Dependence of auditory cortex evoked unit activity on interstimulus interval in the cat.

Publisher Website
Google Scholar
Abstract
The responses of single units in the primary auditory cortex to brief sound stimuli were investigated in anesthetized cats. Each unit was characterized in terms of its sensitivity to tone pips of different frequencies, to bursts of white noise and to short sequences of clicks. For each unit, the stimulus that caused the greatest increase in firing was selected and kept constant in frequency and intensity throughout the remainder of the experiment. The duration of the stimulus and the interstimulus interval (ISI) were varied; 50-ms stimuli were presented at intervals of either 550 or 900 ms, while 100 ms stimuli were presented at intervals of either 900 or 1600 ms. Six sets of the selected stimulus, each comprising 100 presentations, were used to study each unit. Within each set the stimulus duration was kept constant, while the 2 ISI assigned to the 50 and 100 ms stimuli, respectively, were used repeatedly or mixed randomly within the set. The stimulus-evoked activity of each unit was analyzed as a function of stimulus duration, of ISI, and as a function of whether the ISI was mixed or not. Changes in the rate of evoked activity related to any of the above variables were evaluated statistically, and trends in the responses to these variables were analyzed for the entire population of units. A significant relationship between unit activity and ISI was found. Increasing the ISI from 550 to 900 ms or from 900 to 1600 ms resulted in both instances in a 24% increase in the units'' rate of firing. Varying the ISI within each set did not produce a uniform effect over all the units. Mixing the 2 ISI in a single set of stimulus presentations caused a reduction in response in 29% of the units, whereas an increase in evoked activity occurred in 14%. In the remaining 57% of the units no consistent trend was observed. The effect of cycle-to-cycle ISI variations on the response of single units was analyzed in terms of possible neural networks that would produce such effects. About 15% of the units behaved as if related to cortical networks with fixed properties, while most units in our sample appeared to be involved in cortical networks that either change their properties as a function of the ISI or interact with each other in a time-dependent manner.