Response selectivity for multiple dimensions of frequency sweeps in the pallid bat inferior colliculus

Abstract

1. While hunting, the pallid bat uses passive sound localization at low frequencies to find terrestrial prey, and echolocation for general orientation. It must therefore process two different types of acoustic input at the same time. The pallid bat's echolocation pulse is a downward frequency-modulated (FM) sweep from 60 to 30 kHz. This study examined the response selectivity of single neurons in the pallid bat's central nucleus of the inferior colliculus (ICC) for FM sweeps, comparing the response properties of the high-frequency population, tuned to the biosonar pulse, with the low-frequency population, tuned below the pulse. The working hypothesis was that the high-frequency population would exhibit a response selectivity for downward FM sweeps that was not present in the low-frequency population. 2. Neurons were tested for their selectivity for FM sweep direction, duration, frequency range and bandwidth, and rate of frequency change. The extent to which they responded exclusively to tones, noise, and FM sweeps was also examined. Significant differences in the response properties of neurons in the two populations were found. In the low-frequency population, all neurons responded to tones, but only 50% responded to FM sweeps. Only 23% were selective for sweep direction. In the high-frequency population, all neurons responded to FM sweeps, but 31% did not respond to tones. Over one-half of this population was selective for sweep direction, and of those that were selective, all preferred the downward sweep direction of the biosonar pulse. A large percentage (31%) responded exclusively to downward sweeps, and not to tones or upward sweeps. None of the cells in either population responded to noise, or did so only at very high relative thresholds. 3. Both populations contained neurons that were selective for short stimulus durations that approximated the duration of the biosonar pulse, although the percentage was greater in the high-frequency population (58% vs. 20%). In the high-frequency population, 31% of the neurons tested for duration responded exclusively to both the sweep direction and duration of the biosonar pulse. 4. Downward FM-selective neurons, with one exception, were generally insensitive to the rate of frequency change of the FM sweep, as well as the frequency range and bandwidth of the sweep. They responded similarly to both the full 60- to 30-kHz sweep and to 5-kHz bandwidth portions of the full sweep.(ABSTRACT TRUNCATED AT 400 WORDS)