INPUT-OUTPUT PROPERTIES OF A MAPPED SENSORY SYSTEM

点击翻译按钮获取中文摘要

The overall goal of this research is to determine how the neurons in a relatively complex, topographically-mapped sensory system extract information about stimuli and encode that information in their spike trains. The specific aims of the studies are: 1) to determine what parameters of sensory stimuli are encoded in the spike trains of the sensory receptors and projecting interneurons in this system; 2) to determine the accuracy with which that information is encoded; 3) to determine how the information is encoded within different aspects of the spike train patterns; 4) to examine the mechanisms through which the observed coding scheme is actually implemented within this neural network, 5) to determine the extent to which the observed accuracy approaches the theoretical maximum limits, given the constraints imposed by the "physics" of the stimulus environment, and 6) to examine the principles through which features of the system have been optimized to meet those constraints. The preparation to be studied is the cercal sensory system of the cricket. This system mediates the detection and analysis of low velocity air currents in the animal's immediate environment. All relevant sensory information is carried to higher centers by only ten pairs of primary sensory interneurons. All of these output units are identified, and all will be monitored simultaneously with extracellular electrodes. The first three goals listed above will be achieved by carrying out several types of electrophysiological input/output analyses, primarily at a systems level. Principles of information theory will be applied to the data to obtain quantitative, model-independent measures of the amount of information encoded within the spike trains of sensory receptors and primary sensory interneurons. The fourth goal will be achieved through further electrophysiological measurements, the results of which will be embodied in a physiology-based model of the system. The model will be refined, and its validity tested, on the basis of information theoretic analyses similar to the ones that were carried out on the real system. The fifth goal will be achieved through calculations of the constraints on the system's performance imposed by thermal noise in the air current environment. The sixth goal will be achieved through computer simulations of mechanoreceptor and interneuron characteristics. The studies will elucidate general principles related to optimal signal processing within sensory systems and may suggest general algorithms for efficient coding of information by ensembles of neurons.