The Developmental Trajectories of Prosocial Behavior and Empathy Diverge from Childhood into Adolescence

Post by Shireen Parimoo

The takeaway

Prosocial behavior increases as children develop into adolescents, whereas parental reports of empathy show increases until late childhood followed by gradual declines in early adolescence. Brain activity in regions associated with feeling socially included, such as the ventral striatum and medial prefrontal cortex (mPFC), predicts future prosocial behavior in children. 

What's the science?

Empathy and prosocial behavior – social behavior that benefits others more than it benefits us – are important for forming and maintaining meaningful social connections. Prosocial behavior often involves empathy, as adopting someone else’s perspective makes it easier to act prosocially towards them. As empathy and perspective-taking develop throughout childhood, older children exhibit more prosocial behavior than younger children. Adolescents, on the other hand, show more cooperative prosocial behavior but less helping behavior.

The difference between children and adolescents may be explained by the developmental trajectory of brain regions supporting different aspects of prosocial behavior. For instance, a socio-cognitive network underlying perspective-taking, which includes the mPFC, may show a different developmental pattern than a socio-affective network that includes regions like the ventral striatum that support emotional processing. To distinguish between these possibilities, new work in NeuroImage by van der Meulen and colleagues investigated the neural mechanisms underlying the developmental trajectories of prosocial behavior and empathy from middle childhood to early adolescence.

How did they do it?

Children aged 7-13 years old participated in a longitudinal study that took place over five years. Data from three sessions were collected, with each session taking place 2-2.5 years apart. At each session, children played the prosocial cyber ball game which consisted of three other virtual players who tossed a ball to each other. In the Fair Game round, each player received the ball an equal number of times whereas in the Unfair Game round, player two (P2) only received the ball once from the other virtual players. Prosocial behavior was defined as the ratio of tosses from the participant to P2 in the Unfair compared to the Fair Game round. Parents also completed a questionnaire at each session reporting their child’s prosocial behavior and empathy toward others.

Functional MRI was used to record brain activity during the game at each session, except for the Fair Game round during T1. The authors examined whole-brain activity during the Unfair Round as well as activity in the socio-cognitive (temporoparietal junction, precuneus, posterior superior temporal sulcus, and mPFC) and in the socio-affective brain regions (anterior insula, ventral striatum, and dorsal anterior cingulate cortex ). Specifically, they contrasted the difference in brain activity during prosocial behavior (i.e., passing the ball to P2 during the Unfair Round) with non-prosocial behavior (i.e., passing the ball to the other players during the Unfair Round). Additionally, the authors examined brain activity associated with feeling socially included, that is, when the participant received the ball from P1 and P3 compared to when they did not receive the ball.

The authors used mixed effects statistical modeling to study the longitudinal change in 1) parental reports of prosocial behavior and empathy, 2) prosocial behavior during the cyber ball game, 3) brain activity associated with prosocial behavior and social inclusion, and 4) associations between prosocial behavior and brain activity.

What did they find?

Prosocial behavior from both the parental reports and the cyber ball game showed linear increases with age. Empathy, on the other hand, increased from middle to late childhood before showing a gradual decrease into early adolescence. Interestingly, empathy was positively correlated with parental reports of prosocial behavior but not with the prosociality measured by the cyber ball game. At the whole-brain level, no brain area showed greater activation during prosocial behavior at T1 and T2, while mPFC and visual regions showed increased activation at T3 (i.e., in early adolescence). Additionally, there was a gradual increase in ventral striatal activity with age until late childhood, after which it stabilized. However, changes in neural activity over time were not related to changes in prosocial behavior. Together, these results suggest that the neural correlates of prosocial behavior become more prominent during late childhood in socio-affective regions and during early adolescence in socio-cognitive regions.

Social inclusion was linked with more widespread activation in regions of the socio-affective and socio-cognitive networks. The dorsal anterior cingulate cortex, insula, ventral striatum, and the precuneus showed a U-shaped trajectory, with a reduction in activation from middle to late childhood followed by an increase in activity into early adolescence. Thus, in contrast to prosocial behavior, neural correlates of social inclusion are most apparent in middle childhood and adolescence. Notably, there was a negative relationship between changes in neural activity and changes in prosocial behavior. Specifically, a stronger decrease in mPFC and ventral striatal activity during social inclusion was related to greater increases in prosocial behavior over time. Altogether, these findings highlight how the neural and cognitive processes underlying social inclusion interplay with prosocial behavior over the course of development.

What's the impact?

This study found that prosocial behavior and empathy follow different developmental trajectories as children transition into adolescence. The finding that the regions sensitive to social inclusion predict prosocial behavior over time paves the way for future research to investigate the mental processes these regions facilitate in order to support prosocial behavior. 

Access the original scientific publication here.