Abstract
Relative to childhood, peer relationships take on a heightened importance during adolescence. Might adolescents be highly attuned to information that concerns when and how they are being evaluated and what their peers think of them? This review evaluates how continuing brain development—which influences brain function—partially explains and reflects adolescents’ attunement to social evaluation. Though preliminary, evidence is mounting to suggest that while processing information relevant to social evaluation and the internal states of other people, adolescents respond with heightened emotional intensity and corresponding nonlinear recruitment of socioaffective brain circuitry. This review highlights research findings that relate trajectories of brain development to social behavior and discusses promising avenues of future research that will inform how brain development might lead adolescents to be sensitized to social evaluation.
Introduction
When I was 14 years old and in eighth grade, I received an award at the end-of-year school assembly. Walking across the stage, I lost my footing and stumbled in front of the entire student body. To be clear, this was not falling flat on one’s face, spraining an ankle, or knocking over the school principal—it was a small misstep noticeable only to those in the audience who were paying close attention. As I rushed off the stage, my heart pounded with embarrassment and self-consciousness, and weeks of speculation about the consequences of this missed step were set into motion. There were tears and loss of sleep. Did my friends notice? Would they stop wanting to hang out with me? Would a reputation for clumsiness follow me to high school?
Although tripping in public could be embarrassing at any age, the anecdote above illustrates one example of how events that entail social evaluation can be experienced as highly intense, salient, persistent, and emotionally evocative during the adolescent years—perhaps more so than other phases of the life span. Indeed, a defining feature of adolescence is a newfound importance of peer and romantic relationships. A shifting motivation toward social relatedness is thought to intensify the attention, salience, and emotion relegated to processing information concerning social evaluations and social standing, referred to herein as social sensitivity. Understanding the mechanisms and consequences of adolescent social sensitivity and the relationship between these behaviors and brain development has received a surge of scientific interest.
Social sensitivity could “dial up” socioemotional processes at numerous levels of complexity. If adolescents have high social sensitivity, they might be more emotionally reactive to explicit cues indicative of social inclusion or exclusion. They might also be more attuned to instances of real or perceived social evaluation, in which individuals are led to believe that they are under evaluative scrutiny (for me, being onstage and aware of the hundreds of pairs of eyes in the audience). Adolescents might also consider with greater elaboration and emotional import what others are thinking and feeling, supported by mentalizing, or theory-of-mind processes, which enable speculation as to what others’ impressions of them might be. Although these are not the only possible manifestations of social sensitivity, these processes have been subject to experimental inquiry in neurodevelopmental research. As such, these components of adolescent social sensitivity and what is known about their neural bases during adolescence are explored here. This article highlights particularly novel and informative advances from cross-age comparisons in healthy individuals, in the hope of motivating a more complete understanding of adolescent social sensitivity.
What Is Unique About the Adolescent Social Life?
The term adolescence encapsulates a phase of life thought to begin around the onset of puberty and end when an individual achieves adultlike levels of independence. The social life of adolescents takes a central role in daily activities and thoughts. Relative to children, adolescents tend to spend less and less time with their families and more time with their peers (Barnes, Hoffman, Welte, Farrell, & Dintcheff, 2007; Larson, 2001), facilitated in part by greater freedom experienced with age. Digital peer communication also peaks, with a slightly higher frequency of Internet, text messaging, and social media usage in 14- to 17-year-olds than by adults ages 18 to 30 years (Lenhart, Ling, Campbell, & Purcell, 2010). This rise in peer interaction is not unique to humans. For example, “adolescent” rats spend more time engaged in social play than do adult rats (Douglas, Varlinskaya, & Spear, 2004; Primus & Kellogg, 1989).
Not only do adolescents engage in more frequent contact with peers, the quality of peer interactions also changes. The function of social relationships tends to shift away from friends as activity partners and toward intimate platonic and romantic relations. Finally, although adolescent social experiences are mostly positive, they also tend to be in a state of flux (Cairns, Leung, Buchanan, & Cairns, 1995). There is a tendency for adolescent relationships to wax and wane, and thus peer rejection becomes common during this time of life (Wang, Iannotti, & Nansel, 2009).
