-
D
R
T
ENa
b
c
a
ARRAA
KARDSRC
C
n
h1b
ARTICLE IN PRESSG ModelCN-338; No. of Pages 15Developmental
Cognitive Neuroscience xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Developmental Cognitive Neuroscience
j o ur nal ho me pa ge: ht tp : / /www.e lsev ier .com/ locate
/dcn
eview
he dual systems model: Review, reappraisal, and
reaffirmation
lizabeth P. Shulmana,∗,1, Ashley R. Smithb,1, Karol Silvab,
Grace Icenogleb,atasha Duellb, Jason Cheinb, Laurence
Steinbergb,c
Brock University, Psychology Department, 1812 Sir Isaac Brock
Way, St. Catharines, ON L2S 3A1, CanadaTemple University,
Department of Psychology, 1701 N. 13th Street, Philadelphia, PA
19122, USAKing Abdulaziz University, Abdullah Sulayman, Jeddah
22254, Saudi Arabia
r t i c l e i n f o
rticle history:eceived 22 January 2015eceived in revised form 17
July 2015ccepted 19 December 2015vailable online xxx
eywords:dolescentsisk takingual systemsensation-seeking
a b s t r a c t
According to the dual systems perspective, risk taking peaks
during adolescence because activation of anearly-maturing
socioemotional-incentive processing system amplifies adolescents’
affinity for exciting,pleasurable, and novel activities at a time
when a still immature cognitive control system is not yetstrong
enough to consistently restrain potentially hazardous impulses. We
review evidence from boththe psychological and neuroimaging
literatures that has emerged since 2008, when this perspective
wasoriginally articulated. Although there are occasional exceptions
to the general trends, studies show that,as predicted,
psychological and neural manifestations of reward sensitivity
increase between childhoodand adolescence, peak sometime during the
late teen years, and decline thereafter, whereas psychologicaland
neural reflections of better cognitive control increase gradually
and linearly throughout adolescenceand into the early 20s. While
some forms of real-world risky behavior peak at a later age than
predicted,
eward sensitivityognitive control
this likely reflects differential opportunities for risk-taking
in late adolescence and young adulthood,rather than neurobiological
differences that make this age group more reckless. Although it is
admittedlyan oversimplification, as a heuristic device, the dual
systems model provides a far more accurate accountof adolescent
risk taking than prior models that have attributed adolescent
recklessness to cognitive
deficiencies.
© 2016 The Authors. Published by Elsevier Ltd. This is an open
access article under the CC BY-NC-NDlicense
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
ontents
1. The emergence of dual systems models . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
002. The current article . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . 003. Are adolescents particularly prone to
risk taking? . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . 00
3.1. Risk taking in the laboratory . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
004. The development of sensation seeking and reward sensitivity .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . 00
4.1. Sensation seeking . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . 004.2. Behavioral manifestations of reward sensitivity . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . 004.3.
Neuroimaging of reward sensitivity . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . 00
5. The development of self-regulation and cognitive control . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . 005.1. Self-reported
impulsivity . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . 005.2. Behavioral
measures of self-regulation . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . 005.3. Neuroimaging of cognitive
control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . .00
6. Is risk taking during adolescence related to heightened
reward sensitivity and immature cognitive control? . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00
Please cite this article in press as: Shulman, E.P., et al., The
dual sysNeurosci. (2015),
http://dx.doi.org/10.1016/j.dcn.2015.12.010
7. Unresolved questions and future directions . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . .8. Concluding comment .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . .
∗ Corresponding author.E-mail addresses: [email protected]
(E.P. Shulman), [email protected] (A.R. S
[email protected] (N. Duell), [email protected] (J. Chein),
[email protected] (L. Steinber1 These authors made equal contributions
to the article and should be considered co-fi
ttp://dx.doi.org/10.1016/j.dcn.2015.12.010878-9293/© 2016 The
Authors. Published by Elsevier Ltd. This is an open
accessy-nc-nd/4.0/).
tems model: Review, reappraisal, and reaffirmation. Dev.
Cogn.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . 00 . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . 00
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . 00
mith), [email protected] (K. Silva),
[email protected] (G. Icenogle),g).rst authors.
article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/
dx.doi.org/10.1016/j.dcn.2015.12.010dx.doi.org/10.1016/j.dcn.2015.12.010http://www.sciencedirect.com/science/journal/18789293http://www.elsevier.com/locate/dcnhttp://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/mailto:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]/10.1016/j.dcn.2015.12.010http://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/
-
ING ModelD2 ogniti
hatrcsidp
lypolai
1
nbm(nopromtliS2ibtfsaa
psoupsiCiavmtsmasmto
ARTICLECN-338; No. of Pages 15 E.P. Shulman et al. /
Developmental C
Social scientists and casual observers of human developmentave
long noted that the transitional period between childhoodnd
adulthood is a time of heightened risk-taking. Indeed, despitehe
relative absence of illness and disease during this period,ates of
morbidity and mortality increase substantially in adoles-ence,
largely due to risk taking. The question of why adolescentseem
predisposed toward recklessness is age-old; however, workn the
field of developmental psychology, and more recently,evelopmental
neuroscience, has provided new insights into thehenomenon.
For many years psychologists had attempted to explain ado-escent
recklessness as a consequence of cognitive deficiencies inoung
people’s thinking, including irrationality, poor
informationrocessing, and ignorance about risk. As we have noted in
previ-us publications (e.g., Steinberg, 2008), these accounts have
beenargely undermined by available evidence. Generally speaking,
byge 15 or so, adolescents perform as well as adults on tasks
measur-ng logical reasoning, information processing, and risk
perception.
. The emergence of dual systems models
About a decade ago, the budding field of developmental cog-itive
neuroscience began to provide insight into how patterns ofrain
development might explain aspects of adolescent decision-aking
(see, e.g. Dahl, 2004). In 2008, our lab at Temple University
Steinberg, 2008; Steinberg et al., 2008) and Casey’s lab at
Cor-ell (Casey et al., 2008) simultaneously proposed similar
variationsf a “dual systems” account of adolescent decision-making.
Thiserspective attributes adolescents’ vulnerability to risky,
ofteneckless, behavior in part to the divergent developmental
coursesf two brain systems: one (localized in the striatum, as well
as theedial and orbital prefrontal cortices) that increases
motivation
o pursue rewards and one (encompassing the lateral
prefrontal,ateral parietal, and anterior cingulate cortices) that
restrainsmprudent impulses (see e.g., Casey et al., 2008; Duckworth
andteinberg, 2015; Evans and Stanovich, 2013; Luna and Wright,016;
Metcalfe and Mischel, 1999; Steinberg, 2008). Specifically,
t proposes that risk-taking behaviors peak during
adolescenceecause activation of an early-maturing
incentive-processing sys-em (the “socioemotional system”) amplifies
adolescents’ affinityor exciting, novel, and risky activities,
while a countervailing, butlower to mature, “cognitive control”
system is not yet far enoughlong in its development to consistently
restrain potentially haz-rdous impulses.
Several variations on this dual systems model have been
pro-osed. The version that guides our work (Steinberg, 2008) is
veryimilar to that proposed by Casey et al. (2008). Both conceivef
a slowly developing cognitive control system, which contin-es to
mature through late adolescence. However, whereas weropose that the
socioemotional system follows an inverted-Uhaped developmental
course, such that responsiveness to rewardncreases in early
adolescence and declines in early adulthood,asey et al. have
portrayed the socioemotional system as increas-
ng in arousability until mid-adolescence, at which point it
reaches plateau, remaining at this level into adulthood.
Furthermore, ourersion of the dual systems model posits that the
decline in socioe-otional arousability occurs independently of the
development of
he control system, whereas Casey et al.’s model proposes that
thetrengthening of the cognitive control system causes the
socioe-otional system to become less arousable. More recently,
Luna
nd Wright (2016) have proposed another variation on the dual
Please cite this article in press as: Shulman, E.P., et al., The
dual sysNeurosci. (2015),
http://dx.doi.org/10.1016/j.dcn.2015.12.010
ystems model (the “driven dual systems” model), which, like
ourodel, hypothesizes an inverted-U shaped trajectory of
socioemo-
ional arousability, but, unlike our model, hypothesizes a
trajectoryf cognitive control that plateaus in mid-adolescence
rather than
PRESSve Neuroscience xxx (2015) xxx–xxx
continuing to increase into the 20s, as suggested by us and by
Caseyet al. In a similar vein, Luciana and Collins (2012) endorse a
modelthat emphasizes the role of a hyperactive socioemotional
system(“subcortical limbic-striatal systems” in their terminology)
under-mining the regulatory ability of the cognitive control system
(the“prefrontal executive system”) resulting in greater risk-taking
dur-ing adolescence. Like Luna and Wright, Luciana and Collins
arguethat the development of cognitive control is complete by
mid-adolescence, as evidenced by adolescents’ adult-like
performanceon non-affective measures of cognitive capacity. Fig. 1
illustratesthe similarities and differences between these versions
of the dualsystems model.
