Standard economic and evolutionary models assume that humans are fundamentally selfish. On this view, any acts of prosociality—such as cooperation, giving, and other forms of altruism—result from covert attempts to avoid social injunctions against selfishness. However, even in the absence of social pressure, individuals routinely forego personal gain to share resources with others. Such anomalous giving cannot be accounted for by standard models of social behavior. Recent observations have suggested that, instead, prosocial behavior may reflect an intrinsic value placed on social ideals such as equity and charity. Here, we show that, consistent with this alternative account, making equitable interpersonal decisions engaged neural structures involved in computing subjective value, even when doing so required foregoing material resources. By contrast, making inequitable decisions produced activity in the anterior insula, a region linked to the experience of subjective disutility. Moreover, inequity-related insula response predicted individuals’ unwillingness to make inequitable choices. Together, these data suggest that prosocial behavior is not simply a response to external pressure, but instead represents an intrinsic, and intrinsically social, class of reward.

Representation and Understanding Studies in Cognitive Science 1975, Pages 83–102

Reflections on the formal description of behavior

Joseph D. Becker

This chapter focuses on the origins and interconnections of some of the concepts that are generally useful in describing the behavior of animals and machines. These concepts are—hierarchical organization of processes, branch points, information, spheres of influence, goals, resource conflicts, condition conflicts, temporal organization, executive bookkeeping, executive decision-making, and statistical information. It discusses that these concepts must be viewed as descriptive artifices rather than as mechanisms by which the observed behavior is brought about. The chapter also presents the problem of turning this set of concepts into a full-scale mathematics of behavioral systems, finding that this may be impossible owing to the lack of any generally applicable criteria for delimiting what a behavior is. There is no straightforward, absolute, canonical, or true description of a behavioral system; all behavioral descriptions are relative to a particular set of questions they are intended to answer.

http://www.sciencedirect.com/science/article/pii/B9780121085506500082