Search

Bottlenose dolphins (Tursiops truncatus) wore opaque suction cups over their eyes while stationing behind an acoustically opaque door. This put the dolphins in a known position and orientation. When the door opened, the dolphin clicked to detect targets. Trainers specified that Dolphin S emit a whistle if the target was a 7.5 cm water filled sphere, or a pulse burst if the target was a rock. S remained quiet if there was no target. Dolphin B whistled for the sphere. She remained quiet for rock and for no target. Thus, S had to choose between three different responses, whistle, pulse burst, or remain quiet. B had to choose between two different responses, whistle or remain quiet. S gave correct vocal responses averaging 114 ms after her last echolocation click (range 182 ms before and 219 ms after the last click). Average response for B was 21 ms before her last echolocation click (range 250 ms before and 95 ms after the last click in the train). More often than not, B began her whistle response before her echolocation train ended. The findings suggest separate neural pathways for generation of response vocalizations as opposed to echolocation clicks.

Measurements of temporary threshold shift (TTS) in marine mammals have become important components in developing safe exposure guidelines for animals exposed to intense human-generated underwater noise; however, existing marine mammal TTS data are somewhat limited in that they have typically induced small amounts of TTS. This paper presents experimental data for the growth and recovery of larger amounts of TTS (up to 23 dB) in two bottlenose dolphins (Tursiops truncatus). Exposures consisted of 3-kHz tones with durations from 4 to 128 s and sound pressure levels from 100 to 200 dB re 1 μPa. The resulting TTS data were combined with existing data from two additional dolphins to develop mathematical models for the growth and recovery of TTS. TTS growth was modeled as the product of functions of exposure duration and sound pressure level. TTS recovery was modeled using a double exponential function of the TTS at 4-min post-exposure and the recovery time.

Anthropogenic sound is a potential stressor for marine mammals that may affect health, as has been demonstrated in other mammals. Therefore, we have initiated investigations on the effects of intense underwater sounds on nervous system activation and immune function in marine mammals. Blood samples were obtained before and after sound exposures (single underwater impulsive sounds (up to 200 kPa) produced from a seismic water gun and (or) single pure tones (up to 201 dB re 1 µPa) resembling sonar "pings" from a white whale, Delphinapterus leucas, and a bottlenose dolphin, Tursiops truncatus, to measure neural-immune parameters. Norepinephrine, epinephrine, and dopamine levels increased with increasing sound levels and were significantly higher after high-level sound exposures (>100 kPa) compared with low-level sound exposures ( 100 kPa) qu'apres une exposition a un son de basse intensite (

A behavioral response paradigm was used to measure underwater hearing thresholds in two California sea lions (Zalophus californianus) before and after exposure to underwater impulses from an arc-gap transducer. Preexposure and postexposure hearing thresholds were compared to determine if the subjects experienced temporary shifts in their masked hearing thresholds (MTTS). Hearing thresholds were measured at 1 and 10 kHz. Exposures consisted of single underwater impulses produced by an arc-gap transducer referred to as a "pulsed power device" (PPD). The electrical charge of the PPD was varied from 1.32 to 2.77 kJ; the distance between the subject and the PPD was varied over the range 3.4 to 25 m. No MTTS was observed in either subject at the highest received levels: peak pressures of approximately 6.8 and 14 kPa, rms pressures of approximately 178 and 183 dB re: 1 microPa, and total energy fluxes of 161 and 163 dB re: 1 microPa2s for the two subjects. Behavioral reactions to the tests were observed in both subjects. These reactions primarily consisted of temporary avoidance of the site where exposure to the PPD impulse had previously occurred.

Auditory filter shapes were estimated in two bottlenose dolphins (Tursiops truncatus) and one white whale (Delphinapterus leucas) using a behavioral response paradigm and notched noise. Masked thresholds were measured at 20 and 30 kHz. Masking noise was centered at the test tone and had a bandwidth of 1.5 times the tone frequency. Half-notch width to center frequency ratios were 0, 0.125, 0.25, 0.375, and 0.5. Noise spectral density levels were 90 and 105 dB re: 1 microPa2/Hz. Filter shapes were approximated using a roex(p,r) function; the parameters p and r were found by fitting the integral of the roex(p,r) function to the measured threshold data. Mean equivalent rectangular bandwidths (ERBs) calculated from the filter shapes were 11.8 and 17.1% of the center frequency at 20 and 30 kHz, respectively, for the dolphins and 9.1 and 15.3% of the center frequency at 20 and 30 kHz, respectively, for the white whale. Filter shapes were broader at 30 kHz and 105 dB re: 1 microPa2/Hz masking noise. The results are in general agreement with previous estimates of ERBs in Tursiops obtained with a behavioral response paradigm.

A behavioral response paradigm was used to measure masked underwater hearing thresholds in a bottlenose dolphin (Tursiops truncatus) and a white whale (Delphinapterus leucas) before and after exposure to single underwater impulsive sounds produced from a seismic watergun. Pre- and postexposure thresholds were compared to determine if a temporary shift in masked hearing thresholds (MTTS), defined as a 6-dB or larger increase in postexposure thresholds, occurred. Hearing thresholds were measured at 0.4, 4, and 30 kHz. MTTSs of 7 and 6 dB were observed in the white whale at 0.4 and 30 kHz, respectively, approximately 2 min following exposure to single impulses with peak pressures of 160 kPa, peak-to-peak pressures of 226 dB re 1 microPa, and total energy fluxes of 186 dB re 1 microPa2 x s. Thresholds returned to within 2 dB of the preexposure value approximately 4 min after exposure. No MTTS was observed in the dolphin at the highest exposure conditions: 207 kPa peak pressure, 228 dB re 1 microPa peak-to-peak pressure, and 188 dB re 1 microPa2 x s total energy flux.

