What is it about?
People can miss opportunities to cooperate even when everyone stands to gain. We studied 940 people in small online economies where they could produce goods and trade for real money. Participants had different productive strengths, so exchanging goods could make both partners better off. Everyone spoke English. Yet exchange broke down when the greatest gains required trading across racial lines. Many participants tried to produce everything themselves, leaving gains from cooperation unrealized. A brief communication guide helped people speak as a team, explain what they needed and could offer, and keep conversations friendly and focused on progress. It increased exchange and closed the performance gap in these economies. It had little effect where equally profitable same-race partners were available.
Featured Image
Photo by Clay Banks on Unsplash
Why is it important?
The financial reasons to cooperate were already there. Helping people communicate enabled them to act on those incentives and earn more. Language can make needs visible, bring an offer to someone's attention, and help people coordinate a shared effort. For managers, educators, and policymakers, this suggests a practical avenue to explore: pair opportunities to work together with guidance that helps people start and sustain productive conversations.
Perspectives
This is the paper I've always wanted to write. With Simon Siegenthaler and Bart J. Wilson, I brought together Adam Smith, social psychology, and computational linguistics around a question that matters to me: how do people turn opportunities into cooperation? For me, this is the hopeful part: we can do more than document the opportunities people miss. We can find ways to help them seize those opportunities.
Sheen Levine
Stanford University
Read the Original
This page is a summary of: Linguistic intervention unlocks the economic gains of diversity, Proceedings of the National Academy of Sciences, October 2026, Proceedings of the National Academy of Sciences,
DOI: 10.1073/pnas.2534825123.
You can read the full text:
Contributors
The following have contributed to this page
