Sampling of Social Information: Decisions from Experience in Bargaining

Nadine FleischhutMax Planck Institute for Human Development, Berlin, Germany
Florian ArtingerMax Planck Institute for Human Development, Berlin, Germany
Sebastian OlschewskiDepartment of Psychology, University Basel, Switzerland
Kirsten VolzCentre for Integrative Neuroscience, Tübingen, Germany
Ralph HertwigMax Planck Institute for Human Development, Berlin, Germany

Abstract

Whenever people depend on others, information about their likely behavior is important for the pursuit of goals. We investigated how the way social information is learned, by description or experience, affects offers in a bargaining situation. Participants learned how often each offer had previously been accepted or rejected, either as probability information or by experiencing others’ responses. When participants had to draw a representative sample of responses, the proportion of risky offers decreased under social experience, resulting in a gap to the description condition. When participants could terminate sampling whenever they wanted, however, no description–experience gap was observed. The sampling pattern suggests that participants disregarded probability information and relied on the allocation as proxy for risk. Accordingly, a certain amount of social experience seems necessary to overwrite initial expectations and change behavior. Under what conditions people search for social information is crucial for understanding how and when it impacts behavior.

Files

Sampling of Social Information: Decisions from Experience in Bargaining (420 KB)



Back to Table of Contents