Eight bottlenose dolphins Tursiops truncatus (four male, four female) were trained to respond to 100-ms tones. Three male dolphins (ages 23, 26, and 34) exhibited hearing disability at four higher frequencies-70, 80, 100, and 120 kHz even at 111-135 dB re: 1 microPa. Two females (ages 32 and 35) responded to all frequencies as did a male (age 7) and a female (age 11). One female (age 33) responded to all tones at 80 kHz and below; however, she failed to respond at 100 or 120 kHz. One young female dolphin (age 9) exhibited no perception of sound to behavioral or electrophysiological tests. This young female was not only deaf, but mute. The dolphin was monitored periodically by hydrophone and daily by trainers (by ear in air) for 7 years until she was age 16. The animal never whistled or made echolocation pulses or made burst pulse sounds as other dolphins do.

Temporary threshold shift (TTS) was measured in a bottlenose dolphin exposed to a sequence of four 3-kHz tones with durations of 16 s and sound pressure levels (SPLs) of 192 dB re 1 μPa. The tones were separated by 224 s of silence, resulting in duty cycle of approximately 7%. The resulting growth and recovery of TTS were compared to experimentally measured TTS in the same subject exposed to single, continuous tones with similar SPLs. The data confirm the potential for accumulation of TTS across multiple exposures and for recovery of hearing during the quiet intervals between exposures. The degree to which various models could predict the growth of TTS across multiple exposures was also examined.

A behavioral response paradigm was used to measure hearing thresholds in bottlenose dolphins before and after exposure to 3 kHz tones with sound exposure levels (SELs) from 100 to 203 dB re 1 microPa2 s. Experiments were conducted in a relatively quiet pool with ambient noise levels below 55 dB re 1 microPa2/Hz at frequencies above 1 kHz. Experiments 1 and 2 featured 1-s exposures with hearing tested at 4.5 and 3 kHz, respectively. Experiment 3 featured 2-, 4-, and 8-s exposures with hearing tested at 4.5 kHz. For experiment 2, there were no significant differences between control and exposure sessions. For experiments 1 and 3, exposures with SEL=197 dB re 1 microPa2 s and SELor = 195 dB re 1 microPa2 s, respectively, resulted in significantly higher TTS4 than control sessions. For experiment 3 at SEL= 195 dB re 1 microPa2 s, the mean TTS4 was 2.8 dB. These data are consistent with prior studies of TTS in dolphins exposed to pure tones and octave band noise and suggest that a SEL of 195 dB re 1 microPa2 s is a reasonable threshold for the onset of TTS in dolphins and white whales exposed to midfrequency tones.

A behavioral response paradigm was used to measure pure-tone hearing sensitivities in two belugas (Delphinapterus leucas). Tests were conducted over a 20-month period at the Point Defiance Zoo and Aquarium, in Tacoma, WA. Subjects were two males, aged 8-10 and 9-11 during the course of the study. Subjects were born in an oceanarium and had been housed together for all of their lives. Hearing thresholds were measured using a modified up/down staircase procedure and acoustic response paradigm where subjects were trained to produce audible responses to test tones and to remain quiet otherwise. Test frequencies ranged from approximately 2 to 130 kHz. Best sensitivities ranged from approximately 40 to 50 dB re 1 microPa at 50-80 kHz and 30-35 kHz for the two subjects. Although both subjects possessed traditional "U-shaped" mammalian audiograms, one subject exhibited significant high-frequency hearing loss above 37 kHz compared to previously published data for belugas. Hearing loss in this subject was estimated to approach 90 dB for frequencies above 50 kHz. Similar ages, ancestry, and environmental conditions between subjects, but a history of ototoxic drug administration in only one subject, suggest that the observed hearing loss was a result of the aminoglycoside antibiotic amikacin.

The relative contributions of acoustic pressure and particle velocity to the low-frequency, underwater hearing abilities of the bottlenose dolphin (Tursiops truncatus) and white whale (Delphinapterus leucas) were investigated by measuring (masked) hearing thresholds while manipulating the relationship between the pressure and velocity. This was accomplished by varying the distance within the near field of a single underwater sound projector (experiment I) and using two underwater sound projectors and an active sound control system (experiment II). The results of experiment I showed no significant change in pressure thresholds as the distance between the subject and the sound source was changed. In contrast, velocity thresholds tended to increase and intensity thresholds tended to decrease as the source distance decreased. These data suggest that acoustic pressure is a better indicator of threshold, compared to particle velocity or mean active intensity, in the subjects tested. Interpretation of the results of experiment II (the active sound control system) was difficult because of complex acoustic conditions and the unknown effects of the subject on the generated acoustic field; however, these data also tend to support the results of experiment I and suggest that odontocete thresholds should be reported in units of acoustic pressure, rather than intensity.

