流行歌曲教學對英語虛擬語氣的學習與記憶影響效益研究--以臺北市一所國中為例

全文

(1)國立臺灣師範大學英語學系碩士論文 Master Thesis Graduate Institute of English National Taiwan Normal University. 流行歌曲教學對英語虛擬語氣的學習與記憶影響效益研究 —以臺北市一所國中為例. Impact of English Pop Music on Young EFL Learners’ Learning and Retention of English Subjunctive Mood Construction: A Case Study of A Junior High School in Taipei City. 指導教授：劉宇挺博士 Advisor: Dr. Yeu-Ting Liu 研究生：李秋欣 Graduate: Chiu-Hsin Lee. 中華民國一百零二年一月 January, 2013.

(2) 中文摘要第二語言學習的研究者常指出，音樂／歌曲能對學習者產生助益。然而音樂／歌曲與國中學生學習英語之關係的實證研究仍相當缺乏，因此本研究旨在探討英語流行歌曲教學如何促進國中學生的歌詞記憶，以及這些歌曲對虛擬語氣學習之影響。本研究包含為期一個月的實驗，一份問卷，以及研究者在課堂中的觀察記錄，研究對象為三個班共 76 名九年級生。研究結果顯示英語流行歌曲能夠提高學生的歌詞記憶表現，但無法讓學生自然而然地學會虛擬語氣。總結而言，此研究希望能解釋英語流行歌曲如何協助國中生的英語學習，英語教師也可藉此進一步了解在教室中使用英語流行歌曲的相關資訊。. 關鍵詞：英語流行歌曲、記憶表現、英語學習 i.

(3) ABSTRACT Researchers of second language learning have often suggested that music/songs would likely facilitate learners’ language learning. However, research which has empirically documented the link between music/songs and young EFL learners’ English learning is scant. Therefore, the aim of the current study attempts to explore how English pop music can facilitate EFL junior high school students’ text recall performance and the learning of past subjunctive mood. This research involved a onemonth experiment, a questionnaire, and the researcher’s classroom observation. Three classes of 76 ninth graders participated in the study. The results of this study showed that English pop music could facilitate the participants’ text recall performance but failed to facilitate their learning of the target structure. To conclude, this study may be of importance in explaining how English pop music facilitates EFL junior high school students’ English learning, as well as in providing English teachers with a better understanding of adopting English pop music in their language classrooms.. Keywords: English pop music, text recall, English learning ii.

(4) ACKNOWLEDGEMENTS I am very grateful to many people that have contributed in direct or indirect ways to this thesis. First I would like to acknowledge the support of the Department of English at National Taiwan Normal University (NTNU), and in particular the professors in the summer programs for giving me the opportunity to study English teaching and to learn from them. I would also like to thank all my classmates at NTNU who with their support made this thesis possible. I would like to express my indebtedness to Professor Yeu-Ting Liu. His teaching in second language acquisition inspired me so much, and at the same time, his encouragement and support gave me strength and made me keep working on this thesis. I owe most of my skills as a researcher to him. In addition, I would like to thank Professor Yi-Chien Li from NTNU and Professor Mae-Han Chen from National Taipei University of Education for their comments on this thesis. Finally, my deepest gratitude goes to those whose influence may not be so evident but who nevertheless provided the educational and emotional foundation without which this thesis would have never been possible: my parents, my older brother, my younger sister, my brother-in-law, my grandmother, and Doris Lin.. iii.

