Is Normal or Less Than Normal Overall Loudness Preferred by First-Time Hearing Aid Users?

In Brief Objective Most prescriptive methods for nonlinear, wide dynamic range compression (WDRC) hearing aids are based on the assumption that a hearing-impaired listener should perceive amplified sounds at the same overall loudness as would a normal-hearing listener without amplification. However, some previous research on linear amplification has indicated that subjects prefer less overall gain than prescribed by the most commonly used prescriptive method for linear hearing aids, NAL-R, a method that gives close to normal overall loudness for a mid-level input. The current study aims at comparing two prescriptive methods for WDRC hearing aids. The methods differ in the overall loudness they aim to give the hearing aid user. One method, called NormLoudn, is based on a generic method that prescribes gain so that the overall loudness is restored to normal. Another method, called LessLoudn, is based on a hearing aid specific prescription, and gives the hearing aid user less than normal overall loudness. Do first-time hearing aid users prefer the method that restores overall loudness to normal or the method that gives less than normal overall loudness? Design Twenty-one first-time hearing aid users with typical hearing losses for this group of clients participated in a crossover blinded field study where the two fitting methods were compared using a multi-programmable hearing aid, Danalogic 163D. Preference in the field was evaluated using interview, questionnaire, and diary. The field test was accompanied by laboratory tests, which included paired comparison judgments of preference and loudness and a speech recognition test. Loudness calculations were also used when interpreting the results, and a theoretical comparison with other prescriptive methods for WDRC hearing aids was made. Results After necessary adjustments, the measured gain for the two methods was similar in gain-frequency shape, but NormLoudn gave more overall gain than LessLoudn. Generally, NormLoudn fittings led to calculated overall loudness that was close to normal, whereas LessLoudn fittings, in median, led to 3–7 phon less than normal calculated overall loudness according to the loudness model used. During the interview performed after the field test, 19 out of the 21 subjects stated that they preferred LessLoudn. Also the questionnaire and the diary showed a clear preference for LessLoudn in all types of listening situations. Paired comparisons of preference in the laboratory supported the findings in the field. LessLoudn was preferred to NormLoudn in all tested situations, except for soft speech in very soft noise where there was no significant preference for either method. Speech recognition scores were similar for the two fittings. The difference in calculated loudness was clearly distinguishable to the subjects and seemed to govern their preferences. Conclusions LessLoudn, which gave less than normal overall loudness according to the loudness model used, was preferred both in the field and in the laboratory tests. Speech recognition scores were similar for both methods. A comparison between the measured gain for NormLoudn and the gain prescribed by CAMEQ, NAL-NL1, and DSL[i/o], suggests that all three prescriptive procedures (DSL[i/o] in particular) would probably overestimate the required gain for subjects without hearing aid experience and with mild to moderate hearing loss. When providing these clients with hearing aids, it seems important to include a careful adjustment of the overall gain to avoid overamplification if the prescription used aims at restoring overall loudness to normal according to the loudness model used here. Most prescriptive methods for nonlinear hearing aids aim at restoring overall loudness to normal. The study questions this basic assumption by comparing two prescriptive methods, one that provides normal overall loudness and one that gives 3–7 phon less than normal overall loudness. Twenty-one first-time hearing aid users with typical hearing losses participated in a crossover blinded field study and in additional laboratory tests. The method that gave less than normal overall loudness was strongly preferred both in the field and in the laboratory. No significant difference in speech recognition scores was seen. The study concludes that prescriptions based on normal overall loudness might not be appropriate when fitting clients without hearing aid experience and with mild to moderate hearing loss.