Robin Dunbar, professor of evolutionary psychology at Oxford University, has proposed that humans have a cognitive limit on the number of stable social relationships they can maintain, typically cited as approximately 150, with an inner circle of roughly five people who provide the deepest level of support. His research suggests that it is this inner circle, not the broader network, that provides the health-protective benefits of social connection. A person with 500 social media connections but no close confidants is, by every health metric, more isolated than a person with three deep friendships.
The Buffering Hypothesis
The social support buffering hypothesis, developed by Sheldon Cohen and colleagues, proposes that close social relationships protect health primarily by attenuating the physiological stress response to challenging events. When a stressful event occurs, people with strong social support show smaller increases in blood pressure, cortisol, and pro-inflammatory cytokines than those without such support. The buffer is not simply the absence of stress; it is the presence of trusted others whose support reduces the perceived threat magnitude of stressors.
Perceived Support Versus Received Support
A counterintuitive finding in the support literature is that perceived support, the belief that support would be available if needed, is often more strongly associated with health outcomes than received support, the actual support someone has received. This suggests that the health benefit of close relationships comes partly from the psychological safety of knowing support exists, independent of whether you actually use it. Building relationships in which you feel genuinely supported and in which you genuinely support others creates this sense of psychological safety.
Maintaining Friendships in Adulthood
Research on adult friendship development shows that the main barrier to maintaining and developing close friendships in adulthood is not desire but unplanned contact. Childhood and adolescent friendships typically form through repeated unplanned proximity in school settings. Adult friendships must be actively scheduled. Studies by Jeffrey Hall at the University of Kansas found that it takes approximately 50 hours of contact to move from acquaintance to casual friend, and 200 hours to move to close friend. These hours do not accumulate quickly without deliberate investment.
What Friendship Offers That No Other Intervention Does
Close friendship provides a specific form of social support that differs qualitatively from family support or professional support. Friendship is voluntary and non-obligatory, which means it operates on a reciprocity model that generates genuine mutual valuing. Research comparing different sources of social support consistently finds that friendship, specifically, predicts happiness and wellbeing beyond what family relationships alone explain. For older adults, friendship quality is one of the strongest predictors of subjective wellbeing and physical health available.