A bulk of research has been conducted to clarify the effect of speech rate and pause duration on listening comprehension. Some previous studies have shown that a slower speech rate facilitates listening comprehension, while others have yielded opposite results. The conflicting results imply that it is very difficult to specify the most appropriate speech rate for every EFL listener. This study aims to clarify experimentally how pause duration and articulation rates affect EFL learners' listening comprehension by precisely controlling the two parameters: articulation rate and pause duration. Through two listening tests given to Japanese EFL learners, we obtained the result that longer pauses (450 ms) facilitate the listening comprehension of lower intermediate EFL learners, while slower articulation rate does not necessarily have a positive effect on learners' listening comprehension. This means a pause inserted in a passage provides listeners with additional information processing time and thus enhances the comprehensibility of the aural input.

[概要] 本研究では, 現状のリスニングテストでは見過ごされている音声的な要素（音素・音節・語彙の知覚など）を洗い出し, 音声面に特化したリスニングテストを作成するための基礎データを集めることを目的とした. 手法としては, いわゆるリスニングテストで同じスコアの学習者を２グループ各３名ずつ実験協力者に選び, ディクテーション・インタビュー課題を用いた調査を行った. 詳細なデータを集めた. その結果, いわゆるリスニングテストで同程度の習熟度と判定される学習者間であっても聞き間違いのパターンには違いが見られた. [Abstract] Listening comprehension comprises various processes, such as sound perception, phonological decoding, lexical access, syntactical processing, and so on. Because of this complexity, it is improbable that the identical score on one common listening proficiency test marked for different test takers predicts their similar level of sub-listening skills. Based on the above assumptions, the following research question was used to guide the study: (1) Does the typical listening comprehension test, which is composedof multiple choice comprehension questions, predict learners' sub-skills of listening comprehension ? (2) If it does not, what kind of differences can we observe between learners of equivalent listening ability ? Based on the result of dictation tasks and interview with the participants, no predictability of listening test was observed on listening sub-skills.

This study investigated the use of verb subcategorization information during sentence comprehension by Japanese EFL learners, using a self-paced reading task with the embedded anomaly technique. In order to observe when the syntactic structure was determined, filler-gap sentences were constructed as stimuli. Four types of verbs were used: simple transitive verbs, dative verbs, infinitive complement verbs and intransitive preference verbs. The plausibility of the direct object of the embedded verb was also manipulated. The results demonstrate that the high proficiency learners utilize verb subcategorization information of simple transitive verb; however they do not use it automatically as natives. The results also indicate that they cannot utilize that of other types of verbs. The low proficiency learners show the difficulty in processing the sentences with long-distance dependencies.

In order to survey the effects of pause on EFL learners' listening comprehension, two experiments were conducted. The purpose of Experiment 1 is to examine the influence of the following three parameters: (1) The grammatical complexity of the sentence; (2) the location of silent pauses; and (3) the frequency of pauses within the sentence. The results of the experiment show the following effects of pause on listening comprehension: (1) Listeners achieved higher scores when listening to sentences with one pause, two pauses, and no pause, respectively. (2) The shortest sentence (7 syllables in length) with no pause inserted was the hardest to perceive for the EFL listeners. That is, sentence length does not seem to have much effect on perception. (3) The score of sentences which have longer subjects (noun phrase +propositional phrase) was lower than in other sentences with different grammatical constructions. (4) The listening score increasedfor words immediately followed by a pause. The purpose of Experiment 2 is to analyze the effects that pause length have on Japanese EFL learners' listening performance. The result shows that pause length does not affect the overall listening process except for in cases of a certain sentence type. The above findings lead us to conclude that one pause in a sentence of 9 to 14 syllables in length facilitates learner s listening process regardless with the length of each pause.

This project attempted to ecplore the automatization process for langauge comprehension and production by foreign language learners through psycholingusitic and physiological experiments. The major findings are as follows: (1)Sscores on the RST, which places a cognitive load on syntactic processing in particular, was significantly poorer than those of other types of RSTs and correlated significantly with English proficiency scores. English proficiency largely depends on the degree of automatization of syntactic processing. (2)The process required to carry out the sentence building task(syntactic processing) becomes more automatic as the learners' fluency level increases, and the learners with higher L2 fluency had the option of allocating more of their cognitive resources to story comprehension than did the learners with lower L2 fluency.

The purpose of our project is to make an English listening test which can precisely predict the learner's phoneme perception skills. To attain this goal we conducted a preliminary research to see how existing listening tests can predict this ability of the learner's. Through the dictation test and interview to the learners, it was found that the learners who are regarded to have the same level of listening comprehension ability differ in their overall and individual consonant perception skills. The result of a minimal pair test of English initial consonants of mono-syllabic words shows that percentage of correct answers didn't differ much, but reaction time for answering each question differ a lot between the learners and between the test items. From these results it is clear that the existing listening tests cannot predict the learner's consonant perception skills. Based on the findings we obtained through a series of studies, we plan to produce a reliable listening test.