354
P.E. Nachtigall et al.
difference in terms of the magnitude of modulation was systematically
varied using an up-down tracking procedure in which the percentage of frequency deviation of the frequency-modulated signal about its center frequency was decreased following correct responses and increased following
errors. Using pure tone frequencies between 1 and 70 kHz, they found very
good discrimination ability between 6 and 50 kHz. Data were reported using
a common psychophysical index, the Weber ratio. This is a ratio of the
minimum detectable change in a stimulus to the starting level of the stimulus (Gescheider 1997). In this study, the physical property of the stimulus
under evaluation was frequency (f), producing a Weber ratio of Nlf.
Between 6 and 50 kHz, Weber ratios were in the range of 0.002 to 0.003.
Unlike data previously reported by Jacobs (1972), this region of best frequency discrimination corresponded to the dolphin's area of peak hearing
sensitivity (Johnson 1966). The lowest frequency discrimination Weber
ratios for humans range from 0.0025 (at 400Hz) to 0.0016 (at 2,000Hz)
(Weir et al. 1976), and represent the best such capabilites reported for any
mammal.
Additional work with the same animal and a similar procedure
(Thompson and Herman 1975) examined the dolphin's frequency discrimination capabilities up to 140kHz. The ability remained remarkably good
at these higher frequencies with Weber ratios (l::1f/fJ never exceeding 0.008
between 2 and 130 kHz. By comparison, the value reported for a human
subject at 8kHz was 0.0085, with no results reported for higher frequencies
(Weir et al. 1976). Despite the potential importance of this finding in terms
of animal echolocation discrimination, no other odontocete has been
similarly examined.
4.2 Tonal Masking
Masking occurs when the presence of one sound (a masker) interferes with
the perception of another sound (a signal). A simple experiment to examine
the masking effects of tones on the detection of other tonal signals was conducted on T. truncatus by Johnson (1971). In this experiment the masking
tone was a 70-kHz signal held constant in level at 40 and 80dB above
threshold. Signals were tones from 20 to 100 kHz.
Johnson found that signals less than 50 to 60 kHz and greater than 85 kHz
were only weakly masked «10dB) by a 70-kHz tone. The masking effect
was greatest as the difference between the signal and masker frequencies
was reduced. The masker 40dB above threshold masked signals within 5
to 10kHz by as much as 30dB, while the masker 80dB above threshold
masked signals within 5-10kHz by nearly 60dB. As with human listeners,
the dolphin apparently could perceive beats when the signals were within
5 kHz of the masker, and the amount of masking decreased accordingly in
these cases. Overall, the shapes of the masking patterns were similar to
those obtained from humans at much lower frequencies (Wegel and Lane
P.E. Nachtigall et al.
difference in terms of the magnitude of modulation was systematically
varied using an up-down tracking procedure in which the percentage of frequency deviation of the frequency-modulated signal about its center frequency was decreased following correct responses and increased following
errors. Using pure tone frequencies between 1 and 70 kHz, they found very
good discrimination ability between 6 and 50 kHz. Data were reported using
a common psychophysical index, the Weber ratio. This is a ratio of the
minimum detectable change in a stimulus to the starting level of the stimulus (Gescheider 1997). In this study, the physical property of the stimulus
under evaluation was frequency (f), producing a Weber ratio of Nlf.
Between 6 and 50 kHz, Weber ratios were in the range of 0.002 to 0.003.
Unlike data previously reported by Jacobs (1972), this region of best frequency discrimination corresponded to the dolphin's area of peak hearing
sensitivity (Johnson 1966). The lowest frequency discrimination Weber
ratios for humans range from 0.0025 (at 400Hz) to 0.0016 (at 2,000Hz)
(Weir et al. 1976), and represent the best such capabilites reported for any
mammal.
Additional work with the same animal and a similar procedure
(Thompson and Herman 1975) examined the dolphin's frequency discrimination capabilities up to 140kHz. The ability remained remarkably good
at these higher frequencies with Weber ratios (l::1f/fJ never exceeding 0.008
between 2 and 130 kHz. By comparison, the value reported for a human
subject at 8kHz was 0.0085, with no results reported for higher frequencies
(Weir et al. 1976). Despite the potential importance of this finding in terms
of animal echolocation discrimination, no other odontocete has been
similarly examined.
4.2 Tonal Masking
Masking occurs when the presence of one sound (a masker) interferes with
the perception of another sound (a signal). A simple experiment to examine
the masking effects of tones on the detection of other tonal signals was conducted on T. truncatus by Johnson (1971). In this experiment the masking
tone was a 70-kHz signal held constant in level at 40 and 80dB above
threshold. Signals were tones from 20 to 100 kHz.
Johnson found that signals less than 50 to 60 kHz and greater than 85 kHz
were only weakly masked «10dB) by a 70-kHz tone. The masking effect
was greatest as the difference between the signal and masker frequencies
was reduced. The masker 40dB above threshold masked signals within 5
to 10kHz by as much as 30dB, while the masker 80dB above threshold
masked signals within 5-10kHz by nearly 60dB. As with human listeners,
the dolphin apparently could perceive beats when the signals were within
5 kHz of the masker, and the amount of masking decreased accordingly in
these cases. Overall, the shapes of the masking patterns were similar to
those obtained from humans at much lower frequencies (Wegel and Lane