Coding of incremental changes in skin temperature by a population of warm fibers in the monkey: correlation with intensity discrimination in man

Abstract

Human capacity to resolve incremental changes in the intensity of a near-rectangular warming pulse applied to the palmar skin was correlated with the resolution of these same changes in skin temperature that can be achieved by a population of warm fibers responding to the stimulus. The warm-fiber sample used to construct a statistical description of responss of different populations of fibers was obtained by dissection of the monkey''s [Macaca nemestrina] median and ulnar nerves. It was assumed that these warm fibers in the monkey closely resemble those in man. Intensity discrimination in man was estimated as that increment in the intensity of the warming pulse differentiated correctly with a probability of 0.75 (difference limen). The difference limen was estimated for a wide range of stimuli, and the Weber functions relating it to warming pulses of different intensities, presented at different adaptation temperatures (T-base = 29, 34 and 39.degree. C) were defined. Intensity resolution based on the responses of an assembly of warm fibers was similarly analyzed in terms of that stimulus increment that could be resolved correctly with a probability of 0.75 (the discriminable stimulus increment, DSI). If in differentiating between pairs of warming pulses, the brain identifies the more intense stimulus and that which evokes the larger response in the fiber population (paired-comparison strategy), then DSI = 0.67 .sigma..DELTA.p/(d.hivin.p/dI), where .sigma..DELTA.p = standard responses of the fiber population to successive pairs of stimuli, and d.hivin.p/dI = sensitivity of the population response to incremental changes in the intensity of the warming pulse. The statistical properties of these 2 parameters were derived from the properties of the constituent warm fibers, their responses being defined in terms of the cumulative impulse count over successively longer segments of the stimulus period. The effect on the DSI of the number of fibers in the population, the process by which the responses of individual fibers were combined to provide a measure of the responses of the population, and the degree of dependent variability among the simultaneous responses of the constituent fibers evoked by a stimulus were analyzed. The derived Weber functions for different combinations of stimulus intensity and T-base were cross-matched with the corresponding measures of human discriminative behavior. The stimulus information relayed by the whole population of warm fibers engaged by the warming pulse is marginally greater than that needed for the subject''s resolution of incremental changes in stimulus intensity. If these conditions [defined] do not occur, then a more efficient process of combining the inputs of individual responding fibers must be used by the brain in which those fibers signaling the most stimulus information contribute more to the measure of the population response than do other more noisy fibers.