8. Psychoacoustic Studies of Dolphins and Whales
351
the sound is substantially smaller than the diameter of the object casting a
shadow. A dolphin head of approximately 20cm in diameter corresponds
to the wavelength of a 7.5 kHz. signal. Interaural intensity differences
increase at greater frequencies. Using the same estimated 70% correct
thresholds the minimum interaural intensity difference detectable by the
dolphin was 0.5 dB at 10 and 20 kHz. It was 0.6 dB at 40 and 60 kHz, and was
0.7 dB at 80 kHz. There is no simple way to estimate the minimum audible
angle from these thresholds. Nevertheless, the fact that they remain fairly
constant across a range of frequencies implies that they should be increasingly valuable as localization cues as the frequency of the signal increases.
As frequency increases, the density of the sound shadow cast by the head
increases, thereby increasing the interaural intensity difference.
Supin and Popov (1993) investigated the relative spectral sensitivity of
the two ears of an Amazon River dolphin (I. geoffrensis) as a function the
angle of the sound source relative to the dolphin's head. Signals were tonal
pips between 20 and 100kHz. Each pip was a sinusoid with a lO0J.ls rise
time, a 100 J.ls fall time and a 100 J.ls dwell time. They measured the dolphin's
auditory nerve response using scalp electrodes. The response contained a
brief positive-going spike at about 1.6 to 1.7 ms after the onset of the signal.
The response of the ear on the side from which the pip was played (ipsilateral ear) was greater in amplitude and shorter in latency than the
response from other (contralateral) ear. The peak response was obtained
with a signal that was located 5°_10° to the side of the recorded ear. The
higher the frequency, the more steeply the sensitivity of the ear changed as
a function of direction.
The magnitude of the auditory nerve response is an indicator of the
received amplitude of the signal at the measured ear as shown by the
increased amplitude of the response as a function of the amplitude of
the signal presented at a fixed azimuth. Therefore, the difference between
the amplitude of the response at ipsilateral versus the contralateral ear (that
is the ear toward which the sound was located versus the other ear) is a
measure of the interaural intensity difference (lID) detected by the animal.
For high-frequency signals (50kHz and above), the measured lID increased
sharply and then decreased as the azimuth of the sound increased. This
pattern corresponds to the displacement in the location of the peak sensitivity of each ear toward the ipsilateral side. The relatively low responses
obtained to sounds at more extreme angles and the flatness of the curves
beyond about 30° (dependent on frequency) corresponds to the flatness of
the lID at these angles.
While it took some time to verify that the dolphin functionally used two
ears recent data (Mohl 1999) suggested that a simple model with only two
maximum sound entry points may well be simpler than the actual peripheral auditory system of the dolphin. In this work a dolphin was trained to
sit quietly and wear suction cup Auditory Brainstem Response (ABR)
electrodes on the surface of its skin while various sounds were presented
351
the sound is substantially smaller than the diameter of the object casting a
shadow. A dolphin head of approximately 20cm in diameter corresponds
to the wavelength of a 7.5 kHz. signal. Interaural intensity differences
increase at greater frequencies. Using the same estimated 70% correct
thresholds the minimum interaural intensity difference detectable by the
dolphin was 0.5 dB at 10 and 20 kHz. It was 0.6 dB at 40 and 60 kHz, and was
0.7 dB at 80 kHz. There is no simple way to estimate the minimum audible
angle from these thresholds. Nevertheless, the fact that they remain fairly
constant across a range of frequencies implies that they should be increasingly valuable as localization cues as the frequency of the signal increases.
As frequency increases, the density of the sound shadow cast by the head
increases, thereby increasing the interaural intensity difference.
Supin and Popov (1993) investigated the relative spectral sensitivity of
the two ears of an Amazon River dolphin (I. geoffrensis) as a function the
angle of the sound source relative to the dolphin's head. Signals were tonal
pips between 20 and 100kHz. Each pip was a sinusoid with a lO0J.ls rise
time, a 100 J.ls fall time and a 100 J.ls dwell time. They measured the dolphin's
auditory nerve response using scalp electrodes. The response contained a
brief positive-going spike at about 1.6 to 1.7 ms after the onset of the signal.
The response of the ear on the side from which the pip was played (ipsilateral ear) was greater in amplitude and shorter in latency than the
response from other (contralateral) ear. The peak response was obtained
with a signal that was located 5°_10° to the side of the recorded ear. The
higher the frequency, the more steeply the sensitivity of the ear changed as
a function of direction.
The magnitude of the auditory nerve response is an indicator of the
received amplitude of the signal at the measured ear as shown by the
increased amplitude of the response as a function of the amplitude of
the signal presented at a fixed azimuth. Therefore, the difference between
the amplitude of the response at ipsilateral versus the contralateral ear (that
is the ear toward which the sound was located versus the other ear) is a
measure of the interaural intensity difference (lID) detected by the animal.
For high-frequency signals (50kHz and above), the measured lID increased
sharply and then decreased as the azimuth of the sound increased. This
pattern corresponds to the displacement in the location of the peak sensitivity of each ear toward the ipsilateral side. The relatively low responses
obtained to sounds at more extreme angles and the flatness of the curves
beyond about 30° (dependent on frequency) corresponds to the flatness of
the lID at these angles.
While it took some time to verify that the dolphin functionally used two
ears recent data (Mohl 1999) suggested that a simple model with only two
maximum sound entry points may well be simpler than the actual peripheral auditory system of the dolphin. In this work a dolphin was trained to
sit quietly and wear suction cup Auditory Brainstem Response (ABR)
electrodes on the surface of its skin while various sounds were presented