Ingroup bias in conversational memory: the role of nationalism in the saying-is-believing effect.

Abstract

Research on the saying-is-believing effect shows that individuals tune their messages to match the audience's attitude and subsequently exhibit memory bias in the same direction. However, its implications in the context of collective memory remain understudied. Drawing on social identity theory, prior research suggests that people reconstruct memories of their group's past selectively, emphasising positive aspects and minimising negative ones. Using a 2 × 2 factorial design, this study examined the saying-is-believing effect in national history by manipulating audience attitude (positive vs. negative) and message production (with vs. without message generation). Results show that the effect extends to national history, with participants tuning their messages and biasing their memory according to the audience's attitude. However, those high in nationalism showed neither message tuning nor memory bias. These findings suggest that for individuals high in nationalism, social identity motivations override the epistemic and relational motives that typically drive the saying-is-believing effect.