Monaural and dichotic forward masking in the dolphin's auditory system.

Short-latency auditory-evoked potentials (AEPs) were recorded non-invasively in the bottlenose dolphin Tursiops truncatus. The stimuli were two sound clicks that were played either monaurally (both clicks to one and the same acoustic window) or dichotically (the leading stimulus (masker) to one acoustic window and the delayed stimulus (test) to the other window). The ratio of the levels of the two stimuli was 0, 10, or 20 dB (at 10 and 20 dB, the leading stimulus was of a higher level). The inter-stimulus intervals (ISIs) varied from 0.15 to 10 ms. The test response magnitude was assessed by correlation analysis as a percentage of the control (non-masked) response. At monaural stimulation, the test response was of a constant magnitude (5-6% of the control) at ISIs of 0.15-0.3 ms and recovered at longer ISIs. At dichotic stimulation, the deepest suppression of the test response occurred at ISIs of 0.5-0.7 ms. The response was slightly suppressed at short ISIs (0.15-0.3 ms) and recovered at ISIs longer than 0.5-0.7 ms. The relation of parameters of the forward masking to echolocation in dolphins is discussed.

Forward masking in a bottlenose dolphin Tursiops truncatus: dependence on azimuthal positions of the masker and test sources.

Forward masking was investigated by the auditory evoked potentials (AEP) method in a bottlenose dolphin Tursiops truncatus using stimulation by two successive acoustic pulses (the masker and test) projected from spatially separated sources. The positions of the two sound sources either coincided with or were symmetrical relative to the head axis at azimuths from 0 to ± 90°. AEPs were recorded either from the vertex or from the lateral head surface next to the auditory meatus. In the last case, the test source was ipsilateral to the recording side, whereas the masker source was either ipsi- or contralateral. For lateral recording, AEP release from masking (recovery) was slower for the ipsi- than for the contralateral masker source position. For vertex recording, AEP recovery was equal both for the coinciding positions of the masker and test sources and for their symmetrical positions relative to the head axis. The data indicate that at higher levels of the auditory system of the dolphin, binaural convergence makes the forward masking nearly equal for ipsi- and contralateral positions of the masker and test.

Spatial release from masking in a bottlenose dolphin Tursiops truncatus.

The impact of maskers on the receiving beam of a bottlenose dolphin, Tursiops truncatus, was investigated using the auditory evoked potential (AEP) method. The test signal was a train of tone pips with a 64 kHz carrier frequency. The stimulus produced AEPs as a sequence of waves replicating the pip rate: the rate following response (RFR). The masker was band-limited noise, with a passband of 45 to 90 kHz and a level of 105 dB re 1 µPa. Masker azimuths were at 0°, ±30°, ±60°, and ±90° relative to the head midline. The receiving beam was evaluated in terms of the RFR threshold dependence on the signal azimuth. The masked thresholds were higher than the baseline thresholds, which appeared mostly as a shift rather than a deformation in the receiving beam. The largest threshold shift appeared when the masker source was located in the most sensitive direction (zero azimuth); at lateral masker source positions, the threshold shift decreased. When the masker source was not at the head midline, the masked thresholds were higher at signal positions ipsilateral to the masker source than at positions contralateral to the masker source. The largest asymmetry was observed at the 30° masker azimuth in conjunction with the ±30° and ±120° signal azimuths; the asymmetries were 5.6 and 8.1 dB, respectively. This masking asymmetry was lower than expected from the previously found interaural intensity difference, which may be explained by the conflict between the test signal and the masker when it appeared at a binaural level of the auditory system.

Adaptation processes in the auditory system of a beluga whale Delphinapterus leucas.

The effects of prolonged sound stimuli (tone pip trains) on evoked potentials (the rate following response, RFR) were investigated in a beluga whale. The stimuli (rhythmic tone pips) were of 64 kHz frequency at levels from 80 to 140 dB re 1 µPa. During stimulation, every 1000 ms stimulus level either was kept constant (the steady-state stimulation) or changed up/down by 20 or 40 dB. With such stimulus presentation manner, RFR amplitude varied as follows. (i) After a stimulus level increase, the response amplitude increased quickly and then decayed slowly. The more the level increased, the higher the response amplitude increased. (ii) After a stimulus level decrease, the response amplitude was suppressed and then recovered slowly. The more the level decreased, the stronger was the response suppression. (iii) At the end of the 1000 ms window, the response amplitude approached, but did not reach, the amplitude characteristic of the steady-state stimulation. As a result, both after a sound level increase and decrease, the responses were almost stabilized during an analysis time as short as 1 s. This stabilization is attributed to an adaptation process. RFR decay after initial increase could be approximated by an exponent with a time constant of 59.4 ±1.8 (standard error) ms; RFR recovery after initial decrease could be approximated by an exponent with a time constant of 139.2 ±9.9 ms.

