i An update to this article is included at the end

Developmental Cognitive Neuroscience 27 (2017) 19–34

Contents lists available at ScienceDirect

Developmental Cognitive Neuroscience

journal homepage: www.elsevier.com/locate/dcn

Review

Beyond stereotypes of adolescent risk taking: Placing the adolescent brain in T

developmental context☆
Daniel Romera, , Valerie F. Reynab, Theodore D. Satterthwaitec
⁎

a
Annenberg Public Policy Center, University of Pennsylvania, United States
b
Human Neuroscience Institute, Cornell University, United States
c
Department of Psychiatry, Perelman School of Medicine, University of Pennsylvania, United States

ARTICLE INFO ABSTRACT

Keywords: Recent neuroscience models of adolescent brain development attribute the morbidity and mortality of this
Brain development period to structural and functional imbalances between more fully developed limbic regions that subserve re-
Dopamine ward and emotion as opposed to those that enable cognitive control. We challenge this interpretation of ado-
Decision-making lescent development by distinguishing risk-taking that peaks during adolescence (sensation seeking and im-
Cognitive control
pulsive action) from risk taking that declines monotonically from childhood to adulthood (impulsive choice and
Experience
other decisions under known risk). Sensation seeking is primarily motivated by exploration of the environment
under ambiguous risk contexts, while impulsive action, which is likely to be maladaptive, is more characteristic
of a subset of youth with weak control over limbic motivation. Risk taking that declines monotonically from
childhood to adulthood occurs primarily under conditions of known risks and reflects increases in executive
function as well as aversion to risk based on increases in gist-based reasoning. We propose an alternative Life-
span Wisdom Model that highlights the importance of experience gained through exploration during adoles-
cence. We propose, therefore, that brain models that recognize the adaptive roles that cognition and experience
play during adolescence provide a more complete and helpful picture of this period of development.

1. Introduction of these neurobiological imbalance models are illustrated in Fig. 1

(Casey et al., 2008), showing that limbic structures are activated in
Recent theorizing and research regarding the neurodevelopment of excess of prefrontal cognitive control regions during the adolescent
the adolescent brain has generated considerable attention in both the period.
popular media and the scientific literature. The most striking general- A key feature of such imbalance models is the suggestion that a
ization stemming from this work is that the adolescent brain does not developmental deficit in PFC cognitive control limits adaptive decision
fully mature until at least age 25, with the implication that adolescent making by adolescents.1 However, when Giedd et al. (1999) first pre-
decision-making and judgment is similarly limited up to this age (Casey sented evidence of declining PFC gray matter volume in adolescents,
et al., 2008; Giedd, 2004; Steinberg, 2008). This conclusion rests on they attributed the phenomenon to the role that experience plays in
research indicating that the myelination and pruning of the prefrontal sculpting the brain during this developmental period. As they put it, the
cortex (PFC) continues into adulthood, well after ventral limbic regions decline in PFC gray matter “may herald a critical stage of development
that control motivation and reward have achieved these milestones. As when the environment or activities of the teenager may guide selective
a result, it is proposed that adolescents suffer from a structural as well elimination during adolescence.” (p. 863). In other words, gray matter
as functional deficit in the ability of the PFC to exert top-down control decline in the PFC could reflect pruning that results from the experience
over drives that are spurred by the limbic motivational system, leading that adolescents gain during this period rather than a direct marker of
to less than “rational” behavior during adolescence. The basic dynamics increasing behavioral control. As Spear (2010) also noted, pruning may

☆
DR was supported by National Institute on Drug Abuse (R01DA033996); VFR was supported by National Cancer Institute (R21CA149796); National Institute on Nursing Research
(R01NR014368-01) and National Institute of Food and Agriculture (NYC-321423 and NYC-321436); TDS was supported by National Institute on Mental Health (R01MH107703 and
K23MH098130). We thank James Bjork, Joseph Kable, Kathryn Mills, and Flaura Winston for helpful comments on earlier versions of this paper. However, the conclusions reached in this
paper do not necessarily reflect the views of the funding agencies or prior readers.
⁎
Corresponding author.
E-mail address: [email protected] (D. Romer).
1
In a recent review of imbalance research, Casey (2015) prefers not to describe imbalance as a “deficit” but rather a “brain that is sculpted by evolutionarily based biological
constraints….” Nevertheless, these constraints are seen as contributing to a “200% increase in preventable deaths (accidents, suicide, homicide)…” during adolescence (p. 296–297).

http://dx.doi.org/10.1016/j.dcn.2017.07.007
Received 3 May 2017; Received in revised form 24 July 2017; Accepted 24 July 2017
Available online 26 July 2017
1878-9293/ © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/BY/4.0/).
D. Romer, et al. Developmental Cognitive Neuroscience 27 (2017) 19–34

Support for this view has been observed in both humans and other
animals following the onset of puberty. Nevertheless, a focus on adverse
outcomes leaves us with a biased picture that limits our ability to
identify adaptive features of adolescent brain development within the
context of the entire lifespan. Instead, we argue for a more nuanced
interpretation of risk taking and its implications for healthy develop-
ment. In particular, we outline the evidence regarding the role of sen-
sation seeking, which although it peaks during adolescence does not
reflect imbalance, as opposed to forms of impulsivity which either do
not peak or only characterize a subset of youth. Our review of research
regarding structural development indicates that the relation between
brain structure and risk taking has failed to consider the implications of
different forms of risk taking. Our analysis suggests that stereotypes of
adolescents as particularly susceptible to unhealthy risk taking sim-
plifies how adolescents think about risk and ignores the important role
that experience plays in more adaptive forms of risk taking (Reyna
et al., 2015a; Romer, 2010). In what follows, we consider what a
broader perspective on adolescent brain development would suggest,
how that helps to explain the way adolescents make decisions, and how
these decisions can be improved.

Fig. 1. Casey et al. (2008) model of imbalance between prefrontal versus limbic control
over behavior in adolescence. 1.1. The rise in sensation seeking
With permission from Institute of Medicine (2011, p. 38).

Consistent with stereotypes of young people, adolescents exhibit

be “an example of developmental plasticity whereby the brain is on- heightened attraction to novel and exciting experiences despite their
togenetically sculpted on the basis of experience to accommodate en- evident risk (Chambers et al., 2003; Romer and Hennessy, 2007; Spear,
vironmental needs.” Needs could vary dramatically across environ- 2010). This tendency, known as sensation seeking (Zuckerman, 2007),
ments and cultures (Mata et al., 2016), potentially resulting in very rises rapidly during adolescence. As seen in Fig. 2, a nationally re-
different patterns of pruning and brain organization during adolescent presentative U.S. survey of 1800 youth indicates that sensation seeking
brain development (Choudhury, 2010). For example, evidence has ac- peaks around age 19 in males and 16 in females. A similar pattern has
cumulated to suggest that differences in socioeconomic status, which been observed across a wide range of countries (Duell et al., 2016). This
are correlated with cultural influences, are associated with differences rather striking pattern is regarded as a marker of rising dopaminergic
in brain structure (Brito and Noble, 2014; Noble et al., 2015). In par- activation during adolescence (Chambers et al., 2003; Wahlstrom et al.,
ticular, Noble et al. (2015) demonstrated that lower socioeconomic 2010) and may reflect activity in the midbrain dopamine pathway as-
status was associated with diminished cortical surface area and reduced cending from the ventral tegmental region (Ikemoto, 2007; Previc,
hippocampal volume even when controlling for maternal education. 2009). This pathway traverses through the ventral striatum before
Such hippocampal volume reductions have been reported by other branching into the orbital and ventromedial frontal cortex. These re-
studies as well (Hanson et al., 2011; Hueston et al., 2017). Others have gions are heavily involved in recognition and anticipation of reward
observed differences in language-related regions (Piccolo et al., 2016) (Pagnoni et al., 2002; Schultz et al., 1997) and thus suggest a biological
and modular brain organization (Krishnadas et al., 2013). Future re- basis for increased attraction to novel and exciting experience during
search should unpack influences of education, culture, and income adolescence that declines as the brain transitions to adulthood (see
(with concomitant effects on nutrition, access to healthcare, and other Wahlstrom et al., 2010 for a review of evidence linking a peak in ex-
factors that may plausibly affect development) on specific aspects of ploratory behavior during adolescence with changes in dopamine ex-
brain development. pression over the lifespan). A related personality cluster known as the
Rather than emphasizing the important role of culture and experi- behavioral activation system (BAS) is also believed to be related to
ence in shaping the development of the brain, researchers have instead
focused on excess levels of maladaptive risk behavior, such as injury,
drug use, pregnancy, and other unhealthy outcomes, as support for
imbalance (Dahl, 2004; Steinberg, 2008; Casey, 2015). However, the
stereotype of the impulsive, emotional, and distraught adolescent rests
much more on the rise in adverse outcomes during this age period than
on their overall prevalence (Institute of Medicine, 2011; Rivers et al.,
2008). For the vast majority of adolescents, this period of development
passes without substance dependence, sexually transmitted infection,
pregnancy, homicide, depression, suicide, or death due to car crashes
(Institute of Medicine, 2011; Willoughby et al., 2013). Indeed, the risks
of these outcomes are often comorbid with each other (Biglan and
Cody, 2003; Kreuger et al., 2002), leaving the average adolescent
without great risk of life-altering consequences.
We do not question the reality that the adolescent period entails
risk. What we challenge is the interpretation of the brain and behavioral
underpinnings of this risk. Research suggests that the brain is structured
to enhance development by encouraging movement toward in-
dependence and self-sufficiency, a process that supports exploration Fig. 2. Trends in sensation seeking by gender in a national U. S. sample.
and learning (Luna and Wright, 2015; Murty et al., 2016; Spear, 2013). With permission from Romer (2010).

20
D. Romer, et al. Developmental Cognitive Neuroscience 27 (2017) 19–34

dopamine function (Carver and White, 1994). One of the indicators of

the BAS known as fun seeking is highly related to sensation seeking,
while two other related indicators (reward responsiveness and drive)
may be more associated with achievement motivations (Romer et al.,
2016).
What is often neglected in discussion of imbalance is a rise in do-
pamine activity in dorsal and medial PFC (Meng et al., 1999; Weickert
et al., 2007) fed by another pathway originating primarily in the sub-
stantia nigra that ascends through dorsal striatum into dorsal PFC and
parietal cortex, regions that control movement and higher order deci-
sion making (Ikemoto, 2007; Previc, 2009). Dopamine neurons in this
pathway appear to serve more global salience and cognitive processing
functions than the ventral route (Bromberg-Martin et al., 2010; Roeper,
2013). This pathway enables the adolescent brain to exert greater at-
tention and other executive functions that are important abilities for
reasoning and complex decision-making (Cools et al., 2008; Cools and
Robbins, 2004). In particular, dopamine is critical for the maintenance
of activity in working memory (WM) (Arnsten et al., 2012; D’Esposito
and Postle, 2015), a function centered in frontoparietal cortex that is Fig. 3. Data illustrating development of cognitive control during adolescence and early
critical for recruiting experience-based information during decision adulthood.
With permission from Casey and Caudle (2013).
making (Fuster, 2009; Miller and Cohen, 2001; Shamosh et al., 2008).
However, dopamine activation in the dorsal striatum has also been
linked to various cognitive functions, including cognitive control and outcome learning (Matthews et al., 2013). Nevertheless, many im-
episodic memory (Bäckman et al., 2000; Bäckman et al., 2006; Volkow portant cognitive functions that are subserved by the dorsal striatum
et al., 1998). Furthermore, as we describe below, both structural and and its connection with ventral PFC are in place prior to adolescence,
functional dopamine activity in the striatum and PFC declines starting and consistent with the development of cognitive control in humans,
in the third decade of life with associated declines in these cognitive dopaminergic control over cognitive ability centered in the mPFC ap-
functions. Thus, the rise in dopaminergic activity that may underlie pears to be available by early adulthood.
sensation seeking is also accompanied by increased dopaminergic ac- In view of the rise in both limbic and prefrontal dopamine expres-
tivity in corticostriatal pathways that support the ability to exert con- sion during adolescence, the generalization that adolescents lack cog-
trol over rewarding experience and to learn from it (Murty et al., 2016; nitive control relative to limbic activation may have been overstated, a
Whalstrom et al., 2010). conclusion also reached by Crone and Dahl (2012). Following their
review of imaging studies of functional brain development, they found
1.2. Brain development and adolescent self control no pattern of brain activation that consistently distinguished adolescent
from adult performance in cognitive control tasks: Some cognitive
Emphasis on the reward-related functions of dopamine has re- control tasks elicited higher activation in adolescents versus adults,
inforced a focus on impulsive behavior during adolescence. However, if whereas other tasks elicited lower activation. As seen in Fig. 3, by ages
the adolescent brain undergoes development in both ventral motiva- 16 and 17, the variability in executive control as assessed in a go-no
tional and dorsal cognitive capacities, then the hypothesis of structural task is already so large that many adolescents in that age range perform
and functional imbalance as a normative developmental pattern needs at a level that is equal to that of adults. Although early adolescents
reconsideration. Indeed, contrary to structural imbalance models of perform below the average level of adults in go/no-go and similar tasks,
brain development, individual differences in sensation seeking (and most late adolescents are either equal to or better than the average
associated risk taking) have been found to be positively correlated with adult (Williams et al., 1999).
WM and other indicators of executive function (Raine et al., 2002; Similarly, in one of the largest imaging studies of executive function
Romer et al., 2011; Zuckerman, 2007). In one longitudinal study in youth ages 8–22, Satterthwaite et al. (2013a) found that differences
(Romer et al., 2011), individual differences in WM predicted sub- attributable to age were much smaller than individual differences in
sequent levels of sensation seeking even after controlling for age, sug- performance on an N-back WM task (see Fig. 4). Although WM im-
gesting that sensation-based risk taking rises in concert with executive proved with age, individual differences were large, with many late
function. Indeed, executive function rises rapidly during adolescence adolescents exhibiting better WM performance than the average young
(as does sensation seeking) and asymptotes well before age 25 (Gur adult. Furthermore, brain scans demonstrated that WM performance
et al., 2012; Luciana et al., 2005; Luna et al., 2004; Williams et al., was correlated with enhanced activation in PFC executive regions along
1999). Thus, the rise in dopamine expression during adolescence may with reduced activation of the default mode, which includes limbic
play a role in both sensation seeking and executive function. cortex (Buckner et al., 2008; Raichle and Gusnard, 2005). Thus, while
Recent models of dopamine expression in mice and rats suggest that WM and executive function do improve with age in the aggregate, in-
dopamine neurons become active in ventral and dorsal striatum prior to dividual differences are large, such that many late adolescents are as
their emergence in medial PFC (mPFC) (Naniex et al., 2012; Reynolds capable as adults at recruiting performance-relevant activation of the
et al., 2017). Indeed, dopamine pathways between orbitofrontal PFC executive system and deactivation of default mode regions.
and the striatum are in place prior to adolescence in humans (Fareri
et al., 2015). The growth of dopaminergic connections between the 1.3. Sensation seeking vs. impulsivity in adolescent development
striatum and mPFC is associated with improvements in cognitive
functions related to value learning (Naniex et al., 2012; Reynolds et al., A major contention of imbalance models is that adolescents are
2017). However, these gaps are eliminated by early adulthood, perhaps more subject to poor impulse control than either children or adults. For
mirroring what happens in humans. As dopamine function in the mPFC example, Casey (2015, page 299) notes that imbalance is “presumably
grows during adolescence, there is also evidence that activation in the not observed in childhood because there is a relative lack of maturity
dorsal striatum is weaker than in the ventral region, a pattern that may across and between regions within the (corticosubcortical) circuit, and
have the adaptive function of enhancing exploration and action- in adults, there is a relative maturity of the underlying neurocircuitry.”