(5) TABLE OF CONTENTS 中文摘要 ................................................................................................................... i ABSTRACT ............................................................................................................ ii ACKNOWLEDGEMENTS............................................................................... iii LIST OF FIGURES ............................................................................................ vi LIST OF TABLES .............................................................................................. vii CHAPTER ONE INTRODUCTION ...................................................1 Motivation and Background ......................................................................1 Rationale and Purpose of the Study .........................................................5 Significance of the Study ............................................................................7 Organization of the Thesis .........................................................................7 CHAPTER TWO LITERATURE REVIEW ....................................9 Music, Memory, and Language ............................................................... 9 Mnemonic and Memory ................................................................. 9 Music as a Mnemonic Device for Text Recall ............11 Transfer Appropriate Processing (TAP), Language Learning, and Pop music teaching ................................................................................................21 Transfer Appropriate Processing and Language Learning ...23 The Relationship Between TAP and the Studies on Music as a Mnemonic .............................................................................................................25 Pop Music Teaching and Language Learning .........................27 Major Findings and Limitations .............................................................29 Research Questions ....................................................................................29 CHAPTER THREE METHODOLOGY ..........................................31 Participants ...................................................................................................31 Design of the Study ....................................................................................32 Materials .......................................................................................................33 Pop Songs Selected for the Learning Sessions ........................33 Target Structure ...............................................................................34 Procedures for Data Collection ...............................................................34 Assessments .................................................................................................36 Pretest 1: The Singing Task .........................................................36 Pretest 2: The Target Structure Task .........................................36 iv.

(6) Posttest 1: The Text Recall Task ................................................38 Posttest 2: The Final Target Structure Task .............................38 Data Analysis ...............................................................................................39 CHAPTER FOUR RESULTS AND DISCUSSIONS ...................40 Results of the Text Recall Tasks ............................................................40 The Text Recall Tasks: The Posttests ........................................40 Discussions about the Text Recall Tasks .............................................42 An Overall Observation and Reflection ....................................43 Possible Reasons for the Poor Scores in the Second Text Recall Task ...............................................................................................................47 Certain Concerns about Language Learning in Taiwan ........48 Results of the Target Structure Tasks ....................................................50 Discussions about the Target Structure Tasks .....................................51 CHAPTER FIVE CONCLUSION..................................................................53 Summary of the Major Findings .............................................................53 Pedagogical Implications ..........................................................................54 Limitations of the Study ............................................................................55 Suggestions for Future Research ............................................................57 REFERENCES .....................................................................................................59 APPENDIXES .......................................................................................................62 Appendix A Lyrics: Because I Love You .........................................62 Appendix B Lyrics: Tears in Heaven .................................................63 Appendix C Lyrics: One of Us ............................................................64 Appendix D Lyrics: If I Were A Boy .................................................65 Appendix E Target Structure Task 1 ..................................................67 Appendix F Target Structure Task 2 ..................................................69 Appendix G Target Structure Task 3 .................................................71 Appendix H Target Structure Task 4 .................................................73 Appendix I The Final Target Structure Task .................................75 Appendix J Questionnaire ...................................................................77. v.

(7) LIST OF FIGURES Figure 1. Administration Timing of the Text Recall Tests in the Present Study.........35 Figure 2. Administration Timing of the Target Structure Tests in the Present Study...35 Figure 3. Mean Scores on the Four Text Recall Posttests...........................................42. vi.

(8) LIST OF TABLES Table 1. Numbers of Words Recalled in Each song.....................................................41 Table 2. Pretest and Posttests of the Target Structure Tasks………………………....50. vii.

(9) CHAPTER ONE INTRODUCTION Motivation and Background Because it has long been considered that both language and music have sophisticated and intriguing characteristics, there has been a tremendous wave of interest in the relationship between these two subjects throughout the last few decades (see for example Medina, 1993; Besson & Schön, 2001; Gonzalez, 2006; Sposet, 2008). While scientists, musicologists and linguists seem to agree that music and language are two different and separate cognitive domains, there are also certain similarities between language and music. First of all, they both are closely related to human communication that conveys affective meanings, information, or ideas through the combinations of sounds. Besides, music and language have their own specific rule-based systems which consist of basic elements such as sounds, symbols, signs, and grammar (Besson & Schön, 2001). For example, in order to achieve musical harmony and make a pleasant piece of music, composers should arrange notes and sounds based on certain musical syntax (e.g., scale degrees or chord structures); correspondingly, grammatical sentences are constituted by sounds or words that agree with the rules of syntax. Moreover, the patterns of rhythms, stress, and tones/intonations (i.e., prosody of language/musical prosody) all play an essential role in creating music and producing language. In addition to these resemblances, the inherent temporal structure (i.e., speaking/singing speed or pauses in a conversation or “rests” between notes) also lays crucial importance for language and music. Take human speech and musical work, for example, although different people have various speaking speed, a sudden pause or a long delay of utterance may cause peculiarity and attention. Similarly, when a piece of 1.

