Abstract
Building on the foundation of Dr. Whitlow Au’s pioneer and extensive work on the sonar of dolphins, recent research has illuminated the dynamic nature of echolocation involving free-swimming toothed whales. The sophisticated orchestration of movements and biosonar outputs by toothed whales is particularly relevant in the context of developing agile autonomous sonar-guided underwater vehicles. In this study, we test the hypothesis that broad angular coverage, which provide comprehensive echo information of aspect-dependent objects, is necessary for target discrimination via echolocation. We trained a harbor porpoise to actively approach and select a sphere against a spheroid in a two-alternative forced-choice target discrimination task in a net pan. Two spheroids of different aspect ratios representing two overall difficulty levels were used with each presented at diverse orientations when paired with the sphere. We show that the porpoise can discriminate these targets without acquiring aspect-dependent spheroid echo information over broad ranges of angles. Instead, focused, rapid sampling via buzzes appears to aid discrimination in more challenging scenarios. These results suggest that echolocation-based discrimination involving aspect-dependent objects is possible based on subsets of echo signatures without more extensive angular sampling that could build toward an understanding of the target identity. [Work supported by ONR.]