Share-the-load vs bring-your-own-judge: The effects of friends on social evaluative stress in early adolescence

Abstract

Objective: Parental attachment figures effectively buffer their children's cortisol responses to a socially evaluative stressor, the Trier Social Stress Test (TSST), by providing instrumental and emotional support during the preparation period. The effectiveness of parents as stress buffers wanes in adolescence as youth increase their reliance on peers for support. Yet, in a previous study, when peers played the same supportive role as parents, the cortisol response to the TSST was amplified. Here we asked if the role the friend plays matters. We hypothesized that the support of a friend would reduce the cortisol response to the TSST if they were also undergoing the stressor; while friend support would increase the cortisol response if they were only there for support and were not also undergoing the stressor.

Method: 235 youth ages 11.2-15.1 years were tested in the TSST-OL, an on-line version of the TSST task. Saliva samples were obtained for cortisol determination at multiple timepoints. In this pre-registered study, youth were randomly assigned, stratified by age, sex assigned at birth and pubertal status (early/later), to one of the following conditions: 1) Alone, 2) Unfamiliar Peer - Both Tested, 3) Friend as Supporter, 4) Friend - Both Tested.

Results: Area under the curve from intercept (AUCi) for the initial reactivity phase was significantly different by group, matching our pre-registered prediction that friend support would reduce reactivity if both friends were undergoing the stressor, but increase it if the friend providing support did not also undergo the stressor. Recovery did not differ by condition.

Conclusions: The effect of friends as stress buffers varies by the role the friend plays. In the case of the TSST, a social evaluative stressor, support from a friend buffers cortisol reactivity if the friend is also experiencing the stressor; but increases cortisol reactivity if they are just there as a support. Simply being with a supportive friend does not shield the adolescent from reacting with a neuroendocrine stress response to the stress of social evaluation.

Keywords: Adolescence; Cortisol; Friends; Social stress buffering; TSST.

Figure 1. Timeline for TSST-OL.

Sample 1, taken upon arrival T-30, is not analyzed but only used to train the participant to sample. Sample 2 happens after the calming video, T0. Sample 3 happens after 5 minutes of speech preparation, T+5. Sample 4 happens after 10+ minutes of speech and math performance, T+15. Sample 5 happens 15 minutes later, T+30. The peak response to the TSST typically happens at sample 5. Sample 6 and 7 happen at 10 minutes subsequent intervals, T+40, T+50.

Figure 2. Stress Reactivity Means and SEMs for Each Social Buffering Condition.

AUC_I-RESPONSE is computed as the total cortisol from samples 3, 4, 5 above baseline sample 2.