Robert Waldinger stood on the TED stage in 2015 and delivered the conclusion of the longest-running study of human happiness ever conducted. Seventy-five years of data. 724 men tracked from adolescence to old age. Blood samples, brain scans, questionnaires, interviews. And the finding was almost embarrassingly simple: good relationships keep us healthier and happier. Full stop.
What the Research Actually Shows
The Harvard Study of Adult Development, which Waldinger now directs, has been running since 1938. It tracked two groups: Harvard sophomores and boys from Boston's poorest neighbourhoods. The most consistent predictor of health and happiness at age 80 wasn't cholesterol at 50, or career success at 40, or wealth at any age. It was relationship satisfaction at 50.
People in warm, secure relationships lived longer, stayed sharper mentally, and reported greater life satisfaction. People in cold or conflictual relationships declined faster, both physically and cognitively. Loneliness, the study found, is as dangerous as smoking or alcoholism.
This isn't unique to Harvard's data. A 2010 meta-analysis by Julianne Holt-Lunstad at Brigham Young University reviewed 148 studies covering more than 300,000 people. Social connection increased the odds of survival by 50%. The effect size was comparable to quitting a 15-cigarette-a-day habit.
Romantic Love and Well-Being
Romantic partnerships are among the strongest predictors of happiness, but the relationship is complicated. Marriage, on average, boosts life satisfaction, but the boost is temporary. Richard Lucas's longitudinal research found that married people experienced a spike in satisfaction around the wedding, then gradually returned toward their baseline within about two years.
The quality of the marriage matters far more than the fact of it. John Gottman's lab at the University of Washington can predict divorce with over 90% accuracy by observing how couples handle conflict. The critical variable isn't whether couples fight. It's how they fight. Contempt, the single most corrosive behaviour Gottman identified, predicts relationship failure more reliably than any other factor.
Gottman's "magic ratio" is five positive interactions for every negative one. Couples below that threshold drift toward dissolution. Above it, relationships remain stable and satisfying. Love, in the data, looks less like passion and more like a sustained pattern of small kindnesses.
Beyond Romance
Friendship is the undervalued pillar of well-being research. Robin Dunbar's work at Oxford suggests we maintain roughly five close friendships, 15 good friends, and 50 casual friends at any given time. These numbers appear to be constrained by the size of the neocortex, a cognitive limit that social media hasn't expanded despite giving us thousands of "connections."
Close friendships provide something distinct from romantic love: a space for the self that exists outside the pressures of partnership and family. William Rawlins at Ohio University found that the friendships most conducive to well-being shared four qualities: freedom to be oneself, reciprocal affection, mutual trust, and the ability to talk openly.
Community ties matter too. Weak ties, the acquaintances and casual connections we encounter at the corner shop, the gym, the school gate, contribute to a sense of belonging that close relationships alone can't provide. Loneliness often isn't about lacking intimates. It's about lacking a web of peripheral connections that make you feel embedded in a community.
Designing for Connection
The practical implications are clearer than most well-being research delivers. Invest in your relationships with the same seriousness you invest in your career. Schedule time for the people who matter, and protect that time from encroachment. Replace passive socialising (scrolling through feeds) with active socialising (face-to-face conversation, shared activities, collaborative projects).
Learn to repair. Every close relationship involves friction. The difference between relationships that thrive and those that wither is the willingness to acknowledge harm, apologise sincerely, and adjust behaviour. Repair is a skill, not a personality trait. It can be practised.
Masamichi Souzou treats human connection as the central variable in well-being design. Tools, environments, and programmes that strengthen the quality of people's relationships aren't supplements to the core work. They are the core work.