Page 48 - Spring2019

Heptuna and Biosonar
naud and Popper, 1975). The results showed that Heptuna (and other dolphins) could localize sounds in water almost as well as humans can in air.
Because dolphins live in a three-dimensional acoustic world. Donna decided to ask whether Heptuna could localize sound in the vertical plane. The problem was that the study site was quite shallow and there was no way to put sources above and below the animal. To resolve this, Donna decided that if she could not bring the vertical plane to Heptuna, she would bring Heptuna to the vertical plane. She switched the bite bar to a vertical position and trained Heptuna to do the whole study on his side. As a consequence, the same sound sources that were left and right in the earlier experiments were now above and below the animal’s head. Donna found that the MAA for vertical localization was as good as that for hori- zontal localization, suggesting a remarkably sophisticated localization ability in dolphins.
An Overview of Heptuna’s Studies
Shortly after completing the localization study, Heptuna started to “collaborate” with another well-known male dol- phin, Sven. Heptuna and Sven began training to detect targets on a “Sky Hook” device, which moved and placed calibrated stainless steel spheres and cylinders underwater at various distances from the echolocating animals. The purpose of the Sky Hook was to determine the maximum distance at which dolphins could detect objects of different sizes and types (Au et al., 1978).
As training continued, Dr. Whitlow Au, a senior scientist at the NUC, attended the sessions conducted by Ralph Penner or Arthur (Earl) Murchison. Whit recorded the animals out- going echolocation signals (see Au, 2015 for a general history of dolphin biosonar research). These signals were analyzed in an attempt to understand and quantify the echolocation abilities of dolphins. As Whit clearly stated, “In order to bet- ter understand the echolocation process and quantify the echolocation sensitivity of odontocetes, it is important to de- termine the signal-to-noise ratio (SNR) at detection thresh- old” (Au and Penner, 1981, p. 687). Whit wanted to refine his earlier estimates of the animal’s detection threshold based on the transient form of the sonar equation.
Because the earlier thresholds were done in Kaneohe Bay where the background noise was variable and not uniform with respect to frequency, a flat noise floor was needed. Thus, Heptuna was exposed to an added “nearly white-noise source” that was broadcast while he performed the target de- tection task. The results showed that at the 75% detection
46 | Acoustics Today | Spring 2019
Figure 4. A cartoon of Heptuna’s station and the apparatus Earl Mur- chison used to present the targets (see text for details). ΔR, change in preselected target distances. From Murchison (1980).
threshold, Heptuna’s threshold was at a level of 77.3 dB re 1 μPa/Hz, whereas it was 74.8 dB for a second dolphin, Ehiku. Whit went on to speculate that this difference may have been due to the abilities of the two animals to distinguish time separation pitch (TSP). In human hearing, TSP is the sensa- tion of a perceived pitch due repetition rate. For dolphins, the concept suggests that the ripples in the frequency domain of echoes provides a cue, an idea that persists in modeling dol- phin sonar today (Murchison, 1976; Au, 1988).
Heptuna and Echolocation Studies
Heptuna’s biosonar career continued under the tutelage of Earl and Ralph. Earl had begun a long series of experiments on echolocation range resolution of dolphins, and Heptuna was the animal of choice because of his experience. Heptuna faced a new experimental paradigm in this study, requiring him to place his rostrum in a “chin cup” stationing device and echolocate suspended polyurethane foam targets to his left and right (Figure 4). His task was to press a paddle on his right or left corresponding to the closer target. Earl would randomly adjust the targets to one of three preselected ranges (1, 3, or 7 meters). Heptuna was stationed behind an acoustically opaque visual screen so that he could not see or echolocate the targets, and Earl would move one target ever so slightly, moving it a set distance closer or further away in relation to the test range.
Heptuna was the subject for three of Earl’s studies of range resolution. Earl found that Heptuna’s performance indicated that his range resolution conformed closely to the Weber- Fechner function. In human psychophysics, the law relates to the perception of a stimulus; the magnitude of the stimulus when it is just noticeable is a constant ratio (K) of the origi- nal stimulus (∆D/D = K) or conforms to Stevens power law. The results led Earl to speculate how Heptuna’s performance would compare with the results of other echolocation exper- iments. One important observation from Heptuna’s results