21
D. Romer, et al. Developmental Cognitive Neuroscience 27 (2017) 19–34

Fig. 4. Data from Satterthwaite et al. (2013 with permission) illus-

trating the rise in WM ability from ages 8 to 22 that is overshadowed
by individual differences. Panel A illustrates the stimuli used to assess
different degrees of challenge to working memory. B and C show the
increasing difficulty of the task as reflected in behavior. D shows the
overall performance as measured with d’. Red points refer to females
and blue to males.

However, in drawing this conclusion one must distinguish between adverse health consequences, such as addiction or problem gambling,
sensation seeking, which does not appear to reflect imbalance, and than youth with impulsive tendencies (Khurana et al., 2017; Magid
impulsivity, which is a form of decision-making that is overly sensitive et al., 2007; Smith et al., 2007). These findings have remarkable par-
to immediate urges without adequate consideration of consequences. allels in the animal literature where it has been found that sensation-
There are at least two forms of impulsivity that are relevant in regard to seeking lab rats are likely to try addictive drugs, but they are not likely
adolescent behavior (Winstanley et al., 2010). One called impulsive to continue their use when it leads to adverse consequences (Belin et al.,
action reflects tendencies to act without thinking about consequences, 2008; Winstanley et al., 2010). In contrast, rats that act impulsively are
also known as motor impulsivity (Patton et al., 1995; Romer et al., much more likely to develop addictive behavior that persists despite the
2009). Acting without thinking is moderately positively related to maladaptive consequences. Lack of cognitive control, therefore, is more
sensation seeking, as well as the BAS and, also peaks during adoles- clearly characterized by impulsive action than sensation seeking.
cence (Collado et al., 2014; Kasen et al., 2011; Shulman et al., 2015). A Bjork and Pardini (2015) review the evidence regarding develop-
major difference between acting without thinking and sensation mental changes in brain response to rewarding stimuli. Their review
seeking is that, unlike sensation seeking, it is inversely related to WM suggests that youth who exhibit harmful risk-taking tendencies exhibit
ability (Khurana et al., 2012; Romer et al., 2011). This inverse re- brain responses consistent with weak cognitive control. However, this
lationship is understandable in that persons with this form of im- pattern is only representative of a subset of youth. Impulsive youth who
pulsivity lack the attentional control and capacity to consider alter- lack self-control have been observed to display this characteristic at a
natives to strong impulses. young age and to continue to display poor control over behavior well
Another form of impulsivity, known as impulsive choice (e.g., into adulthood (Iacono et al., 2008; Moffitt et al., 2011; Reyna, 2012).
Mischel et al., 1989; Romer et al., 2010), reflects tendencies to choose Indeed, such youth are disproportionately likely to experience the ha-
smaller, immediate rewards over larger but delayed rewards (McClure zards that arise during adolescence and beyond, as examples, higher
et al., 2004; Metcalfe and Mischel, 1999). This preference for im- rates of injuries and illnesses due to automotive crashes, violence, and
mediate reward is also inversely related to WM ability (Shamosh et al., sexually transmitted infections (Moffitt et al., 2011; Sourander et al.,
2008), again suggesting that weak ability to consider alternative 2006). Nevertheless, it is important for both theoretical and pragmatic
courses of action predisposes to this form of impulsivity. Nevertheless, reasons to distinguish risk taking that arises due to interest in exploring
it is largely unrelated to sensation seeking (Cyders and Coskunpinar, the environment from a developmental deficit in cognitive control
2011), which is not surprising given that it involves a choice between during the adolescent period.
two rewards. Although it correlates with impulsive action, it does not We have observed the beginnings of the distinction between ex-
exhibit a peak during adolescence. Rather it declines slowly from ploratory and impulsive risk taking in a longitudinal cohort one of us is
childhood onward, reflecting the increase in executive function during studying in Philadelphia (Romer et al., 2009). Youth ages 13–15 who
adolescence (Green et al., 1994; Romer et al., 2010; Steinberg et al., began to use drugs with increasing frequency were much more likely to
2009; van den Bos et al., 2015). Thus, it is a simplification to assert that be highly impulsive than sensation seeking. Sensation seekers at this
the adolescent period is marked by heightened impulsivity relative to age try drugs, but they do not typically exhibit progression in regular
children and adults considering that impulsive choice does not peak use (Khurana et al., 2015a). A similar pattern was observed for early
during this age period. sexual initiation (Khurana et al., 2012) and unprotected sex (Khurana
Although impulsive action and sensation seeking appear to conform et al., 2015b). Although high sensation seekers may explore novel be-
to the stereotype of the impulsive adolescent, sensation seeking has havior that can lead to harmful outcomes if continued, they appear to
different consequences from impulsive action. Research in both humans learn from these experiences as they age, while youth with impulse
and other animals indicates that sensation seeking is positively corre- control problems do not. These patterns suggest that the increase in
lated with PFC activation, while impulsivity displays the opposite ten- sensation seeking that characterizes adolescence does not necessarily
dency (Jupp and Dalley, 2014). Youth with high sensation seeking lead to maladaptive behavior unless it is accompanied by weak ex-
tendencies gravitate toward potentially risky activities, but in the ab- ecutive function, such as exhibited by acting without thinking or the
sence of acting without thinking, they are less likely to experience desire for immediate reward.

22
D. Romer, et al. Developmental Cognitive Neuroscience 27 (2017) 19–34

As suggested by Reyna and Farley (2006), there appear to be two even though the expected value of the risky option is equivalent to the
divergent routes to heightened adolescent risk taking: one that is as- certain option (e.g., win $2 for sure vs. equal chance to win nothing or
sociated with a greater reliance on executive resources (energized by a $4) (see Levin and Hart, 2003). These tests also demonstrate a mono-
greater drive toward sensation seeking) and one that is associated with tonic decline in risk taking in which children are more risk seeking than
reduced executive capability (impulsivity) (see also Chassin et al., 1989; adolescents who are more risk seeking than adults. When different age
Reyna et al., 2015b). Despite the dominant narrative that adolescents groups are compared on other types of choice tasks in which a certain
are impulsive, Reyna and Farley’s (2006) overview of the literature option is not available, the same decline is evident once IQ is held
suggests that much of adolescents’ risk taking is characterized by a constant (Defoe et al., 2015). This control is important because it is
surprising “rationality” in the conventional economic sense (Institute of difficult to arrange choice tasks that are understandable for children
Medicine, 2011). That is, risk taking across many real-world domains is (e.g., under age 10; see also Reyna and Ellis, 1994).
found to be a function of trade-offs between perceived risks and ben- Like Levin and Hart (2003) and Reyna and Ellis (1994), Paulsen
efits—as contrasted with impulsive or emotional risk taking. If any- et al. (2011, 2012) designed a task that was easily comprehended from
thing, many adolescents can be described as “hyper-rational,” inasmuch childhood to young adulthood and found clear evidence for a decline in
as they rely on the risks and benefits of their behavior even more than risk seeking whether a certain option was available or not. One ex-
adults do, which promotes risk taking when negative consequences are planation for this clear deviation from imbalance models as well as
perceived to be unlikely (as is the case with many public health threats, stereotypes of adolescent impulsivity is that adolescents are more risk
such as contracting HIV). seeking under ambiguity than children or adults (Paulsen et al., 2012).
Brain models that emphasize imbalanced development of limbic That is, given the potential for a reward but lack of information about
versus cognitive control regions suggest that adolescents are resistant to its likelihood, adolescents will be more inclined to explore the risky
information about risks. Because the imbalance is “hard-wired,” there is option than either children or adults. As a result, they may actually
little one can do other than to shield adolescents from their otherwise exhibit a more rational response than adults who are notoriously risk
natural risk tendencies (Steinberg, 2008, 2014). However, trends in the averse when certain rewards are at risk.
use of both legal and illegal drugs, as assessed since 1975 by the In a demonstration of adolescent exploration, Tymula et al. (2012)
Monitoring the Future (MTF) study (Johnston et al., 2015a), indicate showed that compared to adults, adolescents are more likely than
that adolescents are responsive to the harm that drugs can pose. These adults to take risks that are ambiguous. As a result, their behavior was
harms are transmitted through various channels, including media more “rational” in the economic sense of evaluating options based on
campaigns (Elder et al., 2004), school-based education (Faggiano et al., expected value than adults. In their study, adults were so averse to
2008), and parental and peer influences (Bahr et al., 2005). Indeed, use unknown risks that they preferred expected values that were far smaller
of popular substances such as tobacco, alcohol, and marijuana have than adolescents were willing to tolerate. As they conjectured, “such a
declined since the survey began. Furthermore, the correlation in the tolerance may make sense, because it would allow young organisms to
MTF study between annual rates of use of these drugs and perceptions take better advantage of learning opportunities.” Adolescents’ greater
of risk associated with those drugs was r = −0.83 for alcohol, tolerance for ambiguity may also reflect their overly optimistic eva-
r = −0.63 for marijuana, and r = −0.80 for cigarettes. These patterns luation of the rewards of novel behavior (Romer, 2010; Romer and
are suggestive of an adolescent brain that is sensitive to adverse con- Hennessy, 2007). Exploration of novel environments has survival value
sequences despite interest in exploring novel experiences. Contrary to and has been linked to activity in both PFC and subcortical regions
stereotypes about adolescents, Reyna and Farley’s (2006) overview of (Daw et al., 2006), again suggesting that adolescents may not be
the literature showed that much adolescent risk taking was consistent structurally handicapped with respect to specific information-proces-
with sensitivity to both perceived risks and benefits, which is a rational sing abilities that facilitate learning.
rather than impulsive process according to traditional views of ration- The evidence we have reviewed suggests that in characterizing
ality. adolescent risk taking, it is critical to distinguish between different
types of risk behavior, each of which has unique motivational and
1.4. The importance of type of risk cognitive underpinnings. We describe these different patterns in Fig. 5.
Impulsive action is characterized by insensitivity to risk, a form of risk
Research concerning the imbalanced adolescent brain has taken a taking that peaks during early adolescence. However, it is only char-
rather broad brush approach to the assessment of risk taking. In this acteristic of a subgroup of youth with weak executive function and self-
section, we review what is known about developmental trends in risk control, conditions that are present prior to adolescence (Bjork and
taking as assessed in laboratory tasks and how different forms of risk Pardini, 2015; Kreuger et al., 2002; Moffitt et al., 2011). This form of
taking are related to cognitive control. As previously noted, despite risk taking is most clearly associated with the behavior that imbalance
stereotypes of adolescents as more impulsive than either children or models seek to explain. In the absence of intervention, this form of
adults, there is considerable evidence that some risk-taking preferences imbalance can persist into adulthood. Impulsive choice as well as other
(such as impulsive choice) do not peak during adolescence but instead forms of decision making under known risk do not peak during ado-
follow a monotonic decline from childhood to adulthood. A develop- lescence. Indeed, adolescents are more inclined to avoid risks than
mental decline in risk taking is common in tasks in which the gains and children under delay of reward or other forms of decision making under
losses attributable to different choices are explicitly defined or able to known risk. Finally, choice under ambiguity is sensitive to sensation
be learned quickly (Defoe et al., 2015). This kind of task, known as seeking tendencies that encourage exploration, such as use of drugs
decision under risk, is different from ones in which the outcomes and (Romer and Hennessy, 2007). Although it may peak during adoles-
associated probabilities are ambiguous or unknown, commonly known cence, exploration and tolerance of ambiguity is not devoid of cognitive
as decisions under ambiguity (Brand et al., 2006; Volz and Gigerenzer, control and may actually be more adaptive in many circumstances than
2012). the extreme ambiguity avoidance exhibited by adults.
Assessments of impulsive choice fall under the rubric of decision
under risk in that these paradigms explicitly provide information re- 1.5. Do adolescent risk taking tendencies match predictions of imbalance?
garding the magnitude of reward and the likelihood of its occurrence as
denominated by either delay or probability. Other tests of decision If developmental imbalance between cognitive control and limbic
under risk provide choices between two or more alternative options activation were responsible for peaks in adolescent risk taking, one
that differ in reward and probability of outcome. A common task is one would expect those peaks to occur in mid-adolescence when imbalance
in which a certain positive option is contrasted with a riskier option is at its height (Willoughby et al., 2013). However, to the degree

23
D. Romer, et al. Developmental Cognitive Neuroscience 27 (2017) 19–34

Fig. 5. Differences in three types of risk taking ten-

dencies across age. Trends for Known and Ambiguous
risks apply to all adolescents while the trend for
Insensitive risk taking applies to youth with high le-
vels of acting without thinking that precede adoles-
cence and remain elevated into adulthood.

adolescents suffer injury, the period of highest risk occurs rather late in development? Studies of brain structure and risk taking tend to produce
the transition to adulthood when inexperience is still high despite the confusing results, which is not surprising given that risk taking itself is a
nearly complete maturation of brain networks that are thought to en- complex behavior. Some forms of risk taking can be positively related to
able cognitive control. For example, use of either cigarettes or mar- executive function (e.g., decision under ambiguity) and others inversely
ijuana peaks around age 20 in the U.S. (Romer, 2010); while binge related (e.g., impulsive action).
drinking remains high throughout the third decade (Johnston et al., In normal development, gray matter loss in PFC is thought to be a
2015b). Deaths due to overdoses of alcohol peak much later, around marker of maturation, perhaps reflecting fine-tuning of brain structure
age 50 (Kanny et al., 2015), although younger drinkers may be more (Spear, 2010). However, research examining structural brain develop-
likely to overdose and survive. The proportion of driving fatalities at- ment in relation to executive control has found that less prefrontal gray
tributable to alcohol peaks between ages 21 to 34 and continues at high matter is associated with ADHD and forms of impulsivity that emerge
rates until age 54 (US Census Bureau, 2012). Sexually transmitted in- early in development (Shaw et al., 2011; van Ewijk et al., 2012). Such
fections such as gonorrhea and chlamydia peak between ages 20 to 24 persons also exhibit a lower rate of gray matter reduction as they age.
(Centers for Disease Control and Prevention, 2014). With thinner cortical gray matter at the outset of adolescence, there
Conclusions about age trends in risk-taking must take risk oppor- may be less to prune. Thus, simple indices of gray matter reduction are
tunity and other co-occurring factors into account, as Shulman et al. unlikely to be a pure marker of enhanced cognitive control. Indeed, the
(2016) note. In particular, research has shown that adult supervision of development of brain structure varies with IQ in a complex interaction
adolescents declines as they age, especially in males, thereby providing with age. Shaw et al. (2006) demonstrated that higher IQ is associated
increasing opportunity to take risks (Gerard et al., 2008). However, with thinner cortex in childhood, while in adolescents this relationship
with sensation seeking on the decline and cognitive control at its peak is reversed and thicker cortex is associated with higher IQ. A more re-
during early adulthood, any increases in unhealthy risk taking during cent longitudinal study of 504 participants corroborated this interaction
this period would seem less attributable to imbalance than to stable with age, but suggested that the transition point may occur in early
individual differences in impulsivity that emerge prior to adolescence adulthood (age 21) rather than adolescence (Schnack et al., 2015).
and remain evident into adulthood (Iacono et al., 2008; Moffitt et al., Recent research suggests that patterns of gray matter change are
2011). As adolescents enter young adulthood, they are presented with location dependent and underlines the importance of white matter
even greater risk-taking opportunities that will challenge those with expansion occurring as a result of myelination. Vandekar et al. (2015)
weak cognitive control. recently showed that gray matter reduction was maximal in sulci where
In summary, the appeal of the imbalance model rests in part on the white matter organization occurred. Multivariate analyses also revealed
popular stereotype of the adolescent as impulsive and lacking in cog- a second pattern, whereby gyral cortex thickened in early adolescence,
nitive control. Closer examination of this stereotype reveals that only a process that appears to asymptote by age 13. Thus, while this finding
one form of impulsivity (acting without thinking) peaks during ado- awaits replication in longitudinal studies, human neuroimaging re-
lescence and that this form of impulsivity varies significantly across search indicates that cortical thinning may be the result of both myeli-
individuals. The other major form of impulsivity, impulsive choice, nation and pruning, while cortical thickening characterizes a secondary
declines from childhood to adulthood and thus is not likely to be ex- maturation pattern that occurs during adolescence in more localized
plained by peaks in adolescent imbalance. In addition, other decisions parts of gyral cortex. These more complex patterns of gray matter
under conditions of known risk also decline from childhood to adult- change further suggest that indices based on overall gray matter change
hood. Finally, although sensation seeking does peak during adoles- are likely to obscure more complex organizational changes in brain
cence, it is not characterized by the absence of cognitive control. structure as adolescents age. Indeed, it appears that the dominant
pattern of brain development from childhood to adulthood is mono-
1.6. Evidence for structural imbalance tonic decline in gray matter associated with increases in myelination
(Lebel et al., 2012). However, Berns et al. (2009) found that controlling
If imbalance does not characterize all types of risk taking behavior, for age, white matter maturation was positively related to a wide range
what is the evidence regarding structural imbalance in brain of prior real-world risk behavior in adolescents, some of which may

