When witnessing norm violations, initially uninvolved observers often invest own resources to restore justice, either by punishing the offender or compensating the victim. While explicit instructions can influence participants’ cognitive focus on the highlighted party and their subsequent actions, the current study investigated more implicit means of shaping third-party decision-making. In four preregistered experiments (N = 436), participants played an incentivized third-party justice game. We examined how subtle social cues (face visibility, gaze direction, emotional expression, personal information) affect preferences for (i) intervention type (punish, compensate, do nothing) and (ii) its intensity.

Controlling for baseline behavior, justice sensitivity, and empathic concern, Experiment 1 showed that displaying the offender’s portrait photo increased punishment, while showing the victim’s photo increased compensation; gaze direction of the displayed person had no additional effect. In Experiment 2, when both photos were presented simultaneously, gaze direction did not affect decision preferences, but compensation intensity when the victim displayed direct (vs. averted) gaze. Experiment 3 replicated the visibility effect of Experiment 1 and showed that sad (vs. neutral) facial expressions did not sway decision-making. Shifting from visual to verbal cues, Experiment 4 revealed that written information (name and personal fact) about either party did not affect third-party decisions. Across experiments, baseline preferences and transgression unfairness consistently influenced intervention tendencies.

We conclude that social cues, particularly those capturing bottom-up social attention (e.g., faces), can shape third-party decisions. Findings are also discussed in the context of the identifiable victim (and offender) effect.