Page 33 - Spring2020

measures of hearing based on auditory brainstem responses (ABRs) and otoacoustic emissions (OAEs). The initial goal of the work in Gorga’s laboratory beginning in the early 1980s was to demonstrate and improve the accuracy with which the pure-tone audiogram could be predicted from the ABR, an evoked potential generated by the peripheral auditory system, used as a measure of hearing in newborns and others who were unable to be tested behaviorally. To develop normative data that could be used clinically, Gorga focused on calibra- tion issues and the collection of large sets of data from infants and young children including infants who graduated from an intensive care nursery.
At the same time, Neely began working with Susan Norton, a BTNRH postdoctoral fellow at the time, to obtain data on click-evoked and tone burst-evoked OAEs, low-level acoustic signals recorded in the ear canal generated by the cochlea in response to acoustic stimuli (Norton and Neely, 1987). OAEs had recently been discovered by Kemp (1978) and provided Neely with a convenient source of data that could be incorpo- rated into his models of cochlear responses. One of the major issues at that time was whether OAE-response latencies were too long to be accounted for by cochlear mechanics.
To answer the question, Neely and Gorga combined forces on a project that compared the latency of ABR and OAE responses with that of tone-burst stimuli (Neely et al., 1988). They noted that OAE responses could be modeled as having twice the latency of ABR responses, if one assumed that the ABR response consisted of a peripheral component that varied with frequency and intensity and a central component that was independent of both. It was not widely understood at the time, however, that the cochlear traveling wave moved more quickly at higher intensities, making it important to compare OAE and ABR latencies at the same level. It was also not widely accepted that cochlear latencies are longer in humans compared with other species typically used in hearing research.
The collaboration between Gorga and Neely began in earnest following an internal seminar in 1987 where Gorga described click-evoked ABR data from 1,120 infants and children rang- ing in age from 32 weeks gestational age to 3 years of age. After the presentation, Neely asked if he could reanalyze the data, and a few hours later, he came into Gorga’s office, put two figures and an equation on his desk, and stated that the data proved that the cochleas of humans are mature at birth. The key figure is shown here in Figure 4, where Neely was able to fit the data for four presentation levels with an exponential
equation that assumed that the effects of level and age were independent of one another. Neely argued that the latency of the central component was independent of the intensity but decreased with age, completely accounting for the effect of age in the data. The latency of the peripheral component decreased with intensity at the same rate at all ages. Gorga was stunned by the significance of this analysis because the status of the cochlea at birth was hotly debated at the time. The resulting paper by Gorga et al. (1989) was one of 74 of their joint publications in peer-reviewed journals, 48 of which appeared in JASA.
Research in Gorga’s laboratory then shifted to collection of OAE data, specifically distortion-product otoacoustic emis- sions (DPOAEs), where the stimulus consisted of two closely spaced tones and the emission was recorded at the frequency of a combination tone, typically a frequency twice the lower stimulus frequency minus the higher frequency (21- 2). DPOAEs had the advantage of providing frequency-specific data while allowing the separation of the stimulus from the emission in ear canal recordings. In a series of papers, Gorga, Neely, and colleagues determined optimum stimulus param- eters for the generation of DPOAEs and published DPOAE data that provided a framework for interpreting DPOAEs in the clinic (Gorga et al., 1993, 1997). As part of this work,
Figure 4. Auditory brainstem response (ABR) wave V latency as a function of age can be described by the same exponential function at all stimulus presentation levels. Wave V latency = 4.89 + 4.46e−0.0318a + 5.31e−0.0251i, where a is conceptional age in weeks and i is the level in decibels (HLn), where 0 dB HLn is equivalent to a peak sound pressure of 30 dB re 20 μPa. After Gorga et al., 1989, Figure 5.
Spring 2020 | Acoustics Today | 33