Limits on the generalizability of the relationship between state anxiety and categorization: Moderation by feature salience

We investigated the impact of state anxiety on categorization using adapted stimuli from two conflicting previous lines of research. Based on the available literature, we hypothesized that the relationship between categorical width and anxiety would be stimuli-specific. A measure of categorical change was also included in which participants re-categorized stimuli on a repeated trial. State anxiety was quasi-experimentally varied by sampling students in various classroom situations (N = 160) and dental patients awaiting various dental procedures (N = 69). Our results ran counter to hypotheses but nonetheless suggested that the relationship between categorization and state anxiety can vary depending on the stimuli used. State anxiety was associated with more categorical change when using stimuli with salient categorical features, while it was associated with less categorical change when using stimuli with less salient features. Similarly, irrespective of anxiety, participants consistently made more changes to their initial categorizations when using the salient than the non-salient stimuli. Our results failed to corroborate findings of anxiety-induced alterations in categorical width previously reported. The data add to the literature which places limits on the generalizability of category formation across situations and suggests that the relationship between state anxiety and categorization can be moderated by the obviousness of features available for categorical placement.