Responses to social and environmental stress are attenuated by strong male bonds in wild macaques

Abstract

In humans and obligatory social animals, individuals with weak social ties experience negative health and fitness consequences. The social buffering hypothesis conceptualizes one possible mediating mechanism: During stressful situations the presence of close social partners buffers against the adverse effects of increased physiological stress levels. We tested this hypothesis using data on social (rate of aggression received) and environmental (low temperatures) stressors in wild male Barbary macaques (Macaca sylvanus) in Morocco. These males form strong, enduring, and equitable affiliative relationships similar to human friendships. We tested the effect of the strength of a male's top three social bonds on his fecal glucocorticoid metabolite (fGCM) levels as a function of the stressors' intensity. The attenuating effect of stronger social bonds on physiological stress increased both with increasing rates of aggression received and with decreasing minimum daily temperature. Ruling out thermoregulatory and immediate effects of social interactions on fGCM levels, our results indicate that male Barbary macaques employ a tend-and-befriend coping strategy in the face of increased environmental as well as social day-to-day stressors. This evidence of a stress-ameliorating effect of social bonding among males under natural conditions and beyond the mother-offspring, kin or pair bond broadens the generality of the social buffering hypothesis.

Keywords: glucocorticoid metabolites; male social relationships; social bonds; social buffering hypothesis; stress.

Fig. 1.

The interaction effect of social bond strength (CSI) and rate of aggression received on fGCM levels (log nanograms per gram of feces). CSI relates rates and durations of affiliative behaviors of a dyad to the respective group means across dyads. The x axis shows z-transformed values for the added CSI values of a male’s top three social bonds. For illustration data are split into four subsets representing the quartiles of aggression rate: (i) low, 0.0–0.7 bouts per hour (sky blue); (ii) low–mid, 0.8–2.2 bouts per hour (slate blue); (iii) mid–high, 2.3–5.8 bouts per hour (navy blue); and (iv) high, 5.8–13.1 bouts per hour (black); colors run in order of stressor intensity, with darker being higher levels of stressor. This visualization of raw data is not a substitute for the full statistical model results, which are presented in Table 1.

Fig. 2.

The interaction effect of social bond strength (CSI) and mean minimum daily temperature on fGCM levels (log nanograms per gram of feces). CSI relates rates and durations of affiliative behaviors of a dyad to the respective group means across dyads. The x axis shows z-transformed values for the added CSI values of a male’s top three social bonds. For illustration data are split into four subsets representing the quartiles of mean minimum daily temperature: (i) low, −0.9 to 7.5 °C (black); (ii) low–mid, 7.5–15.9 °C (navy blue); (iii) mid–high, 15.9–23.4 °C (slate blue); and (iv) high, 23.4–33.0 °C (sky blue); colors run in order of stressor intensity, with darker being higher levels of stressor. This visualization of raw data is not a substitute for the full statistical model results, which are presented in Table 1.