# Dunbar's Number > [!summary] > Dunbar's number proposes a cognitive limit of approximately 150 stable social relationships that humans can maintain, derived from the correlation between neocortex size and group size in primates. Dunbar's number is a proposed cognitive limit on the number of people with whom one can maintain stable social relationships—relationships in which an individual knows who each person is and how each person relates to every other person. This number was first proposed in the 1990s by British anthropologist Robin Dunbar, who discovered a correlation between primate brain size and average social group size. ## The Core Finding By examining the relationship between neocortex size and group size in 38 non-human primate genera, Dunbar found that the neocortex ratio (measured against the rest of the brain excluding the neocortex) accounted for 76% of the variance in mean group size. Using regression analysis on this primate data, Dunbar predicted a human "mean group size" of 148, casually rounded to 150. Notably, he considered this exploratory because of a large error measure with a 95% confidence interval of 100 to 230. ## Evidence Base Dunbar tested his prediction against ethnographic data from hunter-gatherer societies and found that human communities consistently organized themselves around this number. The 150-person group appeared repeatedly across diverse contexts: - Hunter-gatherer community sizes (the "clan" level between band and tribe) - Neolithic farming villages - Roman military units (the company) - Modern military companies (100-200 members) - Hutterite settlement split points - Anglo-Saxon villages in the Domesday Book - Gore-Tex factory sizes (deliberately capped by company policy) More recently, the pattern has been observed in telephone calling networks, Facebook friend groups, Christmas card lists, church congregations, and online gaming communities. ## The Underlying Mechanism According to the [[Social Brain Hypothesis]], the constraint comes from the cognitive processing required to maintain social relationships. Dunbar theorized that maintaining a group of 150 requires knowing who each person is, how they relate to you, and how they relate to each other. This demands significant cognitive capacity and time investment—Dunbar speculated that maintaining 150 relationships might require up to 42% of a group's time devoted to "social grooming" (conversation, shared activities, mutual support). This time constraint means that only groups under intense survival pressure—such as subsistence villages, nomadic tribes, and military units—typically achieve and maintain the 150-member mark. ## Nested Structure Dunbar's number is not a single threshold but the outermost layer of [[Nested Social Layers]]. Human social networks organize themselves into concentric circles with a consistent scaling ratio of approximately 3: - Support clique: ~5 people (closest relationships) - Sympathy group: ~15 people (good friends) - Affinity group: ~50 people (friends) - Active network: ~150 people (meaningful relationships) - Acquaintances: ~500 people - Recognizable faces: ~1500 people Each layer represents declining emotional closeness and contact frequency, with approximately 40% of social time devoted to the innermost circle of 5. ## Relevance to Institutional Design The number has significant implications for how we design organizations and communities. When institutions grow beyond 150 members without subdividing into smaller units, several problems emerge: - Loss of cohesion and shared identity - Breakdown of informal communication networks - Need for formal hierarchies and bureaucratic structures - Reduced trust and reciprocity - Difficulty maintaining institutional culture However, [[Why Large Institutions Don't Subdivide]] reveals that natural subdivision doesn't occur spontaneously in modern organizations. Unlike traditional communities that split through [[Fission-Fusion Dynamics]], modern institutions face economic incentives, functional specialization, and bureaucratic structures that prevent natural fragmentation. Successful organizations that maintain cohesion either remain under 150 members or deliberately subdivide into smaller units. [[Gore Associates Organizational Model]] famously builds new factories whenever a facility approaches 150 employees. Military organizations worldwide organize around companies of 100-200 soldiers. Business universities illustrate this dynamic: at 150-250 people, institutional identity and social identity overlap—everyone knows everyone. At 2000-3000, this natural cohesion dissolves, and no authentic social unit emerges to replace it. The [[Institutional Subdivision Principle]] proposes that large institutions must deliberately create such units rather than expecting them to form naturally. ## Current Status and Controversies See [[Critiques of Dunbar's Number]] for detailed examination of challenges to this theory. The number remains influential but contested, with recent reanalyses suggesting both larger estimates (290) and such wide confidence intervals (2-336) that some researchers argue no single number can be specified. ## See Also - [[Social Brain Hypothesis]] - [[Nested Social Layers]] - [[Critiques of Dunbar's Number]] - [[Dunbar 1992 - Neocortex Size and Group Size]] - [[Why Large Institutions Don't Subdivide]] - [[Institutional Subdivision Principle]] - [[Gore Associates Organizational Model]]