We Are Bad at Predicting How Talking to Strangers Will Feel
Psychologists have spent more than a decade asking people to talk to strangers and then asking how it went. The results point the same way with unusual consistency.
In short
Across commuter trains, coffee shops and lab conversations, people expected talking to a stranger to be awkward, unwelcome or shallow. When they actually did it, it was usually more pleasant than they predicted, the other person liked them more than they thought, and deeper topics went better than small talk. The catch: most of this research measures short-term mood in face-to-face settings, mostly in North America and Britain.
There is a reason this research exists. Ask people whether they would rather sit quietly on their commute or strike up a conversation with the person next to them, and most will choose silence. That is a reasonable-sounding choice. The experiments below tested whether it is the right one.
What the studies found
-
Commuters who talked had a better journey than those who kept to themselves
In studies on Chicago-area trains and buses, Nicholas Epley and Juliana Schroeder asked some commuters to talk to a stranger, others to keep to themselves, and others to commute as normal. Those who talked reported a more positive journey. Before the experiment, people predicted the opposite.1
The researchers also found a likely reason for the mistake: people assumed others would not want to talk to them. In fact, the strangers were generally receptive.
Field experiments with commuters · Journal of Experimental Psychology: General, 2014 -
A real moment with a barista made people feel better than an efficient one
Gillian Sandstrom and Elizabeth Dunn asked coffee-shop customers either to have a genuine, brief social interaction with the cashier, or to be as efficient as possible. The first group left in a better mood and with a greater sense of belonging.2 The interaction cost a few seconds.
Field experiment · Social Psychological and Personality Science, 2014 -
"Weak ties" count for more than we assume
In a separate study, the same researchers found that on days when people had more interactions with acquaintances and casual contacts, the classmate you nod to or the neighbour you chat with, they tended to feel happier and more connected, over and above their interactions with close friends.3
Diary studies · Personality and Social Psychology Bulletin, 2014 -
After a first conversation, people underestimate how much they were liked
Erica Boothby and colleagues had pairs of strangers talk, then asked each person how much they liked their partner and how much they thought their partner liked them. People consistently believed they were liked less than they actually were. The researchers called it the "liking gap".4 We seem to replay our own awkward moments and miss the other person's warmth.
Lab and field studies · Psychological Science, 2018 -
Deep questions go better than people expect
Michael Kardas, Amit Kumar and Nicholas Epley asked strangers to discuss either shallow or deep questions, such as what they are most grateful for, or a time they cried in front of someone. People expected deep conversations to be more awkward and less connecting than they turned out to be. The researchers argue that this miscalibration keeps many conversations shallower than both people would like.5
Series of experiments · Journal of Personality and Social Psychology, 2022 -
Practice shrinks the fear
In a week-long study, Sandstrom, Boothby and Gus Cooney had participants use an app that set them a "scavenger hunt" of strangers to talk to. Over the week, people's worries about being rejected or not knowing what to say went down, and their predictions about how conversations would go became more accurate.6
Intervention study · Journal of Experimental Social Psychology, 2022
Why our predictions are wrong
The common thread is not that strangers are wonderful. It is that our forecasts are skewed in a predictable direction. We overestimate how awkward the start will be, underestimate how interested the other person is, and assume everyone else prefers silence. Because we act on those forecasts, by staying quiet, we rarely collect the evidence that would correct them. It is a self-sealing mistake.
What this research does not show
It is worth being honest about the limits. Most of these studies measure mood minutes or hours after a conversation, not long-term wellbeing. Most were run in face-to-face settings, in the US, Canada and the UK, with participants who had agreed to take part in a study. None of them show that talking to strangers online produces the same effects, and none of them suggest that every stranger is safe to talk to.
What they do show is narrower and still useful: if your reason for avoiding a conversation is "they won't want to talk" or "it'll be awkward", the evidence says you are probably more pessimistic than you need to be.
Which leaves an interesting question. If our instincts about strangers are this reliably wrong, what other social predictions are we getting wrong in the same direction, and never finding out?
References
- Epley, N., & Schroeder, J. (2014). Mistakenly seeking solitude. Journal of Experimental Psychology: General, 143(5).
- Sandstrom, G. M., & Dunn, E. W. (2014). Is efficiency overrated? Minimal social interactions lead to belonging and positive affect. Social Psychological and Personality Science, 5(4).
- Sandstrom, G. M., & Dunn, E. W. (2014). Social interactions and well-being: The surprising power of weak ties. Personality and Social Psychology Bulletin, 40(7).
- Boothby, E. J., Cooney, G., Sandstrom, G. M., & Clark, M. S. (2018). The liking gap in conversations: Do people like us more than we think? Psychological Science, 29(11).
- Kardas, M., Kumar, A., & Epley, N. (2022). Overly shallow?: Miscalibrated expectations create a barrier to deeper conversation. Journal of Personality and Social Psychology, 122(3).
- Sandstrom, G. M., Boothby, E. J., & Cooney, G. (2022). Talking to strangers: A week-long intervention reduces psychological barriers to social connection. Journal of Experimental Social Psychology, 102.