Rationale Alcohol may be self-administered for its anxiolytic effects to alleviate symptoms of stress, but different types of stressors have varying effects on alcohol intake. Social stress is particularly relevant to alcohol drinking, and a primate model of stress-induced alcohol self-administration would be useful. Objective The objective of the study is to determine if social stresses of different lengths and intensities affect voluntary alcohol intake in monkeys. Materials and methods Subjects were adult male and female squirrel monkeys (Saimiri sciureus) housed in social colonies. Subjects were trained to drink a solution of ethanol and sucrose, alternated daily with a control solution of quinine and an equal concentration of sucrose in 15-min sessions. Drinking was tested during 20-min acute, social separations and 1-week, extended, social separations. Dominance status was quantified using observational records of social interactions within the colonies. Salivary cortisol was sampled in the home colony and during extended social separation. Results Dominance rank was inversely correlated with alcohol intake during social housing but was not correlated with control fluid intake. Acute social separation abolished drinking of both fluids, accompanied by increased anxiety-like behavior. Extended social separation increased salivary cortisol and alcohol drinking but not control fluid intake in males. In females, drinking was unchanged by extended separation. Conclusions The chronic stress of social subordination is correlated with increased alcohol drinking. Acute social separation stress suppresses drinking behavior, while extended separation preferentially increases alcohol intake in a subset of individuals. These findings suggest that social stressors of different time-courses and intensities have opposing effects on alcohol self-administration.

These results confirm the inhibitory effects of 5-HT(1B) receptor stimulation on aggressive performance and drinking. They also reveal an inhibition of voluntary wheel running, contrary to the stimulation of running in a novel, open arena. 5-HT(1B) receptor agonists may be particularly useful for the treatment of aggressive behavioral disorders, but their efficacy and potency appear to be sensitive to the intensity and context of the behavior.

These results support the hypothesis that in rats, 5-HT(1A) receptor stimulation activates somatodendritic receptors at lower doses and postsynaptic receptors at higher doses, each with opposite effects on alcohol intake. The absence of such biphasic dose-effect curves in monkeys suggests a different function of 5-HT(1A) somatodendritic receptors in rats and monkeys, at least with regard to alcohol drinking.