Amplification and Aural Rehabilitation Reliability and lntersubject Variability of the Real Ear Unaided Response

lntratester test-retest reliability of the real ear unaided response (REUR) was determined on 49 ears using the Frye 6500 real ear analyzer. Results revealed mean differences of less than 1 dB for repeat measurements at seven test frequencies between 250 and 4000 Hz. The average peak resonant frequency of the repeated measure was within 16 Hz of the initial measure. In addition, the intersubject variability of the amplitude of REUR was quite large. A range of 7 dB was found at 250 to 500 Hz with the range expanding to 15 to 20 dB at 2000 to 4000 Hz. Also, the peak resonant frequency varied between 2100-4800 Hz. These results are discussed in terms of those dispensers who use the REUR to “custom” order hearing aids (Ear Hear 12 3: 216-220). EACH DAY, HEARING aids are selected having electroacoustic characteristics felt to be appropriate for a given hearing loss. Recently, Martin and Moms (1 989) reported that selection of these characteristics is usually based upon a target of the desired real ear insertion response (REIR) for discrete frequencies recommended by Berger, Hagberg and Rane (1977), Byrne and Dillon (1986), Libby (1985; 1986), or McCandless and Lyregaard (1983). However, the selection of the characteristics may also be based upon a target of the desired real ear aided response (REAR) for discrete frequencies recommended by Cox (1988), Pascoe (1979, Seewald, Ross and Spiro ( 1985) or Skinner ( 1980). Recently, several investigators (Mueller, 1989; Upfold & Bryne, 1988; Valente, Valente & Vass, 1990a; 1990b) have suggested that the real ear unaided response (REUR) of the individual at discrete frequencies should be included in the hearing aid selection process in order to accurately determine the required electroacoustic characteristics necessary to achieve desired REIR. This suggestion is based upon the belief that the natural resonance of the ear canal is eliminated when an earmold or hearing aid is inserted in the ear canal and therefore, significant deviations of the individual REUR from the average REUR may result in difficulty in achieving desired REIR. For example, if the amplitude of the REUR is greater than average, a reduction in the measured REIR (i.e., insertion loss) may occur within that frequency region, and greater coupler gain may be required to obtain desired REIR. On the other hand, if the amplitude of the REUR is less than average, measured REIR may be greater than desired and less coupler gain may be required. Either effect could result in undesirable peaks and troughs in the measured REIR if corrections are not implemented. In determining individual deviations of the REUR from average, some investigators have used the freefield to eardrum transformation data reported by Shaw (1 974) and Shaw and Vaillancourt ( 1985). In addition, software included in some probe tube units contains the Shaw data as a reference for corrections for the individual REUR. However, the data reported by Shaw is an average of a compilation of 12 investigations in which the REUR was obtained in a manner which, in many respects, is significantly different from the way the REUR is measured with many probe tube units. For example, Shaw used the center of the head in an unobstructed free-field as the reference after the REUR was measured with a probe microphone in the ear canal. The use of the center of the head as the reference results in the inclusion of head diffraction and body bame effects in the measured REUR. On the other hand, many probe units use an “at the ear” or “under the ear” location for the reference microphone position. The use of this reference point for “equalizing” or “leveling” the test condition excludes head diffraction and body bame effects from the REUR measure. Exclusion of these effects can result in the measured REUR being different from the Shaw (1974) data from -0.5 to 4 dB (Kuhn, 1979). Bentler (1989), using a Rastronics CCI 10/3 (under the ear reference microphone) at 0” azimuth, reported the REUR in children above the age of 2 yr was reasonably close to the findings of Shaw. However, the mean REUR revealed slightly less gain in the lower and upper frequency regions, Copyright Q 1991 by The Williams & Wilkins Co. Printed in U.S.A. 0196/0202/91 /I 203-021 6$03.00/0 * EAR AND HEARING 216 Valente Ear and Hearing, Vol. 12, No. 3,1991 which she attributes, in part, to the calibration method used. In addition to measuring the REUR for the purpose of “custom” ordering’ hearing aids at discrete frequencies, several investigators (Kruger, 1987; Upfold & Bryne, 1988) have suggested that greater user benefit or acceptance may occur if the peak gain of the REAR matched the peak frequency of the REUR to obtain a “transparent” fit. Currently, several hearing aid manufacturers provide a coupler response which mimics the average REUR. This response should be effective in compensating for the loss of the ear canal resonance if the individual REUR has the same peak amplitude and frequency as the average REUR. However, large intersubject variability of these REUR parameters has been reported in the literature (Bentler, 1989; Upfold & Bryne, 1988). This suggests that the effectiveness ofthis circuit design may be of limited use if a listener has an expected REUR which differs significantly from average. It is expected that technological advances in the near future will result in the ability to select peak coupler gain to match the individual REUR measured in the hearing aid selection process. In addition, these same technological advances may soon allow the dispenser to actively shift the peak frequency during REAR probe measures to match the peak frequency of the individual REUR. Due to an increased interest in the use of the REUR in the hearing aid selection process, it would be beneficial to determine the test-retest reliability of the REUR for one commercially available unit (Frye 6500). More dispensers seem to be using the REUR as a correction factor to determine the coupler response necessary to achieve desired REIR. Also, dispensers may consider using the REUR as a reference of where to adjust or select the peak frequency of a hearing aid. If the REUR measurement is unreliable (i.e., significant differences are present between measures) then using the REUR to customize hearing aids or for matching the peak frequency may not be appropriate. On the other hand, it would be comforting to know that the REUR is a reliable measure for those who choose to use this measure in the hearing aid selection process.