Fairness Lessons from Capuchin Monkeys

brief communications Animal behaviour represented across all trials of both sessions Our study mainly concerned the second Fair refusal by capuchin for Brosnan and de Waal's monkeys, we find ability, showing that capuchin monkeys that there is no overall increase in rejection react negatively when another individual monkeys rate in the inequality-test and effort-control gets a better reward for the same or less effort conditions, and that the rate does not decline on a specific task. This finding suggests that rosnan and de Waal' report that Bcapuchin monkeys show evidence of a across sessions in the food-control condition precursors to inequity aversion are present (results not shown). in animals from which our lineage split mil- sense of fairness or 'inequity aversion' Although explanations of animal behav- lions of years ago. Although capuchins may because they rejected a less preferred reward iour in anthropomorphic terms are notori- be reacting somewhat differently from adult when they saw a partner monkey receive a ously prone to imprecision , if fairness' or humans, we have still learned something preferred reward for the same task. How- 'inequity aversion' mean anything in this about the behaviour's possible evolutionary ever, this does not show that monkeys are context, they surely imply that individuals trajectory. averse to inequity, only that they reject a reject rewards more often when they see Regarding the cross-cultural study, the lesser reward when better rewards are avail- another receive a better reward than when lowest mean offer by a proposer in the ulti- able. There are risks inherent in seeking the better reward is simply in view with no matum game was 26% of the total, whereas anthropomorphic explanations for non- one else there to consume it. The very similar the lowest modal offer was 15%, both by the human behaviour. levels and patterns of cucumber rejection in Machiguenga of Peru . Such relatively high In the 'inequality test', the monkeys the inequality-test and food-control condi- offers would not seem to be consistent with refused to exchange a token for a cucumber tions therefore contradict an account based completely selfish individuals who lack any slice (non-preferred reward) on 43% of trials on fairness or inequality. conception of fairness". when they saw a partner monkey receive a Clive D. L. Wynne As stated earlier , although the mere pres- preferred grape reward for the same effort. Department of Psychology, University of Florida, ence of a higher-value reward affects the However, in the 'food control' condition, in Box 112250, Gainesville, Florida 32611, USA capuchins, their reaction is not the same as which the partner was not present, these e-mail: wynne@ufl.edu when a conspecific receives the higher-value same monkeys were just as likely to refuse the 1. Brosnan, S. F. & de Waal, F. B. M. Nature 425, 297-299 reward. To ignore the differences between cucumber slice when they saw a grape placed (2003). 2. Wynne, C. D. L. Do Animals Think? (Princeton Univ. Press, in the inequality test and the food-control test where the partner normally sat (49% the press). s unwarranted. Our Fig. 1 does not permit refusals). There can be nothing inequitous any conclusions about the effect of the food- about receiving a non-preferred reward if Brosnan and de Waal reply - We have control test and was not used for this pur- nobody is receiving anything better. In the shown' that animals compare their own pose; it is the data in our Fig. 2 that inspired food-control condition, the monkeys are rewards with those of others, and accept or our claim. refusing the non-preferred reward simply reject rewards according to their relative The frequency of refusals across trials because they can see that a better reward is value. Our aim was not to demonstrate increases when a partner receives the reward potentially available. This is therefore the that capuchin monkeys make a human and decreases when a reward is merely visible. most parsimonious explanation for their response to inequality, but rather to eluci- The conservative statistic we chose did not refusal to accept the non-preferred reward date evolutionary precursors to inequity allow significance (P