Global Optimization of Cerebral Cortex Layout
Supplementary Online Material

Christopher Cherniak*,   Zekeria Mokhtarzada,  
Raul Rodriguez-Esteban,   B. K. Changizi

Functional areas of mammalian cerebral cortex seem positioned to minimize finely costs of their interconnections, down to a best-in-a-billion optimality level.  Macaque and cat cortex rank better in connection optimization than the wiring of comparably structured computer chips, but somewhat worse than the economic commodity-flow network among U.S. states.  Cortex wiring conforms to a Size Law better than the macroeconomic patterns, which may indicate cortex optimizing mechanisms involve more global processes.  

Fig. S1  Adjacency vs Wirecost  (C. elegans layouts)  

 Cortex Datasets

Macaque Visual Cortex  

Fig. S2  Cortex map;  Table S1  Connection matrix   
Fig. S3  Size law: optimality results  

 Cat Visual, Auditory, & Somatosensory Cortex  

Fig. S4  Cortex map;  Table S2 Connection matrix     
Fig. S5  Size law: optimality results

 Cat Cortex Meta-Modules  

Table S3  Connection matrix 
Fig. S6  Size law: optimality results 

 Non-neural Datasets  

AMI49 Microchip  

Fig. S7 Layouts;  Table S4  Connection matrix  
Fig. S8  Size law: optimality results  

 Macroeconomic Commodity-Flow Network

Fig. S9  USA (BTS) map;  Table S5  Ex/im matrix  
Fig. S10  Size law: optimality results  

Conclusion 

Table S6  Optimization results summary  
Table S7  Connections vs adjacencies summary  

 References and Notes

 * To whom correspondence should be addressed:

Committee for Philosophy and the Sciences, Department of Philosophy, University of Maryland, College Park, MD 20742, USA. cc27[at]wam[dot]umd[dot]edu

Next

Copyright Christopher Cherniak (2002)
Last Modified 11/19/2002

---