Abstract
An auditory cue improves perception of subsequent visual stimuli that are spatially aligned with the cue as long as the stimulus onset asynchrony (SOA) is short ([[lt]]300ms). Plasticity in these crossmodal links has been shown to persist into adulthood and to be independent of focused attention (Batson, Beer & Watanabe, VSS 2006). Although crossmodal links have been shown to remain vulnerable to perceptual manipulations using a task-irrelevant training paradigm on a mature population of subjects, the location of this crossmodal facilitation remains unclear. We tested the effects of crossmodal cuing before and after task-irrelevant training where target shapes at a visual location proximal to the sound source are weighted to valid (same-side) auditory cueing during training sessions. Test targets were oriented Gabor patches that appeared either at the same or proximal locations to the sound source, all equidistant from fixation, at one of three SOAs; 150, 300 or 1000ms. To investigate the location of this plastic facilitation, the same experiment was repeated with monocular training of one eye and testing on the untrained eye. After training in the binocular condition, discrimination of validly cued targets improved at the newly associated location while discrimination at the location of the sound source shows a decrease in performance compared to the pre-test. Changes in validity effect are statistically significant at the shortest SOA, 150ms. Proposed sites of crossmodal integration extend from primary sensory cortices up to extrastriate and association cortices known to respond to multimodal stimulation. Complete interocular transfer in the monocular condition would suggest that only binocular processes are involved in the observed crossmodal plasticity, while incomplete transfer would imply that monocular processing, such as that of monocular cells in primary visual cortex, is involved. Additional results from tests of object/spatial transfer and retention of these new associations will also be reported.
Supported by: NSF BCS-0345746, NIH R01 EY015980