Techniques for enhancing neuronal evolvability

High dimensionality and interactional complexity, appropriately, introduced , can enhance the evolvability of a pattern processing network. We describe a processor, referred to as the cytomatrix module, that can be used to inv estigate the requisite conditions for such enhancement. The processor is ch aracterized by multiplicity of component types, graded interactions among c omponents, separation of signal integration dynamics from the readout mecha nisms that interpret these dynamics, and multiplicity of parameters open to evolution (including component connectivity). The adaptation procedure is mediated by a multiparameter variation-selection algorithm that acts on the various parameters in an alternating (i.e., phasic) manner. Experiments wi th both structured and unstructured teaming tasks, as well as with difficul t parity problems, demonstrate that opening more parameters to evolution in creases the flexibility exhibited by the processor in response to evolution ary pressure, essentially by loosening the coupling between the local and g lobal aspects of the response. The cytomatrix processor can be thought of a s a highly abstracted representation of signal integration within single ne urons; alternatively, it can be viewed as a collection of cells in a multic ellular organization. (C) 2002 Elsevier Science B.V. All rights reserved.