Neural Darwinism in the Mammalian Spinal Cord

A conceptual framework around which the spinal control of locomotion is viewed and often studied has focused on the property of “Central Pattern Generation”, i.e. the generation of oscillating efferent patterns from the lumbosacral spinal cord in the absence of either supraspinal or sensory oscillating input (Grillner, 1973; 1982; Severston, 1999). An alternative, or at least complementary, concept that may help to understand spinally-controlled locomotion has been presented by Edelman and colleagues (Edelman, 1978; Edelman, 1981; Edelman and Reeke, 1982; Edelman and Finkel, 1984) to account for adaptive events in the brain. They theorized that “the brain is dynamically organized into cellular populations containing individually varied networks, the structure and function of which are selected by different means during development and behavior”. The cellular populations are proposed to be collections of hundreds to thousands of strongly interconnected neurons acting as functional units. The key elements of this Neuronal Group Selection concept, often referred to as Neural Darwinism, are a) “Reentrant” structure and function involving a continuous spatial temporal representation of sensory information and a mechanism for continuous updating of the selection of neuronal groups that can generate a given motor synergy or pattern and, b) “Degeneracy” indicating that there must be variance in the neuronal groups that generate a movement pattern if learning and selective matching of sensory and motor processes are to occur. This concept, therefore, also assumes that different neuronal groups can accomplish the same function.