Selective adaptation of context-conditioned voiceless fricatives

Two fricative-vowel continua were synthesized. One ranged from /si/ to /(sh)i/ and the other from /su/ to /(sh)u/. Listeners’ identification of the consonant portion of the syllables, as either ‘‘s’’ or ‘‘sh,’’ depended on both the frequency of the frication noise and on the vowel quality. Selective adaptation of these continua was investigated in three experiments. In experiment 1, the endpoints of the continua were used as adaptors, and the identification boundary shifted towards the adapting stimulus. In experiment 2, ambiguous frication noises (that were identified as /s/ before /u/ and /(sh)/ before the /i/) adapted the identification boundary in opposite directions, depending on which of the two vowels followed the noise and, therefore, on the perceptual identity of the consonant. In experiment 3, the isolated /i/ and /u/ vowels were demonstrated to be ineffective as adaptors. The selective adaptation effects observed in these experiments were not determined by the acoustical information in the consonant or the vowel alone, but rather the context-conditioned percept of the fricative. [Work supported by NSERC.]