探究虛擬實境英語詞彙學習之效應

全文

(1)國立臺灣師範大學英語學系碩士論文 Master Thesis Department of English National Taiwan Normal University. 探究虛擬實境英語詞彙學習之效應 The Effects of Virtual Reality Mediation on English Vocabulary Learning. 指導教授：曾文鐽博士 Advisor: Dr. Wen-Ta Tseng. 研究生：劉皓娟 Student: Hao-Jyuan Liou. 中華民國 102 年 7 月 July, 2013.

(2) . CHINESE ABSTRACT 幾十年來，以英語作為第二語言之詞彙學習研究，一直不斷地在發展當中。隨著電腦科技的進步，虛擬實境應用的研究越加受重視。雖然虛擬實境帶給我們相當多英語詞彙學習的機會，但同時亦面臨挑戰。其中，學習者之投入程度及雙人合作之學習，是兩個常在研究中討論的項目；不過，很少有學者同時研究這兩個因素對國小學童的影響。因此，本研究旨在探討學習者之投入程度及雙人合作之學習對國小學童於虛擬實境中英語詞彙之學習成效。本研究對象為新北市一所國小三年級的四個班級，每班皆學習二十個詞彙。四個班級分別接受四種不同的學習模式，包含三組虛擬實境實驗組如個人獨立學習組、雙人合作學習組、教師引導組與一組控制組。為了探究與比較因素間的影響，四組在學習單字前皆接受前測、學習後考立即後測、兩週後進行延宕後測，前後測結果以時間為組內因素和組別為組間因素來分析。研究發現：(一) 藉由虛擬實境學習，國小學童能有效地學習英語詞彙；(二) 學習者之投入程度越高，對詞彙學習越有助益；(三) 在學習者之投入程度相同時，個人獨立學習組與雙人合作學習組，在兩次後測成績中並無顯著差異。依此研究成果，提供教師在以英語作為第二語言之詞彙教學的一個參考：將虛擬實境軟體應用在課堂教學，有助於學習者投入程度之提升，亦提高學生詞彙學習的成效。. 關鍵字：虛擬實境、學習者投入程度、英語詞彙學習。. i.

(3) . ABSTRACT Research on vocabulary learning in second language acquisition research has been mounting steadily for a number of decades. With the advancement of computer technology, though a recent surge of research on virtual reality learning has given us new opportunities, little empirical evidence has been gathered to support it. Learner involvement and pair work cooperation were two issues that drew attention but have seldom been investigated in one single study, especially at elementary school level. Therefore, the study investigated effects of the two factors on elementary school students' vocabulary acquisition. Four intact third-grade classes from an elementary school in New Taipei City participated in the study and learned twenty target words in four different modes including three experimental VR groups and one control group. In order to investigate the effectiveness of VR mediation and examine the effects of learner involvement and pair work cooperation on English vocabulary learning, four groups were presented as below: Individual VR Group, Paired VR Group, Teacher-centered VR Group, and Control Group. The results were analyzed across students' performances on the pretest, immediate posttest, and delayed posttest with time as a within-subjects factor and group as a between-subjects factor. The findings showed that (a) elementary school students could learn English vocabulary effectively through virtual reality mediation, (b) more learner involvement loads tended to yield better vocabulary growth and retention, and (c) with equal involvement loads, there was no significant difference between individual and pair work modes on two posttests. Hence, the study concluded that teachers should consider the use of VR technology and incorporate it into classroom instruction.. Keywords: virtual reality, learner involvement, English vocabulary learning ii.

(4) . ACKNOWLEDGEMENTS First of all, my deepest and sincerest thanks go to my advisor, Dr. Wen-Ta Tseng, who never hesitates to provide me with his professional knowledge and kindness. His expertise and insightful understandings enlighten me and sharpen my perceptions of academic studies. During the past years, Dr. Tseng always offers cordial encouragement and great support in guiding me through the challenging and stressful process. Without his suggestions and help, the completion of this thesis would not have been possible. I am also grateful to my committee members, Dr. Hsi-Chin Chu and Dr. Chieh-Fang Hu. Without their careful reading, I could not have improved and enriched my thesis. Thanks to their valuable suggestions and profound comments, the content became better organized and the results could be interpreted from a more precise perspective. My appreciation also goes to my colleagues and students for their participation in the study. And the special gratitude must go to my beloved family and my best friends. Their encouragement and love give me strength to overcome the obstacles that I encountered during thesis writing. It is their tolerance and support that made me strong and faithful to face the difficulties. Therefore, I would like to share all my joy at completing this thesis with them.. iii.

(5) . TABLE OF CONTENTS CHINESE ABSTRACT ................................................................................................ i ABSTRACT ..................................................................................................................ii ACKNOWLEDGEMENTS ...................................................................................... iii TABLE OF CONTENTS ........................................................................................... iv LIST OF TABLES ...................................................................................................... vi LIST OF FIGURES ...................................................................................................vii CHAPTER ONE INTRODUCTION ......................................................................... 1 Motivation and Background of the Study .................................................................. 1 Purposes of the Study ................................................................................................. 7 Research Questions of the Study ................................................................................ 7 Significance of the Study ........................................................................................... 8 Definition of Terms .................................................................................................... 8 Organization of the Thesis ....................................................................................... 10 CHAPTER TWO LITERATURE REVIEW .......................................................... 11 The Vocabulary Knowledge..................................................................................... 11 Learner Involvement ................................................................................................ 12 Learner Autonomy in Language Teaching and Learning ..................................... 12 Involvement Load Hypothesis .............................................................................. 15 The Pair Work Cooperation ..................................................................................... 17 The Virtual Reality Language Learning................................................................... 19 The Multimedia Language Learning .................................................................... 19 The Computer-Assisted Language Learning---CALL applications ..................... 22 The Virtual Reality in Language Learning---VR applications ............................. 24 Summary of Chapter Two ........................................................................................ 27. iv.

(6) . CHAPTER THREE METHODOLOGY ................................................................. 28 Population and Participants ...................................................................................... 28 Instruments ............................................................................................................... 29 Twenty target words ............................................................................................. 30 3D virtual reality computer program .................................................................... 32 Multiple choice tests ............................................................................................. 34 The Data Collection Procedure ................................................................................ 36 Data Analysis Procedure .......................................................................................... 38 CHAPTER FOUR RESULTS AND DISCUSSIONS ............................................. 39 Results ...................................................................................................................... 39 The Tendency of Time within Four Groups ......................................................... 40 Word Gains from VR Vocabulary Learning ........................................................ 43 Learner Involvement in VR Vocabulary Learning ............................................... 46 Pair work Cooperation in VR Vocabulary Learning ............................................ 47 Discussions ............................................................................................................... 48 CHAPTER FIVE CONCLUSIONS ......................................................................... 51 Summary of Major Findings .................................................................................... 51 Pedagogical Implications ......................................................................................... 52 Limitations and Suggestions for Future Research.................................................... 53 REFENRENCES ........................................................................................................ 55 Appendix A The Lesson Plans .................................................................................. 67 Appendix B Pretest .................................................................................................... 83 Appendix C Posttest................................................................................................... 85. v.

