Figure 1

Experimental task structure. (A) The competition with outcome feedback manipulation is represented. In a between-participants design, participants were randomly assigned to receive one of four types of outcome feedback (victory vs. defeat vs. uncertain competitive outcome vs. no competition (control)). We used a standard version of the Deese–Roediger–McDermott (DRM) paradigm as the competitive task. At the end of the competition, participants were immediately given information about their performance. Specifically, they received one of three types of outcome feedback (victory vs. defeat vs. uncertain competitive outcome). After receiving outcome feedback, participants were asked to provide a rating on a nine-point Likert scale of their perceived competition (the intensity they felt about the competition) (1 = not at all, 9 = extremely), self-confidence (1 = not at all confident, 9 = completely confident), the degree of closeness that they felt toward their opponents (1 = not at all close, 9 = completely close) and the level of trust that they felt toward their opponents (1 = no trust at all, 9 = complete trust). (B) After completing the competition stage, participants played a modified version of a two-person public goods game (PGG). Participants were asked to indicate their own willingness to cooperate as well as their predictions about their opponents’ willingness to cooperate on a 9-point rating scale (1 = not at all to 9 = very willing). Meanwhile, participants had to decide how many tokens they wished to contribute to the public good and also predict how many tokens their opponents would contribute. MPCR, marginal per-capita return.