(10) melody is played, if the performers hesitate or break off the music movement, causing a lengthy period of silence between notes, the performance will undoubtedly leave the audience in suspense and astonishment. This kind of expectation, to put it more simply, is the temporal expectancy between interlocutors in a dialogue and among audience in a musical performance. Given the similarity between music and language, the two also subtly differ from each other in several ways. Chomsky (2000, p. 45) pondered over similar issues in his work: “Is [music] a human language? No, of course not; it is not a human language. Is it like human language? Well, sure, in some ways but then the question is: how ‘like’ do you mean?” From his questions, it is important to note that while to some extent, many aspects of music may resemble language, there also exist quite a few differences between music and language. First, based on the knowledge of the brain’s information processing, there is general agreement among neuroscientists that language and music are processed in different brain regions: linguistic functions are processed in the left hemisphere, and the right hemisphere, on the other hand, is responsible for music processing. Secondly, although music and language share the same characteristic of human communication that conveys emotions and feelings, the potential “meanings” or “messages” of a certain melody or a musical piece, if any, are more flexible and subject to different interpretations. On the contrary, readers of different backgrounds generally can reliably extract the same meaning from a given message and understand each other. In addition, unlike music, which is universally comprised of seven tones (sometimes more or fewer) in a scale, a greater number of diverse phonemes appear in the phonotatic spectrum of different languages. As a result, while musicians from 2.

(11) different nationalities are able to successfully decode/perceive music tones (e.g., Do, Re, Mi, etc.), different language speakers may have perceptual problems decoding various phonemes from different languages (e.g., /dʒ/ in English and /ʝ/ in Spanish). Furthermore, while well-organized sounds of different instruments played in chorus can form a beautiful and enjoyable melody, different people speaking at the same time may only produce noises (Besson & Schön, 2001). Granted, music and language separately possess complex, delicate, and distinct systems, but at the same time, some remarkable similar qualities do exist between them. Due to such special features, researchers hence dedicate themselves to study the relationship between language and music. Of these studies, one main stream to be investigated is the combination of melody and texts (e.g., lyrical songs) as a mnemonic device for language learning (Calvert & Tart, 1993; Hébert & Peretz, 2001; Rainey & Larsen, 2002; Wallace & Rubin, 1988). Rainey and Larsen’s (2002) study is a case in point. They recruited adult English native speakers and examined whether unconnected texts (a list of unconnected names) would be better learned in a spoken version, a sung version, or a visual version. Participants were randomly divided into three groups, those who were in the spoken group learned the target names by listening, those in the sung group did it by listening to the names accompanied with a melody, and those in the visual group learned the names by looking at the words on a computer screen without any auditory input/output. The rate of the words appearing either on the earphone or on the screen was all the same. The participants learned the text individually, and the numbers of trials required to complete the task were recorded. The result turned out that although there was no significant difference between the spoken and sung version in initial learning (as shown in the immediate recall test), 3.

