2006 Fiscal Year Final Research Report Summary
Eye-Movement Study on Processing Garden Path Sentences by Japanese EFL Learners
| Project/Area Number |
16520366
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Foreign language education
|
|---|
| Research Institution | Kwansei Gakuin University |
|---|
Principal Investigator |
KADOTA Shuhei Kwansei Gakuin University, Department of Law and Politics, Professor, 法学部, 教授 (20191984)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
HASE Naoya Kwansei Gakuin University, Department of Science and Technology, Associate Professor, 理工学部, 助教授 (50309407)
YOKOKAWA Hirokazu Kobe University, Department of Languages and Communication, Associate Professor, 国際コミュニケーションセンター, 助教授 (50340427)
YOSHIDA Haruyo Osaka Kyoiku University, Department of English Language Teaching, Professor, 教育学部英語教育講座, 教授 (40210710)
|
|---|
| Project Period (FY) |
2004 – 2006
|
| Keywords | Eye Movements / Sentence Processing / Garden Path Sentences / English as a Foreign Language / Syntactic Parsing / Semantic Processing / Japanese Learners of English / Psycholinguistics |
|---|
| Research Abstract |
The study is an attempt to explore the cognitive process in L2 written sentence processing, focusing on the eye-movement data gathered using the EMR-8 (Eye-Mark Recorder). The graduate and undergraduate university students learning English as FL were required to visually process a variety of garden path (GP) and control sentences, while their eye movements were recorded. The major conclusions are as follows:
(1) Japanese EFL learners experience a significant increase of processing difficulty when they are visually given GP-sentences than non-GP control and filler sentences.
(2) The type of GP-sentences affects the relative processing difficulty in terms of the number of fixations and regressions for Japanese EFL learners.
(3) The average duration data per a single fixation in general, do not provide a sensitive measure to differentiate the GP-and the non-GP sentences as well as the types of four different GP sentences.
(4) The syntactic factors such as using an unreduced relative clause or an unambiguous past participle form reduce the processing difficulty of GP-sentences, and thus are important clues to process printed sentences.
(5) Both the semantic variable (i.e.the animacy of the subject NP) and the pragmatic variable (i.e.background knowledge bias on the subject NP) reduce the processing difficulty of GP-sentences, and thus are crucial clues to process printed sentences.
(6) The more detailed, phrase-based eye-movement data on fixations, regressions and total fixation durations also support and confirm the above conclusions.
The fact that both the semantic and pragmatic variables affect sentence processing basically in an incremental fashion seems to support the notion of constraint-based processing model, which assumes the interaction of various linguistic information (e.g.syntactic, lexical, semantic, pragmatic) and is essentially associated with the connectionist architecture of human cognition.
|
Research Products
(6 results)
