The reliability of a modified simplex procedure in hearing aid frequency-response selection.

The reliability of a modified simplex procedure to select the preferred frequency response on a programmable hearing aid was studied. The effect of stimulus materials on the selected frequency response, along with the consistency in which the selected frequency response was chosen in repeated test runs, was examined. Two groups of hearing-impaired elderly adults (from 59 to 88 years of age, with a mean age of 70 years) who had worn hearing aids for at least 1 year, participated in the study. Subjects were seen for six sessions lasting 2 to 3 hours each. During each session, subjects listened through a programmable hearing aid and selected a preferred frequency response for optimal hearing using a modified simplex procedure. Discourse passages and consonant nonsense syllables, presented in quiet and in noise, were used as the test conditions. Subjective judgment of discourse clarity and syllable recognition were used as criteria. Three estimates of preferred frequency response were obtained within a session when subjective judgment was used as a criterion and two estimates were obtained within a session when syllable recognition was used as a criterion. Subjects returned for a retest approximately 1 week after completion of all test conditions. The results of the study revealed that, depending on test conditions, approximately 30% of subjects selected the same frequency response and 80% of subjects showed test-retest deviation of less than one step size (one cell) in the selected frequency response. The consistency was highest for discourse material presented in noise and lowest for discourse material presented in quiet. Subjects preferred more high-frequency response when consonant syllables were used for recognition than when discourse was used for subjective judgment. On the other hand, more subjects preferred more low-frequency response for discourse judgment than for consonant-syllable recognition. These results suggest that the choice of stimulus materials affect the outcome and consistency in which these outcomes are generated with the simplex procedure.