(7) . LIST OF TABLES. Table 1: Four groups .................................................................................................... 29 Table 2: Twenty target words ....................................................................................... 31 Table 3: Results of Two-Way ANOVA ........................................................................ 39 Table 4: Descriptive Statistics of Individual VR Group .............................................. 40 Table 5: Descriptive Statistics of Paired VR Group..................................................... 40 Table 6: Descriptive Statistics of Teacher-centered VR Group ................................... 40 Table 7: Descriptive Statistics of Control Group ......................................................... 41 Table 8: Multiple Comparisons(LSD) of Four Groups within Two Time Levels ........ 45. vi.

(8) . LIST OF FIGURES. Figure 1: Procedures for Conducting the Study ........................................................... 37 Figure 2: Two Time Levels within Four Groups.......................................................... 42 Figure 3: Attrition Rate of Four groups within Two Time Levels ............................... 42. vii.

(9) . CHAPTER ONE INTRODUCTION This chapter offers an overall introduction of the present study. In the first section, the motivation and background of the study are presented. The purposes of the study will be addressed in the second section. Then, the research questions, significance of the study, and definition of terms will be specified in sequence. Finally, the organization of the thesis will be briefly demonstrated in the last section.. Motivation and Background of the Study Learners' lexical knowledge plays an essential role in second language acquisition (Schmitt, 2000). In recent years, the second language vocabulary learning is receiving more and more attention in language acquisition research (Folse, 2006; Lee & Muncie, 2006). Ellis (2003) in his influential book demonstrates the power of task-based learning and teaching in second language acquisition, whereas Nation (2001) proposes nine constructs of word knowledge and the integration of vocabulary learning and classroom instructions. The constructs involve the spoken and written forms of a word, the word parts within a word structure, the meanings, concepts and associations of a word, the grammatical functions, collocations and constraints of a word use in both receptive and productive scales. The issue of children's English vocabulary learning has been tackled over the years. Identifying the overall lexical proficiency of Nation's (2001) nine dimensions at one time may work well with adult learners (Webb, 2005); however, it will raise some concerns and become unsuitable for elementary school students. Especially in Taiwan's EFL context, not all vocabularies are required to be reached at the same. 1.

(10) . degree of mastery in elementary school level. Relying on students' cognitive development and their language proficiency, students are required to learn 300 words for recognition and 180 out of 300 words for production according to the Grade 1-9 Curriculum Guidelines in Taiwan (MOE, 2008). Thus, based upon the guidelines (MOE, 2008), the aim of children's English vocabulary learning is to foster learner-centeredness, help them learn vocabularies effectively, raise their interests, develop their positive attitudes toward learning English, and enhance their overall language proficiency. One of the biggest issues associated with learner-centeredness is the concept of learner involvement. Based on the notion of learner involvement, the research on learner autonomy and Involvement Load Hypothesis has gained its prominence in the realms of language teaching and learning (Holec, 1981; Underwood, 1984; Laufer & Hulstijn, 2001). Referring to learner autonomy, Holec (1981) proposes that learner autonomy is the ability to take charge of one's learning and to hold responsibility for decisions. In a similar vein, Little (1991) notes that learner autonomy is a capacity for detachment, critical reflection, decision-making, and independent action. Little (2002) proposes a more holistic view toward learner autonomy and indicates that the practice of learner autonomy requires insight, the positive attitude, and a readiness to be proactive in self-management and in interaction with others. Also, in the Grade 1-9 Curriculum Guidelines in Taiwan (MOE, 2008), students are promoted to develop their leaner autonomy and attain different levels of learner control over learning management, cognitive processing, and even the learning content in their language learning process (Benson, 2001). In general, students prefer being in control of their learning process and they can feel more competent and motivated when they are empowered to make their own choices (Leonard, 1968; Lepper, 1985). Yet, regarding English vocabulary instruction, it is claimed that the greater levels 2.

(11) . of processing are involved, the better word retention is achieved (Craik & Lockhart, 1972). Craik and Tulving (1975) further interpret their notion of Depth of Processing Hypothesis by stating that the durability of the trace is a positive function of the depth of processing, where greater depth is referred to greater degrees of semantic involvement. Based on the Depth of Processing Hypothesis (Craik & Lockhart, 1972; Craik & Tulving, 1975; Craik, 2002), Laufer and Hulstijn (2001) polish the notion and further propose the Involvement Load Hypothesis, a framework of the three components in their task-induced involvement: need, search, and evaluation. The involvement load of a task is measured by the combination of the presence or absence of the three involvement factors. According to the Involvement Load Hypothesis (Laufer & Hulstijn, 2001), the higher levels of involvement loads will facilitate learners' vocabulary learning. In addition to learner involvement, the notion of pair work cooperation has also been widely addressed in ESL/EFL learning field (Webb, 1982; Johnson & Johnson, 1989, 1990; Nunan, 1992; Liang 2002; Zurita & Nussbaum, 2004a). The cooperative learning is a system of teaching and learning techniques in which students are active agents in the process of learning instead of passive receivers of the product of any given knowledge (Liang, 2002). The characteristics of pair work cooperation involve the positive interdependence, face-to-face interaction, individual and group accountability, quality group processing and explicit teaching of collaborative skills (Johnson & Johnson, 1989, 1990, 1994, 1999; Liang, 2002). In contrast to the individual learning which requires students to work to achieve the learning goals independently without help from others, the cooperative learning emphasizes the importance of co-work to reach the common goals (Johnson & Johnson, 1989, 1993). Furthermore, with the advancement of computer technology, recent years have seen increased attention being given to the computer-assisted cooperation in language 3.

(12) . learning field. And this new direction of research emphasizes on the contribution that can be possibly made by pair work cooperative learning. In the computer-assisted cooperative learning environment, the pair-work learners can search, evaluate and even discuss the learning materials and work out the assigned task together (Ellis et al., 1991; Zurita & Nussbaum, 2004a). Besides, via the cooperative learning platform, learners can get more involved in learning by themselves and cooperate with their partners according to their own pace and individual differences (Liu et al., 2003; Cortez et al., 2004; Zurita & Nussbaum, 2004a, 2004b). Moreover, Penuel (2006) notes that the technology can enhance learners' learning outcomes, bridge the gap from digital divide, help learners get used to the future ubiquitous learning environment. In sum, those technological devices can not only facilitate pair and group interactions but also enhance learners' language learning (Stahl et al., 2006; Johnson, 2007, 2008a, 2008b; Liu & Kao, 2007). Although the issue of pair work cooperation plays a vital role in vocabulary acquisition, more studies are needed to be done to examine whether the pair work cooperation will facilitate third graders' vocabulary learning. Another issue that needed attention is how the multimedia materials can be incorporated in second language vocabulary learning (Secules et al., 1992; Tsai, 2004; Chang, 2006). On the one hand, most of the current multimedia software can provide textual, graphical and auditory input such as audio and pictorial flashcards, visual animations or even the 3D virtual reality (Wang, 2003; Lin, 2009). As supported by the Dual Coding Theory (Paivio, 1986), if information is coded in both verbal and nonverbal systems, learners can retain and recall it more easily. Besides, with great help of visual and contextual clues, students can construct their linguistic knowledge gradually and systematically (Ausubel, 1968; Hadley, 1993). Further, based upon Mayer's (1997, 2001, 2002) Generative Theory of Multimedia Learning, learning is 4.

