Computational Maps in the Visual Cortex
     Figure 4.2
MiikkulainenBednarChoeSirosh
Home    
About the Authors
Back Cover    
Table of Contents 
Sample Chapter 
Figures    
References    
Errata    
Demos     
Talks/Courses 
Software    
Credits    
Purchase online at:

springeronline.com
amazon.com

Click on the image to see a PDF version (for zooming in)

Fig. 4.2. Afferent weights of ON and OFF neurons in the LGN. The receptive fields of a sample ON neuron and a sample OFF neuron, both located at the center of the corresponding 36 × 36 LGN sheet, are shown as gray-scale values on a 54 × 54 retina. The jagged black line traces the anatomical boundary of the RF, that is, connections exist only from retinal receptors located inside the boundary. Medium gray represents zero weight, dark gray stands for inhibitory connections, and light gray represents excitatory connections. Each RF shape is a difference of two Gaussians, the center and the surround. The Gaussians are normalized to have the same total strength, but the center Gaussian concentrates that strength in a much smaller region; in this example, &sigmac = 0.5 and &sigmas = 2. ON cells have an excitatory center and an inhibitory surround (a), and OFF cells have an inhibitory center and an excitatory surround (b), as in Figure 2.2a,b. These RFs perform edge detection at a spatial frequency determined by the width of the center; they highlight areas of the input image that have edges and lines, and do not respond to large areas of constant illumination.