Developmental Properties of Socioaffective Circuitry
Brain imaging complements behavioral science to inform the components and mechanisms of adolescent social sensitivity. Imaging methodology can be used to characterize the magnitude, eliciting conditions, and circuit-based coordination of neural responses to social and emotional cues. A powerful approach to reveal brain-behavior relationships that change across development is to relate shifts in structural, functional, or connectivity-based measures with developmentally mediated differences in laboratory-based or everyday behavior. Such age-related shifts could manifest along simple or complex trajectories (Somerville et al., in press). For instance, it is possible to detect adolescent-specific effects that are maximally engaged during adolescence compared with both older and younger ages, which could parallel a temporary influx of hormones or other shifts that are unique to the adolescent years. Age-linear patterns describe developmental effects that increase or decrease with age at a consistent pace, which could result from developmental progressions or regressions. A third pattern of interest could be termed adolescent emergent, representing a nonlinear asymptotic pattern characterized by a rapid rate of change (and possible peak engagement during adolescence) that persists or partially resolves into adulthood. Effects of this type imply developmentally mediated shifts that are constrained by maturational processes that solidify in early adulthood. Examining linear and nonlinear patterns across transition into and out of adolescence can inform neurodevelopmental theory.
By drawing on a broad understanding of the functioning of neural systems, neuroimaging data can provide clues as to the putative mechanisms of adolescent emotional behavior and serve as an indirect “readout” of emotional and regulatory processes. Of particular relevance are the functional properties of what is termed here socioaffective circuitry, including the amygdala, striatum, and the medial prefrontal cortex (MPFC), relative to regions within the ventromedial and lateral prefrontal cortices thought to flexibly regulate these responses (Blakemore, 2008; Nelson & Guyer, 2011). Broadly speaking, coordinated activity of these regions is critical for (a) detection of salient information; (b) assignment of hedonic, aversive, or emotional value to that information; (c) social cognition; and (d) use of that information to guide learning and behavior.
Theoretical and empirical accounts have proposed that socioaffective function might follow a nonlinear trajectory during adolescence for at least two reasons. For one, regions of the prefrontal cortex continue to reach structural and functional maturity throughout the adolescent years (Shaw et al., 2008; Somerville, Hare, & Casey, 2011), and the connections between subcortical and cortical structures continue to strengthen (e.g., Asato, Terwilliger, Woo, & Luna, 2010; Liston et al., 2006). Given the role of the prefrontal cortex in the regulation of behavior, still-maturing connections between it and subcortical structures might reduce the capacity to exert cognitive or emotional regulation, particularly in emotionally salient contexts (Casey, Jones, & Hare, 2008; Steinberg, 2008). In addition, the influx of hormones during puberty is thought to sensitize functional properties of certain brain circuits because, in part, of the influence of hormones on neurotransmitter systems (see Ernst, Romeo, & Andersen, 2009; Nelson, Leibenluft, McClure, & Pine, 2005; Sisk & Zehr, 2005), potentially resulting in adolescent-specific engagement patterns. Interestingly, the brain regions that are strongly affected by pubertal hormones share a high degree of overlap with the socioaffective circuitry described above. Together, these observations motivate the prediction that sensitized socioaffective circuits of the brain, perhaps less efficiently regulated, could sharpen the detection of and response to salient social cues during adolescence.
Emotional Processing in the Social Context
Information about one’s social standing is inherently laden with emotion. Finding out that a classmate called me “awkward” after tripping onstage is an example of information that is both negative in its valence and relevant to social standing (e.g., “she does not view me favorably”). Do adolescents react with greater emotional intensity when they find out whether others view them in a positive or negative light?
To address this question, researchers have been challenged to develop tasks that deliver self-relevant, salient, and believable social feedback to participants within the confines of the laboratory. Three such paradigms are highlighted in Figure 1. Figure 1a depicts a social-feedback task (adapted from Somerville, Heatherton, & Kelley, 2006) in which participants believe that their photograph was rated by unfamiliar, peer-aged individuals on the basis of how likeable the participant looked. Trial by trial, participants receive feedback indicating whether raters did or did not like the participant’s photo. The “chatroom interact” (Silk et al., 2012; Fig. 1b) and related “chatroom” tasks (Guyer, Choate, Pine, & Nelson, 2012; Guyer, McClure-Tone, Shiffrin, Pine, & Nelson, 2009) indicate whether a peer selected the participant or somebody else to chat online about a topic of mutual interest (in this case, the movies) over a series of trials. In the “cyberball” task adapted from Williams, Cheung, and Choi (2000) and used developmentally (Fig. 1c; Masten et al., 2009; Sebastian et al., 2011), participants believe they are engaging in an online ball-tossing game, and the ball-tossing partners stop passing the participant the ball after a few mutual throws.
