Trust learning in the repeated trust game: A meta-analytic study.

Area of Science:

• Behavioral Economics
• Social Psychology
• Decision Science

Background:

• Trust is fundamental for cooperation, involving vulnerability and expectation of reciprocity.
• The Trust Game (TG) is a standard tool to measure trust in experimental settings.
• The repeated TG allows for studying trust learning through repeated interactions and reputation building.

Purpose of the Study:

• To meta-analytically identify factors influencing trust learning in the repeated Trust Game.
• To examine the impact of partner behavior, participant characteristics, and trustworthiness manipulations on learning.
• To provide a comprehensive overview of trust learning dynamics in repeated interactions.

Main Methods:

• Meta-analysis of 404 effect sizes from 68 studies, totaling over 8000 participants.
• Inclusion of studies utilizing the repeated Trust Game paradigm.
• Statistical analysis to determine the influence of various factors on trust learning.

Main Results:

• Partner's behavioral trustworthiness (reciprocation rate) was the strongest predictor of trust learning (β = 3.0).
• Manipulations of prior information about partner trustworthiness influenced learning, but only when related to morality.
• Trust learning occurred regardless of partner's group membership (ingroup vs. outgroup).

Conclusions:

• Behavioral trustworthiness is paramount for developing trust in repeated interactions.
• Pre-information effects on trust learning are contingent on its moral relevance.
• Group membership does not significantly alter learning from a partner's trustworthy behavior.