SUMMARYHearing is attenuated in the aerial ear of humans and other land mammals tested in pressure chambers as a result of middle ear impedance changes that result from increased air density. We tested the hypothesis, based on recent middle ear models, that increasing the density of middle ear air at depth might attenuate whale hearing. Two white whales Delphinapterus leucas made dives to a platform at a depth of 5, 100, 200 or 300 m in the Pacific Ocean. During dives to station on the platform for up to 12 min, the whales whistled in response to 500 ms tones projected at random intervals to assess their hearing threshold at each depth. Analysis of response whistle spectra, whistle latency in response to tones and hearing thresholds showed that the increased hydrostatic pressure at depth changed each whale’s whistle response at depth, but did not attenuate hearing overall. The finding that whale hearing is not attenuated at depth suggests that sound is conducted through the head tissues of the whale to the ear without requiring the usual ear drum/ossicular chain amplification of the aerial middle ear. These first ever hearing tests in the open ocean demonstrate that zones of audibility for human-made sounds are just as great throughout the depths to which these whales dive, or at least down to 300 m.

A behavioral response paradigm was used to measure masked underwater hearing thresholds in two bottlenose dolphins and one beluga whale before and after exposure to impulsive underwater sounds with waveforms resembling distant signatures of underwater explosions. An array of piezoelectric transducers was used to generate impulsive sounds with waveforms approximating those predicted from 5 or 500 kg HBX-1 charges at ranges from 1.5 to 55.6 km. At the conclusion of the study, no temporary shifts in masked-hearing thresholds (MTTSs), defined as a 6-dB or larger increase in threshold over pre-exposure levels, had been observed at the highest impulse level generated (500 kg at 1.7 km, peak pressure 70 kPa); however, disruptions of the animals’ trained behaviors began to occur at exposures corresponding to 5 kg at 9.3 km and 5 kg at 1.5 km for the dolphins and 500 kg at 1.9 km for the beluga whale. These data are the first direct information regarding the effects of distant underwater explosion signatures on the hearing abilities of odontocetes.

A behavioral response paradigm was used to measure masked underwater hearing thresholds in five bottlenose dolphins and two white whales before and immediately after exposure to intense 1-s tones at 0.4, 3, 10, 20, and 75 kHz. The resulting levels of fatiguing stimuli necessary to induce 6 dB or larger masked temporary threshold shifts (MTTSs) were generally between 192 and 201 dB re: 1 microPa. The exceptions occurred at 75 kHz, where one dolphin exhibited an MTTS after exposure at 182 dB re: 1 microPa and the other dolphin did not show any shift after exposure to maximum levels of 193 dB re: 1 microPa, and at 0.4 kHz, where no subjects exhibited shifts at levels up to 193 dB re: 1 microPa. The shifts occurred most often at frequencies above the fatiguing stimulus. Dolphins began to exhibit altered behavior at levels of 178-193 dB re: 1 microPa and above; white whales displayed altered behavior at 180-196 dB re: 1 microPa and above. At the conclusion of the study all thresholds were at baseline values. These data confirm that cetaceans are susceptible to temporary threshold shifts (TTS) and that small levels of TTS may be fully recovered.

Some indirect evidence exists for multiple sonar signal generators in odontocetes. Direct evidence was collected from two bottlenose dolphins by simultaneously measuring and digitizing internal nasal cavity pressure, nasal tissue motion, and acoustic pressure. Small catheters measured pneumatic pressure changes at the same depth within each bony nasal passage. A high‐speed video endoscopy system revealed tissue motion at both pairs of phonic lips while two hydrophones measured acoustic pressure during biosonar target discrimination. The records clearly demonstrate that acoustic pulses can be generated at the phonic lips on the left and right sides, independently or simultaneously. We have only seen the left phonic lips generate whistles. Air pressure in the bony nasal passages rises and falls together, even if the activity patterns at the two pairs of phonic lips are different. Increasing pulse repetition rate or sound pressure level, and whistle production, are all normally accompanied by increasing nasal air pressure. Whistle production apparently requires greater nasal air pressure than does pulse generation. Discovering multiple sonar sources in a dolphin’s head should cause us to reinterpret previous findings and re‐evaluate physiologic limits on pulse repetition rate, signal bandwidth, frequency composition, and projection beam geometry. [Work sponsored by Office of Naval Research.]

Responses of marine mammals to human-generated sound are of interest due to concern for protection of animals with sensitive hearing from acoustic harassment or injury. [J. Acoust. Soc. Am. 103, 2177–2184 (1998)] compare their limited observations of wild common dolphins (Delphinus delphis) near operating seismic vessels off Wales, with studies of trained bottlenose dolphins (Tursiops truncatus) in San Diego Bay [, Tech. Rpt. 1751, NCCOSC RDTE (1997)]. Because of the considerable difference between numerous seismic pulses and the single 1-s tones employed by , the comparison is not valid. Further, Goold and Fish do not accurately represent the bottlenose dolphin responses.

When cetacean hearing sensitivity at depth versus surface arose among anthropogenic ocean sound issues, Sam Ridgway advanced an approach using the first hearing tests of trained cetaceans in the open sea to provide an answer. Two trained white whales, Delphinapterus leucas, made 885 voluntary dives to a platform at depths from 5 to 300 m, stationing up to 12 min at a time to test the effects of depth on hearing and sound production. As depth increased, hearing sensitivity did not diminish, as in the aerial ears of humans and other terrestrial mammals tested in pressurized chambers. As whale hearing is unchanged by depth to at least 300 m, zones of influence on them are unaffected by depth. The whales were also trained to emit a pulse train, and then whistle to the presence of a small target. Echolocation pulses did not vary statistically in amplitude or frequency at depths between 5 and 300 m. However, the peak frequencies at the open ocean test site were at 4–40 kHz, where in San Diego Bay there were usu...