24
D. Romer, et al. Developmental Cognitive Neuroscience 27 (2017) 19–34

well be associated with exploratory drives. This pattern was recently directly address the conditions underlying adolescent-specific im-
replicated in an experimental context by Kwon et al. (2015). Never- balance.
theless, other research finds that white matter integrity in some brain In total, the findings suggest that white matter development and
regions is weaker in children with ADHD, suggesting that some white associated declines in gray matter are not clearly related to reduced
matter deficits play a role in youth with this form of impulsivity (van risky behavior. Furthermore, connectivity between the striatum and
Ewijk et al., 2012). In sum, there does not appear to be a simple relation vmPFC is established early in development such that adolescents need
between myelination and risky behavior. Developmental differences in not be handicapped by inadequate linkage between these regions.
myelination can be associated with greater rather than less risky beha- Indeed, the evidence appears to be more consistent with the important
vior during adolescence, especially when the risks are ambiguous. On role of the vmPFC in reward-based learning during adolescence, and the
the other hand, conditions such as ADHD which are likely to reflect close connectivity between this region and the ventral striatum (Haber
impulse control problems may be characterized by less white matter and Knutson, 2010; Rolls, 2014).
development.
Analyses of gray matter maturation in limbic regions also fail to 1.7. Other models of risky decision making
conform to expectations of structural imbalance. Rather than reflecting
early maturation in limbic structures, gray matter change continues Other models of risky decision making also focus on the relative
well into adolescence (Dennison et al., 2013; Raznahan et al., 2014). A strength of cognitive control and reward sensitivity processes (Casey,
direct test of the structural imbalance model conducted by Mills et al. 2015). However, these models do not require a structural deficit in the
(2014) examined differential brain maturation in a longitudinal study ability to exercise self-control. For example, McClure et al. (2004) find
of volume changes in the PFC versus the amygdala and the nucleus that within the same individuals, making less impulsive choices is as-
accumbens. Using three scans across childhood, adolescence, and young sociated with greater activity in PFC cognitive control regions, while
adulthood, these researchers found that the amygdala exhibited in- during the same scanning session making impulsive choices is asso-
creased volume up to about age 16, when growth in this structure began ciated with ventral striatal activation. Nevertheless, in a recent devel-
to asymptote. The accumbens exhibited declining volume as adoles- opmentally sensitive study across ages 8–25, van den Bos et al. (2015)
cents aged. Using these limbic regions as indicators of imbalance in found that functional connectivity between medial striatum and cog-
relation to maturation of the PFC, the researchers correlated individual nitive control regions (dorsolateral and ventrolateral PFC) mediated
differences in structural imbalance with reports of real-world risk declines in impulsive choice across age. However, as has been observed
taking. Consistent with the possibility that the risk taking recalled by in other research (Green et al., 1994; Romer et al., 2010; Steinberg
those participants was a mixture of exploratory and impulsive behavior, et al., 2009), discount rates declined rapidly from childhood to early
there was no correlation between the imbalance observed in brain adolescence and showed very little decline from that point onward.
structure and reports of risk behavior during adolescence. Notwith- Thus, apart from individual differences, impatience may not be parti-
standing the study’s sample size (n = 33), the authors “failed to find a cularly relevant for understanding adolescent peaks in maladaptive
relationship between the presence of a mismatch in brain maturation adolescent risk taking. In addition, connectivity change was observed
and risk-taking and sensation-seeking behaviors during adolescence.” with the medial rather than ventral striatum, suggesting greater in-
(p. 147). volvement with cognitive control than motivational functions of the
The imbalance model advanced by Galván et al. (2006) suggested striatum (Cools et al., 2008), a result consistent with the finding that
that ventromedial PFC (vmPFC) matures more slowly than the ventral sensation seeking and discounting are largely uncorrelated (Cyders and
striatum and that greater activation in the striatum relative to vmPFC Coskunpinar, 2011; Romer et al., 2010).
could be the source of greater risk taking in adolescents. This model The model of hot versus cold cognition proposed by Metcalfe and
does not seem to follow from the original observation that brain ma- Mischel (1999) proposes that reducing the appeal of immediate (hot)
turation during adolescence proceeds from ventral to dorsal regions. rewards can be accomplished by flexible allocation of attention (e.g.,
Indeed, a recent study examining resting state connectivity between the thinking about something other than the reward). This model also fo-
striatum and ventral- and medial-PFC found that these regions ex- cuses on the ability to delay gratification, a form of impulsive choice
hibited early and rather stable connectivity from childhood to adult- that does not peak during adolescence. Although the ability to allocate
hood (Fareri et al., 2015). attention may increase with development, it is not a skill that is par-
It is important to note that Galván et al. (2006) based their con- ticularly impaired in adolescence relative to earlier ages, and variation
clusions on a reward learning paradigm in which adolescents have been in tendencies to delay gratification may well be driven by individual
shown to exhibit greater ventral striatal response to reward prediction differences in life experience (McGuire and Kable, 2013).
errors than adults (see also Section 2.2 of this issue later in regard to The Driven Dual Systems model proposed by Luna and Wright
Cohen et al., 2010). Thus, this heightened striatal response may not be a (2015) also focuses on imbalance between cognitive control and do-
particularly sensitive indicator of maladaptive risk taking. Furthermore, pamine driven reward motivation. Unlike the Casey model in Fig. 1,
as participants gained experience in the task, adolescents also showed their model recognizes that cognitive control achieves adult levels by
an anticipatory vmPFC response suggesting that this region “predicted” mid-adolescence. However, they suggest that the rise in dopamine ac-
the outcome of the reward cue, an ability that is associated with healthy tivation during adolescence exceeds the levels experienced by adults,
vmPFC function (Rolls, 2014). Thus, if anything, this study showed thereby predisposing toward immediate rewards in excess of adult le-
greater functional synchrony between these regions in adolescents than vels. Nevertheless, Luna and Wright suggest that the sensation seeking
in either children or adults, a finding potentially indicative of greater that results from this imbalance has adaptive characteristics, such as the
sensitivity to reward learning. A follow-up study by Galván et al. (2007) need to explore the environment. They also note that this imbalance
found that heightened ventral striatal activation in receipt of reward “may make some adolescents vulnerable to risk-taking behavior’ (p.
was a predictor of the likelihood of engaging in hypothetical real-world 107). Luna and Wright use the term risk-taking to characterize mala-
risk-taking; however, this was an individual difference, characteristic of daptive behavior by definition; but as we have noted, exploration is a
both adults and adolescents. form of risk-taking that need not be maladaptive. Thus, their model is
Christakou et al. (2011) found that activation in vmPFC and con- consistent with our suggestion that the rise in maladaptive risk taking
nectivity with ventral striatum was related to age-dependent decline in only characterizes some adolescents and thus accords with the analysis
impulsive choice. Consistent with the cognitive control predictions of presented here.
imbalance models, but not the reward sensitivity predictions, this form Another model that has garnered significant attention in regard to
of risk taking did not peak during adolescence. Thus, this study did not adolescent brain development is the Triadic Model of Ernst and

25
D. Romer, et al. Developmental Cognitive Neuroscience 27 (2017) 19–34

colleagues (Ernst, 2014; Ernst and Fudge, 2009). This model is de- rise in risk-taking during this period. Here we note that this presumed
scribed by Ernst (2014) as a ‘heuristic tool’ for organizing neuroscience linear increase in cognitive control conflates two separate cognitive
research on motivated behavior. The model not only considers im- processes, one based on structural maturation of the cognitive control
balance between cognitive control and reward processing regions but system and the other dependent on increasing connectivity between the
also includes potential imbalance with avoidance processes centered in PFC and parietal, occipital, and temporal cortices that build over time
the amygdala and related regions. Ernst proposes that the three regions with experience (Fuster, 2009, 2013). When these are separated, it
act to achieve an equilibrium that “varies across individuals.” becomes clear that cognitive control also peaks by late adolescence and
The triadic model rightly expands the brain regions that must be early adulthood while experience-based development continues in a
considered in understanding developmental changes during adoles- monotonic fashion well into the aging process.
cence. However, although the amygdala has input to the ventral The distinction between cognitive control and experience-based
striatum, it is sensitive to both rewarding and aversive events. Rather cognitive development is consistent with recent research that has
than serving to balance the ventral striatum, it may actually alert the moved beyond simple models of gray matter change to more nuanced
ventral striatum to salient events that require action (Rolls, 2014). In analyses of brain networks (Ernst et al., 2015; Pfeifer and Allen, 2012).
their reviews of literature regarding reward processing, Richards et al. An important study by Dosenbach et al. (2010) examined the devel-
(2013) also note the wide differences that obtain depending on the opment of functional brain networks from ages 7 to 30 using resting-
laboratory task and the incentives provided to research participants. In state fMRI. Using a machine-learning approach, they showed that
some paradigms, adolescents exhibit control equal to adults, while in measures of functional connectivity could provide an index of brain
others they do not. However, even when adolescents appear to engage network maturation that correlates with age. The most important fea-
in less cognitive control than adults, this deficit can be overcome by tures of this model are enhanced connectivity within large-scale func-
increasing incentives for performance (Richards et al., 2013). In sum, tional brain networks, such as the executive control and default mode
the model may apply more to individual differences due either to ex- networks, but reduced connectivity between such networks during the
perience or tendencies that exist prior to adolescence. adolescent age period (Baum et al., 2017; Stevens, 2016). Interestingly,
Casey (2015) also suggests that models of adolescent risk taking analyses indicate an asymptote in functional network development by
include interconnections between more than just the striatum and PFC. age 22, before presumed maturation of pruning and white matter
She highlights findings suggesting that compared to children and young growth has run its course. However, the dataset was somewhat sparse
adults, adolescents exhibit stronger emotional responses to laboratory in the late adolescent age range, leaving open the possibility that the
stimuli. For example, adolescents commit more errors in a go/no-go asymptote occurred even earlier (e.g., see Vaso et al., 2017). In addi-
task when the no-go cue is a smiling face compared to a neutral face. tion, similar to the pattern of WM development observed by
What is less clear is how these responses relate to real world risk taking. Satterthwaite et al. (2013a,b), the range of maturation of brain net-
It may be that such responses are related to exploratory behavior, works during the resting state varied widely across individuals. These
which is less likely to lead to harmful consequences than high levels of patterns have been subsequently replicated in independent datasets
impulsive behavior. Other examples of emotional responses to emo- controlling for confounds due to head movement (Fair et al., 2013;
tional stimuli suggested that in some paradigms (but not others), ado- Satterthwaite et al., 2013b).
lescents react more strongly to aversive stimuli, such as fearful faces. Rubia (2013) and Luna et al. (2010) summarize the changes in brain
But here again, it is not clear that these responses would lead to activation that occur in cortex from childhood to later adulthood. Their
heightened or harmful risk taking, and in some cases, heightened summaries indicate increasing connectivity within cognitive control
adolescent response only characterized some adolescents, with others networks as children age, which may contribute to greater cognitive
showing emotional responses comparable to children and adults (Hare control during adolescence. This conclusion is consistent with recent
et al., 2008). At this point, without the necessary clarifying information studies indicating that brain networks involved in cognitive control
regarding the type of risk taking that is being examined, it is difficult to versus default mode become more segregated during adolescence
draw conclusions about such evidence. (Baum et al., 2017; Dosenbach et al., 2010; Fair et al., 2008;
In summary, our review of the evidence regarding structural dif- Satterthwaite et al., 2013), but conversely become less segregated
ferences in brain development suggest that the adolescent brain un- during later adulthood, thereby displaying an inverted-U shaped pattern of
dergoes rapid change during this age period, but connections to mala- interconnectivity across the lifespan (Betzel et al., 2014; Chan et al.,
daptive risk behavior depend on both individual differences and the 2014). Furthermore, Chan et al. (2014) found that reduced network
type of risk taking. Evidence linking brain structure and function to segregation at any adult age was associated with an important marker
risky behavior tends to be inconclusive regarding imbalance, and this is of age-related cognitive decline, namely weaker verbatim memory. As
not surprising given the many ways that risk taking can manifest. summarized by Betzel et al. (2014), on the one hand, functional con-
Furthermore, cognitive control reaches maturity by early adulthood nectivity (FC) over the lifespan within resting state networks (RSNs)
when sensation seeking is in decline but the adverse effects of risk “decreased with age, affecting higher-order control and attention net-
taking begin to peak. Thus, the developmental imbalance that is sug- works. On the other hand, FC tended to increase between RSNs, espe-
gested to be at the root of such adolescent risk taking is unlikely to cially among components of the dorsal attentional network, the sal-
explain this rather late appearance of developmental risk. We propose iency/ventral attention networks and visual and attention networks and
instead that for the majority of adolescents, maladaptive risk taking the somatomotor network.” (p. 352).
declines from childhood on. For those with heightened impulsivity, These changes are consistent with a brain that grows in cognitive
risks can continue to grow as opportunities for such behavior increase; ability during adolescence but that increasingly relies on between-
however, this pattern is concentrated in a subset of youth who exhibit network connections as adulthood progresses into aging. For most
impulsive behavior prior to adolescence. adults, the ability to exert cognitive control or behavioral inhibition
eventually declines as indexed by tasks that challenge response speed
2. Cognitive control vs. experience-based cognitive development and attentional skills (e.g., stop-signal and WM) (Lindenberger, 2014).
over the lifespan However, older adults have greater ability to draw from experience,
which is consistent with growing connectivity between networks.
Imbalance models suggest that cognitive control develops linearly
during adolescence while sensation seeking peaks. Furthermore, 2.1. The importance of experience
Shulman et al. (2016) claim that cognitive control continues to grow
well into young adulthood and that this helps to explain the continued Despite the stereotype of adolescents as impulsive risk takers, it is

