Spatial benefit of bilateral hearing AIDS

Abstract

Objectives: To assess the extent to which hearing aids improve spatial benefit by restoring the availability of interaural difference cues, the benefit attributable to spatial separation of speech and babble with and without bilateral hearing aids was measured as a function of low-pass cutoff frequency.

Design: Twenty-one older adults with sloping high-frequency hearing loss were provided commercially available bilateral hearing aids. After a 3 to 6 month acclimatization period, speech levels corresponding to 50% correct recognition of sentences from the Hearing in Noise Test (HINT) were measured in a 65-dB SPL babble, with speech and babble low-pass filtered at 1.8, 3.6, and 5.6 kHz. Sentences were always at 0 degrees azimuth, and babble was at either 0 degrees or 90 degrees . Speech and babble spectra for all conditions were digitally recorded using a probe microphone placed in each ear canal of each subject. Spectra and levels of speech and babble and unaided thresholds for narrowband noises were used to calculate the aided audibility index and provide predictions of unaided and aided thresholds for HINT sentences, hearing aid benefit, and spatial benefit for each cutoff frequency. In addition, subjects' willingness to tolerate background noise with and without amplification was measured in the spatially coincident and spatially separated conditions using the Acceptable Noise Level (ANL) procedure.

Results: Thresholds for HINT sentences in babble and ANL improved significantly when aided and when speech and babble were spatially separated. Specifically, hearing aid benefit improved significantly as cutoff frequency increased from 1.8 to 3.6 kHz but only when speech and babble were spatially separated; likewise, spatial benefit improved significantly from 1.8 to 3.6 kHz but only in the aided condition. No further improvement in hearing aid or spatial benefit was observed when cutoff frequency was increased from 3.6 to 5.6 kHz, although improvement in hearing aid benefit was predicted.

Conclusions: Hearing aid benefit, although significant, was poorer than predicted, suggesting that these older adults with high-frequency hearing loss did not take full advantage of the increase in audible speech information provided by amplification. Hearing aid benefit was also limited because hearing aids for some subjects did not restore speech audibility across the full bandwidth of speech. Unaided and aided spatial benefit was significantly greater than predicted, and aided spatial benefit was greater than unaided. This suggests that these older adults were able to take advantage of interaural level and time difference cues to improve speech recognition in babble and that they benefited from these cues to a greater extent with than without bilateral hearing aids. Finally, in contrast to results of previous studies, ANL may vary for an individual depending on the listening condition.

Fig. 1

Mean audiometric thresholds (±1 SD) for frequencies from 0.25 to 8.0 kHz for 21 subjects who received bilateral hearing aids. Filled circles are for the right ear and open circles are for the left ear.

Fig. 2

Mean (solid line) and range (shaded area) of 2 cc coupler gain (in dB) across frequency for both hearing aids of all subjects. Input was the ICRA composite noise at 65 dB SPL. Arrows denote low-pass cutoff frequencies.

Fig. 3

Observed (filled symbols) and predicted (open symbols) thresholds for HINT sentences (±1 SE) without hearing aids (circles) and with hearing aids (triangles) as a function of low-pass cutoff frequency, with babble at 0° (top panel) and 90° (middle panel). HINT sentences were always at 0°. The bottom panel shows observed and predicted hearing aid benefit (±1 SE) as a function of low-pass cutoff frequency for babble at 0° and 90°. Hearing aid benefit is defined as unaided HINT thresholds minus aided HINT thresholds.

Fig. 4

Top: Aided thresholds for HINT sentences plotted against average high-frequency thresholds (2.0, 3.0, and 4.0 kHz) measured in quiet. Filled and open symbols correspond to thresholds for HINT sentences measured for speech and babble low-pass filtered at 3.6 and 5.6 kHz, respectively. Middle: Same as top panel, but for predicted aided thresholds for HINT sentences. Bottom: Same as top panel, but for differences between observed and predicted aided thresholds for HINT sentences. Pearson correlation coefficients and linear regression functions are included in each panel.

Fig. 5

Data recast from Fig. 3. Observed (filled symbols) and predicted (open symbols) thresholds for HINT sentences (±1 SE) with babble at 0° and babble at 90° as a function of low-pass cutoff frequency, without hearing aids (top panel) and with hearing aids (middle panel). The bottom panel shows observed and predicted spatial benefit (±1 SE) as a function of low-pass cutoff frequency for unaided and aided conditions. Spatial benefit is defined as HINT thresholds with the babble at 0° minus HINT thresholds with the babble at 90°.

Fig. 6

Unaided (solid bars) and aided (striped bars) Acceptable Noise Levels (ANL [±1 SE]) are plotted for two noise-source locations at the low-pass cutoff frequency of 5.6 kHz.

Fig. 7

ANLs plotted against signal-to-noise ratios (SNR) at threshold for HINT sentences (5.6 kHz cutoff). Results are for unaided listening with the babble at 0°. The Pearson correlation coefficient and linear regression function are also included.

Fig. 8

Spatial benefit for ANL plotted against spatial benefit for HINT sentences (5.6 kHz cutoff). Results are for aided listening. The Pearson correlation coefficient and linear regression function are also included.

Fig. 9

Aided benefit for ANL (0°) plotted against hearing aid use as determined by the Glasgow Hearing Aid Benefit Profile (GHABP). The Pearson correlation coefficient and linear regression function are also included.