Off-campus UMass Amherst users: To download dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.
Non-UMass Amherst users, please click the view more button below to purchase a copy of this dissertation from Proquest.
(Some titles may also be available free of charge in our Open Access Dissertation Collection, so please check there first.)
Planning referential expressions in speech production
Abstract
Translating thoughts into sentences involves planning at several levels of representation (e.g., conceptual, syntactic, phonological). An adequate theory of speech production must specify the principles governing planning at different levels of representation, including the scope of planning for a given level and the relations among levels. In three experiments, I investigated referential and phonological planning in a naturalistic speaking task. In the central experiment, subjects described to a listener simple events involving objects displayed on a computer screen. On each trial, line drawings of four objects were presented simultaneously, each on a background square of a different color. One of the pictures jumped next to one of the other pictures and then jumped back to its original position. The subject's task was to describe to a listener what had happened, using the verb "bumped" and having been told that the listener had the same display but without the movement. An example utterance was "The cat bumped the motorcycle." Subjects' utterances were tape-recorded for later analysis. Two factors were manipulated: First, whether a color adjective was needed to distinguish an object involved in the event from an identical, but differently colored, object in the display, and, second, whether the names for the objects were short (one syllable) or long (three or more syllables). The ambiguity manipulation varied the complexity of planning at the message level, and the length manipulation varied phonological encoding (longer words take longer to prepare). The pattern of initiation times and within-sentence delays provided evidence for the following conclusions: (1) The speaker's decision to initiate speech depends on having achieved a certain stage in planning, which can vary from task to task (the Transparency Assumption). (2) In the most natural of the experimental tasks, initiation depended on retrieving the phonology of the subject noun, even when it was preceded by an adjective, consistent with a proposed Lexical Head Principle for initiation. (3) The scope of planning during speech depends on available mental capacity, not on the structure of the utterance--complexity early in an utterance decreased the amount of upcoming material that could be planned concurrently.
Subject Area
Experimental psychology|Language|Psychology
Recommended Citation
Huitema, John S, "Planning referential expressions in speech production" (1993). Doctoral Dissertations Available from Proquest. AAI9408290.
https://scholarworks.umass.edu/dissertations/AAI9408290