Guthrie and Mackersie (2009) report testing word recognition scores (WRSs) at multiple loudness levels is the best practice, yet most audiologists do not obtain WRSs in this manner. The authors examined performance-intensity (PI) functions for 40 subjects with a mean age of 72 years (range 45 to 90 years). Each subject was assigned to one of four groups—mild, moderate, moderately-severe/severe, or steeply sloping. Each group consisted of 10 subjects. The goal was to determine loudness presentation levels that maximized WRSs, while avoiding loudness discomfort.

Guthrie and Mackersie report using a presentation of 40 dB SL (re SRT) violated UCL for 69 percent of subjects with an SRT of 35 dB HL or greater. Using a presentation of 30 dB SL (re SRT) would violate UCLs of subjects with SRTs of 55 dB HL and greater. Maximal speech recognition scores were most often determined at or near UCL, rather than MCL.

For people with mild, moderate gradual sloping, and steeply sloping hearing loss, two methods provided equivalent results.

  1. Set the WRS presentation level 5 dB below speech UCL, or
  2. Set the attenuator dial above the 2 kHz threshold based on:
    1. If 2 kHZ threshold < 50dB HL, add 25 dB SL
    2. If 2 kHz threshold = 60-65 dB HL, add 15 dB SL
    3. If 2 kHz threshold = 70-75 dB HL, add 10 dB SL

For people with moderately/severe to severe hearing loss, UCL minus 5 dB resulted in the highest phoneme recognition score.

Guthrie and Mackersie noted that fixed-level testing (such as 95 dB SPL) and using a fixed-level SL re SRT are not supported, based on this study.

For More Information, References, and Recommendations:
Guthrie LA, Mackersie CL. (2009) A Comparison of Presentation Levels to Maximize Word Recognition Scores. Journal of the American Academy of Audiology (20)6:381-390.