processing at a low-level way for motivationally driven interpretations, motivated
of construal (Fujita, 2011) or because they are intoxi- reasoning may result in self-control failure. For example,
cated and affected by alcohol myopia (J. G. Hull, 1981;a dieter who lapses once at self-control by drinking a
MacDonald, Fong, Zanna, & Martineau, 2000). In thesemilkshake with dinner may think, "oh what the hell, diet-
cases, people fail to experience D-G conflict not because ing is done for today," and then subsequently indulge
desire consumes them but because higher order goals again by eating ice cream for dessert (Herman & Mack,
are not represented when they need to be. 1975). Or, a person may rationalize indulgence in the
here and now by instilling a sense of deservingness or
Control-effort-based failures . These failures com- justify why controlling oneself is unwarranted by aug-
prise control-motivation-based and control-capacity-menting the costs of self-control or downplaying possible
based failures. Their commonality is that desire enactment risks of desire enactment. For example, Kivetz and Zheng
occurs because of a lack of control effort. Their differ- (2006) showed that people are more likely to purchase
ence is in whether this lack of control effort is due to tempting goods when they can point to hard work as
insufficient control motivation or insufficient control justification. More recently, such self-licensing effects
capacity. Either way, potential control effort (Ep), as it is have been shown with regard to tempting food (De Witt
determined by control motivation and control capacity, Huberts, Evers, & De Ridder, 2012). A sense of deserving-
may be critically reduced. Because Ep limits EA, this criti- ness may also stem from the perception that one has
cal reduction increases the likelihood of self-control already committed to a virtuous act in the past (Khan &
failure. Dhar, 2006; Mukhopadhyay & Johar, 2009) or has made
substantial progress advancing a higher order goal
Control-motivation-based failures. Self-control fail- (Fishbach & Dhar, 2005). Motivated reasoning can lead to
ure because of lack of motivation has been a topicbehaviors
of that are arguably insensible. For instance, a
debate in the philosophical literature on weakness of will
dieter may find no reason to control the desire for a ridic-
(Davidson, 1980). It is gaining traction among psycholog-ulously unhealthy entree because he or she got it with a
ical researchers interested in this topic, as evidenced diet
by soda (Chernev, 2011; Chernev & Gal, 2010).
the apparent shift from strictly capacity-based accounts
of self-control failure to broader models that capture the Control-capacity-based failures. Regarding control-
interplay between motivation and capacity (Inzlicht capacity-based
& failures, we assume that limitations in
Schmeichel, 2012; Vohs, Baumeister, & Schmeichel, 2012). control capacity can be attributed to both trait-level
One recent stream of research that serves as an example and state-level effects. A wealth of evidence suggests
of this development shows that one's expectations and that people vary in their trait ability to use the execu-
implicit beliefs of the concept of "willpower" can shape tive functions that subserve self-control (Hofmann, Fri-
self-control outcomes (Job, Dweck, & Walton, 2010; ese, see & Roefs, 2009; Hofmann et al., 2011, 2008; Hofmann,
also Bandura, 1977). Related research has shown thatSchmeichel, & Baddeley, 2012). Additionally, executive
the mere perception of being resource-depleted may functions seem to generally follow a developmental trajec-
sometimes be more impactful than actual depletion in tory such that they tend to mature late (Figner, Mackinlay,
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
632 Kotabe, Hofmann
The ways humans resolve their internal conflicts between 1. A case pertinent to the one described here could be ca
passion and reason have major implications at the per- "dread control." Whereas self-control concerns controlling ap
sonal, social, and societal levels. In this article, we argued titive behaviors driven by desire, dread control concerns cont
that such conflicts may serve as a basis for a clear defini- ling fearful behaviors driven by dread. Biopsychological stud
tion of self-control. From this perspective and within this of rat behavior suggest that the intensely felt motivational st
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
Integrative Self-Control Theory 633
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
634 Kotabe, Hofmann
tions in nucleus accumbens. The Journal of Neuroscience, Godin, G., & Kok, G. (1996). The theory of planned behav-
28, 7184-7192. ior: A review of its applications to health-related behaviors.
Figner, B., Knoch, D., Johnson, E. J., Krosch, A. R., Lisanby, S.American
H., Journal of Health Promotion, 11, 87-98.
Gross,
Fehr, E., & Weber, E. U. (2010). Lateral prefrontal cortex andJ. J. (1998). The emerging field of emotion regulation:
self-control in intertemporal choice. Nature Neuroscience, An integrative review. Review of General Psychology, 2,
13 , 538-539. 271-299.
Figner, B., Mackinlay, R. J., Wilkening, F., & Weber, E. U. M. S., Wood, C., Stiff, C., & Chatzisarantis, N. L. (2010).
Hagger,
(2009). Affective and deliberative processes in risky choice: Ego depletion and the strength model of self-control: A
Age differences in risk taking in the Columbia Card Task. meta-analysis. Psychological Bulletin, 136, 495-525.
Journal of Experimental Psychology: Learning, Memory, Hassin, R. R., Bargh, J. A., & Zimerman, S. (2009). Automatic
and Cognition, 35, 709-730. and flexible: The case of non-conscious goal pursuit. Social
Cognition , 27, 20-36.
Finkel, E. (2013). The I3 model: Meta-theory, theory, and evidence.
Heatherton, T. F., & Wagner, D. D. (2011). Cognitive neurosci-
In M. P. Zanna & J. M. Olson (Eds.), Advances in expenmen-
tal social psychology (Vol. 49, pp. 1-104). Waltham, MA: ence of self-regulation failure. Trends in Cognitive Sciences,
Academic Press. 15 , 132-139.
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
Integrative Self-Control Theory 635
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
636 Kotabe, Hofmann
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
Integrative Self -Control Theory 637
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms
638 Kotabe, Hofmann
This content downloaded from 131.211.12.11 on Sat, 17 Sep 2022 15:08:05 UTC
All use subject to https://about.jstor.org/terms