26
D. Romer, et al. Developmental Cognitive Neuroscience 27 (2017) 19–34

important to consider the role of exploration and learning that occurs risky option provides a critical test of contrasting theoretical predic-
during this period of development. Fuster (2009, 2013) proposes a tions (e.g., see Kühberger and Tanner, 2010). Fuzzy-trace theory pre-
model of brain development across the neocortex involving what he dicts that a gist representation favors the selection of the certain option
calls cognits or networks of neuronal connections between the PFC and for gains, a preference that grows with experience. Experiments on the
other cortical regions that build over time. Cognits provide a bridge development of risk taking confirmed that, in addition to motivational
between “executive memory” in the PFC and sensory and “perceptual and cognitive control factors, risk preference is a function of competing
memory” in other regions. These memories enable a form of what verbatim versus gist mental representations of decision options. From
Goldberg (2006) calls “executive intelligence” built from experience in the perspective of gist-based intuition, risking HIV infection by having
encountering novel problems. These networks are assumed to develop unprotected sex is a bad idea even if the risks are low and the benefits
in a hierarchical manner, such that individual experiences reside at the are high (see Reyna, 2008). These theoretical ideas explain the other-
lowest level of the network. As experience accumulates, more abstract wise puzzling (but predicted and replicated) result that experience,
levels of memory are formed that enable clearer decision rules for ac- both from childhood to adulthood and from novice to expert in a spe-
tion across similar domains. These more abstract memories provide cific domain of decision making, is associated with greater reliance on
experienced actors with shortcuts to decision-making that require less gist-based intuition rather than verbatim reasoning (e.g., Reyna et al.,
cognitive effort than less experienced adolescents might have to exert. 2011; Reyna and Lloyd, 2006).
Fuster’s theory of cognits is also broadly consistent with fuzzy-trace The greater verbatim information-processing efficiency of adoles-
theory advanced by Reyna and colleagues, who highlight the im- cents (relative to children and aging adults) would appear to be a
portance in decision-making of a distributed system of gist in the brain, benefit that compensates for their lack of experience. Adults progres-
as opposed to localized verbatim, memory representations (Reyna et al., sively lose the ability to exert cognitive control over their attention and
2015b; see Reyna and Huettel, 2014, for differences in neural sub- WM capacities (Lindenberger, 2014), leading to what Goldberg (2006)
strates). Fuzzy-trace theory emphasizes the accumulation of experience has termed “The Wisdom Paradox.” With aging, the neocortex con-
that leads to more adult-like decision-making and gist-based intuition tinues to lose gray matter in PFC with associated reductions in the
(Reyna and Brainerd, 2011; Wilhelms et al., 2015). As people gain ability to remember verbatim details of past experience and to hold
experience in a decision domain, they begin to understand patterns in information in WM (Chan et al., 2014; Dennis et al., 2013). Adults
the outcomes that accrue, a process that enables them to rely on more experience a domain-general decline in verbatim cognitive skills
abstract gist principles regarding those decisions and less on the literal starting in the third decade of life (Tucker-Drob, 2011; Lachman et al.,
rewards and costs of a particular decision. This experience encoded in 2014), although gist memory is conserved (e.g., Brainerd et al., 2009;
durable gist memories would be expected to facilitate decision-making Reyna, 2012). During the same period of verbatim decline, the brain is
(Fuster, 2009; Goldberg, 2006; Reyna and Lloyd, 2006; Reyna and estimated to lose about 7% of its striatal dopamine transporters per
Mills, 2014). Although late adolescents and young adults have greater decade (Volkow et al., 1996), with even larger declines in the PFC
cognitive control than the average older adult, they may not have de- (Eppinger et al., 2011). These declines, which begin in the third decade
veloped the insight from experience, or what is conventionally called of life, are associated with reductions in various cognitive and motor
wisdom, that is important for functioning in the world (Reyna et al., functions, including episodic and working memory, inhibitory control,
2011). Such experience would convert many ambiguous risk situations and switching (Bäckman et al., 2010; Li et al., 2010; Volkow et al.,
to ones with known risks that elicit less risk taking with age. 1998). Yet, consistent with conservation of gist-based intuition, older
Research on cognition has shown that people mentally represent adults’ risky decision-making remains largely indistinguishable from
information about decision options in two ways: verbatim representa- that of younger adults when verbatim memory is not required (Mata
tions of details, which are precise enough to support analytical et al., 2011; Samanez-Larkin and Knutson, 2014). Although adults are
thinking, and gist representations, which are less detailed (i.e., fuzzy) able to make good, and perhaps even better decisions than adolescents,
and support impressionistic, parallel, and typically unconscious they rely on their accumulated experience to counterbalance the de-
thinking (similar to characterizations of intuition; Reyna, 2012). The clines in executive function that they once possessed in late adolescence
preference to rely on gist grows with experience, and, for risk and and early adulthood.
probability, the simplest gist is categorical, for example, the categorical From the perspective presented here, experience making risky de-
distinction between some risk or no risk (e.g., Reyna et al., 2014; Reyna cisions during adolescence, as executive functions develop, fosters in-
and Ellis, 1994). As adolescents age, it would be expected that they creased development of gist-based reasoning. This experience is espe-
would also increasingly rely on gist-based reasoning when confronted cially critical because it allows adults to avoid unhealthy risks using
with potentially maladaptive risk taking. The growth of reliance on cognitive capacities (i.e., gist memory) that are preserved over a life-
more abstract gist memories from childhood to adulthood, as predicted time and that are robust in stressful or emotional situations (e.g., Reyna
by fuzzy-trace theory, has been replicated in 53 out of 55 studies on and Brainerd, 2011; see Reyna, 2011, for estimates of verbatim and gist
gist-based “false” memories (Reyna, 2011). memory, as well as cognitive control, across the lifespan). The growth
Consistent with a monotonic decline in risk taking with un- in this ability reflects increasing wisdom, defined as the accumulation
ambiguous risks, there is evidence that adolescents with better execu- of gist-based insight and expert knowledge about the conduct and
tive function perform better on such tasks (Brand et al., 2006; Khurana management of life challenges (Baltes and Smith, 2008; Baltes and
et al., 2015a; Shamosh et al., 2008). This evidence suggests that the Staudinger, 2000; Sternberg, 2001).
decline that occurs with development can be attributed in part to in- From a neurodevelopmental perspective, wisdom most likely in-
creasing ability to store and compare outcomes of risky decisions. Such volves the maturation (including pruning) and interconnection of sev-
ability would also lead to better integration of experience when con- eral brain regions that enable the individual to harness experience in an
fronting risky situations, including reliance on gist-based memories. As adaptive fashion (Meeks and Jeste, 2009; Reyna and Huettel, 2014).
a result, preference for maladaptive risk taking in specific domains These include the executive control and limbic systems. The default
would be expected to decline as experience accumulates and to do so mode network including medial PFC plays an important role by facil-
more rapidly for youth with better executive function. itating self-referential processing, empathy, theory of mind, and future
The meta-analysis by Defoe et al. (2014) (see also Tymula et al., projection (Buckner et al., 2008; Meeks and Jeste, 2009). As noted, this
2012) contrasted the predictions of fuzzy-trace theory versus imbalance system exhibits increasing intra-connectivity during adolescence (Fair
theories in laboratory tests of risk taking. The finding that risk taking et al., 2008; Sherman et al., 2014; Supekar et al., 2010). Nevertheless, it
declines with age, especially when a certain option is available, is not is the integrated functionality between systems across development that
anticipated by imbalance theories. The presence of a certain versus distinguishes wisdom from a simple top-down impulse control system

27
D. Romer, et al. Developmental Cognitive Neuroscience 27 (2017) 19–34

Fig. 6. Hypothesized trajectories of the Life Span

Wisdom Model of cognitive control, exploration, and
experience. Y axis scale is arbitrary.

(Reyna et al., 2015b). of such error feedback as they explore the environment. Although the
We summarize the changes that occur relevant to adaptive decision authors interpreted the finding as consistent with the hypothesis that
making over the lifespan in Fig. 6. This model differs from imbalance adolescents engage in riskier behavior than younger or older persons,
models in several respects but most importantly by including a third the task did not involve risky decision-making and thus was also con-
trajectory representing the accumulation of experience and hence sistent with the conclusion that adolescents are disproportionately
wisdom. Executive function displays an inverted U-shape function that primed to take advantage of positive feedback in a learning situation
peaks in late adolescence and early adulthood (Lachman et al., 2014; (see also Davidow et al., 2016; Satterthwaite et al., 2012).
Lindenberger, 2014; Williams et al., 1999). At the same time as ex- Murty et al. (2016) recently proposed a model of experience-based
ecutive function is improving during adolescence, the rise in sensation brain development termed the Experience-Driven Adaptive Cognitive
seeking and related dopamine expression drives exploration of the en- Model of adolescence that highlights the role of dopamine activation
vironment which peaks earlier than executive function but subsides during adolescence as a modulator of enhanced memory-circuit in-
during later years (Mata et al., 2016). However, as we describe below, tegration between the hippocampus (HPC) and PFC. They review evi-
as a result of these two processes, the brain builds networks of ex- dence indicating the importance of this process for building long-term
perience that foster greater ability to make adaptive decisions in later memory representations that enable the use of experience to further
adulthood despite the decline in executive function (Richards and more adaptive decision making. In particular, abundant evidence from
Hatch, 2011; Webster et al., 2013). Thus, the rise in exploration that studies in animals suggests that dopamine release from neurons in the
characterizes the adolescent brain serves an adaptive purpose of midbrain plays an important role in the coding of reward prediction
building robust representations of experience. errors and uncertainty (Fiorillo et al., 2003; Tobler et al., 2005). In
The model also recognizes that the late adolescent and young adult humans, such signals play an important role in episodic memory for-
brain is still a work in progress during the period when exploration and mation (Shohamy and Adcock, 2010), and tonic levels of midbrain
wisdom are growing despite having reached the peak in cognitive dopamine activation may encourage exploration and acquisition of
control. Thus, late adolescents and young adults will still be exploring long-term memories that support learning and adaptation (Düzel et al.,
their world during this period and taking risks many of which can be 2010). As Murty et al. (2016) say, “…different lines of research provide
adaptive. For those with especially weak cognitive control however, compelling support for adolescence being a unique period of plasticity
this period may produce particularly unhealthy consequences, such as and refinement of HPC-PFC circuits for the establishment of con-
addiction and unintentional injuries, many of which were fore- textually-relevant responses to guide and optimize goal-oriented be-
shadowed by earlier impulsive behavior. haviors.” (p. 54). Their model is consistent with the suggestion that the
exploratory behavior motivated by dopaminergic activation during
2.2. The adaptive adolescent brain adolescence serves adaptive purposes.
A study of adolescent decision-making in the presence of reward
Although heightened sensation seeking makes novel and potentially reversal also supports the adaptive character of the adolescent brain.
risky behavior more common during adolescence, this risk taking may When confronted with changing reward contingencies, adolescents
be motivated by a “rational calculus” (Reyna and Farley, 2006) that exhibited heightened activation of insular cortex, which was associated
may be adaptive for learning that underlies brain maturation (see also with more rapid reversal learning (Hauser et al., 2015). Young adults
Ellis et al., 2012; Luciana, 2016; Telzer, 2015). A study by Cohen et al. were slower to respond to the changes in contingencies. Recognizing
(2010) illustrates the adaptive character of the adolescent brain. In this such changes in contingencies is evidence of engaged executive func-
study, adolescent participants (ages 14–19) showed a stronger dopa- tion. These results suggest a possible mitigating factor (that adolescent
minergic brain response to reward prediction errors when engaging in a brains are quick to recognize changes in reward contingencies), off-
learning task than either younger children (ages 8–12) or adults (ages setting to some degree their heightened sensation seeking or attraction
25–30) (see also Galván et al., 2006, discussed above). Prediction error to novel experience.
is considered important in motivating learning (Pagnoni et al., 2002; Interestingly, youth with higher sensation seeking exhibit less im-
Schultz et al., 1997) and suggests that adolescents can take advantage pulsive choice as they age. In a national sample of adolescents and

28
D. Romer, et al. Developmental Cognitive Neuroscience 27 (2017) 19–34

young adults, Romer et al. (2010) found that high sensation seekers, lower-level (verbatim) representations that encourage risk taking to
who were more likely to engage in risky behavior than low sensation more abstract (gist) representations that support healthier decisions to
seekers, also exhibited higher levels of the ability to delay gratification categorically avoid catastrophic risks (but to take risks when they offer
as they aged, an important indicator of reduced impulsivity and cog- the possibility of a categorically superior outcome relative to less risky
nitive control (Casey et al., 2011). Indeed, they reached higher levels of options). In this regard, the theory has already successfully predicted
patience than youth who were lower in sensation seeking. Thus, ex- self-reported real-world risky behaviors using gist measures (e.g.,
perience gained during risk taking can lead to more adaptive decision- Broniatowski et al., 2015; Fraenkel et al., 2012; Mills et al., 2008;
making over the long term, especially among those with sufficient Reyna et al., 2011; Reyna and Mills, 2014; Wolfe et al., 2015).
cognitive skills, such as WM, to integrate their experience for future Another promising direction for future research is to examine the
behavior. relation between executive functions such as WM and the decline in
Risk taking during adolescence has been described as normative. maladaptive risk taking with age. As the consequences of exploratory
For example, Baumrind (1987) argued that “risk taking behavior risk taking accumulate in experience, those with stronger WM should be
characterizes normal adolescent development.” (p. 98) Furthermore, able to incorporate those experiences more effectively in decisions en-
“…some experimentation – be it with drugs or sex or odd diets or new tailing maladaptive risk. Preliminary evidence for this prediction has
ideas – is typical, and may even be an essential component, of a been observed in a study of late adolescent risk for drug addiction.
healthful adolescent experience and contribute to optimal compe- Those with stronger WM ability were more able to avoid advancing to
tence.” (p. 98) Some studies have shown that adolescents who experi- drug dependence apart from differences in impulsive tendencies
mented with drugs were more socially accepted by peers (Lightfoot, (Khurana et al., 2017).
1997; Maggs et al., 1995) and exhibited better adjustment than those Our model also suggests that we look at risk taking more broadly
who completely abstained from drug use (Shedler and Block, 1990). than just examining behaviors with adverse consequences. For example,
Chassin et al. (1989) observed that youth higher in sensation seeking Romer et al. (2016) showed that both sensation seeking and parts of the
engaged in what they called “constructive” risk taking, characterized by BAS were related to risk behaviors that are considered adaptive, such as
desire for independence and academic success, whereas “destructive” entering scholastic competitions and engaging in sports (see also
risk takers were characterized by impulsivity and antisocial tendencies. Hansen and Breivik, 2001). Many of the risky behaviors that adoles-
In a longitudinal study across grades 5–10, Lewis-Bizan et al. (2010) cents pursue involve potential social conflicts with parents or peers
observed that youth who were characterized as possessing positive (Weber et al., 2002), and these and other forms of risk behavior are also
developmental attributes, such as competent control over behavior, likely to increase during adolescence and should be considered in our
were also likely to engage in risky behavior. However, their risk taking models.
did not continue at high levels later in adolescence. We have said little about sex differences, but as is evident in Fig. 2,
In some statements of imbalance models (e.g., Casey and Caudle, there are gender differences in sensation seeking (Cross et al., 2011),
2013; Spear, 2013), the importance of individual differences in ado- which will have implications for different types of risk taking during
lescent risk taking is acknowledged. Nevertheless, the lower ability of adolescence. The correlation between sensation seeking and impulsive
the adolescent to control socioemotional decisions continues to be cited action is consistent with a smaller but established sex difference in
as a common deficit in adolescent brain function. For example, studies measures of impulsive action (Cross et al., 2011), corresponding to the
using driving simulation tasks by Steinberg and colleagues (e.g., Chein risk insensitive trajectory in our model in Fig. 5. This trajectory helps to
et al., 2011) are interpreted to show that adolescents’ brains respond explain the well-established over-representation of males in ex-
impulsively to the presence of their peers (Steinberg, 2014), whereas ternalizing behavior, a pattern that begins early in development among
adults are less susceptible to these influences. Although peer effects youth with weak cognitive control (Bjork and Pardini, 2015; McGue
may be stronger in adolescents, the direction of such effects appears to and Iacono, 2005; Moffitt et al., 2011). On the other hand, the small
depend on the characteristics of those peers. Simply placing adolescents relation between sensation seeking and decisions under known risk is
behind the wheel with peers in the vehicle does not necessarily produce consistent with the lack of sex differences in decisions under known risk
riskier driving (see Romer et al., 2014, for a review). In particular, (Cross et al., 2011). Nevertheless, the differences in sensation seeking
greater risk taking in the presence of peers is consistent with a group would suggest that females are less inclined to engage in exploratory
polarization effect of peer influence, such that when drivers think peer risk taking. However, many of the rewarding aspects of such behaviors
passengers expect them to drive aggressively, they are more likely to do are likely to be domain specific, such that young women may engage in
so. However, when peers are not expected to hold these preferences, exploration if they perceive the rewards to be sufficiently strong
adolescent drivers are no more likely to drive in a risky manner (Romer and Hennessy, 2007; Santos et al., 2016), for example in social
(Bingham et al., 2016; Simons-Morton et al., 2014). domains (Weber et al., 2002). Future research should examine this
It is likely therefore that youth with relatively good executive possibility as well.
control and peer groups with similar characteristics will be able to Finally, much remains to be learned about the organization of RSNs
experiment with risky behavior without advancing to more serious during the transition to adulthood. It is already known that youth with
outcomes (Lightfoot, 1997). However, as our analysis suggests, some ADHD have weaker ability to suppress the default mode network
youth will experience premature pregnancy, substance use, and other (DMN) than normally developing youth (Kessler et al., 2016; Posner
maladaptive behaviors that adversely affect educational attainment, et al., 2014). This is evident in stronger connectivity between the DMN
health, and other social outcomes (Institute of Medicine, 2011). and task-positive networks in youth with ADHD. Youth with ex-
ternalizing disorder and elevated levels of impulsive action exhibit the
3. Beyond imbalance during adolescence same pattern (Inuggi et al., 2014; Kessler et al., 2014; Shannon et al.,
2011). Future research could identify the neural basis of this deficit and
Despite the valuable insights spurred by imbalance models, it time explore potential interventions that could reduce it (Kelly and
to move beyond these models to consider the role that experience plays Castellanos, 2014; Stevens, 2016). These leads could be followed to
in healthy adolescent development. One potentially fruitful direction in determine the neural basis of harmful forms of impulsivity as opposed
future research would be to compare measures of gist learning and to exploratory forms of risk taking that emerge during adolescence.
decision making to measures that capture the development of wisdom Research regarding the functional roles of RSNs as they respond to
(Sunstein, 2008; see also, Reyna, 2008; Reyna and Huettel, 2014). Such growth in experience and wisdom during the adolescent period would
a direct comparison would test Reyna and Brainerd’s (2011) fuzzy-trace appear to be a fruitful avenue of future research.
theory, which predicts that decision-making shifts from relying on As more is learned about the growth of wisdom over the lifespan, it

