We will argue that the beginnings of words are perceptual "islands of reliability" in connected speech, and that their perceptual and temporal properties allow them to drive critical aspects of spoken word recognition including lexical segmentation. This argument rests on three generalizations derived from research in speech science, phonology, and psycholinguistics. We suggest that word onsets differ from other parts of words in that: (1) they offer more robust and redundant acoustic evidence about phonetic features, (2) they are generally protected from phonological assimilation, neutralization and deletion and therefore show less lawful variation in surface realization, and (3) they may activate lexical representations which facilitate word perception and thus diminish listeners' dependence on veridical acousticphonetic processing of other portions of words. These properties of word onsets allow them to drive lexical segmentation by facilitating the recognition of items that begin with clear onsets. The implications of these findings for several models of lexical segmentation and spoken word recognition are discussed.
The single word repetition of two fluent aphasics was evaluated to explore the nature of their lexical-phonological deficit. Phonological theory postulates that features, subcomponents of segments, are organized hierarchically according to their acoustic and articulatory properties and their role in the phonological system. This study hypothesized that the higher a feature sits in the hierarchy, the more stable it will be following neurological damage, and the less likely it is to be substituted independently of other features. Analyses focused on the status of the feature sonorant, for the following two reasons: (1) this feature is located at the top of the feature hierarchy; and (2) unlike other consonant features, it plays an important role in syllable structure by constraining the ordering of consonants within syllables. The pattern of feature substitutions produced by the two subjects in this study was identical, with the order of least to most affected type conforming to the predictions based on the feature hierarchy: consonant substitution errors due to a change in the value of the feature sonorant were rare, and when this feature changed, it generally occurred in conjunction with changes to other features. By contrast, place of articulation, a feature that sits low in the hierarchy, was the most frequent feature substituted, and was typically altered without changing the other feature values of the affected consonant. Moreover, when sonorant feature substitutions and consonant omissions did occur, they most often produced a syllable with a less complex sonority pattern. These findings strengthen the notion that the relative preservation of sonority, both at the segmental and syllabic levels, contributes to the normal sounding quality of the neologisms produced by even the most severe jargonaphasics.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.