Another perspective, Ernst’s (2014) triadic model, expands onthe
dual systems concept by hypothesizing that a third brainsystem—one
responsible for emotional intensity and avoidance,anchored in the
amygdala—is also important for understanding thedevelopmental
differences in “motivated behavior.” With respectto the type of
reward-seeking risky behavior that the dual systemsmodels seek to
explain, Ernst (2014) speculates that this emo-tion/avoidance
system may serve to boost impulsive decisions inadolescence by
amplifying the perceived cost of delay. She alsoproposes that this
system may become hypoactive—dampeningavoidance impulses—in the
face of a potential reward that acti-vates the socioemotional
system. While this model is intuitivelyappealing, there is not much
evidence to date indicating that theemotion/avoidance system and
its developmental trajectory helpto explain heightened levels of
risk taking in adolescence. Also, therole of the amygdala in
decision-making is not yet clear (see e.g.,Somerville et al.,
2014). Therefore, our review does not address thisthird
hypothesized system.
2. The current article
In this article, we review evidence from both the behavioral
andneuroimaging literatures that has emerged since the dual
systemsmodel was originally articulated in 2008. In particular, we
considerthe degree to which extant research findings support,
extend, mod-ify, and challenge the theory. We focus our discussion
on three mainpropositions of the model: (1) that reward sensitivity
peaks in ado-lescence; (2) that cognitive control increases
linearly during thisperiod; and (3) that heightened risk-taking
during adolescence isthe product of heightened reward-seeking and
relatively weakercognitive control.
We begin by addressing a recent criticism of the basic
premisethat middle adolescence is an especially intensified period
of riskybehavior. We then examine evidence regarding the
trajectoryof sensation seeking across development, the reward
processingcircuitry that might underlie developmental changes in
sensation-seeking behavior, and the extent to which heightened
sensationseeking and reward sensitivity are related to pubertal
develop-ment. Next, we survey evidence on the developmental
trajectoryof the ability to control impulsive behavior through
self-regulatoryprocesses, and on the maturation of the brain’s
cognitive controlnetwork, which is proposed to undergird this
ability. Finally, weconsider evidence concerning the interaction of
the two proposedsystems during risky decision making, identify
several unresolvedissues, and offer some recommendations for how
they might beaddressed in future research.
In examining how recent evidence informs the dual systemsmodel,
we are cognizant of critiques of this viewpoint,
includingcontentions that the model inadequately accounts for
studies
tems model: Review, reappraisal, and reaffirmation. Dev.
Cogn.
that do not find adolescents to be particularly sensitive to
reward(Pfeifer and Allen, 2012; but see Strang et al., 2013 for a
responseto this critique), that cognitive control does not
unequivocallyimprove during adolescence (Crone and Dahl, 2012), and
that
dx.doi.org/10.1016/j.dcn.2015.12.010
-
ARTICLE IN PRESSG ModelDCN-338; No. of Pages 15E.P. Shulman et
al. / Developmental Cognitive Neuroscience xxx (2015) xxx–xxx 3
Dual Systems ModelA.(Steinberg, 2008)
Stre
ngth
Age
Stre
ngth
Age
Stre
ngth
Agesocioemo�onal system cogni�ve control system
Matura�onal Imbalance ModelB.(Casey et al., 2008)
Driven Dual Systems ModelC.(Luna & Wright, 2015)
al (re
arEt
t2bunaep2h
bmb1aiattsacrrpam
pthtrgripobr
atsse
by centuries. And yet, empirical evidence of a mid-adolescent
peakin risk taking (at least in humans) is not unequivocal. As
pointedout in a recent review of epidemiological data, the peak
agefor risk taking varies across different behaviors, and very
often
NEUROBIOLOGICAL
PSYCHOLOGICAL
Reward Sensi�vity
Sensa�on Seeking
Self-Regula�on
Cogni�ve Control
BEHAVIORAL
Decision Making
Risk Taking
Fig. 1. Alternative theoretical models of the development of the
socioemotion
dolescence may not actually be a peak period of vulnerability
toisk-taking (e.g., Defoe et al., 2014; Willoughby et al., 2013;
but seernst, 2014 for a response to Willoughby et al.). We briefly
addresshese critiques here.
We do not disagree with a fourth critique of the dual sys-ems
model—that it is insufficiently nuanced (Pfeifer and
Allen,012)—because this is almost certainly correct. However,
weelieve that even an admittedly simplified model can serve as
aseful heuristic and, more important, can help to motivate
researcheeded to flesh out the details of an initially simplistic
account (for
full discussion see Strang et al., 2013). Moreover, given the
influ-nce this perspective continues to have on legal policy and
practice,ublic health, and popular discourse about adolescence
(Steinberg,014), it is important to ask whether this simplified
account iselpful or misguided.
It may be useful at this juncture to clarify our terminology.
Toegin, the term “adolescence” warrants discussion. Largely as
aatter of convenience, scholars generally agree that
adolescence
egins when pubertal development becomes evident, around age0
(somewhat later among males). The end of adolescence—thettainment
of adult status—is not easily pegged to any single biolog-cal or
social event, however. In research, adulthood is often defineds
beginning at either age 18 or 21, the two ages most often tiedo
legal majority in the developed world. However, given that 18-o
21-year-olds in industrialized societies are rarely regarded
out-ide the legal system as fully mature adults, and typically have
notttained many of the traditional markers of adult status (e.g.,
finan-ial independence, completion of formal education, stable
romanticelationships, full-time employment, parenthood), we prefer
toefer to this age range as “late adolescence.” For purposes of
thisaper, our focus is mainly on the second decade of life, from
aboutges 10 to 21, which we subdivide into early adolescence
(10–13),iddle adolescence (14–17), and late adolescence
(18–21).Another source of confusion in discussions of the dual
systems
erspective concerns levels of analysis, since the perspective
referso overt behaviors (such as risk taking), the psychological
statesypothesized to motivate them (such as sensation seeking),
andhe neural processes believed to undergird these states (such
aseward sensitivity). In an earlier paper (Smith et al., 2013), we
sug-ested that “reward sensitivity” and “cognitive control” be used
toefer to the neurobiological constructs that are measured in
stud-es of brain structure or function (see Fig. 2). These
neurobiologicalhenomena have psychological manifestations (in our
terminol-gy, “sensation seeking” and “self-regulation”) that are
measuredy assessing psychological states or traits through the
subjectiveeports of individuals or their evaluators.
For heuristic purposes, we use “sensation seeking” as an
over-rching label for a number of interrelated constructs that
refer
Please cite this article in press as: Shulman, E.P., et al., The
dual sysNeurosci. (2015),
http://dx.doi.org/10.1016/j.dcn.2015.12.010
o the inclination to pursue “varied, novel, complex, and
intenseensations and experiences and the willingness to take
physical,ocial, legal, and financial risks for the sake of such
experi-nces” (Zuckerman, 1994, p. 26). Recruitment of brain regions
and
ward processing) and cognitive control systems from about age 10
to age 25.
systems implicated in reward-processing (e.g., ventral
striatum,orbitofrontal cortex) has been linked to measures of
sensation seek-ing in humans and other animals (Abler et al., 2006;
Leyton et al.,2002; Lind et al., 2005). In a similar vein, we use
the label “self-regulation” to refer to a group of interrelated but
distinguishableconstructs that refer to the capacity to
deliberately modulate one’sthoughts, feelings, or actions in the
pursuit of planned goals; amongthese constructs are impulse
control, response inhibition, emotionregulation, and attentional
control. Aspects of self-regulation havebeen linked to the
functioning of brain regions and systems thatsubserve cognitive
control (e.g., lateral prefrontal, lateral parietal,and anterior
cingulate cortices) (Luna et al., 2010; Mennigen et al.,2014).
Variations in sensation seeking and self-regulation, in turn,
areassociated with variations in behaviors, including risk taking,
whichcan be measured through objective reports or observations. In
ourmodel, risk taking is a subset of many aspects of decision
makingthat share some, but not all, characteristics in common.
Further-more, as Fig. 2 indicates, all decision making takes place
within abroader context that encourages and enables some acts but
discour-ages and prohibits others. As we discuss, the fact that
adolescents’risk taking is influenced by the broader context in
which it occursmakes it difficult to move seamlessly between
laboratory studiesand the real world.
3. Are adolescents particularly prone to risk taking?
Allusions to adolescence as a time of rash behavior and
poordecision making predate the articulation of the dual systems
model
tems model: Review, reappraisal, and reaffirmation. Dev.
Cogn.
Context
Fig. 2. Constructs implicated in the dual systems model of
adolescent risk-takingarranged by level of analysis.
dx.doi.org/10.1016/j.dcn.2015.12.010
-
ING ModelD4 ogniti
ihed2
saoponcwmCcrrtrttbppm
3
shtttdbfriLBspsle
uddoroda2crrottao
ARTICLECN-338; No. of Pages 15 E.P. Shulman et al. /
Developmental C
t is late adolescents, not middle adolescents, who exhibit
theighest levels of recklessness (see Willoughby et al., 2013).
Forxample, one of the most dangerous forms of substance
use—bingerinking—is most common during the early 20s (Chassin et
al.,002; Willoughby et al., 2013).