29
D. Romer, et al. Developmental Cognitive Neuroscience 27 (2017) 19–34

Table 1
Differences between Imbalance Models and Lifespan Wisdom Model.

Imbalance Model Life-span Wisdom Model

Slower development of PFC and its connection with limbic system results in imbalance Cognitive control and dopaminergic activation rise in tandem during adolescence;
that outweighs cognitive control over impulsive urges during adolescence (Fig. 1). much of adolescent risk taking is exploratory in keeping with the role of dopamine as a
signal for novel reward (Fig. 6).
Rise in risk taking and incidence of health compromising behavior during adolescence Risk taking takes at least three forms, with different developmental trajectories
reflects developmental imbalance. (Fig. 5). The form most closely associated with imbalance reflects insensitivity to risk
and applies primarily to youth with early elevated levels of impulsive behavior.
Peak in sensation seeking during adolescence produces more risk taking than in children Peak in sensation seeking during adolescence motivates greater exploration in
or adults. ambiguous environments, but risk taking declines monotonically from childhood to
adulthood when risks are known, per greater reliance on gist and increasing executive
function (Fig. 5).
Imbalance leads to increased injury and maladaptive outcomes during adolescence. Timing of many maladaptive outcomes occurs in early adulthood when imbalance
should be minimal; maladaptive outcomes are more related to high levels of
impulsivity combined with risk opportunity and inexperience than to developmental
imbalance.
Socioemotional influences excite the dopaminergic system and promote risk taking. Socioemotional influences can promote risk taking, but social experience (interacting
with peers) and positive social influences can promote healthy risk avoidance.
Main emphasis on brain maturation, rather than experience or interventions that can Acknowledges brain maturation that reflects growth in experience and potential
promote adaptive brain development. No predictions about life-span cognitive interventions to promote healthy decision making by increasing reliance on experience
control or increase in wisdom. and wisdom.

is also important not to overplay the wisdom of adulthood. Just as Bäckman, L., Ginovart, N., Dixon, R.A., Wahlin, T.R., Wahlin, Ä., Halldin, C., Farde, L.,
2000. Age-related cogntive deficits mediated by changes in the striatal dopamine
stereotypes regarding adolescence have colored our interpretation of system. Am. J. Psychiatry 157 (4), 635–637.
brain research, it is just as easy to romanticize the experience and Bäckman, L., Nyberg, L., Lindenberger, U., Li, S.-C., Farde, L., 2006. The correlative triad
wisdom of adulthood. Research shows that relying on gist can lead to among aging, dopamine, and cognition: current status and future prospects. Neurosci.
Biobehav. Rev. 30, 791–807. http://dx.doi.org/10.1016/j.neubiorev.2006.06.005.
predictable biases even in experts (see Wilhelms et al., 2015). The in- Bäckman, L., Lindenberger, U., Li, S.-C., Nyberg, L., 2010. Linking cognitive aging to
creasing aversion to risk in ambiguous contexts may also lead to less alterations in dopamine neurotransmitter functioning: recent data and future ave-
than optimal search tendencies (Tymula et al., 2012). A good deal of nues. Neurosci. Biobehav. Rev. 34, 670–677.
Bahr, S.J., Hoffmann, J.P., Yang, X., 2005. Parental and peer influences on the risk of
research in decision making over the past several decades reveals how
adolescent drug use. J. Prim. Prev. 26 (6), 529–551. http://dx.doi.org/10.1007/
heuristics and biases common in adults can produce fallacies in judg- s10935-005-0014-8.
ment (Kahneman, 2013; Stanovich, 2011). This classic research serves Baltes, P.B., Smith, J., 2008. The fascination of wisdom: its nature, ontogeny, and func-
tion. Perspect. Psychol. Sci. 3 (1), 56–64.
as the foundation of more recent approaches, such as fuzzy-trace
Baltes, P.B., Staudinger, U.M., 2000. Wisdom: a metaheuristic (pragmatic) to orchestrate
theory, that account for fallacies in adulthood but also explain the the mind and virtue toward excellence. Am. Psychol. 55 (1), 122–136.
strengths of mature decision making (Defoe et al., 2014; Reyna et al., Baum, G.L., Ciric, R., Roalf, D.R., Betzel, R.F., Moore, T.M., Shinohara, R.T., et al., 2017.
2014). Modular segregation of structural brain networks supports the development of ex-
cutive function in youth. Curr. Biol. 27, 1561–1572. http://dx.doi.org/10.1016/j.
In conclusion, we have presented an alternative model of adolescent cub.2017.04.051.
brain development that emphasizes the accumulation of experience as Baumrind, D., 1987. A developmental perspective on adolescent risk taking in con-
adolescents age and transition to adulthood, with concomitant changes temporary America. In: Irwin, C.E. (Ed.), Adolescent Social Behavior and Health.
Jossey-Bass, San Francisco, pp. 93–125.
in judgment and decision making (see Table 1 for a summary of dif- Belin, D., Mar, A.C., Dalley, J.W., Robbins, T.W., Everitt, 2008. High impulsivity predicts
ferences between the Life-span Wisdom Model and Imbalance Models). the switch to compulsive cocaine-taking. Science 320 (5881), 1352–1355.
The model explains much of the apparent increase in adolescent risk Berns, G.S., Moore, S., Capra, C.M., 2009. Adolescent engagement in dangerous behaviors
is associated with increased white matter maturity of frontal cortex. PLoS One 4 (8),
taking as an adaptive need to gain the experience required to assume 1–12. http://dx.doi.org/10.1371/journal.pone.0006773.
adult roles and behaviors. The risk-taking that reflects lack of control or Betzel, R.F., Byrge, L., He, Y., Goñi, J., Zuo, X.N., Sporns, O., 2014. Changes in structural
excessive sensitivity to immediate rewards is primarily an individual and functional connectivity among resting-state networks across the human lifespan.
Proc. Natl. Acad. Sci. 111, E4997–E5006.
difference that characterizes some persons from an early age that can Biglan, A., Cody, C., 2003. Preventing multiple problem behaviors in adolescence. In:
persist well into adulthood. At the same time, the adolescent brain is Romer, D. (Ed.), Reducing Adolescent Risk: Toward an Integrated Approach. Sage
supremely sensitive to the learning that can occur during this period Publications, Thousand Oaks, CA, pp. 125–131.
Bingham, C.R., Simons-Morton, B.G., Pradhan, A.K., Li, K., Almani, F., Falk, E.B., et al.,
and has cognitive capacities to take advantage of the experience gained.
2016. Peer passenger norms and pressure: experimental effects on simulated driving
The result is a brain with integrated circuits encompassing executive among teenage males. Transp. Res. Part F 41, 124–137.
function (i.e., cognitive control and inhibition), as well as verbatim and Bjork, J.M., Pardini, D.A., 2015. Who are those risk-taking adolescents? Individual dif-
gist memory networks, which can be called upon to negotiate both ferences in developmental neuroimaging research. Dev. Cognit. Neurosci. 11, 56–64.
http://dx.doi.org/10.1016/j.dcn.2014.07.008.
novel and familiar situations. The preservation of robust gist thinking Brainerd, C.J., Reyna, V.F., Howe, M.L., 2009. Trichotomous processes in early memory
maintains wise decision making during later adulthood when cognitive development, aging, and neurocognitive impairment: a unified theory. Psychol. Rev.
control capacities diminish. We believe this approach is more aligned 116 (4), 783–832. http://dx.doi.org/10.1037/a0016963.
Brand, M., Labudda, K., Markowitsch, H.J., 2006. Neuropsychological correlates of de-
with the scientific evidence, including results that challenge stereotypes cision-making in ambiguous and risky situations. Neural Netw. 19, 1266–1276.
about the adolescent brain. http://dx.doi.org/10.1016/j.neunet.2006.03.001.
Brito, N.H., Noble, K.G., 2014. Socioeconomic status and structural brain development.
Front. Neurosci. 8 (276), 1–12. http://dx.doi.org/10.3389/fnins.2014.00276.
Conflict of Interest Bromberg-Martin, E.S., Matsumoto, M., Hikosaka, O., 2010. Dopamine in motivational
control: rewarding, aversive, and alerting. Neuron 68, 815–834. http://dx.doi.org/
None. 10.1016/j.neuron.2010.11.022.
Broniatowski, D.A., Klein, E.Y., Reyna, V.F., 2015. Germs are germs, and why not take a
risk?: patients’ explanations for prescribing antibiotics in an inner city emergency
References department. Med. Decis. Mak. 35, 60–67. http://dx.doi.org/10.1177/
0272989X14553472.
Buckner, R.L., Andrews-Hanna, J.R., Schacter, D.L., 2008. The brain’s default network:
Arnsten, A.F.T., Wang, M.J., Paspalas, C.D., 2012. Neuromodulation of thought: flex-
anatomy, function: and relevance to disease. Ann. N. Y. Acad. Sci. 1124, 1–38.
ibilities and vulnerabilities in prefrontal cortical network synapses. Neuron 76,
Carver, C.S., White, T.L., 1994. Behavioral inhibition, behavioral activation, and affective
223–239. http://dx.doi.org/10.1016/j.neuron.2012.08.038.

