Spinal Processing of Impulse Trains from Sensory Receptors L.M. Mendell· and H. R. Koerber··, ·Dept. of Neurobiology and Behavior, SUNY-Stony Brook, Stony Brook, NY 11794 "Dept. of Neurobiology, Anatomy and Cell Science, University of Pittsburgh, Pittsburgh, PA 15261 Abstract In order to more fully understand the implications of sensory coding, single afferent fibers in anesthetized cats have been stimulated with patterns of impulses similar to those recorded in those afferents in awake behaving animals. Postsyn- aptic activity elicited by individual impulses in such trains are not of constant amplitude. In the case of muscle spindle afferents the postsynaptic potentials can become either monotonically larger or smaller as the train progresses depending on the properties and function of the target motoneuron on which the individual afferent terminates. Cutaneous afferents also evoke similarly heterogeneous effects postsynaptically that correlate with the receptor type that they supply in the skin. In particular, the sensory channels through which slowly- and rapidly adapting receptors project centrally appear to have very different properties. These data indicate that central synaptic action can differ systematically according to func- tion of the neurons involved. These results require consideration in analysis of coding / decoding processes in sensory systems and in the construction of models of sensory processing. 1. Introduction It is well established that the representation of the external world in the central nervous system occurs via transformation of physical stimuli into a spati- otemporal code of impulses in populations of activated neurons (reviewed in Darian-Smith, 1984). The analysis of sensory coding has typically been carried out in terms of measuring the output (discharge frequency) in response to a particular input (the sensory stimulus). This has permitted construction of input- output functions that are characteristic of particular receptor types. Although this approach has been useful in providing estimates of the trans- P. Rudomin et al. (eds.), Neuroscience: From Neural Networks to Artificial Intelligence © Springer-Verlag Berlin Heidelberg 1993