Influence of background noise on auditory evoked responses to rippled-spectrum signals.

The resolution of spectral patterns in adaptation background noise was investigated in a beluga whale, Delphinapterus leucas, using the evoked-potential technique. The resolution of spectral patterns was investigated using rippled-spectrum test stimuli of various levels and ripple densities and recording the rhythmic evoked responses (the rate following response, RFR) to ripple phase reversals. In baseline (no adaptation background noise) experiments, the highest RFR magnitude was observed at signal sound pressure levels (SPLs) of 100-110 dB re 1 µPa; at SPLs both below the optimum (down to 80 dB re 1 µPa) and above the optimum (up to 140 dB re 1 µPa), the RFR magnitude decreased. For high signal levels (above 110 dB re 1 µPa), low-level adaptation background noise (from -10 to -20 dB re signal level) increased RFR magnitude compared to baseline. This effect is considered to be a result of the optimization of the sensation level of the high-SPL signals due to decreasing hearing sensitivity caused by the adaptation background noise.

Influence of fatiguing noise on auditory evoked responses to stimuli of various levels in a beluga whale, Delphinapterus leucas.

The negative impact of man-made noise on the hearing of odontocetes has attracted considerable recent attention. In the majority of studies, permanent or temporary reductions in sensitivity, known as permanent or temporary threshold shift (PTS or TTS, respectively), have been investigated. In the present study, the effects of a fatiguing sound on the hearing of a beluga whale, Delphinapterus leucas, within a wide range of levels of test signals was investigated. The fatiguing noise was half-octave band-limited noise centered at 32 kHz. Post-exposure effects of this noise on the evoked responses to test stimuli (rhythmic pip trains with a 45-kHz center frequency) at various levels (from threshold to 60 dB above threshold) were measured. For baseline (pre-exposure) responses, the magnitude-versus-level function featured a segment of steep magnitude dependence on level (up to 30 dB above threshold) that was followed by a plateau segment that featured little dependence on level (30 to 55 dB above threshold). Post-exposure, the function shifted upward along the level scale. The shift was 23 dB at the threshold and up to 33 dB at the supra-threshold level. Owing to the plateau in the magnitude-versus-level function, post-exposure suppression of responses depended on the stimulus level such that higher levels corresponded to less suppression. The experimental data may be modeled based on the compressive non-linearity of the cochlea. According to the model, post-exposure responses of the cochlea to high-level stimuli are minimally suppressed compared with the pre-exposure responses, despite a substantially increased threshold.

Auditory sensitivity to local stimulation of the head surface in a beluga whale (Delphinapterus leucas).

Using the auditory evoked response technique, sensitivity to local acoustic stimulation of the ventro-lateral head surface was investigated in a beluga whale (Delphinapterus leucas). The stimuli were tone pip trains of carrier frequencies ranging from 16 to 128 kHz with a pip rate of 1 kHz. For higher frequencies (90-128 kHz), the low-threshold point was located next to the medial side of the middle portion of the lower jaw. For middle (32-64 kHz) and lower (16-22.5 kHz) frequencies, the low-threshold point was located at the lateral side of the middle portion of the lower jaw. For lower frequencies, there was an additional low-threshold point next to the bulla-meatus complex. Based on these data, several frequency-specific paths of sound conduction to the auditory bulla are suggested: (i) through an area on the lateral surface of the lower jaw and further through the intra-jaw fat-body channel (for a wide frequency range); (ii) through an area on the ventro-lateral head surface and further through the medial opening of the lower jaw and intra-jaw fat-body channel (for a high-frequency range); and (iii) through an area on the lateral (near meatus) head surface and further through the lateral fat-body channel (for a low-frequency range).

Frequency Tuning of Hearing in the Beluga Whale.

Data on frequency tuning in odontocetes are contradictory: different authors have reported filter qualities from 2 to almost 50. In this study, frequency tuning was measured in a beluga whale (Delphinapterus leucas) using a rippled-noise test stimulus in conjunction with the auditory evoked potential (AEP) technique. The response to ripple reversions was considered to indicate resolvability of the ripple pattern. The limit of ripple-pattern resolution ranged from 20 to 32 ripples per octave (rpo). A model of interaction of the ripple spectrum with frequency-tuned filters suggests that this resolution limit requires a filter quality of 29-46.