Humans are not unique in being social. Ants, bees, rats, parrots, and many other species form social groups, usually because members of the group are close kin or because group members stick with others for their own gain - for example, reducing the risk of predation by hiding in a group (Hamilton, 1971; van Schaik, 1982).

Human sociality, however is radically different from that of other social animals. For while we do form groups for some of the same reasons as other animals, we are also able to engage in forms of social interaction that are rarely seen elsewhere in the animal kingdom. An example is reciprocal altruism, or the mutual exchange of benefits (Trivers, 1971). The potential for gains from this kind of interaction are enormous: for example, it can reduce the risk of starvation if individuals with a food surplus give food to others in exchange for a promise of reciprocation when the tables are turned. This kind of interaction is common in humans, but almost nonexistent elsewhere in the animal kingdom (Hammerstein, 2003), outside of a few species, including some social carnivores (Bekoff et al., 1984).

One reason that has been proposed to account for humans' remarkable skill in social interaction, which in turn might have been a driving factor in the worldwide expansion of human populations, is the ability to make inferences about the goals and intentions and others (Tomasello et al., 2005). In reciprocal altruism, for example, exchange is stabilized if parties can correctly infer each other's intent to cooperate, and quickly unravels otherwise (e.g., Boyd, 1989). Arguably, humans are good at this kind of interaction because we reason beyond actions alone, to the cause and meaning of those actions. Humans’ skill at making inferences about goals and intentions based on the actions of others is pervasive, developing early in childhood (e.g., Lempers 1979; Butterworth, 1991) and extending beyond cooperation to realms such as language learning (Bates et al., 1995) and coordinated social interaction more generally (Desrochers et al., 1995; Tomasello & Camaioni, 1997). However, on tasks that are simple for human children to solve, adult chimps, our closest evolutionary relatives, routinely fail (Povinelli & Eddy, 1996).

In order to understand why humans are so good at inferring the intentions and goals of others, we must do two things:

1) Identify the ecological and social contexts in which these skills evolved; in other words, identify their function in ecological and social context; and

2) Elucidate the information-processing features, or form, of the mechanisms involved. These two goals are mutually interdependent, because aspects of the information-processing form of these mechanisms are likely to reflect the functional contexts in which they evolved. These premises are the foundation for the work for which I am doing.