Research in stress physiology has demonstrated the benefits of receiving social support during stressful conditions. However, recent data have shown that the efficacy of social support in buffering physiological and behavioral responses to stressor agents depends on species, sex, and relatedness among animals. This study investigated whether different kinds of social support (presence of same sex related or nonrelated conspecifics) have the same effect on hormonal (fecal cortisol levels) and behavioral responses (agonistic: scent-marking and individual piloerection; anxiety: locomotion; tension-reducing: autogrooming, allogrooming, and body contact). We used adult male and female isosexual dyads of Callithrix jacchus, a small Neotropical primate from the Callitrichidae family, widely used in the study of stress and related diseases. Following a 28-day baseline phase, dyads faced three challenging situations (phase 1: dyads were moved together from the baseline cage to a similar new cage; phase 2: each dyad member was moved alone to a new cage; and phase 3: dyad members were reunited in the same baseline cage). Type of social support was found to influence the response to stressors differently for each sex. Related male dyads did not change their hormonal or behavioral profile over the three experimental phases, when compared to the baseline phase. For nonrelated male dyads, social support buffered hormonal but not behavioral response. For females, the social support offered by a related and nonrelated animal, does not seem to buffer the stress response, as shown by correlations between agonistic behaviors versus cortisol and locomotion during all three experimental phases and a significant increase in fecal cortisol levels during phases 2 and 3, when compared with baseline levels. The results only partially support the buffering model theory and corroborate other studies reporting that the benefits of social support during a period of crisis arise only when it is adaptive for that species.