(13) . more likely to occur when learners have access to multiple presentation modes of information. With such varied learning materials, learners can be engaged with verbal and visual information, organize the two representations respectively, and finally integrate these two into a new model to foster their language acquisition (Mayer, 1997, 2001, 2002). Hence, as shown in previous studies, the benefits of multimedia software have been elucidated and testified theoretically and empirically (Jones & Plass, 2002; Oxford, 2003; Healey, 2007; Guichon & McLornan, 2008; Lin, 2009). On the other hand, the advances in educational applications of computer technology also provide the rapid growing resource for language classrooms (Brown, 2001). The recent decade has seen a number of projects incorporating computer-enhanced interactive video (Gardner & Garcia, 1996), concordancing (Aston, 1997), computer simulations (Mak, 1994; Ho & Crookall, 1995; Murray, 1999), and e-mail tandem learning (Warschauer et al., 1996). Computer-assisted language learning (CALL) can benefit second language acquisition in the following perspectives (Warschauer & Healey, 1998; Benson, 2001; Schwienhorst, 2002; Egbert et al., 2007). First, the incorporation of computer-based language teaching can offer multimodal practice with feedback. Second, it can not only assign pairs and small-groups to work on projects collaboratively but also promote individualization in a large class. Third, due to the fun factor, the computer component can provide various resources for different learning styles. Fourth, the exploratory learning with large amounts of data is suggested as well. Fifth, the real-life skills can be simulated and built through multimedia-assisted language program. Based on the concept of multimedia and real-life simulations, one of the popular CALL applications in language learning is the virtual reality (VR) innovation. The research and application of VR in language instruction date back to approximately a decade ago (Garcia-Ruiz et al., 2007). As proposed by Moore et al., (2005), a virtual 5.

(14) . environment (VE) is defined as a computer-generated three-dimensional (3D) simulation of a real or imaginary environment. Research topics on VR in educational applications involve multimedia learning, visual cues, learner involvement, etc. (Singhal & Zyda, 1999; Chee, 2001; Jung, 2002; Schwienhorst, 2002; Yeung, 2004; Garcia-Ruiz et al., 2007). In light of multimedia learning, the VR program can provide multimedia learning materials involving textual, pictorial, audio and visual aids to foster learners' ability to visualize, understand and construct their own knowledge (Schwienhorst, 2002; Yeung, 2004; Lin, 2011). Besides, the visual cues in VR technology can simulate the real-life interactions, promote more authentic practices, create new identities for students to lower their anxiety, raise learners' interests in vocabulary acquisition, and further enhance their language proficiency via bountiful communications (Milton & Garbi, 2000; Jung, 2002; Seymour, 2002). In terms of learner involvement, within the bountiful theme-based VR scenarios, learners are not only empowered to be in control of their learning, but immersed in a contextualized language learning environment (Singhal & Zyda, 1999; Chee, 2001; Jung, 2002; Garcia-Ruiz et al., 2007; Chang & Ho, 2009). With those multimedia, hypermedia and interactive technologies, learners will have more control over the selection of materials and strategies of interpretation (Blin, 2005). A variety of CALL and VR applications and artifacts have the potential of establishing an optimal language learning environment, fostering the development of learner involvement and facilitating language acquisition (Blin, 2005; Egbert et al., 2007). Due to the pedagogical potential, the implementation of multimedia, CALL and VR materials in second language learning has gathered great importance in recent years. However, further empirical explorations are still needed. Besides, few studies examine the effects of learner involvement and pair work cooperation on VR-mediated English vocabulary instruction at elementary school level. In response to the previous 6.

(15) . inconclusive research, the present study further explores the effects of virtual reality mediation on Taiwanese elementary school students' vocabulary learning. It is expected that the results of the present study can clarify and offer insights for further pedagogical implications.. Purposes of the Study The purpose of this study is to investigate the effectiveness of virtual reality mediation on English vocabulary learning. A four-group experiment is conducted. To examine whether a stronger involvement (i.e. the loads of need, search, and evaluation) can facilitate students' vocabulary learning, four intact third-grade classes are randomly assigned: three experimental groups through VR mediation and one control group in conventional classroom teaching. In order to investigate the effectiveness of VR mediation and examine the effects of learner involvement and pair work cooperation on English vocabulary learning, four groups are presented as below: 1) Individual VR Group, 2) Paired VR Group, 3) Teacher-centered VR Group, and 4) Control Group. To explore students' learning process, the pictorial, multiple choice test format is adopted in immediate posttest and delayed posttest in a two-week interval. It is expected that the findings of this present study can shed light on English vocabulary instructions for elementary school students in Taiwan.. Research Questions of the Study Based upon the goals of the present study, three research questions are raised to investigate the effects of learner involvement and pair work cooperation on VR vocabulary learning. Research Question 1: Does virtual reality (VR) instruction effectively facilitate learners' vocabulary acquisition? 7.

(16) . Research Question 2: Does different degrees of learner involvement influence learners' vocabulary acquisition through VR mediation? Research Question 3: Does paired mode outperform the individual mode on learners' vocabulary acquisition through VR mediation?. Significance of the Study The significance of this study is two folds. First of all, as noted by Paivio (1986), if information is coded in both verbal and nonverbal systems, learners retain and recall it more easily. By comparing and contrasting three modes of VR condition, the findings attempt to provide English teachers a deeper insight into vocabulary learning and teaching. In addition, by examining and students' learning outcomes of innovative VR vocabulary instruction, teachers may have a better understanding about learners' performance and thoughts toward VR leaning. Hence, English teachers will understand how and when to integrate various multimedia resources into effective and efficient vocabulary instruction. Secondly, based on Laufer and Hulstijn's Involvement Load Hypothesis (2001), by comparing different degrees of learner involvement in VR vocabulary acquisition and investigating the relationship between students' involvement and pair work cooperation, it can provide teachers a more comprehensive view of effective learner engagement and efficient cooperative mode.. Definition of Terms For the purpose of interpretation of this study, the definitions of key words are provided as a frame of reference. 1. Virtual reality (VR) Virtual reality is a kind of computer technology which can simulate environments 8.

(17) . where users can get a sense of immersion and interact with the input devices, such as a mouse or sensor gloves, and the output devices, like a screen or head-mounted displayer (Y. C. Lin, 2010). 2. learner involvement According to the Involvement Load Hypothesis (Laufer & Hulstijn, 2001), there are three major components in their task-induced involvement: need, search, and evaluation. Need is the learners’ drive to accomplish a particular task, and the degrees of need such as “moderate” or “strong” are determined by whether it is externally imposed or self-imposed. Search shows an attempt to find the required information for a specific word. Evaluation is the comparison of different words and their meanings to decide the appropriateness of a given word in particular contexts. And the distinctions between “moderate” and “strong” evaluations are illustrated by the complexity of cognitive processing. In terms of different degrees of the three components, three modes of VR English vocabulary instruction are designed as below. (a) student-centered condition: Without teacher's guidance, it is the mode that learners have a strong need to accomplish a particular task by themselves. And then they have to search the required information for a specific word. Finally, learners have to strongly evaluate and decide the appropriate meaning of a given word in virtual reality contexts. In this study, Individual VR Group and Paired VR Group are two modes of student-centered condition. (b) teacher-centered condition: Under teacher's guidance, learners are guided to accomplish a particular task. Based on this externally-imposed instruction instead of self-imposed learning, learners only have a moderate need to accomplish the assigned task. Besides, under 9.

