Purchase this article with an account.
Stephen Ramanoel, Louise Kauffmann, Nathalie Guyader, Alan Chauvin, Cédric Pichat, Michel Dojat, Carole Peyrin; Effect of RMS contrast normalization on the retinotopic processing of spatial frequencies during scene categorization. Journal of Vision 2014;14(10):1086. doi: https://doi.org/10.1167/14.10.1086.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Since there is considerable evidence suggesting that visual perception is based on spatial frequencies (SF) processing, a growing number of studies investigate the cerebral regions involved in the processing of low and high SF (LSF and HSF) information in complex visual stimuli (such as scenes). LSF and HSF stimuli are created using low- and high-pass filters that respectively attenuates signals with frequencies higher and lower than a cutoff frequency. The contrast is reduced in HSF relative to LSF images. Thus, recent fMRI studies normalized root mean square (RMS) contrast in image in order to avoid that differential cortical activations in LSF and HSF processing might be due to contrast differences. In the present fMRI study, we investigated whether RMS contrast normalization induced change in retinotopic processing of SF during scene categorization. For this purpose, participants performed a categorization task using large black and white photographs of natural scenes filtered in LSF, HSF and non-filtered (NF) scenes, in eight block-designed functional scans. In four functional scans, both mean luminance and RMS contrast of LSF, HSF and NF scenes were equalized, while in the other four functional scans only the mean luminance was equalized. When RMS contrast was not normalized, results showed that LSF (relative to HSF) scenes elicited activation in the anterior half of the calcarine fissures linked to the peripheral visual field, whereas HSF (relative to LSF) scenes elicited activation in the posterior part of the occipital lobes, which are linked to the fovea, according to the retinotopic property of visual areas. However, RMS contrast normalization drastically increased activation for HSF scenes only, such as no significant activation was obtained for LSF scenes compared to HSF scenes. Our study suggests that RMS contrast normalization should be used with caution when investigating the neural basis of SF processing in retinotopic areas.
Meeting abstract presented at VSS 2014
This PDF is available to Subscribers Only