Investigating How Perceptual Organization and Linguistic Memory Processes Interact to Promote Spoken Word Perception

Intro. A full account of speech perception requires explaining how listeners organize the acoustic signal into speech objects (perceptual organization) and how listeners use their memory for language to impart meaning to the speech objects (linguistic memory). Traditionally these mechanisms have been investigated separately, and thus theorized as two independent, sequentially- applied mechanisms. The prominent view of perceptual organization defines it as an early, low-level process, one that occurs prior to linguistic processing of speech. However, recent studies have found influences of organizational cues on linguistic percepts, indicating that the two mechanisms might not be sequential and independent. This dissertation attempts to explicate how perceptual organization and linguistic memory (lexical memory for words and sentential memory for context) interact when organizing and perceiving speech. Methods. I presented participants with complex, un-organized speech scenes, with a stream of [s]s occurring in one ear and a sentence in the other. I asked participants to listen for the last word and report whether it started with a voiceless (p/t) or voiced (b/d) sound. Due to the phonotactic properties of English, the perceived identity of the target word changes from voiced to voiceless if the listener organizes it with a [s] from the s-stream ("[s]+base" perceived as "space"). By manipulating the grouping strength of the [s] stream and the linguistic biases embedded in the sentence (lexical and sentential), I tested whether and how perceptual organization and linguistic memory mechanisms interact during speech perception. The response time and cue weighting strategies of participants were also examined, to give more in depth information about the processing ease and cue tradeoffs. Results. Experiment 1 compared categorization of target words with lexical biases (voiced, unbiased, voiceless) to s-streams with grouping strengths (strong, weak) and found independent effects of both manipulations on perception. Cue weighting analysis found a consistent trade-off in cue reliance by participant, with most but not all relying more on memory cues than organizational cues. The response time data indicated that the speed of response was dependent on the cued organization: it was easier to make responses that agreed with the organization implied by the s-streams. Experiments 2-4 tested the interaction of organization and sentential processing (sentences biased by context) and found an effect of grouping strength that varied (but was never eliminated) based on the presence of sentence. Cue weighting data showed that sentential context cues were almost universally relied upon for categorization, while response time data supported the findings of Experiment 1, that response speed was influenced primarily by organization cues. Conclusions. Taken together, my results imply that that linguistic memory and organizational information are both applied in an overlapping fashion, with an interactive juncture at the level of lexical processing and a much weaker interactive juncture at the sentential level of processing. In all experiments, both linguistic and organizational cues influenced the perceived target word, clarifying the organization of the two overlapping auditory streams. The combined results of the dissertation experiments bring together the two literatures on auditory perception: linguistic and perceptual organization cues are integrated together to inform the speech percept. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]