(18) . teacher's guidance, they don't have to search or evaluate the required information for a specific word. In this study, Teacher-centered VR Group is synonymous with the teacher-centered condition. 3. pair work cooperation As noted by Liang (2002), the cooperative learning is a system of teaching and learning techniques in which students are active agents in the process of learning instead of passive receivers of the product of any given knowledge. In contrast to the individualistic learning which requires students to work to achieve the learning goals independently without help from others, the cooperative learning emphasizes the importance of co-work to reach the common goals (Johnson & Johnson, 1989, 1993). To explore the roles of pair work cooperation within the student-centered condition, the subjects in the Individual VR Group are required to work individually whereas Paired VR Group need to co-work collectively.. Organization of the Thesis This study is composed of five chapters. The first chapter is a general introduction of the current study. The second chapter is a review of the literature of vocabulary knowledge, learner involvement, pair work cooperation, multimedia incorporation as CALL and VR applications in second language vocabulary learning research. This is followed by the detailed method descriptions and procedures for data collection on learners' vocabulary learning outcomes in the third chapter. The results and discussion are presented in Chapter Four, followed by conclusions in Chapter Five.. 10.

(19) . CHAPTER TWO LITERATURE REVIEW This chapter offers a review of literature related to virtual reality English vocabulary learning and teaching. First, the issue of vocabulary knowledge will be presented. The second section introduces the idea of learner involvement. The pair work cooperation will be discussed in the third section. Lastly, the notion of VR language learning will be addressed through three sub-dimensions involving multimedia learning, computer-assisted language learning, and virtual reality applications.. The Vocabulary Knowledge Vocabulary knowledge has been regarded as the core component in second language acquisition and was positively correlated with one's linguistic and academic performance (Laufer & Nation, 1995; Morris & Cobb, 2004). The importance of vocabulary knowledge has been noted by Candy (1988), suggesting that the study of vocabulary need to be placed at the heart of language teaching. Vocabulary knowledge comprised two aspects: vocabulary quality (depth) and vocabulary size (breadth). Vocabulary quality meant how well a word was acquired involving pronunciation, orthography, morphology, syntactic patterning, meaning, collocations, register, and frequency of the word in the target language (Nation, 1990; Qian, 1999). Vocabulary size was the number of words known by an individual learner (Qian & Schedl, 2004). Besides, concerning the assessments of vocabulary knowledge, some researchers (Laufer et al., 2004; Qian & Schedl, 2004) noted the importance of measuring. 11.

(20) . learners’ vocabulary size and strength. Other researchers (Nation, 1990; Qian, 1999; Schoonen & Verhallen, 2008) emphasized on the essentiality of how to assess word knowledge. According to the Grade 1-9 Curriculum Guidelines in Taiwan (MOE, 2008), not all vocabularies were required to be reached at the same degree of mastery at elementary school level. Considering students' cognitive development and their language proficiency level, students were required to learn 300 words for recognition and 180 out of 300 words for production in total throughout the six years (MOE, 2008). Besides, as supported by empirical evidence, the sixth graders had a receptive vocabulary size of about 220 words (Tsao, 2009), and the fourth graders had a receptive vocabulary size of 130 words (Chua, 2007). That is, especially for these young learners with limited cognitive and linguistic knowledge, vocabulary acquisition process needed to start from word recognition and then moved forward to word production and further integrated these lexical items into oral and written communications. Furthermore, for beginners, it was recommended to access learners' lexical knowledge on the written form and meaning association (Schmitt, 2010). In comparison with the written vocabulary assessment, multiple choice vocabulary test tended to have less performance variation and less potential confounding factors with beginners. Therefore, multiple choice lexical assessment appeared to be more suitable than written vocabulary test for elementary school students. To assess the third graders' vocabulary learning outcomes, the multiple choice vocabulary knowledge test was adopted in this present study.. Learner Involvement Learner Autonomy in Language Teaching and Learning Based on the concept of learner involvement, the research on learner control, 12.

(21) . learner autonomy and Involvement Load Hypothesis has gained its prominence in the realms of language teaching and learning (Holec, 1981; Laufer & Hulstijn, 2001). Learner control, originally referred to the opportunities for students to sequence their learning objectives within a particular course of instruction (Mayer & Clark, 1963), could be classified as content control, control of pace, display (strategy) control, and control of internal processing (Merrill, 1984). Based on Merrill's framework (1984), Benson (2001) specified the concept of control in three levels involving control over learning management, control over cognitive processes, and control over learning content. That is, through the three processes, learner autonomy could be developed by beginning with managing the planning, organization, and evaluation of learning at the first level; moving forward directing learners' attention to linguistic input, reflecting upon their learning, and building metacognitive knowledge in the second stage; finally, being given the freedom to decide their goals and learning resources at the ultimate level. Definitions of learner autonomy were presented according to its multidimensional features (Sinclair et al., 2000). Holec (1981) proposed that learner autonomy was the ability to take charge of one's learning and to hold responsibility for decisions including determining the objectives, defining the contents and progressions, selecting methods and techniques to be used, monitoring the procedure of acquisition, and evaluating what has been acquired. In a similar vein, Little (1991) noted that learner autonomy was a capacity for detachment, critical reflection, decision-making, and independent action. In addition, it presupposed and entailed that learner would develop a particular kind of psychological relation to the process and content of his or her learning. The capacity for autonomy could be observed not only in the way learner learned but also in the way learner transferred to wider contexts. (Little, 1991). In Littlewood's model (1996, 1997), autonomy was composed of two elements: one 13.

(22) . was the capacity depending on the level of learners' knowledge and skills about how to carry out the decisions and the other was willingness relying on the level of motivation and confidence. Further, Little (2002) proposed a more holistic view toward learner autonomy and indicated that the practice of learner autonomy required insight, the positive attitude, and a readiness to be proactive in self-management and in interaction with others. And students preferred being in control of their learning process and they could feel more competent and motivated when they were empowering to make their own choices (Leonard, 1968; Lepper, 1985; Cotterall, 1995). In sum, the significance of learner control and learner autonomy has been elaborated and emphasized in promoting effective language teaching (Underwood, 1984; Benson, 2001; Hsu & Xu, 2007). Further, as mentioned by Schwienhorst (2002, 2008), learner autonomy and computer-assisted language learning technology were in fact an ideal combination for language learning. The computer-assisted environment filled with interactions, resources and individualizations could raise students' linguistic awareness, support interaction and collaboration with peers and increase learner involvement in an experimental, learner-centered learning environment (Benson, 2001; Schwienhorst, 2002, 2008). In terms of technical aspect, lots of multimedia software could provide bountiful textual, graphical and auditory input for language learners. Psychological aspects stressed learner characteristics involving motivation, independent style and self-awareness. The socio-cultural perspective was characterized by Vygotsky's (1986) social constructivism and the importance of interpersonal communication. Finally, a political-critical perspective looked into the power shift into learners. To summarize, with the advancement of computer technology, the four dimensions---technical, psychological, socio-cultural and political-critical---were associated with developing and enhancing learner autonomy and learn involvement. (Oxford, 2003; Healey, 14.

