Abstract
While traditional economic models assume that agents are self-interested, humans and most non-human primates are social species. Therefore, many of decisions they make require the integration of information about other social agents. This study asks to what extent information about social status and the social context in which decisions are taken impact on reward-guided decisions in rhesus macaques. We tested 12 monkeys of varying dominance status in several experimental versions of a two-choice task in which reward could be delivered to self only, only another monkey, both the self and another monkey, or neither. Results showed dominant animals were more prone to make prosocial choices than subordinates, but only when the decision was between a reward for self only and a reward for both self and other. If the choice was between a reward for self only and a reward for other only, no animal expressed altruistic behaviour. Finally, prosocial choices were true social decisions as they were strikingly reduced when the social partner was replaced by a non-social object. These results showed that as in humans, rhesus macaques' social decisions are adaptive and modulated by social status and the cost associated with being prosocial.This article is part of the theme issue ‘Existence and prevalence of economic behaviours among non-human primates’.
Highlights
Most economic models rely on the assumption that agents are self-interested [1]
In the following three sessions, the order in which rewards were delivered was altered. In this so-called ‘reverse order’ condition, a prosocial choice resulted in the reward being given to the social partner before the actor; a selfish decision resulted in the reward being given to the actor after an approximately 3–4 s delay
Overall our results confirm that rhesus macaques do make prosocial decisions
Summary
Most economic models rely on the assumption that agents are self-interested [1]. many of our decisions are made in a social environment. The animal was first trained to discriminate between a target associated with a reward for self (referred as self-onlyreward target) and target associated with reward to neither individual (referred to as the reward-for-neither target) Once they reached 80% correct performance for three sessions, the prosocial rewarding target (a target associated with reward both for the actor and the other monkey in the second box—referred to as the reward-for-both target) was introduced, and its value was learned in trials in which it was pitted against the reward-forneither target. In the following three sessions, the order in which rewards were delivered was altered In this so-called ‘reverse order’ condition, a prosocial choice resulted in the reward being given to the social partner before the actor; a selfish decision resulted in the reward being given to the actor after an approximately 3–4 s delay. Repeated-measures correlation was conducted using rmcorr toolbox in R [50]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
