Please use this identifier to cite or link to this item:
|Title:||Perceptual Cue Weighting of Intervocalic Velar Plosives in English|
|Department:||Cognitive Science of Language|
|Abstract:||When producing speech, the acoustic signals contain several extractable parameters (cues) about the intended sound which are reliably mapped to achieve robust phoneme identification in speech perception, despite large acoustic variation in speech signals between and within individuals. This thesis presents findings on the perceptual cue weighting of intervocalic velar stop consonants /k/ and /g/ for English listeners, through use of biomechanical stimuli. Four acoustic cues were systematically manipulated to create the experimental stimuli: voicing maintenance throughout consonantal closure (VM), voice onset time (VOT), the duration of consonantal closure (CL) and the duration of the previous vowel segment (VL). This thesis presents the findings of three experiments. Experiment 1 investigates the cue weighting of three acoustic cues (VM, VL, CL) in the absence of a perceptual VOT cue. Experiment 2 defines a perceptually ambiguous VOT value [for the given biomechanical stimuli] and compares the influences of VOT and VM. Experiment 3 analyses how the perceptual cue weighting of the three acoustic cues (VM, VL, CL) is affected when a perceptually ambiguous VOT value is added to stimuli. The results from the three experiments reveal that VM has a significantly higher influence on voicing perception than the other cues, and only at low VM levels do other cues increase their influence. The cues interacted significantly with each other. The effect of VOT – known in literature as the main cue for consonantal voicing distinctions – was apparent at high VM values but significantly increased at low VM values. Experiment 2 gave additional evidence that individual listeners utilized very different strategies in determining voicing perception. Ultimately the results show that the perceptual cue weighting process is highly complex and cannot be attributed to only one or two perceptual cues (e.g. VOT).|
|Appears in Collections:||Open Access Dissertations and Theses|
Files in This Item:
|Mercier_Olivier_K_201909_MSC.pdf||3.02 MB||Adobe PDF||View/Open|
Items in MacSphere are protected by copyright, with all rights reserved, unless otherwise indicated.