30
D. Romer, et al. Developmental Cognitive Neuroscience 27 (2017) 19–34

responses to impending reward and reward: the BIS/BAS Scales. J. Pers. Soc. Psychol. decision making in human aging. Ann. N. Y. Acad. Sci. 1235, 1–17. http://dx.doi.org/
67, 319–333. 10.1111/j.1749-6632.2011.06230.x.
Casey, B.J., Caudle, K., 2013. The teenage brain: self control. Curr. Dir. Psychol. Sci. 22 Ernst, M., Fudge, J.L., 2009. A developmental neurobiological model of motivated be-
(2), 82–87. http://dx.doi.org/10.1177/0963721413480170. havior: anatomy, connectivity and ontogeny of the triadic nodes. Neurosci. Biobehav.
Casey, B.J., Getz, S., Galvan, A., 2008. The adolescent brain. Dev. Rev. 28 (1), 62–77. Rev. 33 (3), 367–382.
Casey, B.J., Somerville, L.H., Gotlib, I.H., Ayduk, O., Franklin, N.T., Askren, M.K., Shoda, Ernst, M., 2014. The triadic model perspective for the study of adolescent motivated
Y., 2011. Behavioral and neural correlates of delay of gratification 40 years later. behavior. Brain Cognit. 89, 104–111. http://dx.doi.org/10.1016/j.bandc.2014.01.
Proc. Natl. Acad. Sci. 108, 14998–15003. 006.
Casey, B.J., 2015. Beyond simple models of self-control to circuit-based accounts of Faggiano, F., Vigna-Taglinati, F.D., Versino, E., Zambon, A., Borraccino, A., Lemma, P.,
adolescent behavior. In: 66th ed. In: Fiske, S.T., Schacter, D.L., Taylor, S.E. (Eds.), et al., 2008. School-based prevention for illicit drugs use: a systematic review. Drug
Annual Review of Psychology, vol. 66. Annual Reviews, Palo Alto, CA, pp. 295–319 Alcohol Depend. 46, 385–396. http://dx.doi.org/10.1016/j.drugalcdep.2009.11.018.
Retrieved from 10.1146/annurev-psych-010814-015116. Fair, D.A., Cohen, A.L., Dosenbach, N.U.F., Church, J.A., Miezin, F.M., Barch, D.M., et al.,
Centers for Disease Control and Prevention, 2014. 2013 Sexually Transmitted Disease 2008. The maturing architecture of the brain’s default network. Proc. Natl. Acad. Sci.
Surveillance. Atlanta, GA. Retrieved from http://www.cdc.gov/std/stats13/adol. 105 (10), 4028–4032. http://dx.doi.org/10.1073/pnas.0800376105.
htm. Fair, D.A., Nigg, J.T., Dosenbach, N.U.F., Schlaggar, B.L., Mennes, M., Gutman, D., et al.,
Chambers, R.A., Taylor, J.R., Potenza, M.N., 2003. Developmental neurocircuitry of 2013. Distinct neural signatures detected for ADHD subtypes after controlling for
motivation in adolescence: a critical period of addiction vulnerability. Am. J. micro-movements in resting state functional connectivity MRI data. Front. Syst.
Psychiatry 160 (6), 1041–1052. Neurosci. 6 (80), 1–31. http://dx.doi.org/10.3399/fnsys.2012.00080.
Chan, M.Y., Park, D.C., Savalia, N.K., Wig, G.S., 2014. Decreased segregation of brain Fareri, D.S., Gabard-Durnam, L., Goff, B., Flannery, J., Gee, D.G., Lumian, D.S., et al.,
systems across the healthy adult lifespan. Proc. Natl. Acad. Sci. E4997–E5006. http:// 2015. Normative development of ventral striatal resting state connectivity in humans.
dx.doi.org/10.1073/pnas.1415122111. Neuroimage 118, 422–437. http://dx.doi.org/10.1016/j.neuroimage.2015.06.022.
Chassin, L., Presson, C.C., Sherman, S.J., 1989. Constructive vs destructive deviance in Fiorillo, C.D., Tobler, P.N., Schultz, W., 2003. Discrete coding of reward probability and
adolescent health-related behaviors. J. Youth Adolesc. 18 (3), 245–262. uncertainty by dopamine neurons. Science 299, 1898–1902.
Chein, J., Albert, D., O’Brien, L., Uckert, K., Steinberg, L., 2011. Peers increase adolescent Fraenkel, L., Peters, E., Charpentier, P., Olsen, B., Errante, L., Schoen, R., Reyna, V.F.,
risk taking by enhancing activity in the brain’s reward circuitry. Dev. Sci. 14 (2), 2012. A decision tool to improve the quality of care in rheumatoid arthritis. Arthritis
FF1–FF10. http://dx.doi.org/10.1111/j.1467-7687.2010.01035.x. Care Res. 64 (7), 977–985. http://dx.doi.org/10.1002/acr.21657.
Choudhury, S., 2010. Culturing the adolescent brain: what can neuroscience learn from Fuster, J.M., 2009. Cortex and memory: emergence of a new paradigm. J. Cognit.
anthropology. Soc. Cognit. Affect. Neurosci. 5, 159–167. Neurosci. 21 (11), 2047–2072. http://dx.doi.org/10.1162/jocn.2009.21280.
Christakou, A., Brammer, M., Rubia, K., 2011. Maturation of limbic corticostriatal acti- Fuster, J.M., 2013. The Neuroscience of Freedom and Creativity. Cambridge University
vation and connectivity associated with developmental changes in temporal dis- Press, New York.
counting. Neuroimage 54, 1344–1354. http://dx.doi.org/10.1016/j.neuroimage. Galván, A., Hare, T.A., Parra, C.E., Penn, J., Voss, H., Glover, G., Casey, B.J., 2006. Earlier
2010.08.067. development of the accumbens relative to orbitofrontal cortex might underlie risk-
Cohen, J.R., Asarnow, R.F., Sabb, F.W., Bilder, R.M., Bookheimer, S.Y., Knowlton, B.J., taking behavior in adolescents. J. Neurosci. 26 (25), 6885–6892. http://dx.doi.org/
Poldrack, R.A., 2010. A unique adolescent response to reward prediction errors. Nat. 10.1523/JNEUROSCI.1062-06.2006.
Neurosci. 13 (6), 669–671. http://dx.doi.org/10.1038/nn.2558. Galván, A., Hare, T.A., Voss, H., Glover, G., Casey, B.J., 2007. Risk taking and the ado-
Cools, R., Robbins, T.W., 2004. Chemistry of the adaptive mind. Philos. Trans. R. Soc. B lescent brain: who is at risk? Dev. Sci. 10 (2), F8–F14. http://dx.doi.org/10.1111/j.
362, 2871–2888. 1467-7687.2006.00579.x.
Cools, R., Gibbs, S.E., Miyakawa, A., Jagust, W., D’Esposito, M., 2008. Working memory Gerard, M., Gibbons, F.X., Houlihan, A.E., Stock, M.L., Pomery, E.A., 2008. A dual-process
capacity predicts dopamine synthesis capacity in the human striatum. J. Neurosci. 28 approach to health risk decision making: the prototype willingness model. Dev. Rev.
(5), 1208–1212. 28, 29–61. http://dx.doi.org/10.1016/j.dr.2007.10.001.
Crone, E.A., Dahl, R., 2012. Understanding adolescence as a period of social-affective Giedd, J.N., Blumenthal, J., Jeffries, N.O., Castellanos, F.X., Liu, H., Zijdenbos, A., et al.,
engagement and goal flexibility. Nat. Rev. Neurosci. 13, 636–650. http://dx.doi.org/ 1999. Brain development during childhood and adolescence: a longitudinal MRI
10.1038/nm3313. study. Nat. Neurosci. 2 (10), 861–863.
Cross, C.P., Copping, L.T., Campbell, A., 2011. Sex differences in impulsivity: a meta- Giedd, J.N., 2004. Structural magnetic resonance imaging of the adolescent brain. Ann. N.
analysis. Psychol. Bull. 137 (1), 97–130. http://dx.doi.org/10.1037/a0021591. Y. Acad. Sci. 1021, 77–85.
Cyders, M.A., Coskunpinar, A., 2011. Measurement of constructs using self-report and Goldberg, E., 2006. The Wisdom Paradox. Gotham Books, New York.
behavioral lab tasks: is there overlap in nomothetic span and construct representation Green, L., Fry, A.F., Myerson, J., 1994. Discounting of delayed rewards: a life-span
for impulsivity? Clin. Psychol. Rev. 31, 965–982. comparison. Psychol. Sci. 5 (1), 33–36.
Düzel, E., Bunzeck, N., Guitart-Masip, M., Düzel, S., 2010. NOvelty-related motivation of Gur, R.E., Richard, J., Calkins, M.E., Chiavacci, R., Hansen, J.A., Bilker, W.B., et al., 2012.
anticipation and exploration by dopamine (NOMAD): implications for healthy aging. Age group and sex differences in performance on a computerized neurocognitive
Neurosci. Biobehav. Rev. 34, 660–669. http://dx.doi.org/10.1016/j.neubiorev.2009. battery in children age 8–21. Neuropsychology 26, 251–265.
08.006. Haber, S.N., Knutson, B., 2010. The reward circuit: primate anatomy and human imaging.
D’Esposito, M., Postle, B.R., 2015. The cognitive neuroscience of working memory. Annu. Neuropsychopharmacology 35, 4–26. http://dx.doi.org/10.1038/npp.2009.129.
Rev. Psychol. 66, 115–142. http://dx.doi.org/10.1146/annualrev-psych-010814- Hansen, E.B., Breivik, G., 2001. Sensation seeking as a predictor of positive and negative
015031. risk behavior among adolescents. Pers. Individ. Differences 30, 627–640.
Dahl, R., 2004. Adolescent brain development: a period of vulnerabilities and opportu- Hanson, J.L., Chandra, A., Wolfe, B.L., Pollak, S.D., 2011. Association between income
nities. Ann. N. Y. Acad. Sci. 1021, 251–279. and the hippocampus. PLoS One 6 (5), e18712. http://dx.doi.org/10.1371/journal.
Davidow, J.Y., Foerde, K., Galván, A., Shohamy, D., 2016. An upside to reward sensi- pone.0018712.
tivity: the hippocampus supports enhanced reinforcement learning in adolescence. Hare, T.A., Tottenham, N., Galvan, A., Voss, H., Glover, G., Casey, B.J., 2008. Biological
Neuron 92, 93–99. http://dx.doi.org/10.1016/j.neuron.2016.08.031. substrates of emotional reactivity and regulation in adolescence during an emotional
Daw, N.D., O’Doherty, J.P., Dayan, P., Seymour, B., Dolan, R.J., 2006. Cortical substrates go-nogo task. Biol. Psychiatry 63 (10), 927–934.
for exploratory decisions in humans. Nature 441 (15), 876–879. http://dx.doi.org/ Hauser, T.U., Iannaccone, R., Walitza, S., Brandeis, D., Brem, S., 2015. Cognitive flex-
10.1038/nature04766. ibility in adolescence: neural and behavioral mechanisms of reward prediction error
Defoe, I.N., Dubas, J.J.S., Figner, B., van Aken, M.A.G., 2014. A meta-analysis on age processing in adaptive decision making during development. Neuroimage 104,
differences in risky decision making: adolescents versus children and adults. Pychol. 347–354. http://dx.doi.org/10.1016/j.neuroimage.2014.09.018.
Bull. 141 (1), 48–84. http://dx.doi.org/10.1037/a0038088. Hueston, C.M., Cryan, J.F., Nolan, Y.M., 2017. Stress and adolescent hippocampal neu-
Dennison, M., Whittle, S., Yucel, M., Vijayakumar, N., Kline, A., Simmons, J., Allen, N.B., rogenesis: diet and exercise as cognitive modulators. Transl. Psychiatry 7, e1081.
2013. Mapping subcortical brain maturation during adolescence: evidence of hemi- http://dx.doi.org/10.1038/tp.2017.48.
sphere- and sex-specific longitudinal changes. Dev. Sci. 16 (5), 772–791. http://dx. Iacono, W.G., Malone, S.M., McGue, M., 2008. Behavioral disinhibition and the devel-
doi.org/10.1111/desc.12057. opment of early-onset addiction: common and specific influences. Annu. Rev. Clin.
Dosenbach, N.U.F., Nardos, B., Cohen, A.J., Fair, D.A., Power, J.D., Church, J.A., et al., Psychol. 4, 325–348. http://dx.doi.org/10.1146/annurev.clinpsy.4.022007.141157.
2010. Prediction of individual brain maturity using fMRI. Science 329, 1358–1361. Ikemoto, S., 2007. Dopamine reward circuitry: two projection systems from the ventral
http://dx.doi.org/10.1126/science.1194144. midbrain to the nucleus accumbens-olfactory tubercle complex. Brain Res. Rev. 56,
Duell, N., Steinberg, L., Chein, J., Al-Hassam, S.M., Bacchini, D., Lei, C., et al., 2016. 27–78.
Interaction of reward seeking and self-regulation in the prediction of risk taking: a Institute of Medicine and National Research Council, 2011. The Science of Adolescent
cross-national test of the dual systems model. Dev. Psychol. 52 (10), 1593–1605. Risk-Taking: Workshop Report. The National Academies Press, Washington, DC.
http://dx.doi.org/10.1037/dev0000152. Johnston, L.D., Miech, P.M., Bachman, J.G., Schulenberg, J.E., 2015a. Monitoring the
Elder, R.W., Shults, R.A., Sleet, D.A., Nichols, J.L., Thompson, R.S., Rajab, W., Task force Future National Results on Adolescent Drug Use: Overview of Key Findings, 2014.
on Community Preventive Services, 2004. Effectiveness of mass media campaigns for Institute for Social Research, University of Michigan, Ann Arbor, MI.
reducing drinking and driving and alcohol-involved crashes: A systematic review. Johnston, L.D., O’Malley, P.M., Bachman, J.G., Schulenberg, J.E., Miech, R.A., 2015b.
Am. J. Prev. Med. 27 (1), 57–65. http://dx.doi.org/10.1016/j.amepre.2004.03.002. Monitoring the Future National Survey Results on Drug Use, 1975–2014: Volume 2,
Ellis, B.J., Del Giudice, M., Dishion, T.J., Figueredo, A.J., Gray, P., Griskevicius, V., et al., College Students and Adults Ages 19–55. University of Michigan, Ann Arbor, MI.
2012. The evolutionary basis of risky adolescent behavior: implications for science, Jupp, B., Dalley, J.W., 2014. Behavioral endophenotypes of drug addiction: etiological
policy, and practice. Dev. Psychol. 48 (3), 598–623. http://dx.doi.org/10.1037/ insights from neuroimaging studies. Neuropharmacology 76, 487–497. http://dx.doi.
a0026220. org/10.1016/j.neuropharm.2013.05.041.
Eppinger, B., Hammerer, D., Li, S., 2011. Neuromodulation of reward-based learning and Kahneman, D., 2013. Thinking, Fast and Slow. Farrar, Straus and Giroux, New York.

