Chapter

Which Computation Runs in Visual Cortical Columns?

Steven W. Zucker

in 23 Problems in Systems Neuroscience

Published in print January 2006 | ISBN: 9780195148220
Published online May 2009 | e-ISBN: 9780199864676 | DOI: http://dx.doi.org/10.1093/acprof:oso/9780195148220.003.0011

Series: Computational Neuroscience Series

 Which Computation Runs in Visual Cortical Columns?

Show Summary Details

Preview

There is a substantial divide in computational neuroscience, separating the modeling of neurons at a biophysical level from the modeling of function at an information processing level. This chapter focuses on the second, information processing level, and raises a series of questions about it. It considers early vision, and representational structure emerged regarding tangents, curvatures, and continuity. That is, problem instances were formulated using the structure of differential geometry. This is not unexpected, given that differential geometry is the natural mathematics of surfaces and objects. For computation, linear complementarity emerged as a generalization of columnar operations. Algorithms for solving linear complementarity problems provided new insight into finding fast solutions by neural mechanisms. In all of the above cases, it was clear that the resulting networks are quite plausible; but starting only with network components, it seems implausible that all of the above abstract functions would have been inferred.

Keywords: neurons; information processing; function; vision

Chapter.  11796 words.  Illustrated.

Subjects: Neuroscience

Full text: subscription required

How to subscribe Recommend to my Librarian

Buy this work at Oxford University Press »

Users without a subscription are not able to see the full content. Please, subscribe or login to access all content.