Word recognition and the lexicon

Jola Jakimik · The Journal of the Acoustical Society of America · 1979

How the listener uses his knowledge of the words in his language during word recognition was investigated using a listening for mispronunciations task. The critical mispronunciations involved changing the initial consonant of the second syllable of a word to produce a nonword. The amount of constraint on the word's identity provided by its first syllable was manipulated. One set of words had first syllables which begin many words in English, e.g., composer, impossible; a second set had infrequent first syllables, e.g., vampire, whisper. Listeners heard these mispronunciations, along with other items, in the phrase “Say __________ again” and in sentences. Times to detect the mispronunciations were about 100 ms faster for words with infrequent, i.e., more constraining first syllables. In addition, detection times reflected the point at which the mispronunciation became a nonword. These results imply that information provided by the beginning of the word is crucial in word recognition, and that determining the identity of a word involves a sequential interpretation of the input.

Read the paper · More papers on PaperTik