When the National Research Council formed a committee in 1992 to investigate the current knowledge and need for research on the potentially harmful effects of low‐frequency sound on marine mammals, Sam Ridgway embarked on a course of study that is now entering its second decade. Following the committee recommendations and with the goal of providing long‐term protection of the species, Ridgway initiated the first formal investigations of temporary threshold shift (TTS) in dolphins and white whales. Key to that effort was Ridgway’s development of a vocal response paradigm based on the method of free response, which allowed thresholds to be quickly estimated after noise exposure. Initial studies focused on short duration tones. Subsequent studies included single, intermittent, and impulsive sounds and have produced TTSs large enough to track growth and recovery. The resulting data reveal the effects of exposure frequency, sound pressure, duration, and temporal pattern. Recent success using auditory evoked po...

Kramer (1960, 1977) was the first to suggest that dolphins have a compliant skin that enhances their hydrody-namic performance by damping incipient turbulence. Kramer also was the first to develop a synthetic vessel coating based upon dolphin skin. However, his coating contained no mechanism for active vibration or for other adjustments to changing boundary layer conditions. Lang (1966) reviewed the earlier work on dolphin hydrodynamics and evaluated the various theories. Concerning the idea that dolphins might actively change their skin surface to reduce hydrodynamic drag, he stated: “An alternate explanation for low drag with regard to cetaceans is that they actively adjust the flexibility and movement of their skin to damp out the microscopic disturbances in the laminar boundary layer. Betchov showed that the laminar flow might be extended indefinitely by this means.”

Behavioral hearing tests with marine mammals often use single interval experiments with a go/no‐go paradigm, where the subject pushes a response paddle or similar device to indicate the presence of a hearing test tone. This approach is widely accepted, but the time required for the subject to physically move and/or interact with the response device and the requirement of one trial per reinforcement interval limit the speed at which data can be collected (typically 20–30 minutes for a threshold). An alternative to the single interval/paddle press approach is the use of a vocal response technique and the method of free response, where multiple trials are presented within a single reinforcement interval. An acoustic or vocal response eliminates the need for the subjects to physically move, allowing for faster responses and shorter inter‐trial intervals. The method of free response allows multiple tone presentations between reinforcement periods, dramatically increasing the rate at which data may be collected, especially from diving marine mammals. This talk describes the application of these techniques to hearing tests of bottlenose dolphins, white whales, and California sea lions and the advantages and disadvantages compared to the single interval experiment. [Work supported by ONR.]

Auditory evoked potentials (AEPs) offer an alternative to behavioral methods of determining auditory sensitivity in marine mammals. The technique can be performed without the need for animal training, substantially expediting the process, and has the potential for application to stranded and rehabilitating marine mammals, thus providing an opportunity to determine hearing sensitivity in animals not likely to be kept in captivity. As an emerging technology in the field of marine mammalogy, the equivalence of AEP and behavioral thresholds remains to be quantitatively assessed. Human and laboratory animal AEPs are typically −5 to +20 dB of behaviorally determined thresholds and vary by technique and frequency tested. To be an effective tool in the field of marine mammalogy, the expected variation in AEP thresholds relative to behavioral thresholds in marine mammal species needs to be determined. We compare the behavioral and AEP audiograms of several odontocetes covering a range of normal hearing to profound...

The dolphin brain is the central processing computer for a complex and effective underwater echolocation and communication system. Until now, it has not been possible to study or diagnose disorders of the dolphin brain employing modern functional imaging methods like those used in human medicine. Our most recent studies employ established methods such as behavioral tasks, physiological observations, and computed tomography (CT) and, for the first time, single photon emission computed tomography (SPECT), and positron emission tomography (PET). Trained dolphins slide out of their enclosure on to a mat and are transported by trainers and veterinarians to the laboratory for injection of a ligand. Following ligand injection, brief experiments include trained vocal responses to acoustic, visual, or tactile stimuli. We have used the ligand technetium (Tc‐99m) biscisate (Neurolite) to image circulatory flow by SPECT. Fluro‐deoxy‐d‐glucose (18‐F‐FDG) has been employed to image brain metabolism with PET. Veterinarians carefully monitored dolphins during and after the procedure. Through these methods, we have demonstrated that functional imaging can be employed safely and productively with dolphins to obtain valuable information on brain structure and function for medical and research purposes. Hemispheric differences and variations in flow and metabolism in different brain areas will be shown.

A behavioral response paradigm was used to measure pure‐tone audiograms for two white whales (Delphinapterus leucas). Tests were conducted over a 20 month period at the Point Defiance Zoo and Aquarium, in Tacoma, Washington. Subjects consisted of two males, aged 8–10 and 9–11 during the course of the study. Subjects were born in an oceanarium and had been housed together for all of their lives. Hearing thresholds were measured using a modified up/down staircase procedure and acoustic response paradigm where subjects were trained to whistle in response to hearing test tones and to remain quiet otherwise. Test frequencies ranged from approximately 2 to 130 kHz. Best sensitivities ranged from 40 to 50 dB re: 1 Pa. Both subjects had traditional U‐shaped mammalian audiograms; however, one subject exhibited significant high‐frequency hearing loss, above approximately 37 kHz. The experimental setup and procedure will be presented and the measured hearing thresholds compared to those previously measured in white whales. The potential role of ototoxic antibiotics in the observed hearing loss will be discussed. [Work supported by ONR Marine Mammal S&T Program and the U.S. Navy CNO(N45).]

