Agreement between psychophysical tuning curves and the threshold equalizing noise test in dead region identification

Abstract

Objective: Identification of dead regions is currently based on the results of psychophysical tuning curves (PTC) and the threshold equalizing noise (TEN) test. Summers et al (2003) found poor agreement (56%) between PTC- and TEN test results. Kluk and Moore (2005) argued that these results are explained by beat- and/or combination tone detection. The purpose of this study was to examine the diagnostic dead region relationship between PTCs, modified for beats and/or combination tone detection, and the TEN test with levels calibrated in hearing levels (HL), i.e. TEN[HL]. Design: Twenty-four hearing-impaired ears were evaluated using PTCs and the TEN[HL] test. Results: The results show an agreement of about 75%, depending on the criteria applied, between PTC and the TEN[HL] test. Results also show that PTC probe levels affected diagnostic results in PTCs. Conclusions: Recommended criteria based on the highest agreement score included a PTC shift of 20% and an 8-dB probe elevation above TEN masking levels for the TEN[HL] test. Low agreement scores and level dependent effects in PTCs and the TEN[HL] test suggest that identification of dead regions using a single test is not reliable.

Key Words::

• Psychoacoustics/hearing science
• anatomy and physiology
• medical audiology
• noise

Acknowledgements

We would like to thank László Körössy for his help with the hardware and software development needed for this study. We thank Rolph Houben for his comments on some technical issues, and Brian C.J. Moore for supplying the TEN[HL] noise on CD. We also thank Jacek Smurzynski and two anonymous reviewers for their helpful comments on earlier versions of this paper.

Notes

1. In case of HI15, the second highest level TEN (58 dB HL) was used to determine the presence of a dead region. The highest level TEN (68 dB HL) produced lower masking thresholds for the probes with frequencies ≥3000 Hz. It may well be possible that insufficiently masked off-frequency detection below the 1000 Hz cut-off was used to detect high-frequency probes.

2. Not to be confused with PTC tip shift level dependency used for estimations of the best frequency in forward masking experiments, such as Lopez-Poveda et al (2007), Moore et al (2002); or PTC tip shifts in phase dependent tone-on-tone masking experiments (for example: Vogten, 1978).

Declaration of interest: authors reports no conflicts of interest. The authors alone are responsible for the content and writing of the manuscript.