Applying developmental theory to adolescent peer influence processes in the social media context
Many decades of research have documented the critical importance of peer relationships for adolescent development (Choukas-Bradley & Prin-stein, 2014; Rubin, Bukowski, & Bowker, 2015). As youth transition from childhood to adolescence, they spend significantly more time with peers and are highly attuned to peer feedback for identity formation and sense of self-worth (Harter, Stocker, & Robinson, 1996). One key area of study within the field of adolescent peer relations focuses on peer influence — a set of processes by which adolescents’ behaviours and attitudes become more similar to their peers’ over time (for a review, see Brechwald & Prinstein, 2011). The vast majority of research on peer influence has focused on traditional, in-person interactions, with research on online peer interactions still in its infancy. Yet adolescents now spend a significant portion of their time interacting with peers through social media (Anderson & Jiang, 2018; Rideout & Robb, 2018), and rigorous research is needed that examines the unique effects of social media use on adolescents’ interpersonal relationships and psychosocial well-being (Nesi, Choukas-Bradley & Prinstein, 2018a, 2018b).
This chapter applies developmental theory to the study of online peer influence. We focus on social media, or websites and apps that allow users to interact and share content with others (Moreno & Kota, 2013). Our chapter begins with a review of theory and empirical work related to traditional (offline) peer influence processes. We then discuss the transformation framework, a theoretical model for how the features of social media may be transforming traditional peer relations experiences, including peer influence processes. We then provide a review of the burgeoning literature on peer influence via social media. Finally, we discuss methodological advances that will be necessary for a thorough understanding of the role of online peer influence in adolescent development, with an emphasis on the need for more theory-driven longitudinal and experimental studies that integrate peer self-reports, research that rigorously controls for offline peer influence, and a better understanding of mechanisms and moderators.
Traditional peer influence processes: a brief overview of developmental theories and empirical work
Adolescence is a developmental period characterized by identity exploration, increased sensation-seeking, the pursuit of romantic and sexual intimacy, and heightened attention to and valuing of peer approval and peer status (see Dahl, Allen, Wilbrecht, & Suleiman, 2018). These developmental processes are believed to be driven by a complex set of biopsychosocial changes, which are relevant to understanding peer influence. Socially, adolescents become highly attuned to peer status (LaFontana & Cillessen, 2010) and begin to spend more unsupervised time with peers (Lam, McHale, & Crouter, 2014). Based on these and other developmental changes, it is believed that adolescents may aim to match their behaviours and attitudes to those of valued peers and peer groups. Adolescents may seek to emulate relevant social nonns and receive social rewards in the peer hierarchy (e.g., in the fonn of increased social status). Receiving such social rewards is thought to foster a more positive self-identity among adolescents (Brechwald & Prinstein, 2011). With regard to biological changes, research suggests that adolescence is associated with rapid development in the socio-affective circuitry of the brain (i.e., amygdala, striatum, medial prefrontal cortex) and that pubertal honnones may disproportionately affect neurotransmitter activity within these regions, with significant increases in dopamine and oxytocin receptors in the limbic system (Somerville, 2013; Steinberg, 2008). Importantly, these changes may outpace development of the prefrontal cortex, heightening adolescents’ desires for social rewards before they have developed the inhibitory control to modulate these impulses (Somerville, 2013; Steinberg, 2008). These biological changes may combine to create an increased desire among adolescents for affiliation with peers and to heighten sensitivity to interpersonal rewards. The imbalance between socioemotional and cortical control networks may contribute to adolescents’ tendencies to conform to peers, even when peers are engaging in risky behaviours (Steinberg, 2007). Recent theoretical models from developmental neuroscience propose that the social context may be highly relevant to understanding adolescent risk behaviour and that the development of the prefrontal cortex may be less linear than previously thought (Crone & Dahl, 2012; Crone & Steinbeis, 2017). Specifically, recent models propose that cortical-subcortical connectivity and concomitant cognitive control may change flexibly based on the social and motivational salience of a particular goal (e.g., peer approval), and that peer presence and influence may impact these neural processes (Crone & Dahl, 2012; van Duijvenvoorde, Peters, Braams, & Crone, 2016). While these neurobiological models likely will continue to evolve as developmental neuroscience research advances, they dovetail with research on psychosocial factors in highlighting the developmental importance of peers in adolescence.
Given the unique characteristics of the adolescent developmental period, it is not surprising that several decades of research have documented peer influence processes for a broad range of behaviours, including substance use, deviance, depressive symptoms, weight-related behaviours, sexual behaviours, academic engagement and outcomes, and prosocial behaviours (see Brechwald & Prinstein, 2011). However, the body of work on adolescent peer influence has been limited in a number of important ways. One key issue is that many studies use single time-point study designs, which are incapable of distinguishing between peer selection processes versus peer socialization or influence processes. Specifically, if adolescents’ and friends’ behaviours are observed to be similar at one point in time, it is not possible to determine whether they have influenced one another, becoming more behaviourally similar over time, or whether they selected each other as friends based on pre-existing similarities (Kandel, 1978). Longitudinal studies can control for prior levels of behaviour, in order to better account for pre-existing similarities. However, an additional methodological concern is that most longitudinal peer influence studies include only two time points, and two time point designs cannot capture the complex longitudinal trajectories that are believed to underlie development (Curran & Willoughby, 2003), nor the ongoing interplay between selection and socialization processes (Steglich, Snijders, & Pearson, 2010). Another consideration in evaluating the body of work on peer influence is that most studies have assessed participants’ perceptions of their peers’ attitudes and behaviours, which may represent overestimates of peers’ actual behaviours (Helms et al., 2014). A final limitation of the existing peer influence literature is that the vast majority of studies have focused on the influence of friends. However, adolescents are exposed to peer networks that are far broader than their close friends, and several studies suggest that they may be especially influenced by peer nonns related to high-status or popular peers (Brechwald & Prinstein, 2011; Cohen & Prinstein, 2006).