The effect of stimulus duration on preferences for gain adjustments when listening to speech

Abstract

Objectives: In the personalisation of hearing-aid fittings, gain is often adjusted to suit patient preferences using live speech. When using brief sentences as stimuli, the minimum gain adjustments necessary to elicit consistent preferences ("preference thresholds") were previously found to be much greater than typical adjustments in current practice. The current study examined the role of duration on preference thresholds.

Design: Participants heard 2, 4 and 6-s segments of a continuous monologue presented successively in pairs. The first segment of each pair was presented at each individual's real-ear or prescribed gain. The second segment was presented with a ±0-12 dB gain adjustment in one of three frequency bands. Participants judged whether the second was "better", "worse" or "no different" from the first.

Study sample: Twenty-nine adults, all with hearing-aid experience.

Results: The minimum gain adjustments needed to elicit "better" or "worse" judgments decreased with increasing duration for most adjustments. Inter-participant agreement and intra-participant reliability increased with increasing duration up to 4 s, then remained stable.

Conclusions: Providing longer stimuli improves the likelihood of patients providing reliable judgments of hearing-aid gain adjustments, but the effect is limited, and alternative fitting methods may be more viable for effective hearing-aid personalisation.

Keywords: Hearing-aid fitting; duration; fine-tuning; gain.

Figure 1

The left panel shows median pure-tone thresholds as a function of frequency (circles, solid line) and interquartile ranges (error bars), with the individual thresholds for the three lowest and highest average thresholds (dotted lines). The right panel shows median sensation levels (approximated from pure-tone thresholds and applied gain) as a function of frequency (circles, solid line) and interquartile ranges (error bars), with the individual values for the three lowest and highest average sensation levels (dotted lines).

Figure 2

Mean proportion of preferences as a function of gain adjustment for low-frequency (LF; ≤O,5kHz), mid-frequency (MF; 1−2 kHz) and high-frequency (HF; ≥4kHz) bands (left, middle and right panels, respectively) for 2-s, 4-s and 6-s durations (short-dashed, long-dashed and solid lines, respectively; red, green and blue online). Better, worse and no difference preferences are shown as upward triangles, downward triangles and circles, respectively. Grey dotted lines and symbols show results using short sentences from Caswell-Midwinter and Whitmer (2021).

Figure 3

Boxplots of preference thresholds for each stimulus duration: sentences (average duration 1.6 s; Caswell-Midwinter and Whitmer 2021), 2 s, 4 s, and 6 s. Preference thresholds for negative and positive gain adjustments are shown in red and blue, respectively. Circles show means; lines show medians; boxes show interquartile ranges (IQR); whiskers show 1.5·IQR; crosses and pluses show outliers for negative and positive adjustments, respectively.

Figure 4

Proportion of reliable preferences as a function of stimulus duration. Horizontal lines show medians; boxes show interquartile ranges (IQR); whiskers show 1.5·IQR; circles show outliers. Sentence data are from Caswell-Midwinter and Whitmer (2021).