Computational Maps in the Visual Cortex
     Figure 4.6
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. 4.6. Self-organized V1 afferent weights. The final afferent weights of the sample neuron at the center of the V1 network are shown on the (a) ON and (b) OFF sheets as in Figure 4.3a. (c) Their combined effect is shown by subtracting the OFF weights from the ON. From initially random weights, a smooth profile emerged on both LGN sheets, reflecting the Gaussian input patterns. Their combined effect is center-surround, resulting in a focused, edge-enhanced response in V1. Because the anatomical RF (shown with the black outline) of this neuron was positioned slightly left and below its topographically ordered position, the Gaussian weight patterns formed slightly above and to the right of the RF center. As shown in Figures 4.7a and 4.8b, these locations represent the retinotopic order well.