Mathematical modeling of vowel perception by users of analog multichannel cochlear implants: Temporal and channel-amplitude cues

A "multidimensional phoneme identification" (MPI) model is proposed to acco unt for vowel perception by cochlear implant users. A multidimensional exte nsion of the Durlach-Braida model of intensity perception, this model incor porates an internal noise model and a decision model to account separately for errors due to poor sensitivity and response bias. The MPI model provide s a complete quantitative description of how listeners encode and combine a coustic cues, and how they use this information to determine which sound th ey heard. Thus, it allows for testing specific hypotheses about phoneme ide ntification in a very stringent fashion. As an example of the model's appli cation, vowel identification matrices obtained with synthetic speech stimul i (including "conflicting cue" conditions [Dorman et al., J. Acoust. Sec. A m. 92, 3428-3432 (1992)] were examined. The listeners were users of the ''c ompressed-analog'' stimulation strategy, which filters the speech spectrum into four partly overlapping frequency bands and delivers each signal to on e of four electrodes in the cochlea. It was found that a simple model incor porating one temporal cue (i.e., an acoustic cue based only on the time wav eforms delivered to the most basal channel) and spectral cues (based on the distribution of amplitudes among channels) can be quite successful in expl aining listener responses. The new approach represented by the MPI model ma y be used to obtain useful insights about speech perception by cochlear imp lant users in particular, and by all kinds of listeners in general. (C) 200 0 Acoustical Society of America. [S0001-4966(00)00302-7].