(23) . 2007). Therefore, it was claimed that both learner autonomy and their language proficiency level could be elevated with the help of computer-assisted language learning technologies .. Involvement Load Hypothesis Over the past decades, the issues of learner involvement and deep word knowledge has been widely discussed on how to foster learners' vocabulary acquisition (Anderson & Freebody, 1981; Miller & Fellbaum, 1991; Schoonen & Verhallen, 2008). According to Craik and Lockhart (1972), the greater levels of processing were involved, the better word retention was achieved. The basic notion of the Depth of Processing Hypothesis proposed by Craik and Lockhart (1972) was that memory trace could be regarded as a byproduct of perceptual analysis and trace persistence was a positive function of the depth to which the stimulus has been analyzed. They mentioned that trace persistence was a function of depth of analysis. And deeper levels of analysis were associated with more elaborate, longer lasting and stronger traces. Craik and Tulving (1975) further explained that the durability of the trace was a positive function of the depth of processing, whereas greater depth was referred to greater degrees of semantic involvement. Based on the Depth of Processing Hypothesis (Craik & Lockhart, 1972; Craik & Tulving, 1975; Craik, 2002), Laufer and Hulstijn (2001) polished the notion and further proposed the Involvement Load Hypothesis. The involvement load, consisting of three components---need, search and evaluation, was defined as the motivational-cognitive construct which could explain and predict learners' success in the retention of hitherto unfamiliar words (Laufer & Hulstijn, 2001). Need was the learners’ drive to accomplish a particular task and the degrees of need such as “moderate” or “strong” were determined by whether it was externally imposed or 15.

(24) . self-imposed. Search showed an attempt to find the required information for a specific word. Evaluation was the comparison of different words and their meanings to decide the appropriateness of a given word in particular contexts. And the distinctions between “moderate” and “strong” evaluations were illustrated by the complexity of cognitive processing. The involvement load of a task was measured by the combination of the presence or absence of the three involvement components. If the three components were present, then the involvement load of a task was higher than a task which had only one or two components present. Laufer and Hulstijn’s Involvement Load Hypothesis (2001) provided a good direction to examine how the task and material could be designed to facilitate word learning. They claimed that the higher levels of involvement were engaged, the better word retention was achieved. Likewise, Schmitt proposed another similar term---engagement---to show that the more L2 learners were engaged in learning, the more vocabularies they would acquire (Schmitt, 2008). Also, the concept of involvement load was further adopted by Keating (2008) to see whether the task effectiveness and word learning followed the scheme of Involvement Load Hypothesis. Although previous studies indicated the significant role of “involvement” (Laufer & Hulstijn, 2001; Lee, 2003; Keating, 2008), further empirical explorations were still needed. First, the time factor in this hypothesis still remained unjustified. That is, the vocabulary learning effects might be due to the different levels of engagement rather than the potential effects of different types of tasks. Second, the three components of need, search, and evaluation haven’t yet been proved or exactly weighed individually. Third, learners' age factor and different levels of cognitive and linguistic development have not yet been fully considered and addressed. Thus, in this study, based upon the different combinations of the three involvement components, two types of experiments with different levels of involvement loads were designated in two modes. 16.

(25) . One was the student-centered group with higher involvement loads, and the other was the teacher-centered group with lower involvement loads. And the study was aimed to examine whether higher level of learner involvement would facilitate children's vocabulary learning.. The Pair Work Cooperation The idea of pair work cooperation has also been widely addressed in ESL/EFL learning field (Webb, 1982; Johnson & Johnson, 1989, 1990; Nunan, 1992; Liang 2002; Zurita & Nussbaum, 2004a). As noted by Liang (2002), the cooperative learning was a system of teaching and learning techniques in which students were active agents in the process of learning instead of passive receivers of the product of any given knowledge. In contrast to individualistic learning that students worked by themselves to finish learning goals, the cooperative learning emphasized that students worked together to maximize their own and each other's learning. In addition, to reach their common goals, learners in cooperative learning could lead to higher achievement, greater psychological health and self-esteem than competitive and individualistic learning (Johnson & Johnson, 1989, 1993). In pair work cooperation, the characteristics of cooperative learning were presented as follows (Johnson & Johnson, 1989, 1990, 1994, 1999; Liang, 2002). First of all, the positive interdependence was shown when learners had the common goal to work out the project together. Second, through the face-to-face interaction, learners could help and interact with each other to increase their participation. Third, referring to the individual and group accountability, learners not only took charge of their own parts but fulfilled their group task together. Besides, in the pair work, high-achievers could cooperate with low-achievers and help them learn (Slavin, 1978). Fourth, during their group processing, learners could constantly reflect, clarify, 17.

(26) . evaluate, and improve their own works together to elevate their task efficacy. Fifth, the explicit teaching of collaborative skills would have a positive effect on reducing learning anxiety, enhancing mutual trust, and promoting supportive atmosphere to facilitate the accomplishment of pair or group work (Johnson & Johnson, 1989, 1990, 1994, 1999; Liang, 2002). Moreover, the benefits from pair work cooperation could also be explained through the idea of Zone of Proximal Development (Vygotsky, 1978). This term was defined as the distance between the actual developmental level and the level of potential development under adult guidance or in collaboration with more capable peers. This social-interactive notion not only stressed interpersonal relations but also indicated the importance of other people for facilitating self-awareness, reflection and paired cooperation. Furthermore, with the advancement of computer technology, recent years have seen increased attention being given to the computer-assisted cooperation in language learning field. Shifting from seeing the computer as a tool or partner to seeing it as a virtual environment, the research area of computer-assisted language learning has been conducted to explore how learners could collaborate and interact with others (Schwienhorst, 2002). In the computer-assisted cooperative learning environment, the pair-work learners could search, evaluate and even discuss the learning materials and work out the assigned task together (Ellis et al., 1991; Zurita & Nussbaum, 2004a). And the support from their peers could further possess learners the ability to plan, monitor, and examine their own learning process (Schwienhorst, 2002). Besides, via the cooperative learning platform, learners were not merely forced to take responsibility of their pair work, but encouraged to communicate, collaborate with their partners, and participate in the learning process (Schwienhorst, 2002; Liu et al., 2003; Cortez et al., 2004; Zurita & Nussbaum, 2004a, 2004b). In addition, as Penuel (2006) stated, the 18.

(27) . technology could enhance learners' learning outcomes, bridge the gap from digital divide, help learners get accustomed to the future U-learning environment. To summarize, as shown by Liu & Kao (2007), those technological devices could not only facilitate pair and group interactions but also enhance learners' language learning (Stahl et al., 2006; Johnson, 2007, 2008a, 2008b). The issue of pair work cooperation played a vital role in vocabulary acquisition. Hence, more studies needed to be conducted to examine whether the pair work cooperation would foster third graders' vocabulary learning through virtual reality mediation.. The Virtual Reality Language Learning The Multimedia Language Learning Another intriguing topic prevailing throughout the past decades was how the multimedia materials could be incorporated in second language vocabulary learning (Secules et al., 1992; Tsai, 2004; Chang, 2006). According to Mayer and Sims (1994), multimedia learning was referred to the learning in which two or more learning modalities were involved to help learners construct their knowledge. The issue of multimedia learning could be addressed from two dimensions. One focused on the visual and multimedia aid incorporation and the other stressed the computer-assisted language leaning. Technically speaking, most of the current multimedia software could provide learners textual, graphical and auditory input such as audio and pictorial flashcards, visual animations or even the 3D virtual reality in enhancing their vocabulary acquisition. (Wang, 2003; Lin, 2009). According to the Dual Coding Theory (DCT) (Paivio, 1986), there were two internal representations of events. Verbal system units were called logogens; these units contained information that underlied our use of the word. Non-Verbal system units were called imagens. Imagens contained information that generated mental images such as natural objects, holistic 19.

