Abstract
Background The masking-level difference (MLD) can be measured via voluntary behavioral responses (voluntary behavioral MLD [vMLD]) and/or via electrophysiological cortical auditory evoked potentials (CAEPs; electrophysiological MLD [eMLD]). It may be possible to enhance the ecologic validity of the MLD by using nonsense-syllable speech stimuli.
Purpose The aim of this study is to determine the feasibility of measuring both the vMLD and eMLD with speech stimuli. The study also investigates whether certain nonsense-syllable stimuli (/α/, /dα/, /di/, /tα/, /wα/) may be more useful than others in measuring both the vMLD and eMLD.
Research Design This is a descriptive feasibility pilot study.
Study Sample Seventeen young adults (age range 19–26 years; 15 women) with hearing thresholds of 0.25–8.0 kHz ≤ 25 dB HL, bilaterally, were recruited.
Data Collection and Analysis Behavioral and electrophysiological MLDs were measured with similar methods. The MLD was defined as SoNo – SπNo thresholds. Stimuli were natural-sounding nonsense syllables (/α/, /dα/, /di/, /tα/, /wα/), which were presented in 65 dB HL continuous speech-weighted noise. The eMLD was measured with the CAEP P2. Group means, standard deviations, and distributions were presented. The feasibility of using nonsense syllables was evaluated by considering whether measurable vMLDs and eMLDs were produced. Useful nonsense syllables produced vMLDs and eMLDs with (1) comparatively large mean magnitudes, (2) few negligible MLDs, and (3) distributions with adequate spread and few extreme values.
Results The stimuli /α/ (6.0 [1.9]) and /wα/ (7.5 [1.3]) produced vMLDs with the highest average magnitudes, with no vMLDs of 0 dB and with adequate spread. The stimulus /α/ produced eMLDs with the highest average magnitude (9.6 [2.8]), no eMLDs of 0 dB and adequate spread, whereas the stimulus /wα/ produced eMLDs with an adequate magnitude (6.9 [3.9]), no MLDs of 0 dB, but with a right-skewed distribution and an extreme value. The other stimuli produced vMLDs with low mean magnitudes and several vMLDs of 0 dB.
Conclusion These pilot data support the feasibility of using nonsense syllables to record vMLDs and eMLDs. The stimulus /α/ appeared most useful for both behavioral and electrophysiological modalities. Differences in MLDs across modalities may be attributed to low-level audibility of some high-frequency components of the stimuli.
Keywords
masking-level difference - electrophysiology - cortical auditory evoked potentials - audiology
