第19回定例研究会

・ 日時 ３月５日（土）13:30～17:30

・ 場所 法政大学 市ヶ谷キャンパス ボアソナードタワー１１階 BT１１００番教室（下記会場案内をご参照下さい。）

・ 参加 人数把握のため事前に下記問い合わせ先までご連絡をお願い致します。

【問い合わせ，申込み先】

電気通信大学 猪原敬介

E-mail: kei.inohara@gmail.com

・ プログラム

研究発表会（1人25分。5名発表）と特別講演（Doug Roland先生（東京大学））の２つが行われます。

【研究発表会】

★猪原敬介（電気通信大学 ポスドク研究員）

＜タイトル＞

Computational explanation of “fiction text effectivity” for vocabulary improvement: Corpus analyses using latent semantic analysis

（語彙学習における「フィクション本の読書」の優位性についての計算論的検討：潜在意味解析を用いたコーパス分析）

＜アブストラクト＞

本研究では，先行研究より示されてきた「フィクション本の読書頻度はノンフィクション本よりも語彙力と強く関連する」という現象について，「その原因はテキストにおける単語の出現パターンが異なるためである」という仮説を検証した。仮説検証のために，新たに語彙テストを作成し，人間の参加者への実験（参加者実験）と，コーパス中の単語出現パターンを用いたシミュレーション実験（コーパス実験）において使用した。結果として，参加者実験では，フィクション本（物語文）を頻繁に読む参加者が，ノンフィクション本（説明文）あるいはweb上の文章を頻繁に読む参加者よりも，語彙力が高く，先行研究の結果を追試することができた。一方，コーパス実験では，説明文コーパスに基づいたシミュレーションにおいて，物語文およびweb文章に基づいたシミュレーションよりも，語彙テストの成績が高くなった。この結果より，「フィクション本の読書頻度はノンフィクション本よりも語彙力と強く関連する」という現象は，テキストそのものの違いに起因するのではなく，動機づけなどの読書行動と関わる他の要因が原因であることが示唆された。

★上田紋佳（ルーテル学院大学）

＜タイトル＞

e-learningによる英単語学習における成績のフィードバックが動機づけに及ぼす影響：大学生の動機づけスタイルによる検討

＜アブストラクト＞

現代の教育現場では学習者の動機づけを高めることが重要な課題となっている。しかしながら，動機づけを高める方略に関する研究は多くはない。そこで，本研究では，大学生を対象としたe-learningによる継続的な英単語学習において，動機づけを高めることを目的とし，定期的に学習成績のフィードバックを行うことの効果を測定した。研究会では，動機づけスタイルが英単語の学習の成績と動機づけの関係にどのような影響を及ぼすのか，また，学習の継続にどのような影響を及ぼすのかについて，発表を行う予定である。

★Roi Sawm Sumlut（国際基督教大学 大学院生）

＜タイトル＞

Online Adults’ False-belief reasoning in Second-Language Comprehension: Eye-tracking Experiment

＜アブストラクト＞

This experiment is an extension of previous experiment which examined false belief reasoning in English second language learners who read a text in L2. It was showed that in both L1 and L2, egocentric bias occurs, but the effect of the text manipulation was different. Target sentence near to the decision making process showed more bias than target sentence distant position in L2 by estimating higher probability to Green box. It is because participants have limited resource to suppress highly activated information. However, there was no differences between distant and near conditions because participants have enough resource to control in either conditions. The probability estimation judgment is the end product of off-line decision making process. The present experiment is conducted to investigate the on-line decision making process of L2 learners by using eye-tracking method, and whether the direction of text effect will be similar to the previous experiment or not. People tend to look at the thing that they are thinking about. It was predicted that participants who have egocentric bias towards the green box will make first eye fixation movement to the green box. Results of this experiment is consistent with that of previous experiment in which the distant group showed more egocentric bias than the near group. There were more people whose first eye movement fixated at Green box in the distant group. This can be explained by the limitation of cognitive resource availability as egocentric knowledge compromises participants’ ability to reason about protagonist’s beliefs.

