Page 13 - Volume 12, Issue 2 - Spring 2012

Fig. 4. Diffuse-field to ear-canal absorbance is plotted versus frequency for groups of infants at age 1 to 24 months, and adults. [This figure originally published in Keefe et al. (1994)]
ratio of the sound energy absorbed by the ear to the incident energy from a transient sound. The absorbance, which is equal to 1–|R(f)|2 , is insensitive to probe position within the ear canal. As shown in Fig. 3 for ears of adults and children from full-term infants to age 6 months with normal auditory function (Keefe and Abdala, 2007), the median absorbance in each age group varies from 1, when all energy is absorbed by the middle ear, down to 0, when no energy is absorbed by the middle ear (i.e., when all energy is reflected back into the ear canal). The middle ear is most efficient at absorbing sound energy at frequencies between 2 and 4 kHz except for full- term infants showing excessive absorbance at low frequen- cies. Because the ear-canal wall is not yet fully ossified in full- term infants, the ear-canal diameter can increase as much as 70% at air pressures in the tympanometric range (Holte et al., 1991). Increased absorbance at lower frequencies in young infants is likely due to increased viscoelastic wall mobility (Keefe et al., 1993, Qi et al., 2006).
Clinical acoustical testing using insert earphones depends on sound transmission through the ear canal and middle ear as terminated by the input impedance of the cochlea. Under normal listening conditions, sound power is collected by the larger structures of the external pinna, trans- mitted through the ear canal, absorbed by the middle ear, and transmitted into the cochlea. A diffuse-field absorption cross-section AD with units of area (Rosowski et al., 1988; Shaw, 1988) quantifies the ability of the ear to absorb sound power from a reverberant sound field in a room. This meas- ure is obtained by averaging over all directional attributes of the sound source and listener. (A listener’s ability to localize a sound source depends critically on these monaural direc- tional attributes, and this ability is greatly enhanced by the fact that the listener hears binaurally—i.e., based on input from each of two ears).
The AD measured in Keefe et al. (1994) varied with fre- quency with a maximum value of about 800 mm2 for an adult ear, but only about 63 mm2 for the ear of a six-month-old. This age variation is controlled by post-natal growth of exter-
nal ear structures (pinna, concha, ear canal) and maturation
of middle-ear mechanics. The power efficiency of the overall
adult human ear is exemplified by the maximum AD of 800
mm2 from diffuse sound field to ear canal, 50 mm2 for the
ear-canal area, and 3.2 mm2 at the oval window of the
cochlea, an overall areal level difference of 24 dB. An estimate
of the “area” of external pinna comes from Teranishi and
Shaw (1968), who constructed an external-ear model based
on acoustical measurements in real ears and models. Their
model was composed of a smaller cylinder (i.e., a “concha”)
embedded in a rectangular-shaped “pinna” of area 2160 mm2
(i.e., 30 mm wide and 72 mm long). This model matched res-
onance frequencies for various angles of incident sound up to
7 kHz. This area is larger than the maximum A of 800 mm2, D
as would be expected because much of the sound power inci- dent on the ear is reflected.
The dimensionless ratio of the area AD to the cross-sec- tional area of the ear canal at each age is plotted as a diffuse- field to ear-canal absorbance in Fig. 4. The diffuse-field to ear-canal absorbance is always positive and exceeds 1 for each age group, with a maximum of 9 for adults at about 2.7 kHz and 1.6 for six-month-olds at about 3.2 kHz. This is in direct contrast to the ear-canal absorbance plots in Fig. 3 that never exceed 1. This difference between the pair of absorbances illustrates the additional sound power collected by the pinna and other external-ear structures that are more distal to the location in the clinical test of an eartip within the ear canal. The acoustics is more complicated because the spa- tial sound field is inherently three-dimensional near the pinna as influenced by head and body diffraction, but is approximately one-dimensional in the interior of the ear canal a few mm from its entrance at its junction to the con- cha and from its termination at the eardrum. In practice, clinical measurements are considerably simplified by placing the sound source within the ear canal (although headphone presentation of sound is also used in audiological testing).
A wideband tympanometry test was developed based on reflectance measurements (Keefe and Levi, 1996). Based on audiologists’ experience using 226-Hz admittance tympa-
Fig. 3. Measured absorbance at ambient pressure in the ear canal is plotted versus frequency for groups of full-term infant ears, infant ears in age groups from 1.5 to 6 months, and adult ears.
12 Acoustics Today, April 2012