Can we prevent social identity switches? An experimental-computational investigation

Previous studies suggested that social identity switches are rapid and highly effective, raising the question of whether people can intentionally control such switches. In two studies, we tested if participants could exert top-down control to prevent a social identity switch triggered by the experimental context. In Study 1, participants (N = 198) were given a writing task aimed at prompting a switch from their parent identity to their feminist identity. Before the prompt, half of the participants (the experimental group) were instructed to remain in their parent identity, avoiding an identity switch; the control group was not given such instructions. We found no significant difference between the groups in either self-reported salience or the implicit computational measure of salience based on participants' linguistic style, both measures suggesting a switch in both groups. Study 2 (N = 380) followed the same design but included a monetary incentive to prevent the switch in the experimental group. The groups differed significantly in their self-reported salience but not in the implicit measure, which suggests limited ability to avoid the switch even when participants report being able to do so. These results point to limited intentional control over exogenously triggered identity switches, with important practical implications.