(12) surprisingly, the participants in the visual group significantly required fewer trials to recall the texts. At the first glance, this finding seemed to suggest that the results of the recall in the visual condition outperformed the other two learning conditions. However, in a delayed test conducted a week later, participants in the sung group performed significantly greater than those in the spoken group, and more than ten participants in the sung group recalled perfectly even before relearning the text. Based on the results from the experiments of the initial learning and the delayed test, the researchers thus suggested that music might be considered as an effective mnemonic device for text recall because the melody seemed to be “heard” during the week and hence become a beneficial mental driver for retrieving the verbal information. In addition to the behavioral experimental paradigm, some researchers attempt to explore the relationship between music and language using the neuroimaging techniques. In these studies, music and language input was constructed to stimulate the adult participants’ brains in which brain regions and activation patterns were captured and recorded. Among these neuroimaging studies, the processing of prosodic features and syntax of music and language have often been the focus of research interest. In this line of research, researchers generally agreed that the stimuli of musical and linguistic syntax and prosodic features can be both processed in similar brain areas (Levitin & Menon, 2003; Maess, Koelsch, Gunter, & Friederici, 2001; Patel, Peretz, Tramo, & Labrecque, 1998; Schön, Magne, & Besson, 2004). Based on the results of these neuroimaging experiments, researchers provided direct evidence for the claim that music and language have strong association when being processed in the human brain (e.g., syntactic incongruities in a piece of music can be processed in Broca’s area). Findings from neuroimaging evidence therefore suggest that brain’s areas underlying music and language processing are 4.

(13) interconnected. With the support of these neuroimaging and the foregoing behavioral studies, it has been verified that music can have a powerful impact on language learning. As a result, music can be regarded as one of the helpful non-linguistic resources for language acquisition. Indeed, in recent years the research of music/songs on language learning suggested that music/songs may facilitate processes beneficial to second language acquisition. These specific benefits include increased motivation and attention among students (Gonzalez, 2006; Niño, 2010; Schön, 2008), an increased quantity of vocabulary/unconnected texts (Chazin & Neuschatz, 1990; Legg, 2009; Rainey & Larsen, 2002; Schön, 2008), and a better recall/retention of memory. That is, accompanied with certain background music, the adult or younger language learners seem to remember the information previously learned for a longer time than learning just in prose/in a spoken way (Calvert & Tart, 1993; Calvert & Billingsley, 1998; McElhinney & Annett, 1996; Rainey & Larsen, 2002; Wallace, 1994). Moreover, learners may also view music/songs as less threatening, which can result in their increased participation in the traditional classroom setting (Kanel, 1997; Racette, Bard, & Peretz, 2006).. Rationale and Purpose of the Study As shown in the aforementioned existing studies cited above, there are a number of points worth noting. First, most of the previous studies have been conducted using both adults and primary school students as subjects (see for example, Maess, Koelsch, Gunter, & Friederici, 2001; Medina, 1993; Rainey & Larsen, 2002). However, it is surprising that almost no evidence was provided for middle school students (i.e., 7th— 9th graders). While no one denies that adolescence is of crucial importance for 5.

(14) successful second/foreign language learning because at this phase, sophisticated metalinguistic awareness, vocabulary size, and reading ability are rapidly growing, little empirical evidence has been gathered to support this issue. More research is warranted to shed light on the effect of music/songs on ESL/EFL middle school students’ language learning. Second, the majority of relevant research conducted to explore on the effect of music/songs on language acquisition has focused on L1 or ESL contexts (Rainey & Larsen, 2002; Wallace, 1994). However, few studies have been done on the effect of music/songs on foreign language learning. Common sense seems to indicate ESL context may be applicable to EFL ones, but empirical support is still critically lacking. In addition, while several studies have suggested the benefit of music/songs in learning a first or second language, the duration of the experiment time is relatively short-term (Calvert & Billingsley, 1998; McElhinney & Annett, 1996). More existing longitudinal research that targets adolescent EFL learners is warranted to shed light on the relationship between music and language learning. Therefore, in order to help fill the gap in our knowledge, this study tries to investigate the effect of adopting English pop music on language learners’ recall of song lyrics. The reason for selecting text recall as the target focus of this study is that when language learners can naturally memorize the sentences/phrases of the pop songs through. listening. and. singing,. they. may. accumulate. the. prefabricated. sentences/phrases in mind and hopefully, they can more effortlessly extract these sentences/phrases in future communication. Therefore, the aims in this study are (a) to explore how English pop music will influence junior high school students’ text recall of the song lyrics in an EFL context, and (b) to yield theoretical as well as practical implications of incorporating authentic and extralinguistic learning materials such as 6.

