UBC Theses and Dissertations
Perceptual confusions among permissible and impermissible english consonant clusters Newton, Colleen Nora
The present study is an attempt to gain insight into the perceptual mechanism for consonant clusters and to discern its relationship to existing theories of speech perception. Nonsense syllables were constructed, consisting of the vowel /I/ and 18 two consonant clusters composed of one of the fricatives /s/, /f/, and /∫/ and one of the plosives /p/, /t/, and /k/ in either order. Some of these clusters form permissible English clusters, the others impermissible clusters. These clusters appeared in three positions; initial, medial (followed by /in/ ), and final. They were recorded by a trained phonetician. Pink noise at a +3 dB signal-to-noise ratio (determined from peak readings) was used to mask the signal. Nine experimental tapes (three for each cluster position) were constructed; 108 items were presented on each tape. Eighteen subjects (nine males and nine females) each listened to one tape for each position and responded according to a forced choice paradigm to each item by writing the cluster they perceived. Responses for all subjects and for each position were tabulated in confusion matrices and analyzed according 1) to permissibility as English clusters, 2) to manner of articulation, 3) to place of articulation and 4) to differentiation according to the distinctive features of [anterior], [coronal], and [distributed]. Inspection of these confusion matrices indicated that there may be some difference in the perception of permissible and impermissible clusters; however, results are not conclusive. The proposal that clusters are perceived as a unit is refuted by the results obtained. Manner of articulation was identified correctly more frequently than place of articulation. Place of articulation for fricatives was identified correctly more frequently than place of articulation for plosives. Analyses according to distinctive features provides some support for distinctive features. The relationship of the results of this study to four major theories of speech perception is discussed. Three of these can be used to explain some aspect of the responses observed.
Item Citations and Data