Effect of loudness recruitment on the perception of amplitude modulation

Abstract

People with hearing loss of cochlear origin usually display loudness recruitment; the rate of growth of loudness level with increasing sound level is greater than for a normally hearing person. Loudness recruitment has usually been studied with steady sounds of relatively long duration. The present study examines how recruitment affects the perception of dynamically varying sounds, namely amplitude modulated sinusoids. The modulation rates used (4, 8, 16, and 32 Hz) were chosen to span the range of the most prominent modulations present in the envelope of speech. Three subjects with unilateral cochlear hearing loss were used. In experiment 1, subjects were required to make loudness matches between 1-kHz tones presented alternately to the two ears. This was done over a wide range of sound levels. Experiment 2 used 1-kHz carriers that were amplitude modulated. The modulation was sinusoidal on a dB scale. The modulated tones were presented alternately to the two ears and were approximately equally loud in the two ears. The modulation depth was fixed in one ear, and the subject was required to adjust the modulation depth in the other ear so that the modulation depth appeared equal in the two ears. This was done for a range of modulation depths and with the fixed tone presented to both the normal and the impaired ear. A given modulation depth in the impaired ear was matched by a greater modulation depth in the normal ear. To a first approximation, the modulation-matching functions were independent of modulation rate. Furthermore, the functions could be predicted reasonably well from the loudness-matching results of experiment 1, obtained with steady tones. The results are consistent with the idea that loudness recruitment results from the loss of a fast-acting compressive nonlinearity that operates in the normal peripheral auditory system. Possible implications of the results for the use of fast-acting compression in hearing aids are discussed.