Painting of several people talking around a table

· Translated from the Spanish original · 中文 · 日本語

Psychology

The most enjoyable conversations have "pink noise," a study finds

Dartmouth analyzed more than 1,000 chats between strangers: those that balance going deep and changing topic are liked more and predict second dates.

What happened

Scientists at Dartmouth College in the United States found a signature of the conversations participants enjoy most: “pink noise.” The study, published in Science Advances, analyzed more than 1,000 conversations between strangers, who then answered questions about how it went. Chats with more pink noise were liked more, left people wanting to keep talking and created a greater sense of connection.

Why it matters

In physics, pink noise describes a system at the edge of chaos: you can stay at a stable point without being stuck there forever. First author Kathryn O’Nell said that in a conversation this lets you stay on a topic long enough to get to know each other without it going stale. Emily Finn, a neuroscientist and senior author, noted that being “in sync” can be very dull, so the usual metaphor falls short.

The number

More pink noise in a chat predicted which pairs in a speed-dating round met again for a second date.

Context

The team repeated the analysis with conversations between friends about formative experiences, and again more pink noise meant greater enjoyment. The authors admit that defining a good conversation is hard, and open their paper with a humorous comparison: it is hard to define, but you know it when you see it. The finding gives a measurable way to study them. Cicero already advised watching your tone, not showing boredom and not talking about yourself nonstop. On related topics, Mamífero published Coffee and less dementia after 43 years of follow-up; Electrical noise reveals electron pairing in a cuprate; Switching model versions doesn’t improve predictions and Year-long study links AI companions to lower well-being.

What’s next

One possible application is understanding why certain chats flow, but the data show an association between pattern and enjoyment, not a recipe. It remains to be tested whether training that dynamic improves the outcome.

Bottom line

The ideal conversation is neither fully synchronized nor randomly changing, but one that moves between the two.

Sources

  • The Guardian, report on the study published in Science Advances (accessed November 9, 2026).
  • Cover image: Louise Catherine Breslau, Public domain, via Wikimedia Commons

Edited by Rodrigo Cornejo. How we select and verify: who writes these notes.

Related notes

← All notes