Sensitivity to sound-source elevation in nontonotopic auditory cortex.

We have demonstrated that the spike patterns of auditory cortical neurons carry information about sound-source location in azimuth. The question arises as to whether those units integrate the multiple acoustical cues that signal the location of a sound source or whether they merely demonstrate sensitivity to a specific parameter that covaries with sound-source azimuth, such as interaural level difference. We addressed that issue by testing the sensitivity of cortical neurons to sound locations in the median vertical plane, where interaural difference cues are negligible. Auditory unit responses were recorded from 14 alpha-chloralose-anesthetized cats. We studied 113 units in the anterior ectosylvian auditory area and 82 units in auditory area A2. Broadband noise stimuli were presented in an anechoic room from 14 locations in the vertical midline in 20 degrees steps, from 60 degrees below the front horizon, up and over the head, to 20 degrees below the rear horizon, as well as from 18 locations in the horizontal plane. The spike counts of most units showed fairly broad elevation tuning. An artificial neural network was used to recognize spike patterns, which contain both the number and timing of spikes, and thereby estimate the locations of sound sources in elevation. For each unit, the median error of neural-network estimates was used as a measure of the network performance. For all 195 units, the average of the median errors was 46.4 +/- 9.1 degrees (mean +/- SD), compared with the expectation of 65 degrees based on chance performance. To address the question of whether sensitivity to sound pressure level (SPL) alone might account for the modest sensitivity to elevation of neurons, we measured SPLs from the cat's ear canal and compared the neural elevation sensitivity with the acoustical data. In many instances, the artificial neural network discriminated stimulus elevations even when the free-field sound produced identical SPLs in the ear canal. Conversely, two stimuli at the same elevation could produce the same network estimate of elevation, even when we varied sound-source SPL over a 20-dB range. There was a significant correlation between the accuracy of network performance in azimuth and in elevation. Most units that localized well in elevation also localized well in azimuth. Because the principal acoustic cues for localization in elevation differ from those for localization in azimuth, that positive correlation suggests that individual cortical neurons can integrate multiple cues for sound-source location.