Comparison of Nominal and Measured Compression Ratios and Compression Thresholds, in Hearing Aids, Using Electroacoustic Measures

Abstract

Hearing aids incorporate compression in them and this will be useful for individuals having reduced dynamic range. In some hearing aids, the compression ratio (CR) and/ or compression threshold (CT) are programmable through the hearing aid programming software. Various studies highlight the non-compliance of compression parameters when measured by electroacoustic measurements. The CR displayed on the programming screen, i.e., the nominal CR does not always represent the CR that is measured i.e., measured CR. Similarly, the CT that is displayed on the screen during programming, i.e., the nominal CT may not be equal to the CT measured through input-output function of electroacoustic measurement. The present study has two objectives. First, to compare the nominal CR with the measured CR. Second, to compare the nominal CT with the measured CT. Repeated measures research design was used to verify the objectives. Method: Hearing aids were programmed with the National Acoustical Laboratory Non-Linear 2 (NAL-NL2) fitting formula for flat hearing thresholds of 50 dB HL for low-powered and 70 dB HL for high-powered hearing aids, for naïve hearing aid users. The CRs and/or CTs were kept at low-, mid- and high- levels. The other hearing aid features were disabled. The input from the calibrated hearing aid test system was a 2 kHz tone that varied from 50 to 90 dB SPL, in 5 dB steps. The output of each programmed hearing aid (n= 13 in Group 1, low gain hearing aids, n=12 in Group 2, high power hearing aids) was measured in the form of input/output graph. From this, the measured CR and the measured CT were computed. The way in which the nominal CR and nominal CT was set in the two groups of hearing aids slightly differed depending on the programmability feature of CR and CT. Results: Descriptive statistics revealed that the nominal and measured compression parameters were not the same. Hence, Wilcoxon Signed Ranks test was administered since the data were not normally distributed. Measured CR was significantly different from the nominal CR in both the Groups 1 and 2. In Group 1 hearing aids, there was a significant difference between the nominal and measured CR. In Group 2 hearing aids, at low nominal CR, the nominal CR was lower than the measured CR. At mid nominal CR, there was no significant difference between the nominal and measured CR. At high nominal CR, the nominal CR was significantly higher than the measured CR. In Group 1 hearing aids, CT comparisons were not carried out as there was no provision for manipulating the CT, nor was the CT displayed in the programming software. In Group 2 hearing aids, at low nominal CT, the measured CT is significantly higher than the nominal CT. At mid nominal CT, no significant difference was noted between nominal CT and measured CT, except in the condition where the nominal CR was low, where the measured CT was significantly lower than the nominal CT. The measured CT was significantly lesser than the nominal CT at high nominal CT. Conclusions: In the present study, there was a significant difference between the nominal compression parameters and the measured compression. Similar findings were also reported by Verschuure et al. (1996). Hence, this has to be kept in mind while programming CR and CT in hearing aids, i.e., what is displayed on the programming screen is not the same as that measured through electroacoustic verification.