Search

The multidimensional phoneme identification model is applied to consonant confusion matrices obtained from 28 postlingually deafened cochlear implant users. This model predicts consonant matrices based on these subjects' ability to discriminate a set of postulated spectral, temporal, and amplitude speech cues as presented to them by their device. The model produced confusion matrices that matched many aspects of individual subjects' consonant matrices, including information transfer for the voicing, manner, and place features, despite individual differences in age at implantation, implant experience, device and stimulation strategy used, as well as overall consonant identification level. The model was able to match the general pattern of errors between consonants, but not the full complexity of all consonant errors made by each individual. The present study represents an important first step in developing a model that can be used to test specific hypotheses about the mechanisms cochlear implant users employ to understand speech.

In the present study, a computational model of phoneme identification was applied to data from a previous study, wherein cochlear implant (CI) users' adaption to a severely shifted frequency allocation map was assessed regularly over 3 months of continual use. This map provided more input filters below 1 kHz, but at the expense of introducing a downwards frequency shift of up to one octave in relation to the CI subjects' clinical maps. At the end of the 3-month study period, it was unclear whether subjects' asymptotic speech recognition performance represented a complete or partial adaptation. To clarify the matter, the computational model was applied to the CI subjects' vowel identification data in order to estimate the degree of adaptation, and to predict performance levels with complete adaptation to the frequency shift. Two model parameters were used to quantify this adaptation; one representing the listener's ability to shift their internal representation of how vowels should sound, and the other representing the listener's uncertainty in consistently recalling these representations. Two of the three CI users could shift their internal representations towards the new stimulation pattern within 1 week, whereas one could not do so completely even after 3 months. Subjects' uncertainty for recalling these representations increased substantially with the frequency-shifted map. Although this uncertainty decreased after 3 months, it remained much larger than subjects' uncertainty with their clinically assigned maps. This result suggests that subjects could not completely remap their phoneme labels, stored in long-term memory, towards the frequency-shifted vowels. The model also predicted that even with complete adaptation, the frequency-shifted map would not have resulted in improved speech understanding. Hence, the model presented here can be used to assess adaptation, and the anticipated gains in speech perception expected from changing a given CI device parameter.

A simple mathematical model is presented that predicts vowel identification by cochlear implant users based on these listeners' resolving power for the mean locations of first, second, and/or third formant energies along the implanted electrode array. This psychophysically based model provides hypotheses about the mechanism cochlear implant users employ to encode and process the input auditory signal to extract information relevant for identifying steady-state vowels. Using one free parameter, the model predicts most of the patterns of vowel confusions made by users of different cochlear implant devices and stimulation strategies, and who show widely different levels of speech perception (from near chance to near perfect). Furthermore, the model can predict results from the literature, such as Skinner, et al. [(1995). Ann. Otol. Rhinol. Laryngol. 104, 307-311] frequency mapping study, and the general trend in the vowel results of Zeng and Galvin's [(1999). Ear Hear. 20, 60-74] studies of output electrical dynamic range reduction. The implementation of the model presented here is specific to vowel identification by cochlear implant users, but the framework of the model is more general. Computational models such as the one presented here can be useful for advancing knowledge about speech perception in hearing impaired populations, and for providing a guide for clinical research and clinical practice.

In the present study, a computational model of phoneme identification was applied to data from a previous study, wherein cochlear implant (CI) users’ adaption to a severely shifted frequency allocation map was assessed regularly over 3 months of continual use. This map provided more input filters below 1 kHz, but at the expense of introducing a downwards frequency shift of up to one octave in relation to the CI subjects’ clinical maps. At the end of the 3-month study period, it was unclear whether subjects’ asymptotic speech recognition performance represented a complete or partial adaptation. To clarify the matter, the computational model was applied to the CI subjects’ vowel identification data in order to estimate the degree of adaptation, and to predict performance levels with complete adaptation to the frequency shift. Two model parameters were used to quantify this adaptation; one representing the listener’s ability to shift their internal representation of how vowels should sound, and the other representing the listener’s uncertainty in consistently recalling these representations. Two of the three CI users could shift their internal representations towards the new stimulation pattern within 1 week, whereas one could not do so completely even after 3 months. Subjects’ uncertainty for recalling these representations increased substantially with the frequency-shifted map. Although this uncertainty decreased after 3 months, it remained much larger than subjects’ uncertainty with their clinically assigned maps. This result suggests that subjects could not completely remap their phoneme labels, stored in long-term memory, towards the frequency-shifted vowels. The model also predicted that even with complete adaptation, the frequency-shifted map would not have resulted in improved speech understanding. Hence, the model presented here can be used to assess adaptation, and the anticipated gains in speech perception expected from changing a given CI device parameter. [ABSTRACT FROM AUTHOR]

Information transfer analysis [G. A. Miller and P. E. Nicely, J. Acoust. Soc. Am. 27, 338–352 (1955)] is a tool used to measure the extent to which speech features are transmitted to a listener, e.g., duration or formant frequencies for vowels; voicing, place and manner of articulation for consonants. An information transfer of 100% occurs when no confusions arise between phonemes belonging to different feature categories, e.g., between voiced and voiceless consonants. Conversely, an information transfer of 0% occurs when performance is purely random. As asserted by Miller and Nicely, the maximum-likelihood estimate for information transfer is biased to overestimate its true value when the number of stimulus presentations is small. This small-sample bias is examined here for three cases: a model of random performance with pseudorandom data, a data set drawn from Miller and Nicely, and reported data from three studies of speech perception by hearing impaired listeners. The amount of overestimation can be substantial, depending on the number of samples, the size of the confusion matrix analyzed, as well as the manner in which data are partitioned therein. [ABSTRACT FROM AUTHOR]