«

»

Apr 08

CNS 2016 — Using a new category learning paradigm to elicit implicit learning

Using a new category learning paradigm to elicit implicit learning

Catherine Ives-Louter, Ben Reuveni, Paul J. Reber, Northwestern University

Research suggests that both explicit and implicit memory are implicated in visual category learning, which offers a unique opportunity to study the mechanisms and interactions of these memory systems. However, the study of implicit visual category learning is often challenged by the ability of healthy individuals to explicitly remember study stimuli used and consciously extract rules. Here we introduce a new visual category learning paradigm that uses movement and time pressure to reduce the tendency of healthy participants to identify the category structure explicitly. The visual categories to be learned are based on sine wave gratings that vary on two continuous dimensions, orientation and spatial frequency, that are commonly used in studies of rule-based (RB) and information-integration (II) category learning. Participants were shown falling sine wave gratings (velocity of 14.5cm/s, giving 1.6s to respond and total presentation time of 1.9s, similar to traditional static presentation) and were instructed to categorize each stimulus into one of two categories. Participants learned the underlying category structure by trial-and-error with feedback provided after each response, gradually improving performance over 1000 trials and performing reliably above chance. Post-learning interviews indicated low rates of conscious knowledge of the underlying category structure when presented with a traditional II paradigm. In contrast, participants shown stimuli categorized by an RB rule exhibited an early, step-like jump in performance characteristic of explicit learning. By using movement to create a sense of time pressure, we were successfully able to elicit implicit, II category learning with little contamination from explicit rule discovery.

Ives-Louter CNS 2016 -- Final png

PDF — Ives-Louter CNS 2016 — Final

Leave a Reply