Auditory/vocal response time was separated into perception time (PT) and movement time (MT) in trials with bottlenose dolphins (Tursiops truncatus)—two males and one female. Pressure catheters accepted into the nasal cavity by each dolphin recorded the pressure increase that preceded sound production. Time from acoustic stimulus onset to onset of pressure rise was recorded as PT (range 57 to 314 ms) and pressure rise onset to dolphin sound onset was recorded as MT (range 63 to 363 ms). Blindfolded dolphins trained to report a target by whistling often responded before completion of their 200‐ to 800‐ms echolocation click trains. Detection of the target, indicated by whistling, before termination of the animal’s own click train, suggests that dolphins do not voluntarily respond to each successive click but rather set a rhythm such that each click is emitted about 20 ms after the target echo arrives.

Three Atlantic bottlenose dolphins (Tursiops truncatus) were given a target recognition biosonar task. During their performance of the task, both acoustic data in the far field and pressure within the bony nasal passages were digitally recorded (Elsberry et al., 1999). Analysis of over 15 000 biosonar clicks provided new insights into odontocete biosonar sound production and is consistent with acoustic and pressure data taken from white whales (Delphinapterus leucas) during biosonar (Ridgway and Carder, 1988). Our work provides the first evidence for a minimum intranarial pressure during biosonar click production for any odontocete (11.8±0.5 kPascals over basal pressure). All three subjects exhibited nearly the same minimum tranarial pressure difference during biosonar click production. Clicks produced at or near this minimum intranarial pressure exhibited a wide range of acoustic power values. The acoustic power of a biosonar click was not highly correlated with intranarial pressure (R2=0.116). The ra...

A behavioral response paradigm was used to measure underwater hearing thresholds in two bottlenose dolphins (Tursiops truncatus) before and after exposure to intense tones. The fatiguing sounds were 3 kHz, 1 s tones. Hearing thresholds were measured at 3 and 4.5 kHz using a vocal response technique and a modified version of the Method of Free Response. Tests were conducted in a 6 m diam, 1.5 m deep vinyl‐walled swimming pool. Ambient noise spectral density levels in the pool were approximately 44 dB re: 1 Pa2/Hz. Postexposure threshold measurements normally began within two minutes of exposure and continued for 1015 min following exposure. A temporary threshold shift (TTS) was defined as a 6 dB or larger increase in threshold compared to the corresponding pre‐exposure level. Preliminary results indicate no TTS at 4.5 kHz for either subject at received SPLs of 200 dB re: 1 Pa. These data, as well as the test paradigm, absolute hearing thresholds, and observed behavioral reactions, will be presented and compared to existing TTS data for odontocetes.

Intense anthropogenic underwater sound may adversely effect the hearing and behavior of many marine mammals. Exposure to intense sound may produce an elevated hearing threshold, also known as a threshold shift (TS). If the threshold returns to the pre‐exposure level after a period of time, the TS is known as a temporary threshold shift (TTS); if the threshold does not return to the pre‐exposure level, the TS is called a permanent threshold shift (PTS). PTS and TTS data were used to establish noise exposure limits in humans and TTS data are often used to help predict the effects of anthropogenic noise on marine mammals in the wild. Previous studies of TTS in marine mammals conducted at SSC San Diego have examined the effects of exposure to single 1 s pure tones and underwater impulsive waveforms. This paper reviews the TTS studies conducted at SSC San Diego over the last 5 years and summarizes the resulting data. The application of these data to the development of damage‐risk criteria for marine mammals will also be discussed.

A behavioral response paradigm was used to measure masked underwater hearing thresholds in a bottlenose dolphin (Tursiops truncatus) and white whale (Delphinapterus leucas) before and after exposure to single underwater impulses produced by a seismic watergun. Hearing thresholds were measured at 0.4, 4, and 30 kHz using a vocal response technique and a modified version of the Method of Free Response. A pre‐exposure threshold was generally measured 34 min before presentation of the impulse. Post‐exposure threshold measurements began within 2 min of exposure to the impulse and continued for approximately 20 min following exposure, in order to track recovery. A masked temporary threshold shift (MTTS) was defined as a 6‐dB or larger increase in threshold compared to the corresponding pre‐exposure measurement. The MTTS test paradigm will be discussed and preliminary MTTS and behavioral response data presented and compared to existing TTS data for odontocetes. [Work supported by ONR.]

Stebbins (1966) showed that auditory RT of monkeys could be used to measure loudness. Such studies have been replicated in other animals and humans. Although Ridgway et al. (1991) reported that dolphin auditory RT varied with stimulus intensity, curves showing variation in RT with frequency and amplitude were not presented. Choice RT was measured in a female bottlenose dolphin responding to 300 msec bursts of clicks (30 clicks at 10 ms ISI) with a burst of pulses, and to 300 msec tones with a whistle. Responses were received with a hydrophone (B&K 8103), digitized, and stored on disk. The dolphin responded correctly (whistled to the tone, pulsed to the pulses) on more than 98% of presentations. Here we only consider tone presentations. Frequencies, presented randomly, were 5, 10, 20, 40, 60, 80, 100, and 120 kHz and ranged from 85 to 135 dB re 1 Pa rms. At 10 kHz, median RT ranged from 240 msec at 135 dB to 410 msec at 85 dB. At 100 kHz, median RT ranged from 260 msec at 135 dB to 330 msec at 85 dB. Studies must be done over a larger range of amplitude and frequency and with precise control of background noise.

