NotesFAQContact Us
Search Tips
ERIC Number: ED530772
Record Type: Non-Journal
Publication Date: 2010
Pages: 125
Abstractor: As Provided
Reference Count: 0
ISBN: ISBN-978-1-1247-0916-1
Is Word Shape Still in Poor Shape for the Race to the Lexicon?
Hill, Jessica C.
ProQuest LLC, Ph.D. Dissertation, The Florida State University
Current models of normal reading behavior emphasize not only the recognition and processing of the word being fixated (n) but also processing of the upcoming parafoveal word (n + 1). Gaze contingent displays employing the boundary paradigm often mask words in order to understand how much and what type of processing is completed on the parafoveal word before the word is fixated. The word "beach" might be masked by a homophone ("beech"), a semantically related word ("shore"), or a string of random letters ("tcsok"), which can be either similar (b as compared to d) or dissimilar (b as compared to x) to the letters of the original word. Mask use is inconsistent and ill defined across studies. It is common for words to be considered masked with random letters if either all letters in the word are masked with random letters or the same first, second, or third letters of the word are used with random letters masking the rest of the word, thus leading to conflicting results in the body of the reading literature. This work examines the role of word shape and letter discriminability in two experiments in the context of experimental reading research. The first experiment manipulated mask similarity, either similar or dissimilar letters, and whether word shape information was provided or denied, while controlling for lexical frequency (high) and word shape frequency (rare). Results indicate a significant effect of letter similarity on the parafoveal preview benefit, but there was not an equivalent result for word shape. The findings for dissimilar masks impacting parafoveal processing more than similar masks are critical in terms of the methodology of reading experiments. Few researchers control for mask similarity in their studies as there was no indication, before now, that such a control was necessary. The lack of findings for word shape was attributed to the lack of a manipulation that tapped into higher order cognition. A second experiment was conducted that included a contextual constraint manipulation, either high or normal contextual constraint, while controlling for lexical frequency (high), word shape frequency (rare), and mask similarity (dissimilar). The results indicated a significant preview benefit and predictability effect; however, there was no significant finding for word shape. Thus, it seems that word shape still seems to be in poor shape for the race to the lexicon. Results are discussed in terms of models of normal reading. [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:]
ProQuest LLC. 789 East Eisenhower Parkway, P.O. Box 1346, Ann Arbor, MI 48106. Tel: 800-521-0600; Web site:
Publication Type: Dissertations/Theses - Doctoral Dissertations
Education Level: Higher Education
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A