31
D. Romer, et al. Developmental Cognitive Neuroscience 27 (2017) 19–34

Kanny, D., Brewer, R.D., Mesnick, J.B., Paulozzi, L.J., Naimi, T.S., Lu, H., 2015. Vital McGuire, J.T., Kable, J.W., 2013. Rational predictions can underlie apparent failures to
signs: alcohol poisoning deaths–United States, 2010–2012. MMWR 63 (53), delay gratification. Psychol. Rev. 120 (2), 395–410. http://dx.doi.org/10.1037/
1238–1241. a0031910.
Kasen, S., Cohen, P., Chen, H., 2011. Developmental course of impulsivity and capability Meeks, T.W., Jeste, D.V., 2009. Neurobiology of wisdom: a literature review. Arch. Gen.
from age 10 to age 25 as related to trajectory of suicide attempt in a community Psychiatry 66 (4), 355–365.
cohort. Suicide Life Threat. Behav. 41 (2), 180–191. http://dx.doi.org/10.1111/j. Meng, S.Z., Ozawa, Y., Itoh, M., Takashima, S., 1999. Developmental and age-related
1943-278x.2011.00017.x. changes of dopamine transporter, and dopamine D1 and D2 receptors in human basal
Kelly, C., Castellanos, F.X., 2014. Strengthening connections: functional connectivity and ganglia. Cereb. Cortex 843 (1–2), 136–144.
brain plasticity. Neuropsychol. Rev. 24, 63–76. http://dx.doi.org/10.1007/s11065- Metcalfe, J., Mischel, W., 1999. A hot/cool-system analysis of delay of gratification:
014-9252-y. dynamics of willpower. Psychol. Rev. 106 (1), 3–19.
Kessler, D., Angstadt, M., Welsh, R.C., Sripada, C., 2014. Modality-spanning deficits in Miller, E.K., Cohen, J.D., 2001. An integrative theory of prefrontal cortex function. Annu.
attention-deficit/hyperactivity disorder in functional networks, gray matter, and Rev. Neurosci. 24, 167–202.
white matter. J. Neurosci. 34 (50), 16555–16566. http://dx.doi.org/10.1523/ Mills, B.A., Reyna, V.F., Estrada, S.M., 2008. Explaining contradictory relations between
jneurosci.3156-14.2014. risk perception and risk taking. Psychol. Sci. 19 (5), 429–433. http://dx.doi.org/10.
Kessler, D., Angstadt, M., Sripada, C., 2016. Growth charting of brain connectivity net- 1111/j.1467-9280.2008.02104.x.
works and the identification of attention impairment in youth. JAMA Psychiatry 73 Mills, K.L., Goddings, A.-L., Clasen, L.S., Giedd, J.N., Blakemore, S.-J., 2014. The devel-
(5), 481–489. http://dx.doi.org/10.1001/jamapsychiatry.2016.0088. opmental mismatch in structural brain maturation during adolescence. Dev.
Khurana, A., Romer, D., Betancourt, L.M., Brodsky, N.L., Giannetta, J.M., Hurt, H., 2012. Neurosci. 36, 147–160. http://dx.doi.org/10.1159/000362328.
Early adolescent sexual debut: the mediating role of working memory ability, sen- Mischel, W., Shoda, Y., Rodriguez, M.I., 1989. Delay of gratification in children. Science
sation seeking and impulsivity. Dev. Psychol. 48 (5), 1416–1428. http://dx.doi.org/ 244 (4907), 933–938. http://dx.doi.org/10.1126/science.2658056.
10.1037/a0027491. Moffitt, T.E., Arseneault, L., Belsky, D., Dickson, N., Hancox, R.J., Harrington, H., et al.,
Khurana, A., Romer, D., Betancourt, L.M., Brodsky, N.L., Giannetta, J.M., Hurt, H., 2015a. 2011. A gradient of self-control predicts health, wealth, and public safety. Proc. Natl.
Experimentation versus progression in adolescent drug use: a test of an emerging Acad. Sci. 108 (7), 2693–2698. http://dx.doi.org/10.1073/pnas.1010076108.
neurobehavioral imbalance model. Dev. Psychopathol. 27, 901–913. http://dx.doi. Murty, V.P., Calabro, F., Luna, B., 2016. The role of experience in adolescent cognitive
org/10.1017/S0954579414000765. development: integration of executive, memory, and mesolimbic systems. Neurosci.
Khurana, A., Romer, D., Betancourt, L.M., Brodsky, N.L., Giannetta, J.M., Hurt, H., 2015b. Biobehav. Rev. 70, 46–58. http://dx.doi.org/10.1016/j/neubiorev.2016.07.034.
Stronger working memory reduces sexual risk taking in adolescents, even after Naniex, F., Marchand, A.R., Di Scala, G., Pape, J.-R., Coutureau, E., 2012. Parallel ma-
controlling for parental influences. Child Dev. 86 (4), 1125–1141. http://dx.doi.org/ turation of goal-directed behavior and dopaminergic systems during adolescence. J.
10.1111/cdev.12383. Neurosci. 32 (46), 16223–16232. http://dx.doi.org/10.1523/jneurosci.3080-12.
Khurana, A., Romer, D., Betancourt, L.M., Hurt, H., 2017. Working memory ability and 2012.
early drug use progression as predictors of adolescent substance use disorders. Noble, K.G., Houston, S.W., Brito, N.H., Bartsch, H., Kan, E., Kuperman, J.M., et al., 2015.
Addiction 112, 1220–1228. http://dx.doi.org/10.1111/add.13792. Family income, parental education and brain structure in children and adolescents.
Kreuger, R.F., Hicks, B.M., Patrick, C.J., Carlson, S.R., Iacono, W.G., McGue, M., 2002. Nat. Neurosci. 18, 773–778. http://dx.doi.org/10.1038/nn.3983.
Etiological connections among substance dependence, antisocial behavior, and per- Pagnoni, G., Zink, C.F., Montague, P.R., 2002. Activity in human ventral striatum locked
sonality: modeling the externalizing spectrum. J. Abnorm. Psychol. 11 (3), 411–424. to errors of reward prediction. Nat. Neurosci. 5, 97–98.
Krishnadas, R., Kim, J., McLean, J., Batty, G.D., McLean, J.S., Millar, K., et al., 2013. The Patton, J.H., Stanford, M.S., Barratt, E.S., 1995. Factor structure of the Barratt impul-
envirome and the connectome: exploring the sturctural noise in the human brain siveness scale. J. Clin. Psychol. 51 (6), 768–774.
associated with socioeconomic deprivation. Front. Hum. Neurosci. 7, 1–14. http://dx. Paulsen, D., Platt, M.L., Huettal, S.A., Brannon, E.M., 2011. Decision-making under risk in
doi.org/10.3389/fnhum.2013.00722. children, adolescents, and young adults. Front. Psychol. 2 (72), 1–6. http://dx.doi.
Kwon, M.S., Vorobyev, V., Moe, D., Parkkola, R., Hamalainen, H., 2015. Brain structural org/10.3389/fpsyg.2011.00072.
correlates of risk-taking behavior and effects of peer influence in adolescents. PLoS Paulsen, D., Platt, M.L., Huettel, S.A., Brannon, E.M., 2012. From risk-seeking to risk-
One 9 (11). http://dx.doi.org/10.1371/journal.pone.0112780. averse: the development of economic risk preference from childhood to adulthood.
Lachman, M.E., Agrigoroaei, S., Tun, P.A., Weaver, S.L., 2014. Monitoring cognitive Front. Psychol. 3 (313), 1–6. http://dx.doi.org/10.3389/fpsyg.2012.00313.
functioning: psychometric properties of the brief test of adult cognition by telephone. Pfeifer, J.H., Allen, N.B., 2012. Arrested development? Reconsidering dual-system models
Assessment 2 (4), 404–417. http://dx.doi.org/10.1177/1073191113508807. of brain function in adolescence and disorders. Trends Cognit. Sci. 16 (6), 322–329.
Lebel, C., Camicioli, R., Weiler, W., Beaulieu, C., 2012. Diffusion tensor imaging of white http://dx.doi.org/10.1016/j.tics.2012.04.011.
matter tract evolution over the lifespan. Neuroimage 60, 340–352. http://dx.doi.org/ Piccolo, L.R., Merz, E.C., He, X., Sowell, E.R., Noble, K.G., PING Study, 2016. Age-related
10.1016/j.neuroimage.2011.11.094. differences in cortical thickness vary by socioeconomic status. PLoS One 119,
Levin, I.P., Hart, S.S., 2003. Risk preference in young children: early evidence of in- e0162511. http://dx.doi.org/10.1371/journal.pone.0162511.
dividual differences in reaction to potential gains and losses. J. Behav. Decis. Mak. Posner, J., Park, C., Wang, Z., 2014. Connecting the dots: a review of resting connectivity
16, 397–413. http://dx.doi.org/10.1002/bdm/453. MRI studies in attention-deficit/hyperactivity disorder. Neuropsychol. Rev. 24, 3–15.
Lewis-Bizan, S., Lynch, A.D., Fay, K., Schmid, K., McPherran, C., Lerner, J.V., Lerner, http://dx.doi.org/10.1007/s11065-014-9251-z.
R.M., 2010. Trajectories of positive and negative behaviors from early- to middle- Previc, F.H., 2009. The Dopaminergic Mind in Human Evolution and History. Cambridge
adolescence. J. Youth Adolesc. 39, 751–763. http://dx.doi.org/10.1007/s10964-010- University Press, New York.
9532-7. Raine, A., Reynolds, C., Venables, P.H., Mednick, S.A., 2002. Stimulation seeking and
Li, S.-C., Lindenberger, U., Bäckman, L., 2010. Dopaminergic modulation of cognition intelligence: a prospective longitudinal study. J. Pers. Soc. Psychol. 82 (4), 663–674.
across the life span. Neurosci. Biobehav. Rev. 34, 625–630. http://dx.doi.org/10. Raznahan, A., Shaw, P.W., Lerch, J., Clasen, L.S., Greenstein, D., Berman, R.M., et al.,
1016/j.neubiorev.2010.02.003. 2014. Longitudinal four-dimensional mapping of subcortical anatomy in human de-
Lightfoot, C., 1997. The Culture of Adolescent Risk-Taking. Guilford Press, New York. velopment. Proc. Natl. Acad. Sci. 111 (4), 1592–1597. http://dx.doi.org/10.1073/
Lindenberger, U., 2014. Human cognitive aging: corriger la fortune? Science 346 (6209), pnas.131691111.
572–578. http://dx.doi.org/10.1126/science.1254403. Reyna, V.F., Brainerd, C.J., 2011. Dual processes in decision making and developmental
Luna, B., Wright, C., 2015. Adolescent brain development: Implications for the juvenile neuroscience: a fuzzy-trace model. Dev. Rev. 31, 180–206. http://dx.doi.org/10.
justice system. In: Heilbrun, K. (Ed.), APA Handbook of Psychology and Juvenile 1016/j.dr.2011.07.004.
Justice. APA Publications, Washington, DC, pp. 91–116. Retrieved from http://dx. Reyna, V.F., Ellis, S.C., 1994. Fuzzy-trace theory and framing effects in children’s risky
doi.org/10.1037/14643-005. decision making. Psychol. Sci. 5 (5), 275–279.
Luna, B., Garver, K.E., Urban, T.M., Lazar, N.A., Sweeney, J.A., 2004. Maturation of Reyna, V.F., Farley, F., 2006. Risk and rationality in adolescent decision making: im-
cognitive processes from late childhood to adulthood. Child Dev. 75, 1357–1372. plications for theory, practice, and public policy. Psychol. Sci. Public Interest 7 (1),
Luna, B., Padmanabhan, A., O’Hearn, K., 2010. What has fMRI told us about the devel- 1–44. http://dx.doi.org/10.1111/j.1529-1006.2006.00026.x.
opment of cognitive control through adolescence? Brain Cognit. 72, 101–113. http:// Reyna, V.F., Huettel, S.A., 2014. Reward, representation, and impulsivity: a theoretical
dx.doi.org/10.1016/j.bandc.2009.08.005. framework for the neuroscience of risky decision making. The Neuroscience of Risky
Maggs, J.L., Almeida, D.M., Galambos, N.L., 1995. Risky business: the paradoxical Decision Making. American Psychological Association, Washington, DC, pp. 11–42.
meaning of problem behavior for young adolescents. J. Early Adolesc. 15 (3), Reyna, V.F., Lloyd, F.J., 2006. Physician decision-making and cardiac risk: effects of
344–362. knowledge, risk perception, risk tolerance, and fuzzy processing. J. Exp. Psychol. 12,
Magid, V., MacLean, M.G., Colder, C.G., 2007. Differentiating between sensation seeking 179–195. http://dx.doi.org/10.1037/1076-898X.12.3.179.
and impulsivity through their mediated relations with alcohol use and problems. Reyna, V.F., Mills, B.A., 2014. Theoretically motivated interventions for reducing sexual
Addict. Behav. 32, 2046–2061. risk taking in adolescence: a randomized controlled experiment using fuzzy-trace
Mata, R., Josef, A.K., Hertwig, R., 2016. Propensity for risk taking across the life span and theory. J. Exp. Psychol. 143 (4), 1627–1648. http://dx.doi.org/10.1037/a0036717.
around the globe. Psychol. Sci. 27 (2), 231–243. https://doi.org/http://journals. Reyna, V.F., Estrada, S.M., DeMarinis, J.A., Myers, R.M., Stanisz, J.M., Mills, B.A., 2011.
sagepub.com/doi/10.1177/0956797615617811. Neurobiological and memory models of risky decision making in adolescents versus
Matthews, M., Bondi, C., Torres, G., Moghaddam, B., 2013. Reduced presynaptic dopa- young adults. J. Exp. Psychol. 37 (5), 1125–1142. http://dx.doi.org/10.1037/
mine activity in adolescent dorsal striatum. Neuropsychopharmacology 38, a0023943.
1344–1351. http://dx.doi.org/10.1038/npp.2013.32. Reyna, V.F., Chick, C.F., Corbin, J.C., Hsia, A.N., 2014. Developmental reversals in risky
McClure, S.M., Laibson, D.I., Lowenstein, G., Cohen, J.D., 2004. Separate neural systems decision-making: intelligence agents show larger decision biases than college stu-
value immediate and delayed monetary rewards. Science 306, 503–507. dents. Psychol. Sci. 25, 75–84. http://dx.doi.org/10.1177/0956797613497022.
McGue, M., Iacono, W.G., 2005. The association of early adolescent problem behavior Reyna, V.F., Weldon, R.B., McCormick, M.J., 2015a. Educating intuition: reducing risky
with adult psychopathology. Am. J. Psychiatry 162, 1118–1124. decisions using fuzzy-trace theory. Curr. Direct. Psychol. Sci. 24 (5), 392–398. http://

