5. Discussion and conclusion
The findings of this study contribute to the nascent field of research on trust in online communities in general (e.g. Gamboa & Gonçalves, 2014, Shankar et al., 2002, Yang & Wang, 2015) and social trading platforms in particular (Doering et al., 2015; Pan et al., 2012). The first contribution pertains to the complementarity of cognition-based and affect-based signals of trustworthiness. Both configurations that build trust and thereby prompt copying decisions rely on cognitionbased and affect-based signals. This finding echoes previous studies on trust in offline contexts that establish the complementary nature of cognition-based and affect-based signals of trustworthiness in the process of building trust (Lewis & Weigert, 1985; Möllering, 2006). The same finding also suggests that the shift from offline to online contexts does not diminish the importance of this complementarity (O'Sullivan, 2015; Shankar et al., 2002). Clearly, for financial traders wishing to appear trustworthy, financial performance matters, but the study indicates that those traders must complement such cognitionbased signals of trustworthiness with affect-based signals.