Abstract
In visual categorization studies, participants are frequently trained to a specified level of competency with respect to their accuracy (Scott, Tanaka, Sheinberg, & Curran, 2008; Tanaka, Curran, & Sheinberg, 2005). This approach assumes that once the training criterion has been met, participants are equated in performance accuracy. However, less attention has been given to how individual differences in the rate at which a participant learns a visual category predicts subsequent post-acquisition performance. In this study, we investigate the trainability of participants as a predictor of their ability to retain category knowledge.
In Phase 1, participants were trained to categorize images of four species of warblers (Capemay, Townsend, Prairie, and Magnolia) to a 90% learning criterion. The training was administered on a smartphone that recorded both accuracy and response time. Multilevel modeling was used to derive individual acquisition slopes using baseline accuracy and number of trials-to-criterion which allowed participants to be classified into “steep” and “shallow” learners. In Phase 2 participants completed refresher sessions conducted one, two and three days after the initial training. For these sessions, participants categorized new images of the warbler species for a fixed number of trials with feedback. Steep learners showed continued gains during the post-acquisition phase with their accuracy reaching near ceiling levels of 97%. The shallow learners showed the opposite effect. Despite receiving corrective feedback, their accuracy declined to around 80%, significantly below the initial 90% learning criterion that they achieved in the training phase of the study. We speculate that whereas the steep learners were efficient in updating and refining their category representations, the shallow learners were less efficient in utilizing the feedback to modify their category representations. Our results suggest one’s “trainability” on a visual category task predicts future success in their ability to retain and refine visual category representations.