Page 24 - 2016Spring

Evolution of Mammalian Sound Localization
size as possible explanatory factors, was based on data from the 18 species of mammals whose hearing was known at that time (Masterton et al., 1969). The results showed that high- frequency hearing was reliably correlated not with body size, as heretofore expected, but with functional head size.
Over the years, the number of species whose high-frequen- cy hearing is known has expanded by nearly four times to include mammals of extreme sizes and habitats, both wild and domestic, and at different trophic levels (Heffner et al., 2014). For the 69 species for which data are currently avail- able, there remains a strong negative correlation between functional head size and the highest frequency audible at a level of 60 dB (with a correlation coefficient of r = − 0.79). In other words, the smaller an animal’s head size, the higher its upper frequency limit of hearing must be in order for its head to generate a useable binaural intensity difference (Fig- ure 2).
Figure 2. The smaller an animal’s functional head size, the higher it must hear in order to use both the binaural intensity-difference cue and monaural pinna cues. The filled circles are bats. Note that bats that use laryngeal echolocation (labeled as “echolocators”), as well as the echolocating dolphin, hear slightly higher than predicted by their functional head size; the three non-echolocating bats fall among the other mammals. Subterranean rodents, which have relinquished the ability to localize brief sounds, do not hear high frequencies. Modi- fied from Heffner et al. (2014).
Although the initial emphasis on explaining the variation in high-frequency hearing was that it was necessary for small animals to effectively use binaural intensity differences, it soon became apparent that it was also necessary for the ex- ternal ear to generate monaural pinna locus cues, a point to which we will return shortly.
High-Frequency Hearing is a Species Characteristic
We want to emphasize that the relationship between high- frequency hearing and functional head size applies to species differences and not to variation in head size within a species. This was demonstrated in dogs, for which it was shown that the high-frequency hearing of individual dogs did not vary with their functional head size even though the head size of the largest dog tested, a St. Bernard, was twice as large as that of the smallest dog, a Chihuahua. Moreover, the size of the tympanic membranes of the dogs varied from 30 to 55.2 mm2, yet their hearing was similar (Heffner, 1983). There are simi- lar findings for humans (von Békésy and Rosenblith, 1951). Thus, although the size of the auditory apparatus is an im- portant factor in determining the response properties of the ear, it is apparent that the audiogram does not passively fol- low the size of the middle ear.
Echolocating Bats
Although bats require good high-frequency hearing for pas- sive sound localization, echolocating bats that emit tonal signals (laryngeal echolocators) all hear somewhat higher frequencies than similar-sized animals that do not echolo- cate. This suggests that selective pressure for echolocation has caused them to increase their high-frequency hearing beyond that required for passive sound localization by about half an octave, a conclusion supported by the observation that the three bats in Figure 2 that do not emit tonal calls (two non-echolocating bats and Rousettus aegyptiacus, which emits clicks) do not hear higher than predicted by the need to passively localize sound. Thus, the use of echoloca- tion by bats, as well as by toothed whales, is an example of how, once high-frequency hearing evolved for passive sound localization, it was exploited for another function.
Humans
Humans have often been thought to differ from other mam- mals in that we are unable to hear “ultrasound,” which is true by definition, as ultrasound is defined as any high-frequency sound beyond the human hearing range. However, given our large head size, the 60-dB upper limit of 17.6 kHz of young and healthy humans is not significantly different from that predicted by the regression line in Figure 2. Indeed, the observation that the Indian elephant (Elephas maximus) has an even lower upper limit of 10.5 kHz is further indication that our upper limit of hearing is not unique but is expected for a mammal with our large head size.
22 | Acoustics Today | Spring 2016