32
D. Romer, et al. Developmental Cognitive Neuroscience 27 (2017) 19–34

dx.doi.org/10.1177/0963721415588081. Shulman, E.P., Harden, K.P., Chein, J.M., Steinberg, L., 2015. Sex differences in the de-
Reyna, V.F., Wilhelms, E.A., McCormick, M.J., Weldon, R.B., 2015b. Development of velopmental trajectories of impulse control and sensation-seeking from early ado-
risky decision making: fuzzy-trace theory and neurobiological perspectives. Child lescence to early adulthood. J. Youth Adolesc. 44, 1–17. http://dx.doi.org/10.1007/
Dev. Perspect. 9 (2), 122–127. http://dx.doi.org/10.1111/cdep.12117. s10964-014-0116-9.
Reyna, V.F., 2008. Theories of medical decision making and health: an evidence-based Shulman, E.P., Smith, A.R., Silva, K., Icenogle, G., Duell, N., Chein, J., Steinberg, L., 2016.
approach. Med. Decis. Mak. 28 (6), 829–833. http://dx.doi.org/10.1177/ The dual systems model: review, reappraisal, and reaffirmation. Dev. Cognit.
0272989X08327069. Neurosc. 17, 103–117. http://dx.doi.org/10.1016/j.dcn.2015.12.010.
Reynolds, L.M., Pokinko, M., Berrio, A.T., Cuesta, S., Lambert, L.C., Del Cid Pelitero, E., Simons-Morton, B.G., Bingham, C.R., Falk, E.B., Li, K., Pradham, A.K., Ouimet, M.C.,
et al., 2017. DCC receptors drive prefrontal cortex maturation by determining do- 2014. Experimental effects of injunctive norms on simulated driving among teenage
pamine axon targeting in adolescence. Biol. Psychiatry. http://dx.doi.org/10.1016/j. males. Health Psychol. 33 (7), 616–627. http://dx.doi.org/10.1037/a0034837.
biopsych.2017.06.009. Smith, G.T., Fischer, S., Cyders, M.A., Annus, A.M., Spillane, N.S., McCarthy, D.M., 2007.
Richards, M., Hatch, S.L., 2011. A life course approach to the development of mental On the validity and utility of discriminating among impulsivity-like traits. Assessment
skills. J. Gerontol. Ser. B 66B (S1), i26–i35. 14 (2), 155–170.
Richards, J.M., Plate, R.C., Ernst, M., 2013. A systematic review of fMRI paradigms used Sourander, M.D., Elonheimo, H., Niemelä, S., Nuutila, A., Helenius, H., Sillanmaki, L.,
in studies of adolescents vs. adults: the impact of task design and implications for et al., 2006. Childhood predictors of male criminality: a prospective population-based
understanding neurodevelopment. Neurosci. Biobehav. Rev. 37, 976–991. http://dx. follow-up study from age 8 to late adolescence. J. Am. Acad. Child Adolesc.
doi.org/10.1016/j.neubiorev.2013.03.004. Psychiatry 45 (5), 578–586. http://dx.doi.org/10.1097/01.chi.0000205699.
Rivers, S.E., Reyna, V.F., Mills, B., 2008. Risk taking under the influence: a fuzzy-trace 58626.b5.
theory of emotion in adolescence. Dev. Rev. 28 (1), 107–144. http://dx.doi.org/10. Spear, L.P., 2010. The Behavioral Neuroscience of Adolescence. W.W Norton & Co, New
1016/j.dr.2007.11.002. York.
Roeper, J., 2013. Dissecting the diversity of midbrain dopamine neurons. Trends Spear, L.P., 2013. Adolescent neurodevelopment. J. Adolesc. Health 52, S7–S13. http://
Neurosci. 36 (6), 336–342. http://dx.doi.org/10.1016/j.tins.2013.03.003. dx.doi.org/10.1016/j.adohealth.2012.05.006.
Rolls, E.T., 2014. Emotion and Decision-Making Explained. Oxford University Press, Stanovich, K.E., 2011. Rationality and the Reflective Mind. Oxford University Press, New
Oxford, UK. York.
Romer, D., Hennessy, M., 2007. A biosocial-affect model of adolescent sensation seeking: Steinberg, L., 2008. A neurobehavioral perspective on adolescent risk-taking. Dev. Rev.
the role of affect evaluation and peer-group influence in adolescent drug use. Prev. 28 (1), 78–106.
Sci. 8 (2), 89–101. http://dx.doi.org/10.1007/s11121-007-0064-7. Steinberg, L., 2014. Friends Can be Dangerous. New York Times (April 25, p. 12).
Romer, D., Betancourt, L., Giannetta, J.M., Brodsky, N.L., Farah, M.J., Hurt, H., 2009. Sternberg, R.J., 2001. Why schools should teach for wisdom: the balance theory of
Executive cognitive functions and impulsivity as correlates of risk taking and problem wisdom in educational settings. Educ. Psychol. 36 (4), 227–245.
behavior in preadolescents. Neuropsychologia 47 (13), 2916–2926. http://dx.doi. Stevens, M., 2016. The contributions of resting state and task-based functional con-
org/10.1016/j.neuropsychologia.2009.06.019. nectivity studies to our understanding of adolescent brain network maturation.
Romer, D., Duckworth, A.L., Sznitman, S., Park, S., 2010. Can adolescents learn self- Neurosci. Biobehav. Rev. 70, 13–32. http://dx.doi.org/10.1016/j.neubiorev.2016.
control? Delay of gratification in the development of control over risk taking. Prev. 07.027.
Sci. 11 (3), 319–330. http://dx.doi.org/10.1007/s11121-010-0171-8. Sunstein, C.R., 2008. Adolescent risk-taking and social meaning: a commentary. Dev. Rev.
Romer, D., Betancourt, L.M., Brodsky, N.L., Giannetta, J.M., Yang, W., Hurt, H., 2011. 28, 145–152.
Does adolescent risk taking imply weak executive function? A prospective study of Supekar, K., Uddin, L.Q., Prater, K., Amin, H., Grecius, M.D., Menon, V., 2010.
relations between working memory performance, impulsivity, and risk taking in early Development of functional and structural connectivity within the default mode net-
adolescence. Dev. Sci. 14 (5), 1119–1133. http://dx.doi.org/10.1111/j.1467-7687. work in young children. Neuroimage 52, 290–301. http://dx.doi.org/10.1016/j.
2011.01.61.x. neuroimage.2010.04.009.
Romer, A.L., Reyna, V.F., Pardo, S.T., 2016. Are rash impulsive and reward sensitive traits Svatkova, A., Nestrasil, I., Rudser, K., Fine, J.G., Bledsoe, J., Semrud-Clikeman, M., 2016.
distinguishable? A test in young adults. Pers. Individ. Differences 99, 308–312. Unique white matter microstructural patterns in ADHD presentations—a diffusion
http://dx.doi.org/10.1016/j.paid.2016.05.027. tensor imaging study. Hum. Brain Mapp. 37, 3323–3336. http://dx.doi.org/10.1002/
Romer, D., 2010. Adolescent risk taking, impulsivity, and brain development: implica- hbm.23243.
tions for prevention. Dev. Psychobiol. 52 (3), 263–276. http://dx.doi.org/10.1002/ Telzer, E.H., 2015. Dopaminergic reward sensitivity can promote adolescent health: a
dev.20442. new perspective on the mechanisms of ventral striatal activation. Dev. Cognit.
Rubia, K., 2013. Functional brain imaging across development. Eur. Child Adolesc. Neurosci. 17, 57–67. http://dx.doi.org/10.1016/j.dcn.2015.10.010.
Psychiatry 22, 719–731. http://dx.doi.org/10.1007/s00787-012-0291-8. Tobler, P.N., Fiorillo, C.D., Schultz, W., 2005. Adaptive coding or reward value by do-
Samanez-Larkin, G.R., Knutson, B., 2014. Reward processing and risky decision making in pamine neurons. Science 307, 1642–1645.
the aging brain. The Neuroscience of Risky Decision Making. American Psychological Tucker-Drob, E.M., 2011. Global and domain-specific changes in cognition throughout
Association, Washington, DC, pp. 123–142. adulthood. Dev. Psychol. 47 (2), 331–343. http://dx.doi.org/10.1037/a0021361.
Santos, F.J., Roomi, M.A., Liñan, F., 2016. About gender differences and the social en- Tymula, A., Rosenberg Belmaker, L.A., Roy, A.K., Manson, K., Glimcher, P.W., Levy, I.,
vironment in the development of entreprenaurial intentions. J. Small Bus. Manag. 54 2012. Adolescents’ risk-taking behavior is driven by tolerance to ambiguity. Proc.
(1), 49–66. http://dx.doi.org/10.1111/jsbm.12131. Natl. Acad. Sci. 1–6. http://dx.doi.org/10.1073/pnas.1207144109.
Satterthwaite, T.D., Ruparel, K., Loughead, J., Elliott, M.A., Gerraty, R.T., Calkins, M.E., US Census Bureau, 2012. Licensed Drivers and Number in Accidents by age. US Census
et al., 2012. Being right is its own reward: load and performance related ventral Bureau, Washington, DC (Retrieved from http://www.census.gov/compendia/
striatum activation to correct responses during a working memory task in youth. statab/cats/transportation/motor_vehicle_accidents_and_fatalities.html).
Neuroimage 61, 723–729. http://dx.doi.org/10.1016/j.neuroimage.2012.03.060. van den Bos, W., Rodriguez, C.A., Schweitzer, J.B., McClure, S.M., 2015. Adolescent
Satterthwaite, T.D., Wolf, D.H., Erus, G., Ruparel, K., Elliott, M.A., Gennatas, E.D., et al., impatience decreases with increased frontostriatal connectivity. Proc. Natl. Acad. Sci.
2013a. Functional maturation of the executive system during adolescence. J. 112 (29), E3765–E3774. http://dx.doi.org/10.1073/pnas.1423095112.
Neurosci. 33 (41), 16249–16261. Vandekar, S.N., Shinohara, R.T., Raznahan, A., Roalf, D.R., Ross, M., DeLeo, N., et al.,
Satterthwaite, T.D., Wolf, D.H., Ruparel, K., Erus, G., Elliott, M.A., Eickhoff, S.B., et al., 2015. Topologically dissociable patterns of development of the human cerebral
2013b. Heterogeneous impact of motion on fundamental patterns of develpmental cortex. J. Neurosci. 35 (2), 599–609. http://dx.doi.org/10.1523/JNEUROSCI.3628-
changes in functional connectivity during youth. Neuroimage 83, 45–57. http://dx. 14.2015.
doi.org/10.1016/j.neuroimage.2013.06.045. Volkow, N.D., Ding, Y.-S., Fowler, J.S., Wang, G.-J., Logan, J., Gatley, J., et al., 1996.
Schnack, H.G., van Haren, N.E.M., Brouwer, R.M., Evans, A., Durston, S., Boomsma, D.J., Dopamine transporters decrease with age. J. Nucl. Med. 37, 554–559.
et al., 2015. Changes in cortical thickness and surface area of the human cortex and Volkow, N.D., Gur, R.C., Wang, G.-J., Fowler, J.S., Moberg, P.J., Ding, Y.-S., et al., 1998.
their relationship with intelligence. Cereb. Cortex 25, 1608–1617. http://dx.doi.org/ Association between decline in brain dopamine activity with age and cognitive and
10.1093/cercor/bht357. motor impairment in healthy individuals. Am. J. Psychiatry 155 (3), 344–349.
Schultz, W., Dayan, P., Montague, P.R., 1997. A neural substrate of prediction and re- Volz, K.G., Gigerenzer, G., 2012. Cognitive processes in decisions under risk are not the
ward. Science 275, 1593–1599. same as in decisions under uncertainty. Front. Neurosci. 6 (105), 1–6. http://dx.doi.
Shamosh, N.A., Deyoung, C.G., Green, A.E., Reis, D.L., Johnson, M.R., Conway, A.R., org/10.3389/fnins.2012.00105.
et al., 2008. Individual differences in delay discounting: relation to intelligence, Wahlstrom, D., Collins, P., White, T., Luciana, M., 2010. Developmental changes in do-
working memory, and anterior prefrontal cortex. Psychol. Sci. 19 (9), 904–911. pamine neurotransmission in adolescence: behavioral implications and issues in as-
http://dx.doi.org/10.1111/j.1467-9280.2008.02175.x. sessment. Brain Cognit. 72, 146–159. http://dx.doi.org/10.1016/j.bandc.2009.10.
Shaw, P., Gilliam, B.S., Liverpool, M., Weddle, B.S., Malek, M., Sharp, M.S.W., Giedd, 013.
J.N., 2011. Cortical development in typically developing children with symptoms of Weber, E.U., Blais, A.-R., Betz, N.E., 2002. A domain-specific risk-attitude scale: mea-
hyperactivity and impulsivity: support for a dimensional view of attention deficit suring risk perceptiions and risk behaviors. J. Behav. Decis. Mak. 15, 263–290.
hyperactivity disorder. Am. J. Psychiatry 168 (2), 143–151. http://dx.doi.org/10.1002/bdm.414.
Shedler, J., Block, J., 1990. Adolescent drug use and psychological health: a longitudinal Webster, J.D., Westerhof, G.J., Boulmeijer, E.T., 2013. Wisdom and mental health across
inquiry. Am. Psychol. 45, 612–630. the lifespan. J. Gerontol. Ser. B 69 (2), 209–218. http://dx.doi.org/10.1093/geronb/
Sherman, L.E., Rudie, J.D., Pfeifer, J.H., Masten, C.L., McNealy, K., Dapretto, M., 2014. gbs121.
Development of the default mode and central executive networks across early ado- Weickert, C.S., Webster, M.J., Gondipalli, P., Rothmond, D., Fatula, R.J., Herman, M.M.,
lescence: a longitudinal study. Dev. Cognit. Neurosci. 10, 148–159. http://dx.doi. et al., 2007. Postnatal alterations in dopaminergic markers in the human prefrontal
org/10.1016/j.dcn.2014.08.002. cortex. Neuroscience 144, 1109–1119. http://dx.doi.org/10.1016/j.neuroscience.
Shohamy, D., Adcock, R.A., 2010. Dopamine and adaptive memory. Trends Cognit. Sci. 2006.10.009.
14 (10), 464–472. http://dx.doi.org/10.1016/j.tics.2010.08.002. Wilhelms, E., Corbin, J.C., Reyna, V.F., 2015. Gist memory in reasoning and decision

33
D. Romer, et al. Developmental Cognitive Neuroscience 27 (2017) 19–34

making: age, experience and expertise. Reasoning as Memory. Taylor & Francis, New relationship between impulsivity and substance abuse from studies using animal
York, pp. 93–109. models. Alcohol. Clin. Exp. Res. 34 (8), 1306–1318. http://dx.doi.org/10.1111/j.
Williams, B.R., Ponesse, J.S., Schachar, R.J., Logan, C.D., Tannock, R., 1999. 1530-0277.2010.01215.x.
Development of inhibitory control across the life span. Dev. Psychol. 35, 205–213. Wolfe, C.R., Reyna, V.F., Widmer, C.L., Cedillos, E.M., Fisher, C.R., Brust-Renck, P.G.,
Willoughby, T., Good, M., Adachi, P.J.C., Hamza, C., Tavernier, R., 2013. Examining the Weil, A.M., 2015. Efficacy of a web-based intelligent tutoring system for commu-
link between adolescent brain development and risk taking from a social-develop- nicating genetic risk of breast cancer: a fuzzy-trace theory approach. Med. Decis.
mental perspective. Brain Cognit. 83, 315–323. http://dx.doi.org/10.1016/j.bandc. Mak. 35, 46–59. http://dx.doi.org/10.1177/0272989X14535983.
2013.09.008. Zuckerman, M., 2007. Sensation Seeking and Risky Behavior. APA Books,
Winstanley, C.A., Olausson, P., Taylor, J.R., Jentsch, J.D., 2010. Insight into the Washington, DC.

34
 Update
Developmental Cognitive Neuroscience
Volume 45, Issue , October 2020, Page

DOI: https://doi.org/10.1016/j.dcn.2020.100842
 Developmental Cognitive Neuroscience 45 (2020) 100842

Contents lists available at ScienceDirect

Developmental Cognitive Neuroscience

journal homepage: www.elsevier.com/locate/dcn

Erratum
Erratum

The purpose of this publisher correction is to inform readers that the a Declaration of Interest statement which the publisher inadvertently
final version of the articles linked with this correction were replaced omitted from the original version.
with a corrected version in March 2019. The corrected version contains The Publisher apologizes for any inconvenience this may cause.”

DOIs of original article: https://doi.org/10.1016/j.dcn.2017.08.007, https://doi.org/10.1016/j.dcn.2017.09.008, https://doi.org/10.1016/j.dcn.2017.10.007,

https://doi.org/10.1016/j.dcn.2017.11.006, https://doi.org/10.1016/j.dcn.2018.05.004, https://doi.org/10.1016/j.dcn.2018.09.001, https://doi.org/10.1016/j.
dcn.2017.07.002, https://doi.org/10.1016/j.dcn.2017.07.007, https://doi.org/10.1016/j.dcn.2017.09.002, https://doi.org/10.1016/j.dcn.2018.06.003, https://
doi.org/10.1016/j.dcn.2018.07.003, https://doi.org/10.1016/j.dcn.2018.08.001, https://doi.org/10.1016/j.dcn.2017.02.007, https://doi.org/10.1016/j.dcn.2017.
07.006, https://doi.org/10.1016/j.dcn.2017.08.008, https://doi.org/10.1016/j.dcn.2017.10.006, https://doi.org/10.1016/j.dcn.2017.07.001, https://doi.org/10.
1016/j.dcn.2017.08.010, https://doi.org/10.1016/j.dcn.2017.10.001, https://doi.org/10.1016/j.dcn.2017.11.009, https://doi.org/10.1016/j.dcn.2017.12.002,
https://doi.org/10.1016/j.dcn.2017.02.008, https://doi.org/10.1016/j.dcn.2017.02.010, https://doi.org/10.1016/j.dcn.2016.12.001, https://doi.org/10.1016/j.
dcn.2017.07.003, https://doi.org/10.1016/j.dcn.2017.08.003, https://doi.org/10.1016/j.dcn.2017.10.005, https://doi.org/10.1016/j.dcn.2017.09.005, https://
doi.org/10.1016/j.dcn.2018.05.008, https://doi.org/10.1016/j.dcn.2017.01.002, https://doi.org/10.1016/j.dcn.2017.01.004, https://doi.org/10.1016/j.dcn.2017.
01.010, https://doi.org/10.1016/j.dcn.2017.07.004, https://doi.org/10.1016/j.dcn.2017.08.001, https://doi.org/10.1016/j.dcn.2017.10.003, https://doi.org/10.
1016/j.dcn.2017.05.007, https://doi.org/10.1016/j.dcn.2018.03.002, https://doi.org/10.1016/j.dcn.2018.10.002, https://doi.org/10.1016/j.dcn.2017.01.003,
https://doi.org/10.1016/j.dcn.2017.10.009, https://doi.org/10.1016/j.dcn.2017.11.011, https://doi.org/10.1016/j.dcn.2018.05.007, https://doi.org/10.1016/j.
dcn.2018.07.002, https://doi.org/10.1016/j.dcn.2017.11.001, https://doi.org/10.1016/j.dcn.2018.07.004, https://doi.org/10.1016/j.dcn.2017.01.001, https://
doi.org/10.1016/j.dcn.2017.01.005, https://doi.org/10.1016/j.dcn.2017.01.008, https://doi.org/10.1016/j.dcn.2017.08.009, https://doi.org/10.1016/j.dcn.2017.
11.003, https://doi.org/10.1016/j.dcn.2017.11.008, https://doi.org/10.1016/j.dcn.2018.02.004, https://doi.org/10.1016/j.dcn.2018.10.001, https://doi.org/10.
1016/j.dcn.2018.05.010.

https://doi.org/10.1016/j.dcn.2020.100842

Available online 28 August 2020

1878-9293/© 2020 Published by Elsevier Ltd.