The Science of Adult Friendship and Loneliness: Why the Signal Isn't a Verdict

Key findings

1. 2006

   Loneliness fits an evolutionary account: like hunger or thirst, it is an aversive signal that motivated our ancestors to maintain and repair social bonds — a functional mechanism, not a character defect of the person who feels it. ↗

2. 2014

   Perceived social isolation triggers implicit hypervigilance for social threat, producing attentional, confirmatory and memory biases in how social information is processed — which can make the social world look more hostile than it is. ↗

3. 2015

   In high-density EEG work with colleagues, lonely participants' brains differentiated social threat images within about 116 ms of stimulus onset versus about 252 ms in nonlonely participants — direct evidence that the threat bias operates automatically, below deliberate control. ↗

Key findings

1. 2010

   Across 148 studies and 308,849 participants, people with stronger social relationships had a 50% increased likelihood of survival — an influence comparable with quitting smoking and exceeding well-established risk factors such as obesity and physical inactivity. ↗

2. 2015

   In a meta-analysis of 70 studies, loneliness was associated with a 26% increased likelihood of mortality, social isolation with 29%, and living alone with 32% — with the effects strongest in samples under age 65, countering the idea that disconnection only endangers the old. ↗

Key findings

1. 2018

   Tracking adults who had recently moved and first-year students, friendship developed along a dosage curve: roughly 50 shared hours to move from acquaintance to casual friend, about 90 to friend, and more than 200 to close friend. ↗

2. 2018

   Hours alone weren't enough: closeness grew with time spent hanging out, joking around and in meaningful conversation, while hours merely spent together at work or in class did little to turn colleagues into friends. ↗

Key findings

1. 2018

   After conversations, people systematically underestimated how much their partners liked them and enjoyed their company — a gap driven partly by harsh self-judgment of their own conversational performance, present in short, medium and long conversations alike. ↗

2. 2018

   The gap wasn't just a first-impressions artifact: first-year students tracked across an academic year of living with a dormmate continued to underestimate how much that dormmate liked them for months. ↗

3. 2021

   With colleagues, she showed the liking gap extends beyond dyads: in small groups and engineering teams, members underestimated how much their teammates liked them, and larger gaps related to less willingness to speak openly and collaborate. ↗