(28) . parts of objects, and natural grouping of objects. Imagens operated synchronously or in parallel; thus all parts of an image were available at once. Logogens operated sequentially, words came one at a time in a syntactically appropriate sequence in a sentence. Once the verbal and image systems were connected and related, one could not only form the mental image but describe it in words. Paivio (1986) proposed that if information was coded in both verbal and nonverbal systems, learners could retain and recall it more easily. Previous studies indicated that when both verbal and visual materials were shown, learners with low-spatial ability were able to construct referential connections between these two forms of mental representations, and therefore learned more effectively by making use of diverse cognitive resources (Mayer & Sims, 1994; Mayer, 1997; Mayer, 2001). In addition, as mentioned by Sadoski (2005), kindergartners presented with words in together either with pictures or pronunciations outperformed those presented with words alone. Thus, regarding the vocabulary learning, DCT proposed that combining verbal processing and nonverbal processing might effectively contribute to vocabulary acquisition. As Mayer and Sims (1994) proposed, the dual-coding theory of multimedia learning comprised three main processes---verbal encoding, visual encoding, and referential connections between the verbal and visual representations. The theory suggested that when verbal and visual materials were presented simultaneously, students might learn better by building referential connections between the verbal and visual representations in mind. Dubois and Vial (2000) indicated that the simultaneous presentation of visual, textual, and auditory information ensured better learning when the different sources of information were well interrelated. That is, learning within such interrelated contextualized situation would benefit learners in acquiring vocabularies (Folse, 2006; Walters, 2006; Webb, 2008). Even for the learners with relatively lower proficiency, they could still benefit from this visual-based multimedia 20.

(29) . learning (Chang, 2006). Besides, referring to the high-imagery concrete vocabulary learning, it was found that foreign words associated with images or actual objects were learned more easily than those without such additional information (Kellogg & Howe, 1971). According to Terrell's (1986) binding concept, combining an unknown L2 word with a visual representation could not merely bypass a direct translation, but also foster vocabulary learning. It was shown that the binding of form (unknown L2 vocabulary) to meaning (visual representations) was the most effective way for learners to acquire concrete ideas and references. Moreover, based upon Mayer's (1997, 2001, 2002) Generative Theory of Multimedia Learning, learning was more likely to occur when learners had access to multiple presentation modes of information. The theory assumed that learners of a second/foreign language had two separate verbal systems (L1 and L2) and a common imagery system. Translations of words via simultaneous verbal and visual presentation could not only link the two verbal systems, but this storage in the second verbal system would also have an additive effect on learning (Paivio & Desrochers, 1980). With such varied learning materials, learners could be engaged with verbal and visual information, organize the two representations respectively, and finally integrate these two into a new model to foster their language acquisition (Mayer, 1997, 2001, 2002). As shown in previous research, the benefits of multimedia software in facilitating vocabulary instruction have been elucidated and testified theoretically and empirically (Jones & Plass, 2002; Oxford, 2003; Healey, 2007; Guichon & McLornan, 2008; Y. C. Lin, 2009). To sum up, the advantages of application of multimedia in language learning could be illustrated as below (Hoogeveen, 1995; Beatty, 2005; Hong, 2010). The first focused on the promotion of learners' interest in multimedia learning. Second, multimedia could foster autonomous language learning. Third, multimedia could transmit information quickly and effectively to all learners and 21.

(30) . raised their motivation in learning. Fourth, the interactions between multimedia aids and learners could help learners experience the information instead of simply receive it. Fifth, the multimedia-based learning could facilitate learners to develop both technical and research skills. Therefore, due to this convenient access of meanings or various visual and interactive advantages in multimedia environment, instructors could facilitate learners' language acquisition by utilizing this innovative technology (Hong, 2010).. The Computer-Assisted Language Learning---CALL applications The advances in educational applications of computer technology also provided the rapid growing resource for language classrooms (Brown, 2001). And the ways of teaching have been changed a lot with the innovation of technology (Dexter et al., 1999). The development of CALL applications could be divided into three stages: behavioristic, communicative and integrative (Warschauer & Healey, 1998), or restricted, open and integrated classifications in Bax' framework (2003). As summarized by Lin (2010), in the earliest stage, the applications mainly focused on drilling and testing learners' knowledge of grammatical structures through multiple choice tests or matching. This period was characterized by the assumption of computer as tutor and beliefs in habit formation (Fox, 1994). The second stage in the 1980s witnessed changes of CALL applications from drills and practices to more problem-solving activities, including text reconstruction, games, and simulation. The third stage of integrative CALL was characterized by multimedia, hypermedia and interactive technologies. According to the integrative stage nowadays, the role of computers has shifted from a pure tool for drills to an user-friendly environment full of learning interactions (Schwienhorst, 2002). A number of CALL research incorporating computer-enhanced 22.

(31) . interactive video (Gardner & Garcia, 1996), concordancing (Aston, 1997), computer simulations (Mak, 1994; Ho & Crookall, 1995; Murray, 1999), and e-mail tandem learning have been discussed in second language learning and teaching (Warschauer et al., 1996). The benefits of CALL were presented in the following perspectives (Warschauer & Healey, 1998; Benson, 2001; Schwienhorst, 2002; Egbert et al., 2007). To begin with, the incorporation of computer-based language teaching could offer multimodal practice with feedback. Next, it could not only assign pairs and small-groups to work on projects collaboratively but also promote individualization in a large class. In addition, due to the fun factor, the computer component could provide variety in the resources for different learning styles. Besides, the exploratory learning with large amounts of data was suggested as well. Furthermore, the real-life skills could be simulated and built through multimedia-assisted language program. That is to say, the extensive application of CALL software or the Internet was largely based upon the assumption that technologies had the potential of establishing an optimal language learning environment, in which learners could negotiate with meanings, interact with the real audience in the target language, be engaged in authentic task, be exposed to varied and creative language, have enough time and appropriate feedback, be guided attentively to their learning process, learn in an environment of suitable anxiety level, and lastly develop learner autonomy (Egbert et al., 2007). Especially in facilitating learner autonomy and learner involvement, first of all, CALL applications might foster learners' control over pace of learning and mode. Next, ways of text construction, games and simulations might help learners control over path taken. Besides, CALL artifacts involving word processors, desktop publishing packages, concordancers and databases might be beneficial to enhance learner autonomy via control over text creation and interpretation, process of learning, development of metacognitive skills and metalinguistic awareness. Moreover, 23.

(32) . multimedia, hypermedia and interactive technologies promoting integration of skills might provide learners more control over the selection of materials and strategies of interpretation. Finally, through the Internet inclusive of e-mail, on-line discussion, web authoring, large collaborative projects, learners might have better control over access, learning content and interaction (Warschauer & Healey, 1998; Benson, 2001; Blin, 2005). Hence, a variety of CALL applications and artifacts had the potential of establishing an optimal language learning environment and even fostering the development of learner involvement (Blin, 2005; Egbert et al., 2007).. The Virtual Reality in Language Learning---VR applications Based on the idea of multimedia and real-life simulations, the research area of CALL has recently shifted from seeing the computer as a tool or partner to seeing it as a virtual environment (Schwienhorst, 2002). Virtual reality (VR), extensively researched in the 1980s, was not used in educational settings until early 1990s. Yet, the research and application of VR in language instruction dated back to approximately a decade ago (Garcia-Ruiz et al., 2007). According to Pantelidis (1993), VR was a highly interactive, computer-based, multimedia environment in which the user became a participant with the computer in a virtually real world. As proposed by Moore et al. (2005), a virtual environment (VE) was defined as a computer-generated three-dimensional simulation of a real or imaginary environment. With the rapid growth of computer application and advances in developing 3D technology, VR has become more user-friendly and affordable in price in recent years (Yeung, 2004). And the VR technology could be integrated into language learning and teaching (Schwienhorst, 2002; Lin, 2010; Lin, 2011). To begin with, in light of multimedia learning, the VR program could offer multimedia learning materials involving textual, pictorial, audio and visual aids to 24.