Although some argue that these data pose a problem for the
dualystems model, we disagree. The model does not posit that
middledolescents necessarily demonstrate the highest levels of all
formsf risk taking in the real world. Rather, it asserts that
risk-takingropensity is highest in mid-adolescence, but that the
expressionf this propensity is expected to vary depending on the
context (asoted in Fig. 2). Our position is that late adolescents
are less biologi-ally predisposed to risk taking than middle
adolescents (consistentith the dual systems model), but that they
exhibit higher levels ofany forms of real-world risk-taking due to
greater opportunity.
ompared to younger individuals, people in their early 20s
typi-ally experience less supervision from adults, have more
financialesources, and are afforded greater legal access to many
forms ofisk taking (e.g., driving, alcohol, and gambling). Thus, we
contendhat maturational factors predispose middle adolescents to
greaterisk taking, but that social and legal factors constrain
their oppor-unities to realize this predisposition. Simply put, it
is far easier forhe average 21-year-old to take risks with alcohol,
cars, and gam-ling than it is for the average 15-year-old. If
15-year-olds wereermitted to drive, purchase alcohol, and enter
casinos legally, ourrediction is that they would likely crash,
binge drink, and gambleore than people in their early 20s.
.1. Risk taking in the laboratory
In an effort to investigate age differences in risk-taking
propen-ity, unconfounded by age differences in opportunity,
researchersave tested adolescent and adult participants using
artificialasks—typically gambling games and driving
simulations—that givehem the option to take risks in the safety of
a laboratory set-ing. While such tasks are often lacking in
ecological validity, theyo have the advantage of controlling for
contextual differencesetween adolescents and other age groups, as
well as for age dif-erences in behavior preferences. These studies
yield inconsistentesults, with some finding greater risk taking in
adolescence thann adulthood (e.g., Burnett et al., 2010; Mitchell
et al., 2008; Vaneijenhorst et al., 2008, 2010a), others finding no
age effects (e.g.,jork et al., 2007; Eshel et al., 2007; de Water
et al., 2014), andtill others finding that adolescents engage in
less risk taking com-ared to children (Paulsen et al., 2011). These
inconsistent findingsuggest that if there is an increased risk
taking propensity in ado-escence it may only manifest under certain
conditions (see Defoet al., 2014 for a recent meta-analysis).
Recently, researchers have used laboratory tasks and
manip-lations that better approximate certain aspects of real-life
riskyecision-making. These studies have helped to delineate the
con-itions under which adolescents may be more predisposed thanther
age groups to take risks. For example, noting that during
mosteal-world risk taking the actual chances of a positive or
negativeutcome are unknown, researchers recently tested whether
ageifferences in risk taking depend on whether the probabilities
of
successful outcome are known or unknown (Tymula et al.,
2012,013). Tymula and colleagues (2012) had adolescents and
adultsomplete a risk-taking task with two different conditions: a
“knownisk” condition and an “ambiguous risk” condition. In the
“knownisk” condition, participants chose between a sure bet (100%
chancef receiving $5) and a “risky” bet with known reward
probabili-
Please cite this article in press as: Shulman, E.P., et al., The
dual sysNeurosci. (2015),
http://dx.doi.org/10.1016/j.dcn.2015.12.010
ies (e.g., a 50% chance of winning $50, versus $0 if they lost).
Inhe “ambiguous risk” condition, participants again chose
between
sure and risky option, but this time the likelihood of winningr
losing on the risky option was unknown. Compared to adults,
PRESSve Neuroscience xxx (2015) xxx–xxx
adolescents made fewer risky decisions when the probabilities
ofloss were known (i.e., adolescents were less risk tolerant).
However,when the probabilities were unknown, adolescents made
signif-icantly more risky decisions than adults. Thus, under
conditionsthat are more representative of real-life risk-taking
(where riskprobabilities are typically unknown), adolescents evince
a greaterrisk-taking propensity than adults.
Another way in which real-life risk taking differs from risk
tak-ing in the laboratory is with respect to emotional arousal.
Contextsin which risk taking occurs outside the lab are often
thrilling orfrightening; in the lab, both the nature of the risk
taking (the stakesand considerations involved) and the surrounding
environmentare typically less exciting. Scholars have argued that
differencesin arousal give rise to fundamentally different ways of
processinginformation (e.g., Luna and Wright, 2016; Metcalfe and
Mischel,1999). The dual systems model holds that, to the extent
thatdecision-making occurs under conditions that arouse the
socioe-motional system (e.g., conditions that are relatively more
thrilling),differences between adolescent and adult decision-making
and,hence, risk taking will be more pronounced. This pattern
wasobserved in one study that experimentally manipulated the
degreeto which a card game risk-taking task was affectively
arousing(Figner et al., 2009). Consistent with the dual systems
account, ado-lescents evinced greater risk taking and poorer use of
risk-relevantinformation than adults, but only in the more arousing
version ofthe task.
A third difference between most laboratory risk-taking tasksand
real-life risk taking is that, in the laboratory, adolescents
areasked to make decisions when they are alone, whereas the
major-ity of risky behaviors during adolescence occur in groups
(Albertet al., 2013). To mimic this context in the lab, researchers
haveemployed experimental manipulations in which adolescents
com-plete risk-taking tasks in the presence of peers (real or
illusory).Some studies have asked participants to bring same-aged
peers tothe lab (Chein et al., 2011; Gardner and Steinberg, 2005;
Kretsch andHarden, 2014), while others have deceived participants
into believ-ing that they are being observed remotely by a peer
(Smith et al.,2014a). Not only does the “presence” of peers
increase the ecolog-ical validity of the risk-taking task (because
adolescent risk takingoften occurs in groups), but it also appears
to elevate emotionalarousal, which further increases the
comparability to real-worldrisk-taking contexts.
Studies that have manipulated the social context have foundthat
adolescents are more induced by peer presence to take risksthan are
adults (Chein et al., 2011; Gardner and Steinberg, 2005;Smith et
al., 2014a). These findings, which are largely consis-tent with
other studies of peer effects on adolescent driving
[e.g.,Segalowitz et al., 2012; see Lambert et al. (2014) for a
review], sug-gest that adolescents are particularly vulnerable to
the effects ofpeer presence on risk-taking behaviors. Moreover,
neuroimagingdata suggest that the effect of peer presence on risk
taking is dueto increased affective arousal, as evidenced by
greater activationof brain regions within the socioemotional system
(Chein et al.,2011). A recent extension of this line of work in our
lab using arodent model found that adolescent mice, but not adult
mice, con-sume more alcohol in the presence of same-aged
conspecifics thanwhen alone (Logue et al., 2014).
Overall, then, there is evidence for increased risk taking
inadolescence compared to adulthood, though developmental
dif-ferences may only be evident under certain conditions, such
asemotional arousal, ambiguous risk, and the presence of others.The
tendency for adolescents to engage in more risky behaviors
inhighly-arousing contexts together with increased engagement
oftheir socioemotional system during peer observation point to
the
tems model: Review, reappraisal, and reaffirmation. Dev.
Cogn.
importance of reward processing in decision making during
thisperiod of life, a topic to which we now turn.
dx.doi.org/10.1016/j.dcn.2015.12.010
-
ING ModelDogniti
4s
admAiriTst
comoieb2c2m1tedshoriw
4
imoelqiwss(a2onw(e(itaaYti
ARTICLECN-338; No. of Pages 15E.P. Shulman et al. /
Developmental C
. The development of sensation seeking and rewardensitivity
Increased adolescent risk taking in contexts that are
emotion-lly arousing is consistent with one of the central tenets
of theual systems model—that activation and reactivity of the
socioe-otional system reaches its peak during mid- to late
adolescence.
growing literature interrogates this aspect of the model by
exam-ning the psychological and neurological evidence for
heightenedesponsiveness of the socioemotional system during
adolescence,ncluding in situations that do not involve risky
decision-making.his is important because the dual systems model
proposes that theocioemotional system is more responsive generally
in adolescencehan at other ages, not only in the context of risk
taking.
Moreover, the model hypothesizes that the developmentalourse of
the socioemotional system is, unlike the developmentf the cognitive
control system, closely tied to pubertal develop-ent (for review
see Smith et al., 2013). Around age 12 (for boys)
r 11 (for girls), pubertal hormones inundate the brain,
trigger-ng a series of changes in neural structure and function
(Eulingt al., 2008; Schulz et al., 2009), especially in
dopamine-rich lim-ic regions associated with reward processing
(Blakemore et al.,010; Sinclair et al., 2014). It is thought that
these hormone-relatedhanges sensitize the adolescent brain to
reward (Forbes and Dahl,010; Peper and Dahl, 2013), as appears to
be the case in ani-al studies (Alexander et al., 1994; Clark et
al., 1996; Miele et al.,
988). More specifically, the reward system is particularly
sensitiveo the sudden surge of hormones at the start of puberty,
height-ning sensitivity to affective stimuli. Although pubertal
hormoneso not decline into adulthood, we posit that a decrease in
rewardensitivity ensues during later adolescence and into young
adult-ood as the reward system becomes desensitized to the effectsf
these hormones (Smith et al., 2013). While admittedly limited,ecent
evidence integrating measures of puberty into psycholog-cal,
behavioral, and neuroscience studies supports this claim as
ell.
.1. Sensation seeking
One psychological manifestation of socioemotional reactivitys
sensation seeking. As anticipated by the dual systems model,
easures of sensation seeking are often found to be predictivef
self-reported risk taking (e.g., Kong et al., 2013; MacPhersont
al., 2010). True sensation-seeking behavior is difficult to elicit
inaboratory environments (at least, among human subjects);
conse-uently, the vast majority of studies examining age-related
changes
n sensation seeking rely on self-report. As would be
expectedithin the dual systems account, longitudinal and
cross-sectional
tudies generally find evidence of a peak in self-reported
sensationeeking around mid-adolescence and a decrease into
adulthoodHarden and Tucker-Drob, 2011; Peach and Gaultney, 2013;
Quinnnd Harden, 2013; Romer and Hennessy, 2007; Shulman et
al.,014a,b; Steinberg and Chein, 2015; Steinberg et al., 2008).
