Studies of auditory neurophysiology in non-echolocating bats, and adaptations for echolocation in one genus, Rousettus

Abstract

1. Six species of New Guinea megachiropteran bats have been studied electrophysiologically. Five are non-echolocating, one (Rousettus amplexicaudatus) has evolved an echolocation capability independent of echolocating microchiropteran bats. The purpose of the study was to compare auditory responses of non-echolocating bats with those of the echolocating bats studied to date, and to determine what adaptations have occured in Rousettus to permit echolocation. 2. Evoked potentials were recorded from the posterior colliculus and more peripheral levels in anesthetized preparations. 3. Emitted sounds of each species were recorded and are described (Fig. 1). All bats emit sounds at frequencies up to 60 kc/s or higher. 4. All species were found to have similar audiograms, broadly tuned with greatest sensitivity around 45–50 kc/s. Rousettus was unexceptional (Fig. 2). No “off”-responses were seen. 5. All species showed capabilities of frequency resolution comparable to that seen in microchiropteran bats emitting FM pulses (Fig. 3). Rousettus was not conspicuously different in this respect. 6. The non-echolocating bats showed very slow recovery of responsiveness following an initial sound (Fig. 4) and poor ability to follow trains of stimuli at a high repetition rate compared with echolocating bats. Rousettus differed dramatically from its non-echolocating relatives in recovering quickly and in following with little reduction in response amplitude at a frequency of 200/sec. In all species, recovery rate was inversely related to stimulus intensity. In Rousettus even stimuli 20–30 db fainter than an initial sound elicited good responses at 2 msec or greater intervals (Fig. 5). 7. Rousettus and its non-echolocating relatives showed generally smaller changes in sensitivity with changes in stimulus angle than are seen in microchiropteran bats. Greatest sensitivity was uniformly at 60–80° to one side, with relatively reduced sensitivity to sounds coming from directly in front, consistent with pinna position and with the lack of necessity for hearing sounds from directly ahead. Rousettus did not differ in any apparent way from the non-echolocating bats (Figs. 7, 8). 8. It is concluded that non-echolocating bats, like other small mammals, lack several of the auditory adaptations considered to be adaptations for echolocation: sharp restriction of sensitivity to the region of emitted sounds, sensitivity at extremely high frequencies, fast temporal resolution, facilitation of responsiveness to the second of a pair of sounds, and sharply directional hearing in front of the bats. 9. The only conspicuous difference between Rousettus and its non-echolocating relatives was in recovery rate. This suggests that fast temporal resolution is perhaps the most important of the necessary neural adaptations for echolocation.