Detection 131and the lateral superior olive is most sensitive to interaural
level differences. Interaural time and level differences are im-
portant cues for sound localization (Yost & Gourevitch,
1987). The lateral limniscus is primarily a monaural pathway
in many animals. The inferior colliculus appears to be a
major processing nucleus for spatial hearing, modulation
processing, and spectral pattern recognition (Fay & Popper,
1992). In mammalian systems, not a lot is known about the
function of the medial geniculate body (Fay & Popper, 1992).
The auditory cortex in primates is located deep within the
Sylvian fissure, making it difficult to reach for physiological
study. The auditory cortex, as are all parts of the central audi-
tory nervous system, is tonotopically organized: Different
cortical neurons are selective for different frequencies. There
is evidence for modulation processing in the auditory cortex,
and the auditory cortex may provide spatial maps for sound
localization (Altschuler et al., 1989).
The study of animals with special adaptations for hearing,
echo-locating bats (Suga, 1988), and the barn owl (Konishi,
Takahashi, Wagner, Sullivan, & Carr, 1988), have provided
valuable information about the functional role of the central
auditory system. These studies have helped guide the study
of the brain stems and cortices of other animals, including
humans. The auditory nervous system is an anatomically
complex system, perhaps reflecting the amount of neural
computation that hearing appears to require.
DETECTION
Thresholds of Hearing
A basic measure of auditory detection is the threshold of
hearing for pure tones, or the audiogram. The audiogram can
be obtained in two conditions, each requiring its own calibra-
tion procedure. In the minimal audible field (MAF) process,
listeners detect the presence of pure tones presented from
loudspeakers, whereas in the minimal audible pressure
(MAP) process, the sounds are presented over headphones.
Figure 5.10 shows the thresholds of hearing (the audiogram)
for the two procedures. The figure also shows estimates of the
upper limit for hearing, indicating those sound levels that
either are very uncomfortable or yield the sensation of pain.
The thresholds of hearing have been standardized for both the
MAP and MAF procedures, and the two estimates differ by
on average about 6 dB. However, the differences are ac-
counted for by calculating the diffraction of sound around the
head and the resonance properties of the outer ear canal,
which are substantially different in the MAP and MAF pro-
cedures (Yost & Killion, 1993). Figure 5.10 suggests that
young humans can detect sounds from 20 to 20,000 Hz, and
the dynamic range of hearing is about 130 dB in the middle
of the range of the audible frequencies. At 0 dB SPL the pres-
sure in the outer ear canal is 20 Pa, which indicates that the
tympanic membrane at auditory threshold is moving a dis-
tance equal to approximately the diameter of a hydrogen
atom. Females have slightly lower thresholds of hearing than
do males. The thresholds of hearing increase as a function
of age in a frequency-dependent manner (presbycusis), such
that the thresholds for high-frequency sounds increase at an
earlier age than do those for low-frequency sounds. This
frequency dependence is consistent with the operation of the
traveling wave, in which all sounds excite the base of the
cochlear partition, where high-frequencies are coded, but
only low-frequencies sounds excite the apex. Thus, the base
of the cochlear partition, where high frequencies are coded, is
more likely to be fatigued over time than is the apex.
Figure 5.10 shows the threshold levels for tonal detection.
The subjectively perceived loudness of sound is also a joint
function of the sound’s physical frequency and level. Fig-
ure 5.11 shows subjective equal-loudness contours; each
contour describes the tonal levels and frequencies that are
judged, in a loudness matching procedure, equally loud to a
1,000-Hz tone presented at the constant level indicated by the
phon rating of the contour. Thus, all tones that have a loud-
ness of xphons are judged equally loud to a 1,000-Hz tone
presented at xdB SPL.
The thresholds of hearing are dependent on the duration of
the sound—the shorter the sound, the higher the thresholds.
Thresholds for tonal stimuli decrease as duration is increased
until the duration is approximately 300 ms; then threshold
remains constant as duration is increased further. To a firstFigure 5.10 The thresholds of hearing in decibels of sound pressure level
(dB SPL) are shown as a function of frequency for Minimal Audible Field
(MAF) thresholds, MAP, thresholds for pain, and thresholds for discomfort.
The thresholds for pain and discomfort represent estimates of the upper limit
of level that humans can tolerate. Source: These thresholds are based on
national and international standards.