Thisverall pattern is further corroborated by a number of
longitudi-al studies following individuals from childhood into
adolescence,hich find that sensation seeking increases across this
time period
Collado et al., 2014; Lynne-Landsman et al., 2011; MacPhersont
al., 2010). For example, using the Brief Sensation-seeking
ScaleHoyle et al., 2002), Collado and colleagues (2014) found a
linearncrease in sensation seeking in individuals aged 9–13. Fewer
longi-udinal studies of sensation seeking have followed individuals
fromdolescence into adulthood. However, two recent studies
using
Please cite this article in press as: Shulman, E.P., et al., The
dual sysNeurosci. (2015),
http://dx.doi.org/10.1016/j.dcn.2015.12.010
large, longitudinal data set (the National Longitudinal Study
ofouth 1979 Child and Young Adult Survey) have helped to addresshis
gap and clarify the developmental pattern of sensation seek-ng
across adolescence. Harden and Tucker-Drob (2011) found that
PRESSve Neuroscience xxx (2015) xxx–xxx 5
self-reported sensation seeking increased from age 10 to
mid-adolescence, and then decreased thereafter into early
adulthood.Analyzing the same data set, Shulman and colleagues
(2014a,b)found that females demonstrated an earlier peak in
sensation seek-ing (age 16–17) than males (age 18–19), and a
steeper declinethereafter. Overall, these studies suggest that, as
the dual systemsmodel would predict, sensation seeking follows an
inverted-U pat-tern over time, consistent with the proposed pattern
of change inthe socioemotional system.
The hypothesis that pubertal development drives developmen-tal
change in the socioemotional system in adolescence is derivedin
part from older studies linking higher levels of sensation seek-ing
to more advanced pubertal status (Martin et al., 2002; Resnicket
al., 1993). Newer studies have replicated this result
(Castellanos-Ryan et al., 2013; Gunn and Smith, 2010; Quevedo et
al., 2009;Urošević et al., 2014) and have found evidence that the
correla-tion between self-reported pubertal development and
sensationseeking may be stronger for boys than for girls
(Castellanos-Ryanet al., 2013; Steinberg et al., 2008). Also, as
would be expected basedon the link between puberty and sensation
seeking, recent studieshave found that more advanced pubertal
status in adolescents isassociated with greater involvement in
behaviors that are closelyrelated to sensation seeking, such as
substance use (Castellanos-Ryan et al., 2013; de Water et al.,
2013; Gunn and Smith, 2010),law-breaking (Collado et al., 2014;
Kretschmer et al., 2014), andrisk taking in laboratory contexts
(Collado et al., 2014; Kretsch andHarden, 2014; Steinberg et al.,
2008; but see van Duijvenvoordeet al., 2014 who did not find a
correlation between pubertal statusand performance on a gambling
task).
4.2. Behavioral manifestations of reward sensitivity
Compared to self-report studies of sensation seeking, there
aremarkedly fewer behavioral studies examining the development
ofreward sensitivity, and these have heterogeneous methodologiesand
findings, which makes it difficult to draw firm conclusionsabout
age differences. One large-scale study utilized the IowaGambling
Task (IGT; Cauffman et al., 2010) to explore age-relatedchanges in
reward sensitivity. In the standard version of the IGT,participants
are presented with four decks of cards, two that willwin them money
over repeated play (advantageous decks) and twothat will lose them
money over repeated play (disadvantageousdecks); participants are
permitted to choose freely from the fourdecks (e.g., Smith et al.,
2011b). However, Cauffman et al. (2010)modified the task such that
the computer pseudorandomly selecteda deck on each trial and the
participant was asked to decide whetherto play or pass. This
modification allowed the researchers to disen-tangle affinity for
the advantageous decks—a measure of rewardsensitivity—from
avoidance of disadvantageous decks. The resultsindicated that
mid-adolescents aged 14–17 and older adolescentsaged 18–21 learned
to play from advantageous decks faster thaneither younger
adolescents (ages 10–13) or adults (ages 22–25),a finding that was
recently replicated in an international sampleof more than 5000
individuals (Steinberg and Chein, 2015). Thisoutcome suggests that
ages 14–21 are a period of heightened sen-sitivity to reward. Using
the same data set, Steinberg (2010) alsofound that self-reported
sensation seeking, but not impulsivity, wasassociated with overall
rate of plays from rewarding decks at theend of the task.
Another way researchers have examined developmental differ-ences
in reward sensitivity is by substituting neutral stimuli
(e.g.,letters) with rewarding ones (e.g., happy faces) in
traditional behav-
tems model: Review, reappraisal, and reaffirmation. Dev.
Cogn.
ioral tasks (e.g., measures of impulse control), and then
observingthe extent to which the presence of rewarding stimuli
impactsperformance. Two such studies have examined age
differences(comparing children, adolescents, and adults) in
performance on
dx.doi.org/10.1016/j.dcn.2015.12.010
-
ING ModelD6 ogniti
aaui(aet[l2
(Tdwrataf(taitefovnmtta
4
attuaawamtcsbal
fssrdseaiat
ARTICLECN-338; No. of Pages 15 E.P. Shulman et al. /
Developmental C
n “Emotional Go/No-Go” task. In all Go/No-Go tasks,
participantsre presented with a rapid sequence of target and
non-target stim-li, both of which are typically emotionally
neutral. Participants are
nstructed to press a button when a target stimulus is presenteda
“go” trial) and to withhold the button press (do nothing) when
non-target stimulus is presented (a “no-go” trial).
Non-targetvents occur relatively infrequently, making it
challenging for par-icipants to restrain the impulse to press the
button on no-go trials.As with most measures of self-regulation,
performance improvesinearly with age on traditional Go/No-Go tasks
(see Casey et al.,002 for a review).]
In the Emotional Go/No-Go task employed by Somerville et
al.2011), the stimuli were photographs of either happy or calm
faces.hey found that for adolescents (ages 13–17) more than for
chil-ren (ages 6–12) or adults (ages 18–29), withholding a button
pressas more difficult when the no-go stimulus was a happy
face—a
ewarding stimulus—than when it was a calm face. In fact,
onlydolescents showed impaired impulse control in the happy,
rela-ive to the calm, no-go condition. The researchers proposed
thatdolescents’ greater emotional response to the (rewarding)
happyace made it harder for them to restrain the impulse to
“approach” iti.e., press the “go” button). If so, these results
support the proposi-ion that adolescents are particularly sensitive
to reward. However,nother study (Tottenham et al., 2011) using a
similar, but notdentical, Emotional Go/No-Go task did not find this
pattern (i.e.,hey found no emotion by condition by age group
interaction forrroneous button presses). Though there were
methodological dif-erences between these two studies, the failure
to find the effect inne of the two underscores the need for further
research in thisein. It also highlights the benefits of being able
to incorporateeuroimaging methods. Engagement of the socioemotional
systemay not always be robust enough (especially in laboratory
settings)
o consistently bias behavior. Neuroimaging enables researcherso
detect age differences in the engagement of this system, evenbsent
behavioral consequences.
.3. Neuroimaging of reward sensitivity
In recent years, many neuroimaging studies have asked
whetherdolescents are particularly sensitive to reward. Beyond
iden-ifying differences between adolescents and other age
groups,hese studies help address questions about the neural
mechanismsnderlying adolescents’ heightened reward-seeking. For
example,mong those who agree that adolescents are more inclined
thandults to seek out rewards, there has been disagreement
overhether this results from the fact that rewards are
experienced
s exceedingly pleasurable during adolescence (and are
thereforeore enticing) or because they are experienced as less so
(and are
herefore less satisfying). Indeed, one early notion, now largely
dis-redited, posited that adolescents suffer from a “reward
deficiencyyndrome” which impels them to seek out exciting
experiencesecause mundane ones are not sufficiently rewarding, much
likeddicts who seek out drugs because quotidian experiences noonger
excite them (for a discussion, see Spear, 2002).
To date, most of the developmental neuroscience literature
hasocused on developmental differences in the striatum, and
morepecifically in the ventral portion of the striatum, which is
con-idered one of the main regions involved in the calculation
ofeward (Knutson et al., 2001; Luciana and Collins, 2012). In
ourual systems account, increases in risk taking and other
reward-eeking behaviors are thought to be a consequence of
increasedngagement of the striatum during decision-making, thus
biasing
Please cite this article in press as: Shulman, E.P., et al., The
dual sysNeurosci. (2015),
http://dx.doi.org/10.1016/j.dcn.2015.12.010
dolescents toward more rewarding choices. Heightened sensitiv-ty
to rewarding outcomes of prior decisions may contribute todolescent
risk-taking as well. There is evidence, for example, thathe volume
of the nucleus accumbens (part of the ventral striatum
PRESSve Neuroscience xxx (2015) xxx–xxx
and presumed to be the central structure in the reward
system)increases during the first part of adolescence and then
shrinksthereafter (Luciana and Collins, 2012).
