| Abstract | Presented dichotic CV syllables contrasting in 2 features of the initial stop consonant (voicing and place) to 8 university students for identification in a single-response paradigm without selective attention instructions. The acoustic structure of the syllables was varied within categories on both dimensions (voice onset time and formant transitions). The variations (especially those in voice onset time) had a clear influence on the pattern of responses (including blends), which ruled out a simple phonetic feature-recombination model. Rather, the auditory properties of the stimuli seemed to be preserved at the stage of dichotic interaction. A prototype model, which assumed that
dichotic integration of information takes place at a stage intermediate between auditory and phonetic processing, was only moderately supported by the data. Nevertheless, some arguments are presented for maintaining this model as a working hypothesis. A new procedure for estimating the dichotic ear advantage is applied for the first time in conjunction with the single-response requirement. Most Subjects showed unusually large right-ear advantages, making the present methodology interesting for the study of hemispheric asymmetry. |