Response variability of single cells in the dorsal lateral geniculate nucleus of the cat. Comparison with retinal input and effect of brain stem stimulation.

1. We studied the degree and source of response variability in different classes of cell in the dorsal lateral geniculate nucleus (dLGN). The response of single cells to a series of contrasts of a stationary flashing light spot was measured. The variability analyses were based on the mean and SD of the response to a number of repeated stimulus presentations. The relative variability was expressed by the coefficient of variation (Cv; SD/mean). 2. At a given contrast, the Cv for lagged cells was larger than for nonlagged cells. No difference was found between the Cv of X and Y cells. The magnitude of the Cv was about the same as previously found for cells in striate cortex. Accordingly, little variability is added at the cortical level. The Cv decreased with increasing contrast showing that the reliability of response and the signal-to-noise ratio was improved with increasing contrast. 3. For some cells, the retinal input was determined by recording S potentials in addition to action potentials. The Cv of the retinal input was smaller than the Cv of the dLGN cells at a given contrast. Thus in the paralyzed and anesthetized preparation, variability was added at the geniculate relay. 4. The additional variability was related to modulatory input from the brain stem. This was shown by comparing Cv versus contrast curves for the dLGN cells obtained during electrical stimulation of the peribrachial region of the brain stem (PBR) with corresponding curves obtained without PBR stimulation. During PBR stimulation, which presumably mimics the effects of arousal on the dLGN cell, the Cv at a given contrast was reduced toward the value for the retinal input to the cell. Furthermore PBR stimulation increased the signal-to-noise-ratio of the cell to the level of the retinal input. 5. When Cv was plotted against response rather than against contrast, approximately the same function was found for the various dLGN cell classes. This indicated that the variability basically depended on firing rate rather than on stimulus contrast. No difference of Cv was seen between lagged and nonlagged cells at a given level of response. The difference found at a given level of contrast reflected differences in firing rate of the two cell classes. During PBR stimulation, there was no clear difference between the Cvs of the dLGN cell and its retinal input at a given level of response.(ABSTRACT TRUNCATED AT 400 WORDS)