The coding of vowel identity by patients who use the Ineraid cochlear implant.

The aim of this experiment was to determine whether patients who use the Ineraid cochlear implant code formant frequency by means of the information in the time waveforms presented to the electrodes, or by the information in the distribution of energy among the electrodes. Six patients were presented three vowels for identification. The vowels were presented in two configurations. In one, the temporal cues and distribution of energy cues were both appropriate for a given vowel. In the other, the temporal information specified one vowel while the distribution of energy specified another vowel. The results of the identification task indicated that patients relied on information contained in the distribution of energy for vowel identification. These results can be interpreted as indicating that the Incraid overcomes the limited spatial resolution afforded by four monopolar electrodes by using the balance of energy among electrodes to produce a continuous, rather than discrete, coding of cochlear place of stimulation.