Speech-in-noise screening tests by internet, part 3: test sensitivity for uncontrolled parameters in domestic usage

The online speech-in-noise test 'Earcheck' is sensitive for noise-induced hearing loss (NIHL). This study investigates effects of uncontrollable parameters in domestic self-screening, such as presentation level and transducer type, on speech reception thresholds (SRTs) obtained with Earcheck. Subjects performed 26 Earchecks that differed regarding presentation level (65, 71, and 77 dBA), presentation mode (monotic or diotic), and masking noise (two different low-pass filtered noises) in the lab. To investigate effects of test environment, participants conducted eight additional Earchecks at home using different transducer types (headphones or loudspeakers). Thirty noise-exposed workers, either normal-hearing (n = 10), or with different degrees of NIHL (n = 20), participated. There was a minor effect of presentation levels exceeding 65 dBA in severely impaired listeners. Diotic presentation mode yielded lower SRTs compared to monotic presentation mode. Normal-hearing test results at home were poorer than in the laboratory, whereas hearing-impaired subjects performed better in domestic testing. Using loudspeakers deteriorated SRTs significantly in comparison to headphones, but only in hearing-impaired subjects. A monotic presentation mode using headphones is recommended for domestic screening. Since domestic testing affects SRT results, a follow up study using a large study population should assess Earcheck's validity when performed at home