(15) English pop songs into foreign language classrooms.. Significance of the Study Language learning is a life-long journey. As a junior high school teacher in Taiwan, I perceive that it is quite difficult for students to actively and authentically initiate and maintain contact with English because of lack of appropriate occasions and interlocutors outside the classroom in an EFL context. The current study is done with the hope that it may contribute to a better understanding of the relationship between English pop music and EFL middle school students’ foreign language learning. First, the results may demonstrate whether English pop music can serve as an effective and motivating learning material for EFL junior high school students. Second, by learning and listening to English pop songs, it is hoped that students can regard English as a practical tool, rather than a subject for study. Finally, this study may provide an alternative teaching material (i.e., song lyrics) to facilitate students’ language learning for Taiwanese high school teachers. It is hoped that the experimental results can shed more light on the effect of English pop music, offering pedagogical implications for high school teachers and educators in Taiwan.. Organization of the Thesis The thesis is organized as follows. Chapter One provides background information about the relationship between language and music, accompanying noteworthy findings about behavioral as well as neuroimaging studies. Chapter Two will provide a literature review of the research topics, including (a) music, memory, and language, and (b) transfer appropriate processing (TAP), language learning, and pop music teaching. Chapter Three will describe the methodology adopted in this 7.

(16) study. The results will be presented and discussed in Chapter Four. Finally, Chapter Five will summarize the major findings and offer pedagogical implications and suggestions for further research.. 8.

(17) CHAPTER TWO LITERATURE REVIEW This chapter reviews relevant literature on exploring the major factors in the efficacy of music as mnemonics on ESL learning. Two main bodies of literature on text recall and the integration of music into language classrooms are presented. In section one, studies on music as a mnemonic device for text recall will first be reviewed. Then, findings of studies on familiar versus unfamiliar tunes and on connected versus unconnected texts will also be reviewed and discussed. In section two, the tenets of transfer appropriate processing theory and its relationship with language learning are reviewed, together with language researchers’ viewpoints of incorporating pop music into classrooms as alternative/supplementary teaching and learning materials. The final section will briefly summarize the major findings and limitations in the previous studies. Research questions will be proposed at the end of the section.. Music, Memory, and Language Mnemonic and Memory The term mnemonic, as defined by Ashcraft (1994), means “ ‘to help the memory.’ It … refers to an active, strategic kind of learning device or method, a rehearsal strategy (p. 199).”. According to Ashcraft, memory involved three. elements/levels: “the acquisition, retention, and retrieval of information.” Anyone who tried to have solid memory traits and to successfully retrieve the learned information should accomplish all the three sequences—to learn and acquire the information, to retain it, and to be able to retrieve it. He then proposed three principles of effective mnemonic devices that could improve memory. First, a mnemonic device 9.

(18) should provide “a structure for learning, for acquiring the information.” The structure could be either complex or simple, as long as it could build a connection between the to-be-remembered information and the preestablished set. To achieve such a beneficial effect, Ashcraft suggested that the materials should be “practiced repeatedly.” Secondly, through “rhymes, or other kinds of associations, the mnemonic device [could ensure] a durable record of the material in memory, one that won’t easily be forgotten.” In other words, after the learners had acquired the information, there would be the integration of the existing memory framework and the newly-learned information. By the combination of this memory framework, the information learned might be retained longer or more clearly. Finally, an effective mnemonic device should equip the learners with “effective cues for recalling the information” during the retrieval process. As Ashcraft claimed, “this function of the mnemonic device is critically important” because much of the forgetting often resulted from “retrieval difficulty.” That is to say, while the stages of acquisition and the retention are reached, an effective mnemonic device should play a role to guide the learners and lend support during retrieval. From the perspective of Ashcraft (1994), songs, for example, can serve as one of the effective mnemonic devices. The reasons are as follows. First, if the sentences or the phrases of the song lyrics are considered to be the learning target, the delicate combination of the melody with the song lyrics (i.e. the text) can form a structure for language learners to acquire. Moreover, because the “verse” and “chorus” of songs are usually repeated, this repetitive characteristic is consistent with Ashcraft’s previous suggestion for improving the acquisition stage (i.e. the materials should be practiced repeatedly). 10.

