Epley · Journal of experimental psychology. General 2014 · randomized controlled field and laboratory experiments · n=?

Mistakenly seeking solitude.

Cited 309 times in the scientific literature.

Level 2 - randomized trial

Series of randomized behavioral field and laboratory experiments (by design analogy, not clinical CEBM).

PubMed 25019381 · doi:10.1037/a0037323 · record verified 2026-08-26

What was done

Commuters on trains and buses were instructed to connect with a nearby stranger, remain disconnected, or commute as normal (Experiments 1a and 2a). Separate groups predicted the experience of these conditions (Experiments 1b and 2b). Further experiments assessed beliefs about others' interest in interacting (Experiments 3a and 3b) and how those beliefs impede learning (Experiments 4a and 4b). A laboratory waiting room experiment tested whether the recipient of an interaction had as positive an experience as the initiator (Experiment 5).

What was found

No numerical data, sample sizes, or statistical tests are reported in the abstract. Participants assigned to connect reported more positive (and equally productive) experiences than those assigned to solitude. Conversely, forecasters predicted the reverse outcome. In the laboratory, recipients of interaction experienced equal positivity compared to initiators.

Why it matters

This paper identifies a widespread forecasting error: people avoid casual interactions because they underestimate others' interest in connecting, thereby forfeiting a reliable source of well-being.

Limits

Sample size is not reported in the abstract. No quantitative effect sizes, variance measures, or demographics are provided. Results depend on adherence to experimental instructions and may reflect specific cultural norms regarding public transit interactions.

Cited by