If you have any problems related to the accessibility of any content (or if you want to request that a specific publication be accessible), please contact (email@example.com). We will work to respond to each request in as timely a manner as possible.
An fMRI study of object recognition and cerebral laterality
StatisticsView Usage Statistics
The ventral visual cortex houses a hierarchical system for object processing and exhibits category-specific processing, especially for fine-grained discrimination of words and faces. The degree to which different regions in ventral visual cortex, within and between the two cerebral hemispheres, are involved in visual processing of different kind of visual objects is not known. In two fMRI studies of words and faces, we compare univariate and multivariate analysis results for neural differences between and within categories, at different visual hemifield locations (center, left and right hemifield). In the functional localizer experiment, opposite lateralization between words and (to a lesser degree) faces were observed. However, between-category univariate fMRI results were not consistently lateralized to the left hemisphere for words or oppositely lateralized to the right hemisphere for faces in the two main experiments. Multivariate fMRI results were less restricted to category-specific areas of ventral visual cortex and showed considerable lateralization for discrimination of objects at the exemplar-level. Between-category multivariate fMRI results showed less lateralization, despite strong evidence of neural discriminability. This means that within-category discrimination fMRI multivariate results are more consistent with laterality effects in behavior as measured in divided field experiments. Such results are important because they highlight a fundamental discrepancy in the hemifield-hemisphere relationship between visual object recognition behavior and fMRI results. In the visual hemifield experiments, both category-specific regions for words and faces showed contralateral biases of within-category decoding for preferred stimulus categories. More interestingly, a posterior object-sensitive region exhibited successful within-category decoding for both words and faces presented in the non-preferred location. We offer a novel explanation of this apparent discrepancy revealed by the influence of object (face or word) location on within-category multivariate fMRI results: Ipsilateral visual hemifield deficits observed in divided field studies of words and faces reflect a reliance on feature-based processing by generic object recognition circuits posterior to category-specific regions in the ventral visual cortex.