(19) Second, one of the most outstanding features of songs is that they are mostly in rhyme. The rhymes in each verse or the chorus may therefore offer a “durable record” in memory and make the remembered information retain longer. Finally, because of the rhythm, easily-caught melody, and the rhymed words, the phrases of the song lyrics may more easily be recalled by the learners. However, despite the fact that language learners may benefit from songs, there are few empirical studies established to prove the effectiveness of songs as a mnemonic device. In this case, can music really facilitate learning? Fortunately, there exists some research that concerns this issue—music as a mnemonic device. In the following sections, related studies are reviewed and discussed.. Music as a Mnemonic Device for Text Recall Music is ubiquitous and influential. Rainey and Larsen (2002) claim that it is generally believed that the process of learning and memorizing information can be enhanced through the employment of music. Try to recall the lyrics of the national anthem and ponder the thinking process. Do you recall the lyrics by speaking or singing them? Many people may express that the melody behind the lyrics facilitates the recalling/retrieving process. This kind of phenomenon is not surprising since most people have the experience of learning nursery rhymes when they are children. The songs are sung and repeated again and again during the learning period, and although many years pass without practicing, when the familiar melody is heard again, the lyrics of the song can be recalled almost effortlessly. Such kind of retrieval process, therefore, can be considered the employment of music as an effective mnemonic device. Sloboda (1985) once claimed that “music is indeed of immense benefit as a 11.

(20) mnemonic aid” (p. 268). Also, Wallace (1994) declared that music facilitated word and phrase chunking because it provided a “rich mnemonic structure.” However, even though many educators and researchers agree with this contention, little empirical research was conducted to establish such an argument. The following sections, therefore, try to review the few existing studies which adopt music as mnemonic devices to present that music and familiar melody can positively facilitate learning in general and in language acquisition.. Studies on familiar tunes and unfamiliar tunes About thirty years ago, educational theorists and researchers started to investigate the effects of music as a mnemonic device. Gfeller (1983) found that musical mnemonics could bring great benefit to both normal and learning disabled students during the learning process. In the study, the researcher included 30 normal and 30 learning disabled primary school students into two experiments of learning multiplication tables. In experiment one, the target memory task for the participants was to learn and memorize “the multiplication tables 7 x 5 through 7 x 9 and 8 x 5 through 8 x 9 (p. 181).” All the students took individual pretests, followed by two rehearsals, and then the posttests. It is worth noting that even though learning the multiplication tables appears to be learning mathematics at the first glance, it is related to language learning because they both involve mental processes, including learning the basic and systematic rules, digesting them, synthesizing the learned information, and then gradually understanding the whole system more entirely. In the pretest, the participants read flash cards and answered the multiplication problems in the written/visual modality. During the rehearsal process, for the purpose of the study, the participants listened to the problem two times; that is, they repeated 12.