Cetaceans use sound in foraging, communication, and navigation. There is growing concern that human‐made sound is a potentially serious auditory problem for cetaceans. Available data were inadequate to allow confident predictions regarding the levels of sound that should be of concern. An accepted method of determining the level of sound that produces a temporary reduction in the ears ability to respond fully is called temporary threshold shift (TTS). In the open waters of San Diego Bay, masked hearing thresholds were estimated for dolphins, Tursiops truncatus, and white whales, Delphinapterus leucas, before and after exposure to louder 1‐s tones at five sonar frequencies. Masking noise was used to create a floor effect, thereby eliminating threshold variability in the ever‐changing ambient noise environment in the natural waters where tests were conducted. The criterion for masked TTS (MTTS) was a 6‐dB increase in threshold over baseline. Subjects exhibited alterations in their behavior starting at levels around 180 dB, and experienced MTTS at levels between 192‐202 dB. Maximum levels of sound exposure without risk of hearing damage can be determined with the MTTS approach. More research employing different sound sources is needed. [Supported by Office of Naval Research.]

Masking noise is often used in hearing tests to create a floor effect in the presence of ambient noise or to examine specific features of the auditory system (e.g., the critical bandwidth). One of the chief requirements of the masking noise is that it possess a flat frequency spectrum within some user‐defined bandwidth. Generation of suitable masking noise is complicated by the frequency response of the sound projector, which may possess a frequency‐dependent transmitting sensitivity and/or exhibit resonances within the desired frequency range. At low frequencies acoustic standing waves may also alter the noise frequency spectrum. To overcome these limitations, a technique has been developed to generate bandlimited noise whose frequency content is compensated in order to flatten peaks or valleys in the measured frequency spectrum. Compensation is performed by passing white noise through a digital filter whose coefficients are determined from previous measurements of the acoustic system frequency response. The system has been implemented using a personal computer with commercial hardware and custom software. The method has been used to quickly generate bandlimited Gaussian and uniform white noise for studies of masked underwater hearing thresholds in marine mammals. [Work supported by ONR and the NRC Research Associateship Program.]

In an effort to develop acoustic safety criteria for marine mammals that may be exposed to impulsive sounds a masked temporary threshold shift (MTTS) experimental approach was used. Two dolphins, Tursiops truncatus, and two white whales, Delphinapterus leucas, were systematically exposed to increasing levels of impulsive sound (S‐1) under water, then tested to threshold at three pure‐tone frequencies (S‐2) under water in the presence of controlled masking noise. One hearing frequency tested was at the pulse peak frequency, with the other two S‐2 test tones at higher frequencies. Behavioral responses were monitored and hearing thresholds were ascertained to determine if criterion (6‐dB increase over baseline) was reached. Thresholds were estimated using the free response paradigm method and an up–down staircase psychophysical procedure for stimulus presentation in open water in San Diego Bay. These data suggest that the rapid rise time of the leading edge in 1‐s tones used in a previous study is not responsible for producing MTTS. Results of the present study add to the existing knowledge of MTTS for marine mammals and provide the first data on MTTS in cetaceans for impulsive sounds. [Work supported by the Office of Naval Research.]

Critical bandwidth is an important measure of the frequency selectivity within the auditory filters of the ear. In studies of masked hearing thresholds, the bandwidth of the masking noise must be greater than the subject’s critical bandwidth for the frequency of interest. There have been few direct measurements of critical bandwidths for dolphins and none exist for white whales. Underwater pure‐tone hearing thresholds were estimated for three dolphins, Tursiops truncatus, and one white whale, Delphinapterus leucas, at 20 and 30 kHz. Band‐limited Gaussian masking noise projected at 80 and 90 dB re: 1 μPa2/Hz was centered around the test frequencies. Six different noise bandwidths were used in threshold estimation and were presented in a quasi‐random order. Bandwidths had Q values of 0.8, 1, 1.3, 2, 4 and 10, where Q is the ratio of the center frequency to bandwidth. Thresholds were estimated using the method of free response paradigm and an up–down staircase psychophysical procedure for stimulus presentation. Results of the present study add to the existing knowledge of critical bandwidths for dolphins and provide the first data on critical bandwidths for white whales. [Work supported by the Office of Naval Research.]

Determining the site of the delphinid sonar signal generation has eluded cetologists for decades. Activities within the pharyngeal and nasal cavities of two bottlenose dolphins (Tursiops truncatus) were studied during sound production and echolocation. A high‐speed dual‐camera video system provided synchronized windows for recording two concomitant events: (1) movements visible through an endoscope and (2) oscilloscope traces of acoustic pressure at a hydrophone placed near the animal’s head. Dolphins have two tissue complexes, one located on either side, and just above, the membranous nasal septum [Cranford et al., J. Morph. 228, 223–285 (1996)]. They apparently generate acoustic pulses by pushing air across sets of internal ‘‘lips.’’ The acoustic pulse occurs coincident with one oscillatory cycle of the lips. Changes in the acoustic pulse repetition rate and the vibration cycles of the lips are simultaneous, indicating that their rates and periods are synchronous. No other structures were found to vibrate in synchrony with each acoustic pulse generation event. The palatopharyngeal muscle complex compresses air for the system. These observations settle a long‐standing controversy over the site of biosonar signal generation in odontocetes and open a vista of potential avenues for future investigations. [Work supported by the Office of Naval Research.]