As discussed by Steinberg (2008), the neuroscience
literatureincludes both studies that support and challenge the dual
systemsaccount of heightened striatal engagement during
adolescence(e.g., Bjork et al., 2004; Galvan et al., 2006). Since
that 2008 publi-cation, several reviews have discussed
methodological differencesacross these studies that may have
contributed to the inconsistentfindings (see Galvan, 2010; Richards
et al., 2013). Indeed, the devel-opmental neuroimaging literature
on reward processing has grownsubstantially over the last several
years, and we believe there arepatterns to be noted, and
conclusions to be drawn, that help explainwhat appear to be
contradictory findings.
In its current state, the literature provides considerable
evi-dence that when developmental differences in striatal
activationare present during reward processing (both during the
anticipa-tion and the receipt of reward) adolescents engage the
striatum toa greater extent than both children and adults
(Barkley-Levensonand Galvan, 2014; Christakou et al., 2011; Galvan
and McGlennen,2013; Galvan et al., 2006; Geier et al., 2010;
Hoogendam et al., 2013;Jarcho et al., 2012; Padmanabhan et al.,
2011; Silverman et al., 2015;Van Leijenhorst et al., 2010b). For
example, a recent longitudinalstudy found that, across
mid-adolescence (roughly ages 15–18),ventral striatal activation in
response to “risk taking” on the bal-loon analogue task (which also
reflects reward-seeking) declinesintra-individually over time, and
that striatal activation during thetask is correlated with
self-reported risk taking outside the lab-oratory (Qu et al.,
2015). However, a handful of studies find theopposite
pattern—dampened striatal response during adolescencerelative to
adulthood (Bjork et al., 2004, 2010; Hoogendam et al.,2013; Lamm et
al., 2014)—and others fail to demonstrate any agedifferences (Krain
et al., 2006; Teslovich et al., 2014; Van Leijenhorstet al.,
2006).
In trying to explain this inconsistency, it is important to
notethat disparate findings emerge only for contrasts that focus on
theanticipation of a reward. Studies focusing on striatal
engagementduring the receipt of a reward consistently find that
adolescentsengage the striatum to a greater extent than adults
(Galvan andMcGlennen, 2013; Hoogendam et al., 2013; Van Leijenhorst
et al.,2010b), suggesting that adolescents are—as the dual systems
modelclaims—more sensitive to rewarding outcomes.
The fact that striatal engagement is relatively higher among
ado-lescents than among children or adults during receipt of
rewardsbut not necessarily during reward anticipation potentially
chal-lenges our conception of adolescent risk taking as being
drivenby the prospect of a reward. However, nuances in task
design,modeling of the anticipatory event in imaging analyses, and
therelationship between striatal engagement and behavioral
rewardsensitivity may account for these seemingly inconsistent
results,for several reasons. First, the time points at which events
are mod-eled, and the specific trial periods that are included
within themodel, can dramatically affect the observed neural
response (e.g.,Geier et al., 2010). One factor that seems to
differentiate studiesthat do and don’t report increased adolescent
engagement of thestriatum during reward anticipation is the degree
to which antic-ipatory cues reliably predict the delivery of the
reward. Studiesusing a task design in which the reward cue signals
not only theopportunity for reward, but also an increased
likelihood of earningthat reward, tend to find increased adolescent
activity in the stria-tum during anticipation (e.g.,
Barkley-Levenson and Galvan, 2014;Van Leijenhorst et al., 2010b).
Meanwhile, studies using tasks for
tems model: Review, reappraisal, and reaffirmation. Dev.
Cogn.
which the anticipatory cue signals the possibility to earn a
reward,but is equivocal with respect to the likelihood of
succeeding inobtaining the reward (as in typical implementations of
the Mon-etary Incentive Delay task), do not yield a consistent
pattern of
dx.doi.org/10.1016/j.dcn.2015.12.010
-
ING ModelDogniti
d2
iirrWidoeeiacdsfi
etpIlvMctFgricwso(
tcSi(Fdpimapeti
wbiite2L(
n
ARTICLECN-338; No. of Pages 15E.P. Shulman et al. /
Developmental C
evelopmental differences (e.g., Bjork et al., 2007; Teslovich et
al.,014).
Second, developmental findings regarding striatal outputs dur-ng
reward anticipation are more consistent in studies where theres
also concomitant behavioral evidence that the adolescents
areelatively more sensitive to the rewards being presented (e.g.,
fastereaction times on rewarded trials, more reward-related errors,
etc.).
hile most reward processing tasks used in neuroimaging stud-es
do not include a behavioral measure or control for
behavioralifferences in reward sensitivity across development, the
handfulf studies that do (Barkley-Levenson and Galvan, 2014;
Christakout al., 2011; Geier et al., 2010; Padmanabhan et al.,
2011; Somervillet al., 2011) all report both greater recruitment of
the striatum dur-ng anticipation of reward and higher reward
sensitivity amongdolescents compared to adults, reflected in the
behavioral out-omes. Unfortunately, the majority of reward tasks
used in theevelopmental literature lack a useful behavioral index
of rewardensitivity—an issue that may also account for variability
in striatalndings across development.
Our lab recently explored how socioemotional arousal influ-nces
adolescents’ neural responses to reward by testing whetherhe
presence of peers increased striatal activation during a
reward-rocessing task in which no risk was involved (Smith et al.,
2015).
n this study, we examined the effects of peer observation on
ado-escents’ and adults’ neural response to reward using a
modifiedersion of the High/Low Card Guessing Task (Delgado et al.,
2003;ay et al., 2004). During the receipt of reward, adolescents
who
ompleted the task in the presence of their peers recruited the
stria-um to a greater degree than when they completed the task
alone.urthermore, only when peers were present did adolescents
evincereater striatal activation than adults. These findings
provide cor-oborating evidence that, during adolescence, social
context is anmportant modulator of reward processing, even when
this pro-essing is uncoupled from risk taking. Consistent with this
claim,e have shown that, in the presence of peers, adolescents
evince a
tronger preference for immediate (as opposed to delayed)
rewardsn a Delay Discounting task that does not involve risk
takingO’Brien et al., 2011; Weigard et al., 2014).
Recent neuroimaging studies also support the idea that, in
addi-ion to having profound effects on brain structure [a topic
notovered in the present article; see Blakemore et al. (2010)
andmith et al. (2013) for reviews], pubertal development plays a
rolen developmental change in the sensitivity of the striatum to
rewardBraams et al., 2015; Forbes et al., 2010; Op de Macks et al.,
2011).or example, a landmark, longitudinal neuroimaging study of
chil-ren, adolescents and young adults (N = 299, ages 8–27) found,
asrevious studies have, that activation of the nucleus
accumbens
n response to monetary reward (relative to loss) was higher
inid-adolescence than at other ages (Braams et al., 2015).
Moreover,
ctivation of this region was related both to greater
self-reportedubertal stage and higher levels of salivary
testosterone (Braamst al., 2015). This finding provides strong
support for the claim thathe heightened responsiveness of the
socioemotional system dur-ng adolescence is, at least in part, a
result of pubertal development.
Thus far, we have discussed reward sensitivity specificallyith
respect to striatal activation. However, there also have
een advances in how we understand developmental changesn the
functioning of other regions hypothesized to participaten reward
processing, including the dorsal portion of the stria-um
(Benningfield et al., 2014; Hoogendam et al., 2013; Lammt al.,
2014), mPFC (Christakou et al., 2011), OFC (Galvan et al.,006;
Galvan and McGlennen, 2013; Hoogendam et al., 2013; Van
Please cite this article in press as: Shulman, E.P., et al., The
dual sysNeurosci. (2015),
http://dx.doi.org/10.1016/j.dcn.2015.12.010
eijenhorst et al., 2010b), and the anterior insular cortex
(AIC)Galvan and McGlennen, 2013; Van Leijenhorst et al.,
2010b).
In a recent paper, we posited that continuing maturation of
con-ectivity between the striatum and the AIC, which appears to
act
PRESSve Neuroscience xxx (2015) xxx–xxx 7
as connective hub that influences the engagement of both the
con-trol and reward processing networks, may account for
inconsistentrecruitment of the striatum in adolescent reward
processing (Smithet al., 2014b). Because reward processing entails
the coordinatedaction of a network of regions, developmental
studies examiningthe reward system as a whole, rather than focusing
on activa-tion of specific regions considered in isolation, will
likely yieldgreater insight into changes in reward processing
during adoles-cence, including the reasons for the inconsistent
recruitment of thestriatum in adolescent reward processing (Smith
et al., 2014b).
One study has already demonstrated the potential value ofsuch a
network-based approach. Using resting state data, Cho andcolleagues
(2012) examined functional connectivity between thestriatum,
thalamus, and AIC as adolescents and adults completeda reward
processing task. They found that during anticipation ofreward
(i.e., during cue presentation) adolescents and adults didnot
differ in the functional connectivity between these regions.
Fur-ther, they observed that activity in the AIC and thalamus
precededVS activation in both adolescents and adults. These results
sug-gest that the bottom-up processing of rewards (as demonstrated
bycommunication between these three regions) is adequately
devel-oped by adolescence. Therefore, it may be that
developmentaldifferences between adolescents and adults in reward
sensitivityare not due to immature connectivity, but rather to
differencesin top-down influences on the subjective valuation of
reward.More studies considering the socioemotional system as a
coordi-nated network are needed to inform our understanding of how
thedevelopment of this system relates to age-differences in
rewardprocessing.
