Impedance Matching in Sound Production and Hearing: a Comparative Study
53
mismatch, the auditory sensitivity is close to the level of ambient noise in the
environment.
Within the inner ear, impedance matching between the liquid medium and the
sensory cells is associated with frequency analysis. The sensory cells are attached
to the basilar membrane which, near to the oval window is narrow and stiff so it
can only vibrate at high frequencies but, towards the distal end of the cochlea, the
basilar membrane is both broad and compliant so it vibrates at lower frequencies:
this oversimplified account of cochlea mechanics highlights the fact that good
impedance matching may be highly frequency-dependent. More detailed accounts
of cochlear mechanics are found in von Bekesy (1960) or Fletcher (1992).
Other evidence for frequency-dependent impedance matching comes from
locust ears (Stephen and Bennet-Clark 1982), which are situated on either side of
the abdomen. Each ear consists of a tympanic membrane with a thick flexible
anterior region and a thin posterior dorsal region which bears three sclerites
(Fig.5A) to which the four different groups of the sensory cells that make up
MUller's organ are attached (Gray 1960). MUller's organ is suspended by its
sensory cells on the tympanum and connected to the rest of the insect only by the
tympanal nerve.
The thickness of the tympanal membrane varies from region to region: around
the folded and styliform bodies, it is between 2 and 4 11m thick, but there is a thin
compliant arc around the posterior edge of the tympanum that is less than 1 )!ill
thick. The different regions show independence of movement when driven by high
pressure sound (Stephen and Bennet-Clark 1982; Breckow and Sippel 1985).
Calculations of the resonant frequencies (Eqn. 3) of the different regions (Fig. 5B),
using an estimate of the mass of the different regions based on the thickness of the
membrane, and its measured compliance (Stephen and Bennet-Clark 1982), gave
values close to the measured frequencies of maximal amplitude of vibration of the
membrane (Breckow and Sippel 1985). The best frequencies of the C, B and D
groups of sensory cells (Michelsen 1971) suggest that the mechanics of the
membrane determines the different responses of the different regions (Stephen and
Bennet-Clark 1982; confirmed by Breckow and Sippel 1985).
The adequate stimulus requires mechanical strain of the sensory cells. The
attachment of MUller's organ to the tympanum is highly compliant and its
calculated resonant frequency is lower than that of any region of the tympanum
(Fig. 5B; Stephen and Bennet-Clark 1982). When the tympanum vibrates, the
relatively heavy MUller's organ vibrates at a lower amplitude and at a different
phase to the tympanum, so there is relative movement between the ends of the
sensory cells by which MUller's organ is suspended; this thereby provides
frequency-dependent adequate stimuli.
The locust ear also provides evidence of close-to-optimal impedance matching
between the tympanum and the sensory cells. When the vibration of the folded
body region was compared before and after removal of MUller's organ, ftrst, the
resonant frequency was unchanged but, second, the amplitude of vibration of the
unloaded membrane increased by 4 dB (Stephen and Bennet-Clark 1982). The
53
mismatch, the auditory sensitivity is close to the level of ambient noise in the
environment.
Within the inner ear, impedance matching between the liquid medium and the
sensory cells is associated with frequency analysis. The sensory cells are attached
to the basilar membrane which, near to the oval window is narrow and stiff so it
can only vibrate at high frequencies but, towards the distal end of the cochlea, the
basilar membrane is both broad and compliant so it vibrates at lower frequencies:
this oversimplified account of cochlea mechanics highlights the fact that good
impedance matching may be highly frequency-dependent. More detailed accounts
of cochlear mechanics are found in von Bekesy (1960) or Fletcher (1992).
Other evidence for frequency-dependent impedance matching comes from
locust ears (Stephen and Bennet-Clark 1982), which are situated on either side of
the abdomen. Each ear consists of a tympanic membrane with a thick flexible
anterior region and a thin posterior dorsal region which bears three sclerites
(Fig.5A) to which the four different groups of the sensory cells that make up
MUller's organ are attached (Gray 1960). MUller's organ is suspended by its
sensory cells on the tympanum and connected to the rest of the insect only by the
tympanal nerve.
The thickness of the tympanal membrane varies from region to region: around
the folded and styliform bodies, it is between 2 and 4 11m thick, but there is a thin
compliant arc around the posterior edge of the tympanum that is less than 1 )!ill
thick. The different regions show independence of movement when driven by high
pressure sound (Stephen and Bennet-Clark 1982; Breckow and Sippel 1985).
Calculations of the resonant frequencies (Eqn. 3) of the different regions (Fig. 5B),
using an estimate of the mass of the different regions based on the thickness of the
membrane, and its measured compliance (Stephen and Bennet-Clark 1982), gave
values close to the measured frequencies of maximal amplitude of vibration of the
membrane (Breckow and Sippel 1985). The best frequencies of the C, B and D
groups of sensory cells (Michelsen 1971) suggest that the mechanics of the
membrane determines the different responses of the different regions (Stephen and
Bennet-Clark 1982; confirmed by Breckow and Sippel 1985).
The adequate stimulus requires mechanical strain of the sensory cells. The
attachment of MUller's organ to the tympanum is highly compliant and its
calculated resonant frequency is lower than that of any region of the tympanum
(Fig. 5B; Stephen and Bennet-Clark 1982). When the tympanum vibrates, the
relatively heavy MUller's organ vibrates at a lower amplitude and at a different
phase to the tympanum, so there is relative movement between the ends of the
sensory cells by which MUller's organ is suspended; this thereby provides
frequency-dependent adequate stimuli.
The locust ear also provides evidence of close-to-optimal impedance matching
between the tympanum and the sensory cells. When the vibration of the folded
body region was compared before and after removal of MUller's organ, ftrst, the
resonant frequency was unchanged but, second, the amplitude of vibration of the
unloaded membrane increased by 4 dB (Stephen and Bennet-Clark 1982). The