Rules and theoretical prediction
In the experiment by Berg, Dickhaut and McCabe (1995), each of two participants received 10 dollars. The investor could send any portion to an anonymous trustee, and the amount sent was tripled on arrival. The trustee then decided how much to return to the investor. If both were rational and purely self-interested, game theory predicts that the trustee would return nothing and that the investor, anticipating this, would send nothing in the first place. Yet if both trust and reciprocate, both earn more, so the game captures situations in which trust raises efficiency.
Experimental results and measurement
In the original experiment, 30 of 32 investors sent money, on average about half. Trustees' returns varied: many returned more than was sent, while some returned nothing. In a meta-analysis of over 160 experiments, Johnson and Mislin (2011) reported that investors send about half their endowment on average and trustees return about a third of what they receive. Cox (2004) ran a dictator game alongside to show that part of sending reflects simple altruism rather than trust. Glaeser and colleagues (2000) reported that agreeing with the survey statement that most people can be trusted relates more to trustworthy behavior, that is, returning, than to sending.
Biological research and debate
Kosfeld and colleagues (2005) drew wide attention by reporting that participants given intranasal oxytocin sent more money in the trust game. Fehr (2009) argued that trust differs from mere risk-taking and that aversion to being betrayed shapes trusting behavior. However, Nave, Camerer and McCullough (2015), in a critical review, found that later oxytocin studies replicated poorly and cautioned against concluding that oxytocin increases trust. The trust game is widely used alongside the ultimatum game and research on reciprocal altruism to study human cooperation and social capital.
FAQ
- What is the trust game in simple terms?
- You send money to a partner, the money triples, and the partner decides how much to give back. How much you send measures your trust, and how much your partner returns measures reciprocity.
- How much do people actually send in the trust game?
- Meta-analysis shows that investors send about half their money on average and trustees return about a third of what they receive, far from the theoretical prediction of sending nothing.
Sources reviewed: 2026-10-26
References
- Berg, J., Dickhaut, J., & McCabe, K. (1995). Trust, reciprocity, and social history. Games and Economic Behavior, 10(1), 122–142. The original investment game — start here for the design and core results.
- Glaeser, E. L., Laibson, D. I., Scheinkman, J. A., & Soutter, C. L. (2000). Measuring trust. Quarterly Journal of Economics, 115(3), 811–846. Compares survey trust with game behavior — what trust measures really capture.
- Cox, J. C. (2004). How to identify trust and reciprocity. Games and Economic Behavior, 46(2), 260–281. Separates trust and reciprocity from altruism using dictator-game controls.
- Kosfeld, M., Heinrichs, M., Zak, P. J., Fischbacher, U., & Fehr, E. (2005). Oxytocin increases trust in humans. Nature, 435(7042), 673–676. The famous oxytocin study — the biology of trust.
- Fehr, E. (2009). On the economics and biology of trust. Journal of the European Economic Association, 7(2–3), 235–266. Argues trust is distinct from risk preference — economics and biology combined.
- Johnson, N. D., & Mislin, A. A. (2011). Trust games: A meta-analysis. Journal of Economic Psychology, 32(5), 865–889. A meta-analysis of over 160 experiments — typical sending and returning rates.
- Nave, G., Camerer, C., & McCullough, M. (2015). Does oxytocin increase trust in humans? A critical review of research. Perspectives on Psychological Science, 10(6), 772–789. A critical review showing oxytocin–trust findings replicate poorly.