Cortical units in AI have topographically ordered receptive fields with a range of bandwidths and asymmetries that make them selectively responsive to different scales and local shapes of the input spectral profile. Thus, the basic organization of processing in AI appears to be analogous to that of visual cortex. One aspect of this project will be to explore more completely these similarities. Our primary goal will be to apply to AI the methods that we have used to explore the microcircuitry of the primary visual cortex (VI). The question here is whether the superficial functional similarities between AI and VI are reflected also in the more detailed operation of the neuronal circuits in these two regions of cortex. This would be a major project in itself, and so in this MURI proposal we will make only the first steps in this direction. We will begin by exploring the possibility of making intracellular electrophysiological recordings from neurons in different layers of AI and examining their response to the auditory stimuli used by Shamma [.Shamma kowalski mov1, Shamma rip1 1995.]. The intracellular recordings will be used to examine the biophysical mechanism of gain control by the neuronal circuits of AI, and to compare this mechanism with what we have found in VI. Gain control adjusts the sensitivity of cortical circuits to the intensity range of the prevailing input - and so is an important basic process on which more elaborate cortical computations depend. Contrast gain control in VI is a multiplicative mechanism and was thought to be mediated by shunting inhibition. Our recent intracellular recordings have shown that, unlike the spike discharge, the membrane potential does not exhibit gain control, and that shunting inhibition does not occur in this context [. Allison 1996 .]. We are currently exploring other possible sources of the gain control in VI.
Our goal is to examine whether gain control properties in AI are similar to those found in VI. The significance of this cross-modality comparison is that for the first time it will be possible to make direct comparisons of the synaptic events underlying the processing of equivalent stimuli in quite different cortical areas. This will be an important test of the hypothesis that fundamentally the same processing strategy is used in different neocortical areas.