Identifier
etd-04132015-125534
Degree
Doctor of Philosophy (PhD)
Department
Psychology
Document Type
Dissertation
Abstract
Within the field of psycholinguistics there are those who argue for a close relationship between working memory capacity (WMC) and syntactic processing (Just and Carpenter, 1992) and those who argue that there is no such relationship (Waters and Caplan, 1996b; 2004). Despite years of research, empirical data has yet to settle this disagreement, perhaps because a number of methodological differences between studies from each side make direct comparisons of data nearly impossible. The current study was designed to partially replicate three previous studies using their own experimental sentence types in a self-paced word-by-word reading paradigm in order to examine the effects of several methodological factors, including judgment type and object-relative clause construction, on performance in the syntactic processing task. In Experiment 1 we used Just and Carpenter’s (1992) true/false judgment and found data theoretically consistent with their main results which supported a relationship between WMC and syntactic processing, but only for sentence sets constructed in the manner of Waters and Caplan (1996b; 2004). In Experiment 2 we used Waters and Caplan’s (1996b; 2004) acceptability judgment with the same stimulus sets and found no support for a WMC-syntactic processing relationship. Finally, in Experiment 3 we used a grammaticality judgment with the same stimulus sets and once again found no support for a WMC-syntactic processing relationship. Together, the results of all three experiments suggest that a number of methodological factors that have been previously considered irrelevant in the syntactic processing task in fact produce significant changes in the results that in turn alter the conclusions drawn. In particular, we found evidence that judgment type can alter overall reading times, suggesting that task demands may cause participants to alter their processing strategies in the task. Our results also illustrate that the pattern of results can depend largely upon the method of object-relative clause construction used for the task.
Date
2015
Document Availability at the Time of Submission
Release the entire work immediately for access worldwide.
Recommended Citation
Horn, Rebecca Ann, "Single or Dual Resources: The Role of Working Memory in Syntactic Processing" (2015). LSU Doctoral Dissertations. 4072.
https://repository.lsu.edu/gradschool_dissertations/4072
Committee Chair
McDonald, Janet
DOI
10.31390/gradschool_dissertations.4072