Adolescents (10–24 years)1 come of age within the context of changing social environments2. As a developmental period, adolescence is characterised by a shift in focus away from the immediate family towards interactions with peers and other members of society3. Social interactions during this period are often novel and less stable, playing a crucial role in adolescents’ developing sense of self, well-being and cognitive abilities4. Frequently, these interactions now occur in online contexts as adolescents spend a daily average of 6-hours online for non-school purposes5, with the majority of this time spent on social media sites6. Yet, the impact of online social interactions on adolescent emotional and cognitive functioning remains little understood.
Work on the mental health impact of social media use has shown mixed results7. Interestingly, a survey in 743 13–17 year-olds found that 31% reported social media having a mostly positive effect on their lives, whereas 24% of respondents thought it had a mostly negative impact6. Individual differences then may help account for previous research’s mixed findings7. The present study therefore takes an individual differences approach examining whether offline social risk and protective factors moderate the impact of online social interactions on adolescent emotional and cognitive functioning.
Offline social risk and protective factors: Social rejection sensitivity and social support
Social relationships have been shown to foster resilience against emotional disorders during adolescence8,9. To form and maintain social relationships, adolescents become sensitive to various cues of social acceptance and rejection10. Excessive sensitivity to social rejection, however, is a known risk factor for emotional disorders11. Social rejection sensitivity refers to the tendency to expect social rejection by others12, be hypervigilant to social rejection cues and interpret ambiguous social situations negatively13. In addition to its association with adverse mental health outcomes, social rejection sensitivity may also negatively impact cognitive functioning and learning in adolescence. The attentional control theory posits that anxiety and uncertainty inhibit higher order cognitive functions and learning by placing cognitive demands on individuals14. The elevated levels of stress and anxiety experienced by individuals with higher social rejection sensitivity may therefore lead to higher attentional load and consequently poorer cognitive functioning.
In contrast, social support has been shown to be a protective factor in adolescents at risk for mental ill health15,16. While less is known about the effect of social support on adolescent cognitive functioning, evidence from the aging literature suggests it has a protective effect on cognition17. Conversely, the absence of social support, experienced through social isolation, is associated with poor mental health and reduced cognitive functioning in young people18 and related mid-life outcomes (incl., lower income)19.
Risk and protective factors in digital interactions
Importantly, in the digital age, these risk and protective factors can operate anywhere and at any time. That is, the advent of online social interactions on mobile devices has eradicated non-social spaces. There are at least two potential implications. First, the perpetual potential for social rejection that ensues from online social interactions may negatively impact young people high in social rejection sensitivity. Second, the easy availability of social interactions may potentiate the beneficial effects of perceived social support by being more continuously available to booster resilience in young people.
Preliminary research shows that online social support is associated with increased well-being20 and contributes to improvements in low mood even after accounting for offline social support21. Cross-sectionally, both online and offline social rejection sensitivity are associated with symptoms of depressed mood22. The continuous threat of rejection may also impact other areas of functioning such as learning. Vocational and academic learning, especially homework and study, which now take place in the context of online social interactions, may be impacted differently as a function of individuals’ social rejection sensitivity and perceived social support. That is, building on attentional control theory, the anxiety caused by the continuous potential for social rejection may reduce the capacity to learn in individuals high on social rejection sensitivity, whereas young people with good social support may be less affected.
The Current Study
This pre-registered study (https://osf.io/e8wyp) had two aims. The first was to investigate the effects of online social evaluative threat on adolescent mood and learning. The second aim was to explore whether these effects varied as a function of social risk (i.e., social rejection sensitivity) and protective (i.e. social support) factors, as well as self-reported mental health symptoms (primarily depressive symptoms) that are associated with these risk/protective factors.
To address these aims we designed a novel online social evaluative threat paradigm. Evaluative threat was evoked by asking participants to disclose personal information, which they were told would potentially be rated by peers. Under online social evaluative threat, participants then completed a perceptual learning task that has shown to be sensitive to social stress23. Performance on the task was compared to performance under no threat. 255 individuals aged 11–30 years completed the paradigm, which allowed us to investigate age-related differences across adolescence and early adulthood. Specifically, given research showing that adolescents are most sensitive to social rejection10, the impact of social rejection sensitivity in the current study was expected to be greatest in adolescents compared to young adults.
The study design allowed us to investigate the hypothesis that individuals with greater social rejection sensitivity would show increased negative mood (Hypothesis 1a) and poorer learning performance (Hypothesis 1b) following online social evaluative threat compared to the control condition, and the opposite would be shown by individuals with increased social support. The magnitude of the impact of social evaluative threat (relative to control) on mood (Hypothesis 2a) and learning (Hypothesis 2b) would be increased with self-reported mental health symptoms. Given the proposed role of social rejection sensitivity in the onset and maintenance of depression13, we proposed that this would be specifically the case for depressive symptoms. Specifically, we hypothesised that the effect of mental health (again depressive symptoms in particular) on mood and learning would be partially accounted for by social rejection sensitivity (Hypothesis 3).