Dunbar's Number

Dunbar's Number refers to the cognitive limit on the number of stable social relationships a human can maintain — estimated at approximately 150 by evolutionary anthropologist Robin Dunbar based on his research throughout the 1990s. The figure is not arbitrary: Dunbar derived it from the correlation between neocortex size and typical social group size across primate species. Humans have the largest neocortex relative to brain size, and 150 represents the social group size that our cognitive architecture can support.

The number describes a specific kind of relationship: one in which you know who a person is, how they relate to others in your network, and enough about their circumstances to interact with them meaningfully without requiring re-introduction. Dunbar identified concentric circles of relationship quality around this core number — roughly five intimate relationships, fifteen close friends, fifty regular contacts, and 150 meaningful acquaintances. Beyond 150, the research suggests, our ability to model other people's situations and maintain coherent social knowledge degrades significantly.

The challenge social media poses to this framework is that it creates the infrastructure for maintaining apparent contact with vastly more people. Facebook's average user has several hundred connections; influencers and public figures have millions. The question Dunbar's research raises is not whether you can follow these people, but whether following them constitutes a relationship in any meaningful sense, and what the cognitive cost of attempting to maintain the illusion of relationship with so many people actually is.

The evidence suggests that digital connection does not expand the underlying cognitive capacity. Studies examining the social networks of heavy social media users find that their number of close, emotionally supportive relationships is not larger than those of lighter users — and in some cases is smaller. What expands with platform use is not the number of genuine relationships but the ambient noise of peripheral connections, notifications, and passive content from people whose lives are observed but not engaged with.

The practical implication is a form of social opportunity cost. Cognitive and emotional energy available for relationships is finite. Time and attention spent managing hundreds of weak-tie digital connections — monitoring feeds, responding to notifications, maintaining the performance of sociability — is time and attention not spent deepening the 150 relationships that, by Dunbar's account, represent the actual capacity of human social cognition. Social media may not be making us more connected in the sense that matters. It may be distributing a fixed relational resource more thinly across a larger surface area.