Poster A121, Saturday, March 25, 5:00 – 7:00 pm, Pacific Concourse
Tracking the Time Course of Visual Prediction: Graded Effects of Preactivation Shift Earlier Given Extended Preview Time
Cybelle Smith1, Kara D. Federmeier1; 1University of Illinois, Urbana-Champaign
When and how is the brain able to make use of newly-learned information about objects and their associated contexts? We recorded EEG as 48 participants learned paired associations between visual scenes and novel objects from novel object categories. At test, participants indicated whether an object matched a previously viewed scene. Critically, at test, the scene was previewed for either 200ms (N=24) or 2500ms (N=24), prior to object onset. ERPs time-locked to object onset at test displayed a graded pattern of facilitation contingent on how closely the test object matched that presented with the scene at study. Critically, the time-course of this sensitivity varied with the amount of preview time, such that fine-grained distinctions based on object similarity and category structure were observed earlier in the long preview condition. With long previews, graded facilitations emerged during the N300 time window, as early as 200-300ms, suggesting that participants were able to anticipate structural features of the objects and object categories. Instead, when participants had little time to develop predictions, fine-grained distinctions emerged only later, beginning at ~300-400 ms. In addition, a later positivity (400-600ms) exhibited a similar graded pattern of effects, and may reflect integrative processing used to assess the degree of match between the presented object and the scene. With short previews, this later positivity was numerically larger and exhibited a larger contrast between matching and mismatching objects. Our findings suggest that the availability of predictive information systematically affects how processing unfolds when an object is encountered in a given context.
Topic Area: LONG-TERM MEMORY: Priming