 |
15th International Congress of Phonetic Sciences (ICPhS-15)Barcelona, Spain |
 |
|
15th International Congress of Phonetic Sciences (ICPhS-15)Barcelona, Spain |
|
We examined the effect of a "mismatch" between a listener's phonological representations and speech input on spoken word recognition. Native Spanish (NS) and native English (NE) listeners were asked to write down 80 English words that were presented in noise. The words differed in neighborhood density (ND: dense versus sparse); half were Spanish-accented (produced by a NS talker), the other half were unaccented (i.e., produced by a NE talker). We hypothesized that phonological mismatches would occur when NS listeners responded to unaccented words and when NE listeners responded to Spanish-accented words. Further, the effect of the mismatch was expected to be greater for words from dense versus sparse neighborhoods because these words can be confused with many minimally-paired neighbors. The results supported the mismatch hypothesis. NS listeners showed a larger ND effect for unaccented than Spanish-accented stimuli, whereas NE listeners showed a larger ND effect for Spanish-accented than unaccented stimuli.
Bibliographic reference. Imai, Satomi / Flege, James E. / Walley, Amanda (2003): "Spoken word recognition of accented and unaccented speech: lexical factors affecting native and non-native listeners", In ICPhS-15, 845-848.