Abstract
When people make perceptual or evaluative judgments on a series of stimuli, the judgment on one trial is often influenced by the trials that came before (e.g. Matthews and Stewart 2007, Kondo et al. 2012). In such cases, it is difficult to characterize the separate influences of the previous response, the previous stimulus, or both, since these two effects are usually highly correlated. We attempted to solve this problem by creating an experimental design that allowed us to independently measure the effects of the previous stimulus and the previous response. 112 female faces were displayed sequentially for 4 seconds each. For each face, participants either made a face attractiveness or hair darkness rating on a 1-8 Likert scale. Importantly, these judgments alternated such that all attractiveness judgments were preceded by hair darkness judgments, and vice versa. Because hair darkness was uncorrelated with face attractiveness, we were able to regress the attractiveness rating in a given trial on both the previous response and previous mean face attractiveness (values for which were determined based on data from 28 independent raters). Across 30 subjects, we found that the mean attractiveness of the previous face negatively predicted the current face attractiveness judgment; in contrast, we saw no influence of the previous response. Our results suggest that the attractiveness of the previous face has a contrastive effect on the current attractiveness judgment given to a face, and that assimilative effects due to the previous response may only operate across trials of the same judgment type. The alternating design we use here offers a method for researchers to independently characterize influences due to the previous response and the previous stimulus in sequential judgment tasks.
Meeting abstract presented at VSS 2014