| Presentation | 2010-07-27 High-order receptive fields of V1 neurons as mechanisms to achieve efficient coding of contrast distribution Hiroki Tanaka, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Understanding cortical representation of the external world is one of primary goals of systems neuroscience. At the initial stage of cortical visual processing, luminance distribution of images, which provides us with the strongest cues for shape, motion and depth perception, is efficiently coded based on its decomposition into local sinusoidal components approximated by Gabor functions. On the other hand, such perception is also possible with the contrast components of the images alone. Therefore, there is a possibility that spatial distribution of the contrast may also be efficiently represented based on its decomposition into Gabor-like features. In this study we examined whether such representation for contrast is conducted at the early stage of conical processing. Many neurons in the primary visual cortex have suppressive surround regions outside the excitatory classical receptive fields (CRFs). These neurons show the strongest response when higher contrast region of an optimal sinusoid matches the CRF region and its lower contrast region matches the surround region. Therefore the two regions can be viewed as a high-order receptive field that acts as a basis function for the decomposition of contrast distribution. We developed a new method to map the CRF and suppressive surround regions using amplitude modulated gratings and applied it to the cat primary visual cortex (1). We found that the majority of neurons we examined had the CRF and suppressive surround regions that are elongated in a parallel manner like Gabor functions. The orientation of elongation and the width of the two regions widely varied across cells. The results indicate that activities of a population of neurons with surround suppression in V1-allow an efficient decomposition coding of contrast distribution of images. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Image representation / High-order receptive field / Contrast / Primary visual cortex (V1) / Gabor function |
| Paper # | NC2010-41 |
| Date of Issue |
| Conference Information | |
| Committee | NC |
|---|---|
| Conference Date | 2010/7/20(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Neurocomputing (NC) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | High-order receptive fields of V1 neurons as mechanisms to achieve efficient coding of contrast distribution |
| Sub Title (in English) | |
| Keyword(1) | Image representation |
| Keyword(2) | High-order receptive field |
| Keyword(3) | Contrast |
| Keyword(4) | Primary visual cortex (V1) |
| Keyword(5) | Gabor function |
| 1st Author's Name | Hiroki Tanaka |
| 1st Author's Affiliation | Graduate School of Frontier Biosciences, Osaka University() |
| Date | 2010-07-27 |
| Paper # | NC2010-41 |
| Volume (vol) | vol.110 |
| Number (no) | 149 |
| Page | pp.pp.- |
| #Pages | 2 |
| Date of Issue | |