Poster presented at the 43rd Annual Meeting of the Psychonomic Society, Kansas City, KS.

Modality Constrained Statistical Learning of Spatial, Spatiotemporal, and Temporal Input



Christopher M. Conway & Morten H. Christiansen


Abstract

Vision appears to process input best when the input is spatially rather than temporally distributed, whereas audition is the opposite (Mahar, Mackenzie, & McNicol, 1994). Here we explored whether such modality constraints also affect statistical learning. Participants were exposed to statistically-governed visual input sequences--distributed spatially, spatiotemporally, or temporally--and then tested on their ability to appropriately classify novel sequences. At moderate presentation rates (4 elements per second), there was little difference across the three conditions. However, at faster presentation rates (8 elements per second), performance declined such that it was worst with temporal and best with spatial input. Additionally, in an auditory (temporal) version of the experiment, run at both presentation rates, performance was better than in any of the visual conditions. These results suggest that statistical learning processes are affected by modality constraints: vision is biased toward processing spatial input whereas audition is biased toward temporal input.


rule.gif (155 bytes)

Home | People | Research | Links | Contact | Publications | Presentations
Cognitive Neuroscience Laboratory

Please email suggestions/errors to mhc27@cornell.edu