Pack and Herman tested Elele by showing her an assortment of plastic tubes twisted and bent in odd shapes. To see if she could recognize with sonar an object she had first seen with her eyes, an assistant stood at the edge of her tank, holding one of the objects, and let the dolphin look at it for a few seconds. (Dolphin sonar doesn’t work in the air.) Then, out of Elele’s sight, another assistant put a matching object in a black Plexiglas box underwater. A different object was placed simultaneously in a second submerged box. The dolphin’s sonar could penetrate the Plexiglas and reflect off the object inside, but Elele couldn’t see into the black box.
In 16 tests, Elele made 14 correct matches, for which she was rewarded. To reverse the test--that is, to see if Elele could recognize objects visually that she had examined with her sonar--the researchers had her first probe a single object in the underwater box with her sonar. Then, above water, one assistant held a matching object in one hand, while a second assistant held a different one. Elele indicated her choice by swimming and holding herself in front of one of the objects. In these tests she successfully matched 13 of 16 objects.
Since the dolphin’s performance was virtually equal in both cases, says Herman, it suggests the animal’s sonar is functionally equivalent to its vision--and far more acute than anyone had thought. The stimulus the dolphin gets back is a stream of echoes, spatially and temporally arranged, he says. The analogy to vision is that what it receives is a stream of photons, and the brain constructs a perception out of that. So the brain is performing the same end result, although the original inputs differ.
But this was not intuitive to anybody, he adds. I think the community working in dolphin sonar was quite surprised by these findings.