In examining the potential impact of human‐made sound on sea mammals, it was considered that whale hearing sensitivity might diminish with increasing ambient pressure. To test the effect of depth, two white whales made 885 dives to a platform at 5, 100, 200, or 300 m in the Pacific Ocean. At each stationing on the platform up to 12 min at a time, whales whistled when they heard a 500‐ms tone from a hydrophone. With increasing depth, air density increase in the middle ear, sinuses, and nasal cavity changed each whale’s whistle response, but did not attenuate hearing as it does in the aerial ear (of humans and other land mammals tested in pressure chambers) due to middle ear impedance changes. The findings support theories that sound is conducted through whale head tissues to the ear without the usual ear drum/ossicular chain amplification of the aerial middle ear. These first ever hearing tests in the open ocean demonstrate that whales hear as well at depth as near the surface, therefore, zones of influence for human‐made sounds are just as great throughout the depths to which whales dive, or at least to 300 m. [Work supported by Office of Naval Research N0001496WK30349.]

After seizing a fish, small odontocetes often produce a rapid series of pulses that we call a ‘‘Victory squeal.’’ Animals trained to dive to depth and turn off a switch that was emitting a continuous tone [S. H. Ridgway et al., Science 166, 1651–1654; Can. J. Zool. 62, 2349–2351] made the victory squeal (Vs) within 1 s of tone termination (reinforcement signal or Sr). When dolphins (Tursiops truncatus) or white whales (Delphinapterus leucas) were trained for other tasks, a Vs was commonly produced after the Sr (in these cases Sr was a whistle, tone, or underwater buzzer activated by the trainer for correct responses). At a depth of 100 m or more, Vs peak frequency of the white whale was lower than at the surface (p

The echolocation signals of the same beluga whale (Delphinapterus leucas) were measured first in San Diego Bay, and later in Kaneohe Bay, Oahu, Hawaii. The ambient noise level in Kaneohe Bay is typically 12 – 17 dB greater than in San Diego Bay. The whale demonstrated the adaptiveness of its biosonar by shifting to higher frequencies and intensities after it was moved to Kaneohe. In San Diego, the animal emitted echolocation signals with peak frequencies between 40 and 60 kHz, and bandwidths between 15 and 25 kHz. In Kaneohe, the whale shifted its signals approximately an octave higher in frequencies with peak frequencies between 100 and 120 kHz, and bandwidths between 20 and 40 kHz. Signal intensities measured in Kaneohe were up to 18 dB higher than in San Diego. The data collected represent the first quantitative evidence of the adaptive capability of a cetacean biosonar system.

We recorded the auditory brainstem response (ABR) in four dolphins (Tursiops truncatus and Delphinus delphis). The ABR evoked by clicks consists of seven waves within 10 msec; two waves often contain dual peaks. The main waves can be identified with those of humans and laboratory mammals; in spite of a much longer path, the latencies of the peaks are almost identical to those of the rat. The dolphin ABR waves increase in latency as the intensity of a sound decreases by only 4 microseconds/decibel(dB) (for clicks with peak power at 66 kHz) compared to 40 microseconds/dB in humans (for clicks in the sonic range). Low-frequency clicks (6-kHz peak power) show a latency increase about 3 times (12 microseconds/dB) as great. Although the dolphin brainstem tracks individual clicks to at least 600 per sec, the latency increases and amplitude decreases with increasing click rates. This effect varies among different waves of the ABR; it is around one-fifth the effect seen in man. The dolphin brain is specialized for handling brief, frequent clicks. A small latency difference is seen between clicks 180 degrees different in phase--i.e., with initial compression vs. initial rarefaction. The ABR can be used to test theories of dolphin sonar signal processing. Hearing thresholds can be evaluated rapidly. Cetaceans that have not been investigated can now be examined, including the great whales, a group for which data are now completely lacking.

1. 1. In the laboratory, eye movements in the horizontal and vertical planes of a normal human and two adult dolphins, Tursiops truncatus , were analyzed and compared. Several of the visual conditions included structured, unstructured and moving stimulus fields. 2. 2. Power spectral density analyses of the dolphin eye movements showed maximal power around 0.1 Hz, lower than the human counterpart. 3. 3. Coefficients of determination ( r 2 ) and correlation coefficients ( r ) derived from autocorrelation and crosscorrelation of dolphin eye signals produced typically small values scattered randomly around zero. 4. 4. From 51,200 correlations, the greatest r 2 value indicated that movements of one dolphin eye predict no more than 30% of the movement variance of the fellow eye. 5. 5. Although dolphin eyes are mobile at lower fundamental frequencies than in humans, there is a low level of synchrony between the two eyes.

Indirect evidence for multiple sonar signal generators in odontocetes exists within the published literature. To explore the long-standing controversy over the site of sonar signal generation, direct evidence was collected from three trained bottlenose dolphins ( Tursiops truncatus ) by simultaneously observing nasal tissue motion, internal nasal cavity pressure, and external acoustic pressure. High-speed video endoscopy revealed tissue motion within both sets of phonic lips, while two hydrophones measured acoustic pressure during biosonar target recognition. Small catheters measured air-pressure changes at various locations within the nasal passages and in the basicranial spaces. Video and acoustic records demonstrate that acoustic pulses can be generated along the phonic fissure by vibrating the phonic labia within each set of phonic lips. The left and right phonic lips are capable of operating independently or simultaneously. Air pressure in both bony nasal passages rose and fell synchronously, even if the activity patterns of the two phonic lips were different. Whistle production and increasing sound pressure levels are generally accompanied by increasing intranarial air pressure. One acoustic “click” occurred coincident with one oscillatory cycle of the phonic labia. Changes in the click repetition rate and cycles of the phonic labia were simultaneous, indicating that these events are coupled. Structural similarity in the nasal apparatus across the Odontoceti suggests that all extant toothed whales generate sonar signals using the phonic lips and similar biomechanical processes.

