WITHDRAWN: Four studies Yield limited evidence for prepared (disgust) learning via evaluative conditioning

Abstract

Prepared learning accounts suggest that specialized learning mechanisms increase the retention of associations linked to ancestrally-prevalent threats. Few studies have investigated specialized aversion learning for pathogen threats. In four pre-registered studies (N's = 515, 495, 164, 175), we employed an evaluative conditioning procedure to test whether foods (versus to non-foods) are more readily associated with negative content associated with pathogens than negative content not associated with pathogens. Participants saw negatively valenced (either pathogen-relevant or -irrelevant), neutral or positively-valenced stimuli paired with meats and plants (in Studies 1 and 2) and with meats and abstract shapes (in Studies 3 and 4). They then evaluated each stimulus explicitly via self-reports (Studies 1–4) and implicitly via an Affect Misattribution Procedure (Studies 3 and 4). Linear mixed models revealed general evaluative conditioning effects, but little evidence for specialized (implicit or explicit) learning for a food-pathogen association. However, across Studies 2–4, participants who saw pathogen-relevant (vs. -irrelevant) negative stimuli at any point during evaluative conditioning evaluated the conditioned stimuli more negatively, regardless of whether that stimulus had previously been paired with an unconditioned stimulus. Results speak against the hypothesis that evaluative conditioning effects are stronger for foods paired with disgust-eliciting images than for foods paired with other negative images.