Computational Maps in the Visual Cortex
     Figure 5.20
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. 5.20. LISSOM model of orientation and direction selectivity. The architecture is similar to the LISSOM OR network, except the ON and OFF channels consist of sheets of neurons with different lag times (from 0 to 3 in this case). Moving input patterns are drawn on the retina in discrete timesteps, like frames of a movie. At the first timestep, the ON and OFF LGN cells with time lag 3 compute their activity. At each subsequent timestep, the input pattern is moved slightly and LGN cells with lags 2, 1, and 0 each compute their activity in turn. Once all LGN cells have been activated, the initial V1 response is computed based on the responses on the eight LGN sheets. The activity then spreads laterally within V1 as usual in LISSOM.