(33) . foster learners' ability to visualize, understand and construct their own knowledge (Yeung, 2004; Lin, 2010; Lin, 2011). In Lin's (2010) study, the collaborative virtual environment was integrated into vocabulary learning and teaching for elementary students. Through the multi-user interactive network learning platform, five high-grade participants in eastern Taiwan were recruited to learn three sets of target words. Based on learners' performance on word recognition and spelling throughout tests, interviews, instructor's observation logs, learners' reflection logs, questionnaires and attitude scales, positive results were presented as follows. Through the multimedia learning materials, participants obtained at least 96% retention rate of vocabulary identification and made progress in word spelling thirty days after the training ended. Besides, all the participants showed positive attitude toward collaborative VR English learning. In addition, the visual cues in VR technology could simulate the real-life interactions, promote more authentic practices, create new identities for students to lower their anxiety, raise learners' interests in vocabulary acquisition, and further enhance their language proficiency via bountiful communications without being constrained by time or the other restrictions (Milton & Garbi, 2000; Jung, 2002; Seymour, 2002, Chang & Ho, 2009). In a study that investigated the effects of learner control of one Web-based learning program on their comprehension, Chang and Ho (2009) showed that learners with flexibility to choose the learning sequences and to have access to external links have better learning achievement and self-efficacy than their counterparts who were limited by restricted learning content. Furthermore, regarding the different functions for users, single-user virtual environment and multi-user collaborative virtual environment platforms were designed for dissimilar purposes respectively (Moore et al., 2005; You, 2007; Schwienhorst, 2008). A study by Schwienhorst (2008) explored how a virtual 25.

(34) . environment, Multi-user Object Oriented systems (MOO), could support learner autonomy and enhance learners' language proficiency. Results indicated that most learners preferred MOO over traditional classrooms, as learners could have immediate feedback and enjoy much more engagement in the process. Moreover, synchronous MOO tandem partnership was found to compel learners to assume responsibility for their learning and allowed them to interact and collaborate. Lastly, in terms of learner involvement, within the bountiful theme-based VR scenarios, learners were not only empowered to be in control of their learning according to their individual differences, but immersed in a meaningful, contextualized and first-person language learning environment (Esch, 1996; Singhal & Zyda, 1999; Milton & Garbi, 2000; Chee, 2001; Jung, 2002; Garcia-Ruiz et al., 2007; Chang & Ho, 2009). As proposed by Milton and Garbi (2000), they developed a collaborative virtual environment program for elementary students to learn foreign languages in various 3D settings including 3D towns, a zoo, and a shopping center. Via the platform, learners could meet and interact with a pre-programmed robot. And throughout the practices in 3D activities, learners could use different linguistic items in various VR contexts. This study showed that VR environment could provide more authentic practice opportunities to facilitate language acquisition. In sum, VR technologies could not only empower and immerse students in a more learner-centered environment, but also lead students to have a higher level of cognitive engagement than traditional classroom learning (Jung, 2002; Schwienhorst, 2002; Blin, 2005; Garcia-Ruiz et al., 2007). Next, without being constrained by time or other restrictions, the 3D VR environment allowed for greater self-awareness and encouraged learners to experiment with different roles through the use of virtual representations (Milton & Garbi, 2000; Schwienhorst, 2002; Seymour, 2002). Besides, within diverse scenarios, individual learners in VR mediation were encouraged and 26.

(35) . enabled to actively participate in the creation and organization of their own learning. Furthermore, via the interactive platform, VR could also enhance group or pair work by allowing for learner interaction and collaboration in a contextualized language learning environment (Jung, 2002; Schwienhorst, 2002; Garcia-Ruiz et al., 2007). Though implementation of technology in second language learning has gathered great importance in recent years (Warschauer & Healey, 1998; Milton & Garbi, 2000; Jung, 2002; Schwienhorst, 2002; Seymour, 2002; Moore et al., 2005; Garcia-Ruiz et al., 2007; You, 2007; Schwienhorst, 2008), more studies were needed to be conducted to explore the effects of learner involvement and pair work cooperation on VR English vocabulary learning at elementary school level. In response to the previous inconclusive research, the present study further explored the effects of virtual reality mediation on Taiwanese elementary school students' English vocabulary learning. It is expected that the results of the present study can clarify some theoretical issues and offer some insights for further pedagogical implications.. Summary of Chapter Two In this chapter, the issues related to learner involvement and pair work cooperation in virtual reality English vocabulary learning have been reviewed and discussed. Although it was assumed that VR technology and both two factors were influential for language learning and teaching, there was little research examining the relationship of learner involvement and pair work cooperation in VR vocabulary learning. Therefore, more works on these topics were desired.. 27.

(36) . CHAPTER THREE METHODOLOGY This chapter comprises four sections. The first section delineates the background information and distribution of the participants in this study. The instruments used in the study are introduced in the second section. The third section briefly describes the data collection procedures of the study. Finally, the techniques employed to analyze the collected data are presented in the fourth section.. Population and Participants The present study was to investigate the effectiveness of virtual reality mediation on English vocabulary learning in Taiwan's EFL context. The participants of the research were recruited from one elementary school in New Taipei City. Four intact third-grade classes at this school were randomly selected as the participants. Originally, two classes consisted of 26 students and two classes consisted of 25 students. However, 6 special cases were further removed from statistical calculation because 3 participants were from special education and 3 participants were absent in one or two sessions of the study. Hence, after the removal, each class was composed of 24 students with half males and half females and there were 96 third-grade elementary school students in total in the present study. With the same mother tongue in Mandarin Chinese, the participants shared similar background in education. Besides, students with different achievement levels, or dissimilar socio-economic status have been equally arranged in each class. Therefore, each class was basically the same. A four-group experiment was conducted as follows (Appendix A---lesson plans).. 28.

(37) . To examine whether the more learner involvement (i.e. the loads of need, search, and evaluation) would facilitate students' vocabulary learning, four intact third-grade classes were randomly assigned as three experimental groups through VR mediation and one control group in conventional classroom teaching. Within three experimental groups, Individual VR Group and Paired VR Group were in student's autonomous condition while Teacher-centered VR Group was in teacher's facilitating condition. Further, to explore the role of pair work cooperation within the student's autonomous condition, Individual VR Group was assigned to work individually whereas Paired VR Group was assigned to work cooperatively. Thus, in order to investigate the effectiveness of VR mediation and examine the effects of learner involvement and pair work cooperation on English vocabulary instruction, four groups were presented as below (Table 1): 1) Individual VR Group, 2) Paired VR Group, 3) Teacher-centered VR Group, and 4) Control Group. Table 1: Four groups. Need Search Evaluation. G1 (Individual VR Group) ＋＋＋＋＋. G2 (Paired VR Group) ＋＋＋＋＋. G3 (Teacher-centered VR Group) ＋－－. G4 (Control Group) －－－. Instruments Quantitative methods were employed to investigate and analyze the effects of VR vocabulary instruction on the third graders' word acquisition. Research instruments utilized in this study included pictorial flashcards and 3D virtual reality computer program (Institute for Information Industry, 2011). As for the data collection instruments, the multiple choice test with one L1 translation and four L2 word options was used in pretest for 20 target word selection (Appendix B---pretest). To explore 29.

