Esprits non-humains: cognition animale, artificielle ou autre
Cognitio 2011
Colloque jeunes chercheurs en sciences cognitives
Montréal, les 3, 4, et 5 juillet 2011.
[ Retour à la liste des présentations ]
Social Evaluation by Nonhuman Primates
Mark Sheskin, Amy Skerry, Kimberly Ashayeri et Laurie Santos
Abstract:
Recent research has established that complex social evaluations may be performed by beings with minds far simpler than those of human adults. For examples, Hamlin et al. (2007) demonstrated that six month-old infants form preferences and expectations about others based on observed third-party interactions, and Call et al. (2004) found that chimpanzees distinguish between a human who fails to give a treat due to malice and one who fails due to incompetence (see Phillips et al. (2009) for a similar result with capuchin monkeys).
In this paper, we present results of a study to determine whether capuchin monkeys socially evaluate others based on how fairly they behave. Previous research has provided (controversial) evidence that capuchins may respond negatively to unfair outcomes (Brosnan & de Waal, 2003), which makes it plausible that they might negatively evaluate the social agents that provide unfair outcomes.
Our study first introduced subjects to pairs of experimenters who traded food for tokens with both the subject and another “model” monkey. One of these experimenters traded fairly, providing both the subject and the model with an identical reward. The other experimenter traded unfairly, providing the subject and the model with unequal rewards. After the training period, and without the model monkey present, each subject was given a new opportunity to trade with either the fair experimenter or the unfair experimenter. We found no tendency to avoid the unfair trader, indicating that previous evidence of nonhuman fairness concerns is best interpreted conservatively: nonhuman primates may respond negatively to unfair outcomes, but they do not seem to negatively evaluate the agents that provide those unfair outcomes. Thus, fairness concerns represent an area in which nonhuman minds show, at most, a simpler capacity than human minds.
Whereas previous research has focused on social behaviors produced by primates, or responses to specific instances of social behaviors by others, this study represents one of the first investigations of primate social evaluation (i.e., long-term preferences of others formed in response to social behaviors). Such research is an important part of charting the ways in which human and nonhuman minds are similar, and areas in which they might diverge. Future research should investigate additional instances of primate social evaluation, notably regarding helpful/harmful behaviors and dominant/submissive behaviors. These are domains in which nonhuman primates show strong concerns regarding individual behaviors (e.g., Warneken & Tomasello, 2006; Seyfarth et al., 2005), and so are likely candidates for domains in which they may form long-term preferences.
In this paper, we present results of a study to determine whether capuchin monkeys socially evaluate others based on how fairly they behave. Previous research has provided (controversial) evidence that capuchins may respond negatively to unfair outcomes (Brosnan & de Waal, 2003), which makes it plausible that they might negatively evaluate the social agents that provide unfair outcomes.
Our study first introduced subjects to pairs of experimenters who traded food for tokens with both the subject and another “model” monkey. One of these experimenters traded fairly, providing both the subject and the model with an identical reward. The other experimenter traded unfairly, providing the subject and the model with unequal rewards. After the training period, and without the model monkey present, each subject was given a new opportunity to trade with either the fair experimenter or the unfair experimenter. We found no tendency to avoid the unfair trader, indicating that previous evidence of nonhuman fairness concerns is best interpreted conservatively: nonhuman primates may respond negatively to unfair outcomes, but they do not seem to negatively evaluate the agents that provide those unfair outcomes. Thus, fairness concerns represent an area in which nonhuman minds show, at most, a simpler capacity than human minds.
Whereas previous research has focused on social behaviors produced by primates, or responses to specific instances of social behaviors by others, this study represents one of the first investigations of primate social evaluation (i.e., long-term preferences of others formed in response to social behaviors). Such research is an important part of charting the ways in which human and nonhuman minds are similar, and areas in which they might diverge. Future research should investigate additional instances of primate social evaluation, notably regarding helpful/harmful behaviors and dominant/submissive behaviors. These are domains in which nonhuman primates show strong concerns regarding individual behaviors (e.g., Warneken & Tomasello, 2006; Seyfarth et al., 2005), and so are likely candidates for domains in which they may form long-term preferences.