Background: The decision to fit one or two hearing aids in individuals with binaural hearing
loss has been debated for years. Although some 78% of U.S. hearing aid fittings are
binaural (Kochkin , 2010), Walden and Walden (2005) presented data showing that 82%
(23 of 28 patients) of their sample obtained significantly better speech recognition
in noise scores when wearing one hearing aid as opposed to two.
Purpose: To conduct two new experiments to fuel the monaural/binaural debate. The first experiment
was a replication of Walden and Walden (2005), whereas the second experiment examined
the use of binaural cues to improve speech recognition in noise.
Research Design: A repeated measures experimental design.
Study Sample: Twenty veterans (aged 59–85 yr), with mild to moderately severe binaurally symmetrical
hearing loss who wore binaural hearing aids were recruited from the Audiology Department
at the Bay Pines VA Healthcare System.
Data Collection and Analysis: Experiment 1 followed the procedures of the Walden and Walden study, where signal-to-noise
ratio (SNR) loss was measured using the Quick Speech-in-Noise (QuickSIN) test on participants
who were aided with their current hearing aids. Signal and noise were presented in
the sound booth at 0° azimuth under five test conditions: (1) right ear aided, (2)
left ear aided, (3) both ears aided, (4) right ear aided, left ear plugged, and (5)
unaided. The opposite ear in (1) and (2) was left open. In Experiment 2, binaural
Knowles Electronics Manikin for Acoustic Research (KEMAR) manikin recordings made
in Lou Malnati's pizza restaurant during a busy period provided a typical real-world
noise, while prerecorded target sentences were presented through a small loudspeaker
located in front of the KEMAR manikin. Subjects listened to the resulting binaural
recordings through insert earphones under the following four conditions: (1) binaural,
(2) diotic, (3) monaural left, and (4) monaural right.
Results: Results of repeated measures ANOVAs demonstrated that the best speech recognition
in noise performance was obtained by most participants with both ears aided in Experiment
1 and in the binaural condition in Experiment 2.
Conclusions: In both experiments, only 20% of our subjects did better in noise with a single ear,
roughly similar to the earlier Jerger et al (1993) finding that 8–10% of elderly hearing
aid users preferred one hearing aid.
Key Words
Hearing aids - hearing loss - speech recognition
