Sunday, May 6, 2007

"Without A Theory The Facts Are Silent"

-- Friederich A. von Hayek

Hayek was the Nobel Prize winner in Economics in 1974 and co-originator of The Hayek/Hebb Synaptic Model. Haven't heard of the Hayek/Hebb Synaptic Model (it's just Hebbian learning, right)? Greg Ransom explains why.

Dr. Joaquin Fuster, distinguished Professor of Psychiatry and Biobehavioral Sciences at UCLA, won the George A. Miller Prize in Cognitive Neuroscience at the CNS 2007 Annual Meeting. He began his lecture with the above quote by Hayek and summarized his work on the network model of cortical cognition (e.g., Fuster, 1999, 2003), providing an historical overview of work on the Perception-Action Cycle (Fuster, 2004).



Fig. 1 (Fuster, 2004). Representational map of the human lateral cortex. (a) Schema of the hierarchical organization of memory and knowledge. (b) Approximate topographic distribution of memory networks, using the same color code as in (a).

In contrast to the outdated (Fuster, 2000) Modular Paradigm (in which cognitive functions and the contents of cognition are localized in discrete regions dedicated to the specific functions and domains), Fuster has long supported the Network Paradigm where higher cortical functions are distributed across brain regions, showing extensive intersection and overlap. In this scheme, one neuron can be part of many networks.

The thing about a lot of contemporary cog neuro research (including some presented at this conference) is that it conforms to the Modular Paradigm of extreme localization of function:
Cognitive neuroscience is not a new field, though its precise origin is difficult to trace. It was born whenever and wherever scientists began to ponder the logical relations between brain and mind and to explore those relations by observation and experiment. Most certainly, it was not born in Squaw Valley, California, where in the summer of 1993 a group of neuroscientists met to give it a name and to give themselves an agenda. [NOTE: this is not a snide remark, oh no]

. . .

...they presumably wanted the all-encompassing coverage to support their main agenda, which was the cognitive science of the cerebral cortex. That extended coverage diluted the agenda, but not enough to obscure a central concept that appears in various forms in many of the chapters. That concept is the cortical module of cognition: a discrete and continuous piece of cortical tissue specialized to serve one cognitive function or to represent one essential aspect of the information processed by it. ...

. . .

In the higher levels of that theoretical edifice, anatomists and physiologists met neuropsychologists, who were eager to localize in the brain their own constructs of cognitive function—mostly derived from studies of the effects of cortical damage. The result of the intellectual alliance of those three groups of scientists was, in the humble opinion of this reviewer, something not too distant from a new phrenology more or less legitimized by the scientific method. One of my purposes in this review is to caution the reader about some of the problems with the neuromodular principle of cognition and the methods used to support it. Another is to point out the pressing need for a more apt and useful model of cognitive organization in the cortex.


There is no greater impediment to a unified cognitive neuroscience than our inveterate Aristotelian tendency to consider cognitive functions as separate entities...
References

J.M. Fuster - Memory in the Cerebral Cortex: An Empirical Approach to Neural Networks in the Human and Nonhuman Primate. Cambridge, MIT Press, (paperback), 1999.

J.M. Fuster - The module: crisis of a paradigm (book review, "The New Cognitive Neurosciences" Second Edition, M.S. Gazzaniga, Editor-in-Chief, MIT Press). Neuron 26:51-53, 2000.

J.M. Fuster - Cortex and Mind: Unifying Cognition, Oxford University Press, 2003.

J.M. Fuster - Upper processing stages of the perception–action cycle. Trends in Cognitive Sciences, 8:143-145, 2004.

No comments:

Post a Comment