A new study explains why stress makes us both cooperative and defensive

People under acute stress may cooperate with potential allies while becoming more defensive toward the people responsible for the stressful situation, according to research published in Psychoneuroendocrinology. The findings suggest that stress does not have one universal effect on social behavior; whom a person is dealing with may shape the response.

Two theories have traditionally offered different predictions about behavior under pressure. The “tend-and-befriend” account proposes that people seek affiliation and cooperate with others who could provide support. The “fight-or-flight” account proposes that threats can instead encourage withdrawal, aggression or protection of one’s own resources. Previous studies have found evidence for both patterns, prompting researchers to examine whether features of the situation can explain the inconsistency.

Laura Oswald of the University of Freiburg led a team that investigated whether the social role of an interaction partner changes behavior following acute stress. The researchers expected people to treat another, unstressed participant as a possible ally but respond more defensively to an evaluator who had helped create the stressful experience.

The final sample contained 121 healthy young adults (62 women, average age of 23.64 years). Participants first completed the Trier Social Stress Test for Groups. After five minutes of preparation, they gave a 12-minute speech and completed an eight-minute mental-arithmetic task before a jury wearing white coats that provided no encouraging feedback. Saliva samples, heart-rate recordings and subjective ratings confirmed whether the procedure produced a stress response.

Participants then made decisions involving real money in four established games measuring trust, trustworthiness, sharing and punishment. They completed decisions involving another participant who had not been subjected to the stress test and decisions involving a jury member who had been part of the stress test. This allowed the researchers to compare how the same stressed person treated a peer and a source of stress.

Free daily newsletter

The stress hormone cortisol, heart rate and perceived stress all increased substantially during the procedure. Participants made an average of 2.52 trusting choices toward peers, compared with 2.02 toward jury members. They also made more trustworthy choices toward peers, 3.06 compared with 2.40, and more sharing choices, 2.83 compared with 2.27. Conversely, punishment was more frequent toward jury members than peers, averaging 0.74 compared with 0.54. Each difference was statistically significant.

The behavioral pattern did not differ significantly between men and women. The size of participants’ cortisol, heart-rate or perceived-stress response also did not clearly predict how differently they treated peers and jury members. The findings therefore point more strongly to the interaction partner’s role than to the measured intensity of an individual’s stress response.

As the researchers explained, “individuals adapt their behavior toward others depending on the social role of their interaction partner.” Cooperating with a peer may increase the possibility of receiving reciprocal help, while withholding cooperation from an evaluator may protect resources or help restore a sense of control.

Important caveats remain. Everyone completed the stress procedure, so the study cannot show whether stress increased cooperation with peers, decreased cooperation with jury members or produced both changes. Also, participants always dealt with peers before jury members, leaving possible order and comparison effects. Studies including a non-stress control group, friendlier evaluators and more natural interactions are needed.

The study, “Tend-and-befriend toward peers, fight-or-flight toward stressors: A dual response to stress,” was authored by Laura Oswald, Fabian Dvorak, Jeremy P. Jamieson, Markus Heinrichs and Bastian Schiller.