Unfakeable Facial Configurations Affect Strategic Choices in Trust Games with or without Information about Past Behavior

Background: Many human interactions are built on trust, so widespread confidence in first impressions generally favors individuals with trustworthy-looking appearances. However, few studies have explicitly examined: 1) the contribution of unfakeable facial features to trust-based decisions, and 2) how these cues are integrated with information about past behavior. Methodology/Principal Findings: Using highly controlled stimuli and an improved experimental procedure, we show that unfakeable facial features associated with the appearance of trustworthiness attract higher investments in trust games. The facial trustworthiness premium is large for decisions based solely on faces, with trustworthy identities attracting 42% more money (Study 1), and remains significant though reduced to 6% when reputational information is also available (Study 2). The face trustworthiness premium persists with real (rather than virtual) currency and when higher payoffs are at stake (Study 3). Conclusions/Significance: Our results demonstrate that cooperation may be affected not only by controllable appearance cues (e. g., clothing, facial expressions) as shown previously, but also by features that are impossible to mimic (e. g., individual facial structure). This unfakeable face trustworthiness effect is not limited to the rare situations where people lack any information about their partners, but survives in richer environments where relevant details about partner past behavior are available.