Acoustic behavior of echolocating Atlantic Bottlenose Dolphins

Abstract

An echolocation click detector was used to monitor the acoustic emission of two dolphins performing a target detection task in the presence of artificially induced white noise. The time of each click emission along with the time of response was measured for every trial of the entire psychophysical experiment, which involved the detection of a 7.62-cm sphere located at a range of 16.5 m from the animals. From the data, the number of clicks, the click interval and its distribution, the response latency, and the total latency as a function of the masking noise level were determined. The average number of clicks emitted per trial increased with the masking noise level, until a particular level was reached after which the number of clicks decreased with further increases in the noise level. The dolphins seemed to have found the task insoluble beyond a 77 dB re 1 μPa2/Hz noise level, and reacted by decreasing the number of clicks emitted per trial. Dolphin E also did not emit any detectable clicks during 20% of the trials at 82 dB and 41% at 87 dB, while dolphin H did not emit any detectable clicks in 14% of the trials at the 87-dB noise level. The response latencies for correct rejections were always greater than for correct detections at all noise levels for both dolphins. The click interval results indicated that the animals seem to have good control of the click intervals used in a click train, and may preprogram a desired click interval even before starting on an echolocation search, based on prior knowledge of the target range. The use of the click detector in conjunction with a microprocessor has provided a convenient technique by which to study the acoustic behavior of echolocating marine mammals.