Though not always observed, adolescents have shown evidence of heightened positive and negative emotional responses in these experimental contexts. After experiencing negative social feedback, adolescents report a greater drop in mood and a greater increase in anxiety than do adults when excluded from the virtual ball-tossing game (Sebastian, Viding, Williams, & Blakemore, 2010) and expect less frequent positive feedback when their picture is supposedly judged by unfamiliar peers (Gunther Moor, van Leijenhorst, Rombouts, Crone, & Van der Molen, 2010). Silk and colleagues (2012) used eye tracking, an indirect measure of salience processing, to target implicit emotional and motivational responses to social feedback in the “chatroom interact” task depicted in Figure 1b. Although all participants (9–17 years) showed a pupil difference to rejecting compared with accepting trials, this response was exaggerated in older adolescents. Further, heightened pupillary responses to rejecting social feedback predicted less connectedness in participants’ real-life social relationships.
Interestingly, adolescent emotional reactivity is not limited to instances of social rejection. Adolescent participants report a boost in positive affect when experiencing social acceptance from a desirable peer (Guyer et al., 2012), though it is not yet known whether the magnitude of this effect differs relative to older or younger ages. Taken together, the findings suggest that these tasks are effective at indexing adolescent social sensitivity, that emotional responses in adolescents are particularly robust, and that heightened social sensitivity in the lab predicts less satisfaction in real-life social relationships.
Do adolescents show distinct recruitment of socioaffective circuitry when processing explicit positive and negative social feedback? Using the task depicted in Figure 1a, Gunther Moor and colleagues (2010) found that adolescents engaged similar regions of the brain as did children and adults when processing positive and negative social feedback, but activity increased with age within the striatum and subgenual anterior cingulate cortex—regions thought to support emotional valuation—while anticipating positive feedback. Sebastian and colleagues (2011) observed that adolescents recruited the medial prefrontal cortex more strongly relative to adults and recruited “regulatory regions” of the lateral prefrontal cortex less strongly while being actively excluded from the ball-tossing game. The latter study suggests that adolescents robustly engage socioaffective processes and reduce recruitment of regulatory circuitry while processing emotional qualities of social feedback compared with other ages. More research will be required to specify the implications of these activity patterns to brain maturation and to adolescent social sensitivity.
Vigilance to Social Evaluation
As I walked across the stage at the school assembly, I was painfully aware that others were watching and forming impressions of me. Yet I was not privy to what they were actually thinking. Might the real or perceived experience of being evaluated contribute to adolescent social sensitivity? Historical accounts have proposed that adolescents are hyperaware of others’ evaluations and feel they are under constant scrutiny by an imaginary audience (Elkind & Bowen, 1979). Though this characterization has been challenged (Vartanian, 2000), adolescents do report a greater day-to-day tendency to feel self-conscious (Westenberg, Drewes, Goedhart, Siebelink, & Treffers, 2004), and in laboratory-based social stressor tasks, compared with children, adolescents respond with greater release of cortisol (a stress hormone) when under social scrutiny (Gunnar, Wewerka, Frenn, Long, & Griggs, 2009; Stroud et al., 2009). Thus, social evaluative situations—even those devoid of feedback—induce self-consciousness and engage stress systems of the body in adolescents.
Following these observations, a recent study (Somerville et al., in press) sought to assess whether adolescents demonstrate heightened self-consciousness and exaggerated engagement of socioaffective neural circuitry under minimal conditions—just knowing that someone is looking at them. We informed participants that, at times, they would be watched by a peer via live video feed during portions of a brain-imaging scan (Somerville et al., in press). Relative to both children and adults, adolescents experienced greater self-reported embarrassment (Fig. 2a) evoked by ostensible video monitoring, which partially subsided into adulthood, and uniquely heightened responding of the autonomic nervous system (indexed by skin conductance). The MPFC (a key structure for integrating emotional and social information; see Fig. 2b) showed robust age differences, such that activity drastically increased during adolescence and partially subsided into adulthood. Though preliminary, these findings suggest that adolescents’ social sensitivity extends to subtle evaluative contexts. Adolescent-emergent engagement of the MPFC could reflect, or perhaps result in, social evaluative situations being assigned a high degree of salience, emotional arousal, and self-relevance.
Thinking About the Thoughts of Others
The night after I tripped onstage, I could not sleep because I was thinking about my friends and classmates, predicting who would sympathize with me in my state of embarrassment and who would poke fun at me for my clumsiness. A third feature of adolescent social sensitivity is a tendency to speculate about the thoughts and feelings of peers, a cognitive ability referred to as mentalizing or theory of mind. Although adolescents excel at simple mentalizing tasks, they perform significantly worse than adults when the theory-of-mind tasks are made highly complex (Dumontheil, Apperly, & Blakemore, 2010), which suggests that mentalizing abilities continue to mature through adolescence.
