A comparison of the effect on consonant discrimination of combining low- and high-frequency passbands in normal, congenital, and adventitious hearing-impaired subjects.

To investigate consonant information in low frequencies, a low-frequency band which contains negligible consonant information in isolation was added to a high-frequency band. There was a significant increase from the high-frequency band-alone score for normal-hearing subjects when the bands were presented at favorable low-frequency/high-frequency bands sensation levels either to the same ear or opposite ears. For congenitally and adventitiously hearing-impaired subjects, there was a significant increase only when the bands were presented to opposite ears. A clinical application of these results is split-band amplification whereby low frequencies were added to one aid and removed from the other aid. Removal of low frequencies from one aid reduces internal noise, attenuates unwanted background noise, reduces cochlear masking, and thus optimally amplifies that part of the speech spectrum which conveys most of the consonant information. The low-frequency response aid adds additional information, including pitch, rhythm, and intonation.