★上野泰治（名古屋大学 日本学術振興会特別研究員PD）

＜タイトル＞

The causal role of the anterior temporal lobe in word reading: A rTMS investigation

＜アブストラクト＞

Cognitive neuroscience aims to understand how activities and interactions from specific brain areas give rise to human functions, thereby bridging a brain-mind gap. Different disciplines have different advantages to achieve this aim. For example, psychological experiments and computational modeling have clarified the mechanisms of human functions. Neuroscience and neuropsychological investigations have contributed to map these mechanisms into specific brain regions/fasciculi. Finally, brain stimulation techniques have demonstrated the causality of these identified regions/pathways for the target mechanisms. One of the missing links between these disciplines is regarding the role of the anterior temporal lobe in reading a word with an atypical pronunciation (e.g., pint). Psychology and modelling clarified that the reference to lexical-semantic information is necessary for reading atypical words. Patients who have a damage in the anterior temporal lobe show impairments in semantic tasks, and actually can not read words with atypical pronunciations. Also, fMRI evidence suggest this area is actually more activated when reading atypical words. However, somehow surprisingly, there have been no demonstration of causality of this region in reading atypical words. In this study, we applied repetitive transcranial magnetic stimulation on the left anterior temporal lobe, and demonstrated the significantly impaired reading performance in atypical words. Simultaneously, a control stimulation on supramarginal gyrus demonstrated the causal role of this area in single-word auditory repetition for the first time. These findings are discussed in terms of the dorsal-ventral dual-pathway framework of language processing.

★Xiao Jing（国際基督教大学 大学院生）

＜タイトル＞

Japanese Kanji Word Recognition Processing for Chinese Learners of Japanese: A Study of Phonological Primed Lexical Decision Tasks

＜アブストラクト＞

This study addresses the question of whether recognition of a Japanese kanji, which originates in Chinese, for semantic representation is performed without phonological mediation by advanced Chinese learners of Japanese. We investigated whether L2 kanji recognition is affected by whether a Japanese kanji word is shared with Chinese, and cause different word processing routes for Chinese learners. We also examined the effects of word frequency. This study used a computerized lexical judgment task with 2 types of Japanese kanji to investigate the question: (a) cognates (sharing the same Chinese characters), (b) non-cognates (not sharing the same Chinese characters). In addition, we have two types of non-words in the experiment. In this study, we examined phonological processing in Japanese kanji recognition by measuring the response time to a two-kanji compound word presented after a sound prime that had the same sound from the first kanji and to the two-kanji compound word presented after a sound prime which had a different sound from the first kanji. The results showed that there were significant effects on sound prime (target words with same sound primes had shorter RTs) as well as form similarity (cognates had shorter RTs). In addition, there was an interaction effect between sound primes and form similarity. That is, non-cognates were facilitated by the sound prime that had the same sound from the first kanji of the non-cognates. But the facilitation effects did not occur when the target words were cognates. We conclude that advanced Chinese learners of Japanese do not activate phonological representations for cognates in the word recognition process whereas they do activate phonological representations for non-cognates.

【特別講演】

★Doug Roland先生（東京大学）

＜タイトル＞

What role does discourse play a role in relative clause processing?

＜アブストラクト＞

Based on work by Fox and Thompson (1990), a series of studies looking at the processing of relative clauses (Mak, Vonk, & Schriefers, 2002, 2006, 2008; Reali & Christiansen, 2007; Roland, Mauner, O’Meara, & Yun, 2012) have suggested that discourse-based expectations play an important role in determining the difficulty incurred while processing such relative clauses. However, these previous studies all relied on reading time data from the self-paced moving window paradigm. While reading time data from studies relying on the self-paced paradigm generally corresponds with reading time data from eye movement studies, some differences have been noted (e.g., Smith & Levy, 2013; Staub, 2010). In particular, recent data (Roland, Hirose, Mauner, & Foraker, 2015) suggest that some of the previous results that were attributed to discourse effects may in fact be due to the nature of the self-paced reading paradigm. In this talk, I will discuss the evidence for the role of discourse context in relative clause processing, focusing on the relationship between the underlying causes of processing difficulty and the techniques used to measure this difficulty.

会場案内：

法政大学市ヶ谷キャンパス

ボアソナードタワーは，市ヶ谷キャンパス内の２６階建てのビルです。

エレベーターはオレンジ（低層階用）と青（高層階用）の２種類あります。

どちらもご利用できますが，青いエレベータの方が便利です。

そのまま直接，11階のBT1100教室までお越しください。

守衛所等で受け付けなどは必要ありません。

なお，緊急の連絡先は，０３-３２６４-９３８２（福田由紀研究室直通），

０３-３２６４-５０２４（実験室付属の準備室）です。

http://www.hosei.ac.jp/hosei/campus/annai/ichigaya/access.html

http://www.hosei.ac.jp/hosei/campus/annai/ichigaya/campusmap.html