A subset of the socioaffective circuitry described earlier is thought to support theory-of-mind processes. These regions, sometimes termed the social brain (Blakemore, 2008), include the MPFC, temporoparietal junction, superior temporal sulcus, and other regions that are consistently engaged across a wide variety of tasks that carry mentalizing demands. Such tasks include those in which participants are asked to reason about moral scenarios (Blakemore, Den Ouden, Choudhury, & Frith, 2007) and about social emotions, such as guilt and embarrassment, relative to less socially guided emotions, such as disgust and fear (e.g., Burnett, Bird, Moll, Frith, & Blakemore, 2009). Developmental investigations have demonstrated that MPFC recruitment is more robust in adolescents than in adults during tasks that share the common process of considering the thoughts and intentions of others (see Burnett, Sebastian, Kadosh, & Blakemore, 2011, for a review). Burnett and Blakemore (2009) also observed heightened task-based functional connectivity between brain regions important for social cognition in adolescents compared with adults. Given that the prefrontal cortex continues to undergo changes in structural maturation throughout adolescence, heightened activity during adolescence has been proposed to reflect the MPFC’s continuing developmental course (Blakemore, 2008). It has also been proposed that adolescents use strategies for theory of mind that more strongly engage the MPFC. Alternatively, adolescents might consider thoughts and feelings of others to be more important or salient than adults do, which might strongly engage the processing resources of socioaffective neural circuitry through heightened motivation to understand others (and thus greater effort allocated to mentalizing).
Conclusions
Convergent evidence suggests that adolescents display heightened sensitivity to social evaluation at various levels of complexity and continue to refine their capacity to represent the thoughts and feelings of others. These features of social sensitivity appear to be instantiated by robust response properties in neural circuitry important to assigning value to social-affective information during adolescence. Though the present report focused on brain function, it should be acknowledged that adolescent social sensitivity is a product of multidirectional interactions of brain development, experience, and sociocultural factors (Crone & Dahl, 2012). Indeed, the greater independence afforded by adolescence poses a unique set of challenges that require adolescents to navigate their social worlds in ways that are new and challenging (Pfeifer & Peake, 2012), thus shaping brain development (Crone & Dahl, 2012; Pfeifer & Peake, 2012). Though speculative, it is possible that sensitized responding in socioaffective brain circuitry enables a heightened capacity to detect, learn from, and adapt to the myriad social challenges characteristic of adolescence, ultimately facilitating mature social competence.
It is important to consider that the studies presented above represent findings that have emerged within just the past few years. As a result, models pinpointing the neural mechanisms that give rise to adolescent-specific social sensitivity remain tentative. Studies that sample a large range of ages, including both pre- and postadolescents, or track individuals longitudinally over time provide the most compelling means to ask whether socioaffective sensitivity is specific to adolescence, emerges during adolescence, or is comparable to older or younger ages. With a greater corpus of data, we will also be capable of refining the putative functional roles for subcomponents of socioaffective circuitry, which are much more complex than sketched here. Future research will also address the role of factors that likely influence adolescent social sensitivity in complex ways, including pubertal hormones, cultural norms, gender, social status, and self-esteem, among many others (Crone & Dahl, 2012).
The goal of understanding the neural mechanisms of adolescent social sensitivity is not just compelling from a basic science standpoint. Social concerns are a primary source of anxiety for adolescents, and social exclusion during this time of life predicts risk for mood and anxiety disorders throughout the life span (Lev-Wiesel, Nuttman-Shwartz, & Sternberg, 2006). Initial clinical studies of social sensitivity have implicated biased responses in socioaffective circuitry in clinical and at-risk adolescent samples (Guyer et al., 2008; Masten et al., 2011). Thus, exaggerated responding in socioaffective brain circuitry might represent a common feature of adolescent-specific social sensitivity and emergent mood and anxiety dysregulation during adolescence, a connection that will be further specified with the study of clinical and at-risk samples.
If readers were to reflect on the formative events of their own adolescent years, I would bet these would involve friends, love interests, or events experienced in social groups (in my case, embarrassing ones). The field of adolescent science is just beginning to uncover some of the causes and consequences of adolescent social sensitivity. Ultimately, this line of work will reveal the biological underpinnings of this core feature of the adolescent experience.