Supplemental Information:
Spontaneous human speech mimicry by a cetacean
Sam Ridgway, Donald Carder, Michelle Jeffries, and Mark Todd
Supplemental Experimental Procedures:
Before we observed spontaneous human speech mimicry, the whale was exposed to human speech not only from speakers on the surface. He was sometimes nearby when divers used underwater communications equipment. (“Wetphone” Sound-Wave Systems Inc., Costa Mesa, CA 92626; 200 Hz to 3.2 kHz bandwidth).
Underwater sound recordings were with two B & K 8103 hydrophones (Brüel and Kjaer, Denmark) amplified with a B&K 2635 charge amplifier. The amplified signal was connected to a broadband instrumentation tape recorder (Racal store 7D). In air, recordings used a B&K Model 4135 ultrasonic condenser microphone recorded on a separate channel of the same tape. Human voice descriptions of events were also recorded on the same tape (Fig 1.).
All recordings were made when the whale was at the surface with his blowhole in air. The recording microphone in air was directly above the whale's head. Hydrophones were in the water at 1 m depth and within 1 m of the whale's head. Speak commands were given by the trainer directly or near directly above the whale's head. Airborne sound enters the water and presumably the whale's head with very little attenuation within a 13-degree angle from air to water [S1].
After the initial recordings of speech like sounds (Fig. 1), the whale was trained [6] to accept a micro-tip pressure transducer (Millar Instruments, Houston, Texas, USA) catheter into his nasal system. The catheter has a frequency response up to 35 kHz so that even the most rapid pressure changes could be followed as the whale vocalized.
Supplemental Results:
Our observations of human voice mimicry began in May 1984 when a male white whale. about 9 years of age at our facility in San Diego Bay spontaneously began to make such speech-like sounds. The whale called "NOC" was the youngest of three that had been collected at Churchill, Manitoba, Canada in August of 1977. At times, the speech-like sounds were heard from a distance when no humans were nearby, however most episodes were observed when humans were nearby on the surface or underwater. Speech-like sounds were never observed in the two older whales that lived with him and they were not heard from NOC when he was interacting with other whales. Observations of white whales in our care indicate that they are very vocal animals consistent with the many observations of these "sea canaries" in the wild. The usual sounds of our whales were similar to those presented by others [S2, S3].
With the sealed pressure catheter in place, the whale was signaled to "speak." Nasal cavity pressure increased with each vocal burst and fell after each burst (Fig. S1). In the nasal cavities, below the phonic lips, peak pressures were as high as 500 mmHg during the speech like sounds. Pressures in the vestibular sacs above the phonic lips were under 200 mmHg. These cetaceans are capable of producing high pressures in the nasal cavities due to the action of strong muscles below and above the nasal cavities. During whistle-like sounds, pressures of 800 mmHg or greater that one atmosphere may be produced in the nasal cavities [6].
Figure S1. An underwater recording of a segment of whale speech-like behavior. The heavy dark line in the lower part of the figure shows variations of nasal pressure in mmHg. Pressure is indicated on the right ordinate and acoustic frequency is indicated on the left ordinate. The waterfall spectrum is shown below and the acoustic amplitude rhythm is shown above.
Supplemental Discussion:
Between collection in Canada and in 1977 and May 1984 when speech-like sounds were observed, the whale lived continuously with other whales. NOC was always in contact with other whales except during training sessions lasting several hours each day. Along with an older female, he was trained to cooperate in the open sea for deep diving studies [S4]. For the seven years before speech-like sounds were observed, NOC was continuously in training for one project or another. He had several different trainers and no particular individual bond appeared to develop.
Two juvenile killer whales separated from their natal pods were shown to mimic barks of California sea lions (Zalophus californianus) [S5]. Perhaps the young age of the whales had something to do with such vocal learning.
Marine mammals suggested for spontaneous speech-like behavior have all been young males. These include the white whale at Vancouver Aquarium, our whale NOC and a common seal called Hoover [S6]. This suggests that hormonal changes in the attainment of sexual maturity might be involved in development of spontaneous speech-like behavior. It has also been suggested that isolation from conspecifics could be a factor in speech imitation [S5]. This could not be the case with NOC. However, he did have close association with many trainers and attendants.
Supplemental References:
S1. Young, R. W. (1973) Sound pressure in water from a source in air and vice versa. J. Acoust. Soc. Amer. 53,1708-1716.
S2. Vergara, V. and Barrett-Lennard, L. (2008) Vocal Development in a Beluga Calf (Delphinapterus leucas) Aquat. Mamm. 2008, 34, 123-143.
S3. Sjare, B. L. and Smith, T. G. (1986) the vocal repertoire of white whales, Delphinapterus leucas, summering in Cunningham Inlet, Northwest Territories. Can. J. Zool. 64, 407-415.
S4. Ridgway SH, Bowers CA, Miller D, Schultz ML, Jacobs CA, et al. (1984) Diving and blood oxygen in the white whale. Can. J. Zool. 62: 2349-2351.
S5. Foote AD, Griffin RM, Howitt D, Larsson L, Miller PJO, et al. (2006) Killer whales are capable of vocal learning. Biol. Lett. 2: 509–512.
S6. Ralls K, Fiorelli P, Gish S (1985) Vocalizations and vocal mimicry in captive harbor seals, Phoca vitulina. Can. J. Zool. 63: 1050-1056.
Figure 2S. An MP3 recording of the unusual speech-like sounds from white whale NOC.