(38) . students' learning process, the pictorial, multiple choice test format was adopted in immediate posttest and delayed posttest in two weeks later (Appendix C---posttest). The following introduced these instruments respectively. Twenty target words The use of pictorial vocabulary cards has been widely advocated by many scholars to benefit language learners in two perspectives: vocabulary memorization and self-directed learning (Mueller, 1980; McCarthy, 1990; Teng, 1994; Schmitt & Schmitt, 1995; Hulstijn, 2001; Nation, 2001). According to the studies summarized by Nation (1990), a word required five to sixteen or more repetitions to be learned. Via the repetitions and a variety of activities and techniques, learners could gradually set up their own target vocabularies according to their individual differences and preferences (McCarthy, 1990; Schmitt & Schmitt, 1995; Hulstijn, 2001; Nation, 2001). Therefore, through pictorial flashcards, the visual aids could not merely establish learners' association between a word's form and meaning (Schmitt & Schmitt, 1995; Nation, 2001), lower learners' anxiety (Ye, 2000), provide cultural information, but also offer contextual cues to make comprehension easier (Teng, 1994). Namely, the visual aids acting as the contextual cues could help students construct their linguistic knowledge gradually and systematically (Ausubel, 1968; Hadley, 1993). Before the actual treatment, the pretest containing 30 items was carried out for 20 target word selection (Appendix B---pretest). In the principles of word removal, three target words without audio input in the VR program were deleted. Another seven words were removed because learners obtained some scores in pretest and might have partial vocabulary knowledge on them. It was crucial for word selection to ensure learners had no background knowledge of those target words. Eventually in this study, twenty target words were utilized in three experimental groups and one control group. This study employed these 20 target words (Table 2) in an experimental design: 30.

(39) . Table 2: Twenty target words One-syllable words:. clerk seat sink tie. Two-syllable words:. luggage mirror sidewalk sofa T-shirt. Three-syllable words:. boarding gate cosmetics luggage cart monitor passenger. Four-syllable words:. check-in counter flight attendant X-ray machine. Five-syllable or more than five-syllable automatic door words: metal detector overhead compartment. Based upon both the research and pedagogical value, twenty target words in this study were all actual words. For motivational and educational purposes, four words were from simple one-syllable category, five were from two-syllable words, five were from three-syllable types, three were from four-syllable items, and three were from five-syllable or more than five-syllable vocabularies. According to the suggested wordlist in Grade 1-9 Curriculum Guidelines in Taiwan (MOE, 2008), the following theme-based word selection criteria were shown as follows: (1) all were associated with airport setting from the suggested wordlist of transportation category, (2) all were from the same part of speech (i.e., all were nouns); (3) all were without other grammatical clues; (4) all carried concrete meanings and could be illustrated by 31.

(40) . distinctive pictorial flashcards. On these criteria, within each syllable category, the target words remained roughly equal at the same difficulty level and worth learning and teaching. And all the twenty words above were covered in pictorial flashcard teaching for Control Group and VR English learning for Individual VR Group, Paired VR Group, and Teacher-centered VR Group.. 3D virtual reality computer program Since 2008, the Institute for Information Industry has been working on integrating the idea of Second Life into VR computer program called Second Classroom designed for foreign learners in Taiwan learning Chinese as a second language. Originally, the program aimed to help foreigners learn Chinese autonomously. Due to the program effectiveness and positive feedback from foreign learners, the Institute for Information Industry further cooperated with Kang Hsuan Corporation and developed a new 3D VR interactive program---3D Classroom for English learners. Therefore, 3D Classroom, the VR computer program examined in this study was designed by Institute for Information Industry and published by Kang Hsuan Educational Publishing Corporation in 2011. In line with the benefits in VR applications pointed out by the previous studies, the characteristics of 3D Classroom were illustrated as below. First, as Yeung (2004) noted, this 3D and VR technology could foster learners' ability to visualize, understand and construct their own knowledge. For instance, as an interactive platform, 3D Classroom offered multimedia learning materials involving textual, pictorial, audio and visual aids to facilitate language learning and teaching. Second, the visual cues in the program could not only simulate the real-life interactions, but also create new identities for students to lower their anxiety (Jung, 2002). For example, in the 3D Classroom, learners could choose their own avatar and have their 32.

(41) . new identity to travel around the Welcomeland. On this land, learners could shuttle back and forth by any transportation. Third, as mentioned by Seymour (2002), through the simulations in VR condition, learners could feel flexible to learn without the constraints of age, stage, or location. As indicated by some teachers or researchers, Taiwan seemed to be the relatively limited English learning environment due to the lack of authentic input and real conversation practices. Yet, this VR program could both raise learners' interests in vocabulary acquisition, and further enhance their language proficiency via bountiful communications without being constrained by time or the other restrictions. Fourth, in light of language learning, VR could help learners to immerse themselves in a contextualized language learning environment (Garcia-Ruiz et al., 2007). And the platform could also empower learners to choose from different VR scenarios according to their needs and individual differences in computer-based VR environment (Jung, 2002). In this program, the theme-based virtual reality scenarios could be divided into four types (Lin, 2011). Home setting was the first type including house, apartment and countryside. Second one was associated with community involving supermarket, restaurant, park, stadium, MRT station, hotel, post office, hospital, clinic, movie theater, etc. School scenario was the third setting inclusive of teaching building, gym, playground, dorm, library, and so on. The fourth setting focused on the airport, factory, harbor, railway and farm. Through this program, students could not merely learn language with teacher's guidance, but explore the specific scenario to acquire their vocabulary knowledge according to their personal need and preference autonomously. Within these diverse scenarios, learners were empowered to be in control of their learning at their own pace in the VR learning environment. Besides, they could act as the active participants and explore the environment by themselves (Garcia-Ruiz et al., 2007). For example, in the airport 33.

(42) . setting, learners could begin with choosing their avatar to launch their journey in the VR world. Fifth, as Milton and Garbi (2000) noted, through lots of authentic practices in 3D virtual reality settings, learners had better opportunities to communicate in a relatively naturalistic way. In terms of authenticity, either through projection for the whole class or installation in students' personal computers, this first-person experience could not only foster learners' engagement but also create meaningful interaction and retention about what they've already learnt (Singhal & Zyda, 1999; Chee, 2001). Especially, for learners with different language proficiency levels, this 3D Classroom could provide suitable supplementary materials. For beginners like low or mid-graders, they could learn to recognize and acquire vocabularies, whereas advanced learners such as high-graders or junior highs could interact with others based on some words and daily conversational patterns in a communicative way.. Multiple choice tests A thirty-item multiple choice questions with one L1 translation as the stem and four L2 word options were used in pretest for 20 target word selection (Appendix B---pretest). After the target word selection, a twenty-item pictorial, multiple choice questions were designed according to the twenty target words. These questions aimed to evaluate students' learning process on word acquisition in different stages. To examine the effects of VR English vocabulary learning, a 20-question vocabulary test was utilized for immediate posttest, and delayed posttest in two weeks later (Appendix C---posttest). All options in each question remained identical but would be reordered throughout these two tests. There were one picture as the cue and four choice options involving one correct answer and three distracters in each multiple-choice item. Each item would be scored 0 or 1 to show wrong or correct 34.