Acoustic Cue Weighting in Children Wearing Cochlear Implants

The purpose of this study was to determine how normal hearing adults (NHA), normal hearing children (NHC) and children wearing cochlear implants (CI) differ in the perceptual weight given cues for fricative consonant and voiceless stop consonant continua. Ten normal-hearing adults (NHA), eleven 5-8-year-old normalhearing children (NHC) and eight 5-8-year-old children wearing cochlear implants (CI) were participants. For fricative consonant perception, the /su/-/∫u/ continua were constructed by varying a fricative spectrum cue in three steps and by varying a F2 onset transition cue in three steps. For voiceless stop consonant perception, the /pu/- /tu/ continua were constructed by varying a burst cue in three steps and a F2 onset transition cue in three steps. A quantitative method of analysis (ANOVA model) was used to determine cue weighting and measure cue interaction. For the fricative consonant, both NHC and NHA gave more perceptual weight to the frication spectral cue than to the formant transition. NHC gave significantly less weight to the fricative spectrum cue than NHA. The weight given the transition cue was similar for NHC and NHA, and the degree of cue interaction was similar between two groups. The CI group gave more perceptual weight to the fricative spectrum cue than to the transition. The degree of cue interaction was not significant for CI. For the voiceless stop consonant, both NHC and NHA gave more perceptual weight to the transition cue than to the burst cue. NHC gave proportionately less weight to the transition cue than NHA. The weight given the burst cue and the degree of cue interaction were similar between NHC and NHA. The CI group gave more perceptual weight to the transition cue than to the burst cue, and there was no significant difference between children wearing cochlear implants and normal hearing children group; however, the degree of cue interaction was not significant for CI. These results indicated that all groups favored the longer-duration cue to make phonemic judgments. Also there were developmental patterns. The CI group has similar cue weighting strategies to agematched NHC, but the integration of the cues was not significant for either fricative or voiceless stop consonant perception.