Dishonest collaboration in an intergroup context

Collaborations may sometimes increase the likelihood of engaging in dishonest behaviour. As yet, it remains unknown what factors contribute to this phenomenon. Here, we investigate whether it matters with whom people are collaborating for the extent to which they are dishonest. We aim to (I) replicate dishonest collaboration effects and (II) examine whether collaborating with an ingroup member (vs. outgroup member) amplifies dishonesty. In three pre-registered studies (N = 782), we used the sequential dyadic die-rolling paradigm in both student (Study 1: Lab context) and community samples (Studies 2-3: Field context). In this paradigm, two players form a dyad (either with an ingroup or an outgroup member) and earn winnings depending on both their own and the other's dice roll. Crucially, dice rolls are private, and players can inflate their winnings by misreporting their dice roll. Collectively, our studies replicate the dishonest collaboration effect, showing the robustness of collaborative dishonesty over samples and contexts. However, our studies suggest that dishonesty is similar when collaborating with an ingroup versus an outgroup member. Alternative processes are discussed that may play a role in collaborative dishonesty resulting in the absence of an intergroup effect.