In summary, despite occasional inconsistencies in the
literature,self-reported sensation seeking, behavioral measures of
rewardsensitivity, and neuroimaging studies of reward processing
sup-port the contention that reward sensitivity reaches its apex
duringadolescence (e.g., Barkley-Levenson and Galvan, 2014;
Christakouet al., 2011; Collado et al., 2014; Galvan and McGlennen,
2013;MacPherson et al., 2010; Shulman et al., 2014a,b; Somerville
et al.,2011; Van Leijenhorst et al., 2010b). The bulk of
developmen-tal research on this topic provides evidence for a
mid-adolescentpeak in reward sensitivity, and although the
neuroimaging litera-ture does not allow for a precise estimation of
age of peak striatalresponse, the weight of the evidence indicates
that adolescentsengage the striatum (and other components of the
reward net-work) to a greater extent than adults, particularly
during receiptof reward and when differences in reward sensitivity
are reflectedin decision-making behavior. Also consistent with the
dual systemsaccount, studies that have incorporated measures of
puberty typi-cally find that sensation seeking and reward
sensitivity are higheramong those (particularly boys) who are more
pubertally advanced.
5. The development of self-regulation and cognitive control
5.1. Self-reported impulsivity
A second major claim of the dual systems model is that
cognitivecontrol increases linearly across adolescence and does not
reachfull maturity until several years after the peak period of
rewardsensitivity. In the developmental literature, impulse control
(or itsinverse, impulsivity) is the psychological variable most
often usedto assess self-regulation (or its absence).
Impulsiveness—actingin an unplanned and reactive, or less thought
out, fashion—isoften considered a quintessential adolescent
characteristic that
tems model: Review, reappraisal, and reaffirmation. Dev.
Cogn.
predisposes adolescents to engage in reckless behaviors
(Romer,2010). To date, studies examining age differences in
self-reportedimpulsivity—both cross-sectional (Leshem and
Glicksohn, 2007;Steinberg et al., 2008) and longitudinal (Harden
and Tucker-Drob,
dx.doi.org/10.1016/j.dcn.2015.12.010
-
ING ModelD8 ogniti
2d
bovsflQssf
5
rio2(crastct122
dhsccasti(itritrmngantbii2
rdrpa2
ARTICLECN-338; No. of Pages 15 E.P. Shulman et al. /
Developmental C
011)—find that impulsivity decreases with age across the
secondecade of life.
Importantly, the protracted maturation of impulse control
iselieved to continue into young adulthood, where even 18–19
yearlds report higher impulsivity (i.e., less impulse control) than
indi-iduals in their early twenties (Vaidya et al., 2010). Although
adultsometimes engage in impulsive acts, by the early-to-mid 20s
therequency of impulsive behavior appears to stabilize at levels
muchower than those exhibited by adolescents (Steinberg et al.,
2008;uinn and Harden, 2013). For example, using a three-item
impul-
ivity scale, Quinn and Harden (2013) found a linear decrease
inelf-reported impulsivity between the ages of 15 and 21, but
nourther age differences among individuals between 21 and 25.
.2. Behavioral measures of self-regulation
Self-regulation is commonly assessed in behavioral tasks
thatequire response inhibition, a form of cognitive control
thatnvolves overcoming automatic or inappropriate responses in
favorf goal-relevant information processing and actions (Casey et
al.,002). The most widely used measures of response inhibitione.g.,
Go/NoGo, antisaccade, and Stroop paradigms) are typicallyonfigured
to assess “reactive inhibition,” which refers to the out-ight
restraint of motor and perceptual impulses in response ton external
stimulus (e.g., canceling a prepotent response uponeeing a signal,
or maintaining attention in the presence of dis-ractions). A wealth
of behavioral evidence on reactive inhibitoryontrol demonstrates
that self-regulation improves from childhoodo adulthood (Bezdjian
et al., 2014; Bunge et al., 2002; Casey et al.,997, 2002; Durston
et al., 2002; Marsh et al., 2006; Paulsen et al.,015; Rubia et al.,
2006, 2013; Smith et al., 2011a; Tamm et al.,002; Velanova et al.,
2009; Veroude et al., 2013).
Within this literature, adolescents and adults
consistentlyemonstrate better inhibitory control compared to
children;owever, differences between adolescents and adults are not
con-istently found unless the behavioral paradigm is
particularlyhallenging. For example, studies that use the
traditional Stroopolor-word task find no differences in cognitive
control betweendolescents and adults (e.g., Andrews-Hanna et al.,
2011), whereastudies that use an emotional version of the Stroop to
assesshe effect of emotional interference in cognitive control
reportmprovements in self-regulation from adolescence to
adulthoode.g., Veroude et al., 2013). Thus, while adolescents’
ability to inhibitmpulses appears to be comparable to adults’ in
relatively simpleasks, the sort of self-regulatory skills necessary
to appropriatelyespond to more cognitively demanding situations
continue tomprove from adolescence to adulthood. This developmental
pat-ern is also observed in measures of proactive (as opposed
toeactive) inhibitory control, which involves advance planning
andonitoring in anticipation of the need to stop a response or
to
ot engage in a future action (e.g., slowing down responses
too-stimuli in anticipation of a no-go signal approaching,
thereforellowing more time to appropriately cancel a response when
theo-go signal appears; Vink et al., 2014). These findings
suggesthat while basic response inhibition mechanisms may be
maturey adolescence, self-regulatory mechanisms underlying
challeng-
ng reactive response inhibition tasks and proactive
responsenhibition (e.g., planning) may still be developing into the
early0s.
The proposition that the prolonged development of self-egulation
is more evident under challenging conditions has beenemonstrated
using the Tower of London task. In this task, which
Please cite this article in press as: Shulman, E.P., et al., The
dual sysNeurosci. (2015),
http://dx.doi.org/10.1016/j.dcn.2015.12.010
equires strategic planning, participants must rearrange objects
onegs (either real or depicted on a computer monitor) to
produce
specific pattern in the fewest possible moves (De Luca et
al.,003; Steinberg et al., 2008). Researchers manipulate the
difficulty
PRESSve Neuroscience xxx (2015) xxx–xxx
of trials by increasing the number of moves required to
completethe rearrangement successfully. The amount of time a
participantspends deliberating before making his or her first move
(latencyto first move) is used as a measure of impulse control
(becausemaking an initial move too rashly can extend the number of
movesneeded to solve the problem). A study from our lab found no
dif-ferences between children, adolescents, and adults in latency
tofirst move or in the number of moves taken to complete the
trialon easy trials (i.e., those that can be solved in 3 moves)
(Albertand Steinberg, 2011). However, on difficult trials (i.e.,
those thatrequired 5 or more moves to be solved), performance
improvedwith age from childhood to adulthood, and this trend
coincidedwith greater deliberation time prior to the initial move.
These find-ings suggest that when difficult tasks are used, such as
those thatrequire strategic planning, improvement in
self-regulation contin-ues throughout adolescence and into the
early 20s, consistent withthe dual systems model.
Importantly, the ongoing development of self-regulation
intoearly adulthood is also in line with the idea that the
develop-ment of self-regulation is independent of pubertal
development(Smith et al., 2013). In the most comprehensive test of
the rela-tionship between self-regulation, and pubertal status
Steinbergand colleagues (2008) found that self-reported and
behavioral self-regulation was correlated with age but not pubertal
status. Instead,pubertal status was more closely tied to sensation
seeking. Whilethis is the only study we know of that simultaneously
examinesthe relationship between age, pubertal status,
self-regulation, andsensation seeking, thus far the findings
support the notion that thedevelopment of the socioemotional system
is dependent on puber-tal status while self-regulation seems to
develop independently.
Other findings suggest that adolescents’ ability to exert
adult-like self-control also may vary depending on whether rewards
areoffered for better performance (Luna et al., 2001). In several
stud-ies that have rewarded participants for better performance on
anantisaccade task, researchers have found that incentives boost
ado-lescents’ performance to adult levels (Geier et al., 2010;
Jazbec et al.,2006; Padmanabhan et al., 2011). For example, using
an antisac-cade task where some trials were rewarded and some were
not,Geier and colleagues (2010) found that adolescents performed
bet-ter on rewarded trials, compared to non-rewarded trials,
thoughtheir overall task performance did not differ from that of
adults.At first blush, it may appear that these results are
incompatiblewith the dual systems model, since its basic claim is
that height-ened awareness of the availability of rewards should
underminecognitive control in adolescents, not strengthen it. It is
importantto note, though, that this proposition of the dual systems
modelis posited specifically with respect to situations in which
reward-seeking impulses conflict with self-regulatory efforts, as
do mostinstances of risk taking. In contexts where increased
sensitivityto the opportunity for reward serves to motivate faster
and moreattentive responding, without disturbing relevant cognitive
pro-cesses, adolescents’ relatively heightened sensitivity to
reward maybe helpful rather than harmful. For that matter, even in
certain risk-taking scenarios—in particular, those in which
increased risk takingresults in more optimal performance, such as
in certain gamblingtasks for which risky choices have a higher
expected value—greaterreward sensitivity can offer an
advantage.
Overall, the self-report and behavioral literatures on
self-regulation suggest that this capacity improves with age
acrosschildhood, adolescence, and into adulthood. Furthermore, it
maybe that adolescents’ ability to self-regulate is more dependent
thanadults’ on contextual factors, such as task difficulty, the
prospect
tems model: Review, reappraisal, and reaffirmation. Dev.
