Chen et al. (1) raise several issues with the data and interpretations published in De Dreu et al. (2). We appreciate their constructive remarks. Here, we address their concerns and explain why our data do, in fact, show that oxytocin promotes human ethnocentrism.

For starters, Chen et al. (1) believe we oversimplify past work by referring to oxytocin as an “indiscriminate love drug” (1). Although one may take issue with the catchy wording, the view of oxytocin as motivating general prosocial tendencies still dominates the scholarly literatures: “oxytocin is not only the moral molecule, but also a primary element of the physiology of love” (p. 63 in ref. 3). We hope that our finding that oxytocin promotes in-group favoritism and not widespread goodwill will deter future use of such generic labels.

Oxytocin’s nicknames aside, we agree with Chen et al. (1) that goodwill is not a fixed pie. For precisely that reason, we designed our experiments so that expressions of goodwill to the in-group were independent of expressions of goodwill to the out-group. We observed that oxytocin increased in-group favoritism in all five experiments. In experiment 1 and to a lesser extent, experiment 2, we also observed that oxytocin increased out-group derogation. We interpreted this latter observation with caution, because in subsequent experiments, oxytocin neither increased nor decreased out-group derogation.

In all, oxytocin-induced in-group favoritism is robust. Oxytocin increased in-group favoritism in all five experiments and never increased out-group favoritism (at times, it even increased out-group derogation). We emphasize that these findings permit the conclusion that oxytocin promotes ethnocentrism—the tendency to view one’s group as centrally important and superior to other groups. Ethnocentrism does not necessarily involve out-group derogation and has been shown to surface more commonly through in-group favoritism than out-group derogation (2, 4).

Chen et al. (1) correctly note that we did not show that oxytocin creates intergroup conflict. However, by extending goodwill to the in-group but not the out-group, one creates intergroup bias: in-group members get relatively better treatment and receive benefits more readily than out-group members (2, 4). Extensive literature shows that even very subtle forms of in-group favoritism create intergroup tension and conflict (4). It thus follows that, because oxytocin promotes in-group favoritism and intergroup bias, oxytocin indirectly contributes to intergroup tension, conflict, and perhaps, even violence.

Goodwill is not a fixed pie, but oxytocin-induced goodwill is limited to one’s in-group and not extended to out-groups. We posit that this has important thought-provoking implications for intergroup tension and conflict and invites new research into the neuroendocrinology of human conflict and cooperation within as well as between groups. Without downplaying the possible roles of vasopressin and testosterone (1), our findings (2, 5) showed that oxytocin’s role in intergroup competition and conflict should not be underestimated.