| Popis: |
The hearing aid algorithm “BioAid” mimics two basic principles of normal hearing: the instantaneous compression of the basilar membrane and the efferent feedback of the medial olivocochlear (MOC) reflex. The design of this algorithm aims at restoring those parts of the auditory system, which are hypothesized to dysfunction in the individual listener. In the initial stage of this study individual computer models of three hearing-impaired listeners were constructed. These computer models reproduce the listeners’ performance in psychoacoustic measures of (1) absolute thresholds, (2) compression, and (3) frequency selectivity. Subsequently, these computer models were used as “artificial listeners.” Using BioAid as a front-end to the models, parameters of the algorithm were individually adjusted with the aim to ‘normalize’ the model performance on these psychoacoustic measures. In the final stage of the study, the optimized hearing aid fittings were evaluated with the three hearing-impaired listeners. The aided listeners showed the same qualitative characteristics of the psychoacoustic measures as the aided computer models: near-normal absolute thresholds, steeper compression estimates and sharper frequency selectivity curves. A systematic investigation of the effect of compression and the MOC feedback in the algorithm revealed that both are necessary to restore performance. [Work supported by DFG.] |
