Purchase this article with an account.
Christiane Wiebel, Matteo Valsecchi, Karl Gegenfurtner; Differential Processing of Material and Object Images: Evidence from ERP Recordings. Journal of Vision 2012;12(9):861. doi: https://doi.org/10.1167/12.9.861.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
It has been recently proposed on the basis of psychophysical experiments that material categorization is less efficient than object categorization (Wiebel, Valsecchi & Gegenfurtner, VSS 2011). Here we set out to find electrophysiological evidence for the differential processing of material and object images measuring ERPs while subjects performed a Go/NoGo task on either material images (wood vs. stone) or object images (people vs. animals). Each of the four categories was used as a target and as a nontarget in eight alternating sessions. For each category we presented 80 images. Mean luminance and contrast were normalized. Object images were from the COREL database, whereas material images were taken by ourselves. The subject's task was to press a button as fast as possible if the image belonged to the target category and to suppress any response otherwise.
The results showed that in line with previous findings, subjects responded significantly slower on material trials compared to objects trials (Adelson, Sharan & Rosenholtz, VSS 2011). Besides, the target effect (Go minus NoGo activity at frontal sites) which had been described for objects (Johnson & Olshausen, JOV 2003) was extended to materials. Furthermore, using a linear classifier we were able to decode each specific category independent of its target- status based on the spatial pattern of the ERPs. Go vs. NoGo trials could also be classified above chance. Crucially, we managed to classify material and object images as early as 110ms. Moreover, at this point in time we found the spatial distribution of weights assigned to each electrode in the material vs. object classification to be similar for Go and NoGo trials. This suggests differential processing of the low-level properties of material and object images independent of their target status very early in time.
Meeting abstract presented at VSS 2012
This PDF is available to Subscribers Only