Social Brain Hypothesis

The social brain hypothesis is the claim that primate brain evolution, particularly the enlargement of the neocortex, was driven primarily by the computational demands of social living rather than by ecological or technical problem-solving. The hypothesis was developed by Robin Dunbar and Leslie Aiello in the early 1990s as an alternative to the prevailing "technical intelligence" view that emphasized tool use and foraging complexity.

The core argument is that social groups are information-processing systems of extraordinary complexity. To function effectively in a group, an individual must track alliances, obligations, reputations, and hierarchies — a cognitive load that scales nonlinearly with group size. The hypothesis predicts that species with larger social groups should have larger neocortices relative to body size, a prediction that has been confirmed across primates and, more tentatively, in other social mammals.

Critics note that the hypothesis struggles to explain solitary primates with large neocortices, such as orangutans, and social mammals like elephants and cetaceans whose brain architecture diverges from the primate pattern. The hypothesis may therefore describe a primate-specific pathway to social cognition rather than a universal law of brain evolution. The relationship between Dunbar's number and the social brain hypothesis is direct: if social complexity drove brain expansion, then brain size should predict social group size, which is precisely the correlation Dunbar documented.

The social brain hypothesis is often presented as a theory about brain size. It is better understood as a theory about computational constraints: the brain is not merely larger in social species; it is differently organized, with expanded regions dedicated to face recognition, intention-reading, and emotional regulation. Size is a proxy for a structural transformation that we are only beginning to map. The real question is not whether social life made brains bigger, but whether it made them into social computers — and if so, what that means for the design of artificial systems that must navigate social worlds.