At the Jersey conference on Animal Sonar Systems in 1979 we gave strong evidence that dolphins produce sound in the nasal system rather than in the larynx as most mammals do (Ridgway et al. 1980). With electromyography (EMG), we studied the activity of laryngeal muscles and nasal muscles, making comparisons between the two groups of muscles during sound production. Certain muscles of the nasal system were active during all dolphin whistles and click trains while muscles of the larynx were active during respiration but not during sound production. Perhaps more importantly, we measured pressure in the nasal cavities and in the trachea adjacent to the larynx. During sound production, intranasal pressure increased markedly but intratracheal pressure remained unchanged. Subsequently, Amundin and Andersen (1983) replicated the EMG and pressure monitoring aspects of our study in Tursiops and Phocoena.

HEART rate is usually reduced when aquatic mammals dive. The degree of this reduction, known as bradycardia, is remarkable1–3, and immersion in water has been regarded as important in its initiation3,4. The heart of the phocid seal beats slower in forced or restrained dives than during trained or voluntary dives5,6, and bradycardia occurs during respiratory pauses in air as well as in water7–9. Central nervous system (CNS) control of bradycardia has been suggested5,6,8,9. We have investigated whether bradycardia can be conditioned in the California sea lion Zalophus californianus since cardiovascular responses can be conditioned in man and many other animals10.

Summary Although dolphins ( Tursiops truncatus ) have been trained to match numbers and durations of human vocal bursts [1] and reported to spontaneously match computer-generated whistles [2], spontaneous human voice mimicry has not previously been demonstrated. The first to study white whale ( Delphinapterus leucas ) sounds in the wild, Schevill and Lawrence [3] wrote that "occasionally the calls would suggest a crowd of children shouting in the distance". Fish and Mowbary [4] described sound types and reviewed past descriptions of sounds from this vociferous species. At Vancouver Aquarium, Canada, keepers suggested that a white whale about 15 years of age, uttered his name "Lagosi". Other utterances were not perceptible, being described as "garbled human voice, or Russian, or similar to Chinese" by R.L. Eaton in a self-published account in 1979. However, hitherto no acoustic recordings have shown how such sounds emulate speech and deviate from the usual calls of the species. We report here sound recordings and analysis which demonstrate spontaneous mimicry of the human voice, presumably a result of vocal learning [5], by a white whale.

The echolocation signals of a Beluga whale (Delphinapterus lecus) were measured, first in San Siego Bay, and later in Kaneohe Bay, Oahu, Hawaii, while the animal was performing a target detection task. The ambient noise of both bays is dominated by snapping shrimp, with the noise level in Kaneohe Bay typically 20 dB greater than in San Diego Bay. In San Diego Bay, the animal emitted clicks with peak frequencies between 40 and 60 kHz, and bandwidths between 15 and 25 kHz. When the animal was moved to Kaneohe Bay, it used higher frequency clicks with peak frequencies between 100 and 120 kHz, and bandwidths between 20 and 40 kHz. The shift in the peak frequencies was probably caused by the difference in the ambient noise environment between the two sites, and may represent an attempt by the animal to optimize the signal‐to‐noise ratio during the target detection task. The whale also emitted signals with higher source levels and source energy flux densities in Kaneohe Bay than in San Diego Bay, even though the detection task in Kaneohe Bay was relatively simple and the task in San Diego Bay was relatively difficult.

In the great majority of cetaceans, the sense of hearing has not been studied. There are no audiograms on any of the great whales. To us, the ABR (Averaged Brainstem Response) technique seems the only way to get such data on the larger species. We have used the ABR in smaller cetaceans such as Delphinus, Tursiops, and Delphinapterus to record audiometric data. Fine wires about the size of a human hair are inserted into the head through a hypodermic needle. Bare tips of the insulated wire over the vertex and mastoid region pick up electrical potentials (electroencephalogram or EEG). The EEG is amplified, filtered, and tape recorded during periods when sound pulses are presented through a hydrophone. We observe the ABR by synchronous averaging of the EEG for 10 to 25 ms after each pulse. The sound intensity and frequency are varied to determine the threshold at each frequency, thereby producing an audiogram for the animal. Audiometric data would be helpful in assessing the impact of human generated acoustic events on whales. We have developed a portable system that we hope will enable us to make measurements on whales trapped in fishermen's nets, cetaceans stranded in tidal basins, or animals that beach themselves.

Little is known about the ontogeny of echolocation in dolphins. A bottlenosed dolphin calf born at our facility in San Diego Bay was observed continuously for two hours each day and randomly at other times. Squeals were heard about ten seconds after birth and whistlelike calls soon thereafter but we did not notice head scanning movements concurrent with high‐frequency pulses until the dolphin's sixtieth day of life. On this day we noticed that the calf would examine items such as a human foot inserted in the water by approaching, head scanning, and pulsing. A hydrophone placed in the water elicited the same behavior. Seven pulse trains were recorded at distances of 1–3 m. Peak frequencies ranged from 33 to 120 kHz with 3‐dB bandwidths of 28–81 kHz.