Cogn.
of a reward for better self-control, and the manner in
whichrewards are presented. Although adolescents may exhibit
adult-like self-regulation under ideal circumstances by around age
15,this capacity is still tenuous, and maturation of
self-regulation may
dx.doi.org/10.1016/j.dcn.2015.12.010
-
ING ModelDogniti
bs
tviiaaltf(
5
diitwmcecws
isw2oamw2Tnatbf2R
aarcBefnaihfh“ofcau
ARTICLECN-338; No. of Pages 15E.P. Shulman et al. /
Developmental C
e best indexed by the consistency with which individuals
demon-trate self-control across different contextual
circumstances.
As we have noted previously (Strang et al., 2013), it is
imprudento conclude that heightened reward sensitivity is
inherently disad-antageous, or that impulsivity is always
problematic. In situationsn which greater attentiveness to reward
or more impetuous behav-or is desirable, adolescents may enjoy a
distinct advantage overdults. Indeed, one of the tenets of the dual
systems model is thatdolescence evolved as a period during which
individuals are moreikely to engage in sensation seeking and less
likely to restrain urgeso pursue immediate rewards because this
combination may con-er a reproductive advantage during a period of
heightened fertilitySteinberg, 2014).
.3. Neuroimaging of cognitive control
In recent years, developmental neuroimaging has helped eluci-ate
the neural mechanisms underlying age-related improvements
n cognitive control. Continuing maturation of response
inhibitions often examined in terms of development of the
prefrontal cor-ex, and particularly the lateral prefrontal cortex
(lPFC). In lineith the dual systems framework, we postulate that
develop-ental improvements in cognitive control are supported by
the
oncurrent maturation of these underlying neural regions and
bynhancements in top-down connectivity between frontal
cognitiveontrol regions and other cortical and subcortical areas
associatedith motor processing, affective processing, and the
execution of
elected actions.Irrespective of age, individuals who perform
better on response
nhibition tasks (i.e., Go/No-Go, Flanker, Stroop, Stop Signal,
anti-accade) exhibit greater activation of the lPFC compared to
thoseho perform poorly (Durston et al., 2006; Rubia et al.,
2006,
013; Velanova et al., 2009). Across adolescence, performancen
response inhibition tasks improves with age—a pattern thatppears to
be explained by continuing maturation of the lPFC, withost studies
finding either a linear increase in lPFC recruitmentith age
(Adleman et al., 2002; Bunge et al., 2002; Durston et al.,
006; Marsh et al., 2006; Paulsen et al., 2015; Spielberg et al.,
2015;amm et al., 2002; Velanova et al., 2009; Vink et al., 2014) or
sig-ificantly increased engagement of the lPFC from adolescence
todulthood (Rubia et al., 2000, 2006, 2013; Veroude et al., 2013).
Fur-hermore, several studies have demonstrated a direct
relationshipetween age-related increases in lPFC engagement and
success-ul cognitive control (Adleman et al., 2002; Andrews-Hanna
et al.,011; Bunge et al., 2002; Casey et al., 1997; Durston et al.,
2006;ubia et al., 2006, 2007, 2013; Velanova et al., 2009).
Whereas the behavioral and neuroimaging literatures gener-lly
indicate a relationship between increases in cognitive controlnd
engagement of the lPFC from adolescence into adulthood,
theelationship between age, behavior, and neural engagement
fromhildhood to adolescence is not as consistent (Alahyane et al.,
2014;ooth et al., 2003; Braet et al., 2009; Casey et al., 1997;
Durstont al., 2002). In fact, some studies find that children
utilize morerontal regions than adults—in terms of overall volume
and/or mag-itude of activity—in order to successfully withhold a
prepotentction. These findings have led researchers to posit that
increasesn self-regulation from childhood to adolescence and into
adult-ood may be due to a developmental progression from diffuse
to
ocal activation (Durston et al., 2002). In this account, during
child-ood and early adolescence, the brain is inefficient and needs
towork harder,” recruiting neurons across a larger frontal area
inrder to successfully inhibit a response (though see Poldrack,
2014
Please cite this article in press as: Shulman, E.P., et al., The
dual sysNeurosci. (2015),
http://dx.doi.org/10.1016/j.dcn.2015.12.010
or a critique of the explanatory value of the term “efficiency”
in thisontext). As the brain undergoes continued reorganization
acrossdolescence, necessary neural connections are strengthened
andnnecessary ones are pruned, creating a more efficient brain
and
PRESSve Neuroscience xxx (2015) xxx–xxx 9
leading to more focal recruitment of regions within the lPFC
duringsuccessful inhibition.
Cognitive control encompasses the integration of several
(oftensimultaneous) processes that support planning behavior in
accordwith one’s intentions (Miller, 2000). The effective
integration ofthese processes relies not only on the functional
recruitment ofimplicated brain regions, but also on the strength of
connectiv-ity among them (Hwang et al., 2010; van Belle et al.,
2014). Forexample, a study by Hwang and colleagues (2010) examined
devel-opmental changes in connectivity underlying inhibitory
controland found that connectivity between the prefrontal cortex
andother cortical areas increased from childhood into
adolescence,with some connections continuing to strengthen from
adoles-cence to adulthood. The increases they observed in the
numberand strength of frontal connections to both cortical and
subcorti-cal regions during the transition from adolescence into
adulthoodsuggest that developmental improvements in cognitive
controlmay be supported by age related enhancements in the
top-downregulation of task-engaged regions. Results such as these
under-score the potential benefit to the field of fMRI studies
movingbeyond simplistic models of regional activation toward more
elab-orate models that consider connectivity among regions
throughoutdevelopment, as well as the strength and efficiency of
thoseconnections, which likely support age-related increases in
theacquisition and execution of complex cognitive control skills
(see,e.g., Satterthwaite et al., 2013). This is particularly true
because, asnoted earlier, there is reason to believe that
continuing changes inconnectivity account for the observation that
some aspects of cog-nitive control continue to strengthen into
early adulthood, insteadof plateauing in adolescence.
6. Is risk taking during adolescence related to heightenedreward
sensitivity and immature cognitive control?
As reviewed above, research largely supports the dual
systemsmodel’s characterization of adolescence as a time of
heightenedsocioemotional reactivity (relative to earlier and later
periods) andstill maturing cognitive control. Moreover, there is
considerableevidence consistent with the proposition that the
developmen-tal trajectories of reward sensitivity and cognitive
control (and,by extension, sensation seeking and self-regulation)
differ, withthe former following an inverted U-shaped pattern and
the latterevincing protracted, linear improvement that extends into
the thirddecade of life.
How well does the literature support the claim that
devel-opmental change in these two systems explains heightened
risktaking during adolescence? The model posits that it is the
con-fluence of the developmental patterns of the socioemotionaland
cognitive control systems—relatively high responsiveness toreward
combined with relatively weak self-regulation—that ren-ders
adolescents particularly vulnerable to risk taking. If the
twosystems contribute to risk taking in an additive manner, we
shouldfind independent correlations between the functional state of
eachsystem and risk-taking propensity. Indeed, there is evidence
for thispattern in the literature.
In order to serve as a test of the dual systems model in
predict-ing risk taking, a behavioral study must include measures
of bothsocioemotional reactivity and cognitive control.
Unfortunately,constructs reflecting the functional status of the
socioemotionaland cognitive control systems, like sensation seeking
and impulsiv-ity, tend to be highly correlated (e.g., Shulman and
Cauffman, 2014;
tems model: Review, reappraisal, and reaffirmation. Dev.
Cogn.
Steinberg et al., 2008), despite being theoretically and
empiricallyseparable (Duckworth and Kern, 2011; Duckworth and
Steinberg,2015). Thus, for studies that examine the relationship
betweenonly one of these constructs and risk taking, the
correlation will
dx.doi.org/10.1016/j.dcn.2015.12.010
-
ING ModelD1 ogniti
bSmseeaotgrmiscthlwe
rstov(Fsu2dohadsu2a2teoeapolra
srcdfecCDftc(f
ARTICLECN-338; No. of Pages 150 E.P. Shulman et al. /
Developmental C
e contaminated by the contribution of the unmeasured
construct.ome of this overlap between sensation seeking and
self-regulationay be artifactual—a result of the difficulty of
developing mea-
ures that cleanly assess one construct and not the other.
(Forxample, items like “I often get myself into trouble” could
reflectither sensation seeking or impulsivity.) But some of the
observedssociation between these constructs may be attributable to
anngoing dynamic interplay between the socioemotional and cogni-ive
control systems; for example, when the socioemotional
systemenerates an impulse to pursue an intrinsically rewarding
expe-ience and the cognitive control system counters with a
signaleant to restrain the impulse. Although a large number of
stud-
es have examined risky behavior in relation to measures of
eitherensation seeking or impulse control, very few have examined
theoncurrent contributions to risk taking of psychological
manifesta-ions of socioemotional activation and cognitive control.
