Second Biennial Hearing Aid Research and Development Conference

September 22-24, 1997
National Institutes of Health
Bethesda, Maryland

Estimation of Loudness Growth Using Thresholds and LDLs

Clinical measures of loudness growth are currently available. An alternative to making these measures is to estimate loudness growth based on conventional audiologic data. The purpose of this research was to compare how well loudness growth could be estimated using thresholds plus frequency-specific Loudness Discomfort Levels (LDLs).

Audiologic and loudness growth measures were completed on 25 hearing-impaired subjects (34 ears). The range of hearing loss varied in degree from mild to severe and included sensorineural and mixed types. Loudness growth was measured using a computerized, adaptive procedure developed at Beltone Electronics that defines the range of comfort by determining the Lower Level of Comfort (LLC) and Upper Level of Comfort (ULC). This procedure first obtains thresholds and frequency-specific LDLs to define the range for LLC and ULC testing. Both narrowband noise and warble tones at 250, 500, 1000, 2000, 3000, and 4000 Hz were used as stimuli.

Estimations of LLCs and ULCs were made using formulas that included data for the average of normal-hearing listeners. These data were collected using the Beltone procedure for 30 normal-hearing subjects. Estimations were made based on threshold plus LDL data. These estimations were then compared to the measured LLCs and ULCs.

The differences between measured and estimated values were larger for LLCs than for ULCs. The average difference for the 34 ears was within +/-5.2 dB for LLCs and within +/-3.4 dB for ULCs. However, the differences for some individual ears were quite large. Of the 34 ears, 84% had LLCs that could be estimated within 10 dB while only 57% could be estimated within 5 dB. The ULC values could be estimated within 5 dB for a greater number earsó79%. Test frequency was found to have an effect on the differences for LLCs in that the differences decreased for the higher frequencies.

These data support the assumption that on the average, loudness growth can be estimated within an "acceptable range" of 5 dB using a combination of thresholds and LDLs. However, estimated measures will not be accurate for a large number of individuals. In addition, the range of what is considered an "acceptable" difference decreases as the dynamic range of the individual decreases. Therefore, analysis of the data that have been normalized for dynamic range will show larger differences between the estimated and measured values. In addition, the capabilities of the hearing aid circuitry will dictate what precision is needed for loudness growth measures. As hearing aid technology advances, more precise measures will be required. Therefore, loudness growth measures should be obtained directly until a method is developed for predicting which individuals have "average" loudness growth and which do not.

Return to Conference Abstracts List

Top