Even fewerave examined this question in a sample spanning
childhood, ado-
escence, and adulthood, which would be necessary to fully
testhether variation in the functional status of these two
systems
xplains age-related patterns in risk taking.In the few studies
that have simultaneously assessed constructs
eflective of the socioemotional and cognitive control systems
(e.g.,ensation seeking and impulse control) along with measures of
riskaking, the anticipated correlations are found. Both higher
levelsf sensation seeking and lower levels of impulse control
explainariation in risk taking, over and above the effects of one
anotherCyders et al., 2009; Donohew et al., 2000; Quinn and Harden,
2013).or example, one study of college students found that
sensationeeking uniquely predicted increases in the frequency of
alcoholse, over and above several measures of impulsivity (Cyders
et al.,009). Another found that both sensation seeking and
“impulsiveecision making” were independently associated with
greater oddsf ninth-graders engaging in sex, non-coital sexual
behavior, alco-ol use, and marijuana use (Donohew et al., 2000).
Moreover, thesessociations were comparable in magnitude, except
that impulsiveecision-making was more strongly associated with
having sex andensation seeking was more strongly associated with
marijuanase. Similarly, unpublished data from our lab—based on a
sample of83 10–30-year olds and using a scale that surveyed
involvement in
wide range of risk-taking behaviors (see Shulman and
Cauffman,014)—suggests that impulse control and sensation seeking
con-ribute equally (betas = −.21 and .21, respectively) to
self-reportedngagement in risky behaviors (controlling for age,
sex, and eachther). An obvious shortcoming of these studies is that
they relyxclusively on self-report. However, if common method
variancelone were driving the findings, one would not expect to see
inde-endent relations between risk taking and either sensation
seekingr impulsivity once the other predictor was controlled.
Anotherimitation of these studies is that it is not yet clear how
well self-eport measures reflect the functional status of the
socioemotionalnd cognitive control systems.
Neuroimaging studies have the advantage over behavioraltudies of
being able to measure activation within distinguishableegions
thought to correspond to the socioemotional and cognitiveontrol
systems (although heightened activity in these regionsuring a
laboratory task does not constitute a direct measure of
theunctional status of these systems). A few studies have
examinedngagement of regions associated with the socioemotional
and/orognitive control systems during adolescent decision making
(e.g.,ascio et al., 2015; Kahn et al., 2015; Paulsen et al., 2011;
vanuijvenvoorde et al., 2015b). However, only Chein et al.
(2011)
ound increased engagement of structures within the socioemo-
Please cite this article in press as: Shulman, E.P., et al., The
dual sysNeurosci. (2015),
http://dx.doi.org/10.1016/j.dcn.2015.12.010
ional system and decreased activation of structures within
theognitive control system simultaneously within a risk-taking
taska driving simulation). One additional study (Paulsen et al.,
2011)ound age effects in both the PFC and striatum during risk
taking.
PRESSve Neuroscience xxx (2015) xxx–xxx
During risky (i.e., varying expected values) compared to
sure(i.e., guaranteed reward) decisions, several PFC regions
showedincreased activation with age, consistent with Chein et al.
(2011).On the other hand, striatal activation was inconsistent,
makingthese results difficult to interpret.
Another recent study examined the extent to which age
differ-ences in impatience during a temporal discounting task—in
whichparticipants choose between a smaller immediate reward and
alarger delayed reward—are explained by variations in
self-reported“present hedonism” (i.e., reward sensitivity) and
“future orien-tation” (i.e., impulse control) and engagement of
neural regionsand networks during decision making (van den Bos et
al., 2015).Though the decision making task did not involve risk,
the studyis nonetheless relevant to the dual systems model because
it wasdesigned to probe the degree to which adolescents’ tendency
todiscount the future is due to greater reward sensitivity or
weakerself control. The results indicated that choices to delay
gratifica-tion in the decision task were associated with greater
self-reportedfuture orientation and increased engagement of
frontoparietal con-trol circuitry, but not with variation in self
reported hedonism.Also, improvements in frontostriatal connectivity
mediated thelink between age and willingness to wait for a larger
reward. Theauthors interpreted these results as showing that weak
cognitivecontrol, rather than heightened reward sensitivity,
explains ado-lescents’ tendency to discount the future. However,
limitations ofthe methodology (e.g., limited range on the present
hedonism scale,lumping immediate rewards together with rewards to
be receivedin two weeks for analysis of the discounting data, and
the unemo-tional context of the laboratory) may have biased the
study againstfinding linkages between reward sensitivity and
impatience (seeSteinberg and Chein, 2015).
Whereas other studies have not demonstrated
simultaneouslyheightened socioemotional activation and dampened
cognitivecontrol within the same task, a few recent ones have
observedheightened striatal activation when adolescents receive a
rewardfollowing a decision (Braams et al., 2014, 2015). In one
furtherrelevant study (Cascio et al., 2015), researchers recruited
recentlylicensed drivers (∼age 16) to complete a response
inhibition taskand, one week later, a driving simulation in the
presence of a peerconfederate. The peer either encouraged risky
driving or safe driv-ing. In the latter condition (encouragement of
safe driving), greaterengagement of cognitive control circuitry
(i.e., IFG and BG) duringthe response inhibition task (indicative
of better cognitive control)predicted safer driving behavior in the
simulated driving task. Par-ticipants who exhibited higher
cognitive control also showed noincrease in risky driving in the
condition in which the peer encour-aged risk taking, suggesting
that individuals who evince greaterengagement of cognitive control
circuitry may be more resistantsocioemotional arousal. These
findings indicate that poor cogni-tive control, as expected, also
plays a role in risk-taking behavior.However, because the study did
not compare age groups, it cannotaddress whether maturation of
cognitive control helps to accountfor developmental patterns in
risk taking.
In another recent study, van Duijvenvoorde and colleagues(2015b)
had children, adolescents, and adults complete a risk-taking task
(Columbia Card Task). While overall risk-takingtendency did not
differ by age, adolescents showed greater acti-vation of control
circuitry (including the dmPFC) as the riskinessof the decision
increased. This effect was not seen in children oradults. The
authors suggest that heightened recruitment of controlcircuitry was
necessary due to the heightened emotional responseto risk during
this age. Although there is good reason to believe
tems model: Review, reappraisal, and reaffirmation. Dev.
Cogn.
that the functional status of both the socioemotional and
cognitivecontrol systems during adolescence contribute to
heightened risktaking during this stage of development, the dual
systems modelstill awaits a comprehensive study that confirms (or
disconfirms)
dx.doi.org/10.1016/j.dcn.2015.12.010
-
ING ModelDogniti
ttb
7
fFbtrUriitimtvncrpt
taeoemcceswerhensoNedocaan2r(tflmm
ao
ARTICLECN-338; No. of Pages 15E.P. Shulman et al. /
Developmental C
he purported joint effects of the developmental trajectories
ofhe socioemotional and cognitive control systems on
risk-takingehavior.
. Unresolved questions and future directions
There are many unresolved issues in the literature that
awaiturther research attention. Here we highlight just a few of
them.irst, because of differences in opportunity to engage in
riskyehavior outside the laboratory environment, the effects of
matura-ion of the socioemotional system and cognitive control
system oneal-world risk taking are likely to be modest and
difficult to detect.ndoubtedly, contextual constraints on the
behavior of adolescents
elative to adults overwhelm any putative effect on actual risk
tak-ng. To take an obvious example, even if 15-year-olds are
highern sensation seeking and lower in self-regulation than people
inheir 20s, these differences will not be reflected in age
differencesn reckless driving in a country where 15-year-olds are
not per-
itted to drive. Thus, tests of the dual systems model will
requirehe continued development of laboratory tasks that are
ecologicallyalid but that afford individuals of different ages
equal opportu-ity to take risks. Future efforts to test the dual
systems model’slaims also would benefit from collaboration between
behavioralesearchers and neuroscientists to develop measures that
morerecisely reflect the functioning of the neural systems
underlyinghe socioemotional and cognitive control systems.
Second, still unresolved is the question of why the
socioemo-ional system declines in responsiveness between
adolescence anddulthood. Luciana and Collins (2012) have speculated
that experi-nce with rewards and learning lead to lower background
levelsf dopamine, a proposition that has not yet been tested.
Caseyt al.’s (2008) model implies that decreases in risk taking
after theaturation of the socioemotional system, which in their
view is
omplete by mid-adolescence, are ultimately attributable to
theontinued strengthening of the cognitive control system. Given
thevidence of reduced reward responsiveness in the key node of
theocioemotional system in adulthood (relative to adolescence),
itould seem that their version of the model suggests that
strength-
ning of the cognitive control system prospectively dampens
theeactivity of the socioemotional system. One study from our labas
tested this hypothesis: Shulman et al. (2014b) examined theffects
of self-reported impulse control (a reflection of the cog-itive
control system) and sensation seeking (a reflection of
theocioemotional system) on one another over time in a large
samplef youth, ages 10–25, who were assessed biennially as part of
theLSY79 Children and Young Adults Study. The analysis failed to
findvidence that increases in impulse control prospectively
predictecreases in sensation seeking; in general, these two traits
devel-ped independently. Recently, a neuroimaging study using
intrinsiconnectivity found that increases in dlPFC-subcortical
(thalamusnd striatum when not controlling for age2) connectivity
acrossdolescence were associated with increases in cognitive
control butot with decreases in reward sensitivity (van
Duijvenvoorde et al.,015a). Instead, decreases in reward
sensitivity were related to age-elated decreases in connectivity
within the socioemotional systemvmPFC, OFC, and striatum). Together
these findings lend supporto the notion that these traits develop
independently. However,urther investigation of this question is
warranted; in particular,
Please cite this article in p