(21) the problem, wrote the answer, and then repeated the problem again. Next, the students were randomly assigned into two groups. Those in the spoken group rehearsed the problems verbally, and for those in the music group, by singing them. Immediate posttest was conducted. The multiplication problems in the posttest were the same as those in the pretest. Participants’ recall performance was graded by the number of correct answers. The result indicated that the recall performance was better in the verbal-rehearsal condition than the musical rehearsal for both groups. At the first glance, the music mnemonic appeared to fail to provide an effective retrieval cue for the target information, and the verbal repetition appeared to be more beneficial for the participants to remember the multiplication problems. The researcher thus pondered the result and created a second experiment to further explore the question about whether and how the music mnemonic could facilitate learning. In experiment two, however, while the participants and the materials remained intact, by increasing students’ rehearsal session, lengthening the learning duration up to three days, and combining instructions on modeling and cuing the learning materials, the result showed a significant effect of musical rehearsal, both for the normal or learning disabled groups. Based on the outcome, the author argued that the primary school students, regardless of their learning potentials, could achieve significantly greater recall through the musical mnemonic device and appropriate guidance from the teacher. To conclude, Gfeller (1983) contended that music could facilitate students’ retention, which plays an important role in second/foreign language acquisition, when the students were sufficiently exposed to the music and became familiar with it. That is, when the tune was familiar enough to the learners, the to-be-remembered verbal information could be connected with the music, forming a solid mnemonic frame for 13.

(22) later retrieval. In addition, the researcher also argued that the teachers’ instructions of modeling and cuing, accompanied with the encouragement of adopting the learning strategies, could empower both normal and learning disabled students to benefit more from the mnemonic device (i.e. music). In order to deepen our understanding of the effect of music as a mnemonic device on text recall, Wallace (1994) further investigated the advantages and disadvantages of familiar tunes and unfamiliar ones as mnemonics. In the study, she conducted a series of experiments to examine the effect of music on the recall of different text materials. In experiment one, 64 undergraduates were informed to listen to three verses of a novel ballad in either a sung or spoken way, and then verbatim recalled the words in a written mode. For the participants in the sung condition, they listened to the three verses sung to the same melody. After listening to the ballad five times, the participants’ recall performance (i.e., the number of words recalled exactly from the original text) was scored after the first, second, and the fifth trials, respectively. The results showed that the participants in the sung condition significantly outperformed the ones in the spoken condition. Moreover, in a 20minute delayed test, participants in the sung condition also had better recall than in the spoken condition. The researcher, therefore, suggested that familiar melody might facilitate initial learning and short-term retention because the familiar musical structure could serve as an effective encoding and retrieval cue to improve recall. In a subsequent experiment, to determine whether unfamiliar tunes would also facilitate text recall, the researcher extracted only the first verse from the previous three-verse ballad in experiment one. By this experiment design, the researcher expected that participants in the spoken group and sung group would have equal exposure to the text and the melody. More specifically, after the participants in both 14.

(23) groups finished listening to the verse once, those in the spoken group would have exposed themselves to the text once and those in the sung group would have also heard the melody once (hence were not familiar with the melody). In other words, regardless of the input modality (spoken versus sung), the participants would receive the same amount of the spoken input and the singing melody. The other procedure in experiment two remained identical to that in the first experiment. 39 new undergraduates were invited to participate in the experiment. Again, the recall performance was scored after trial one, two, and five, in which the numbers of correct words recalled by the participants were counted. The result, contrary to the first experiment, showed a main effect of condition, revealing that participants in the spoken condition had better recall than those in the sung condition. The researcher hence suggested that without melody repeated enough or sufficiently learned, music might not facilitate recall and might become distraction because the basic melodic structure was not entrenched in the memory during the learning phase and therefore, failed to provide adequate retrieval cues for later use—a finding consistent with Gfeller (1983). The findings of the two experiments from Wallace’s (1994) study, along with Gfeller’s (1983), clearly support the premise that music can facilitate text recall only when the participants have ample opportunity to learn the melody. As a result, in order to confirm the findings in the above experiments, and to further substantiate that the melodic repetition is necessary for text recall, Wallace (1994) conducted a fourth experiment. In experiment four, 48 masters students were divided into three groups, including: spoken, sung to the same melody, and sung to three different melodies. To be more specific, the participants in the spoken group listened to the three-verse ballad in the 15.

(24) spoken mode. Those who were in the “sung to the same melody” group would hear the text sung to the same melody three times, and those in the “sung to three different melodies” group listened to the three verses sung with three different melodies. By this design, the researcher hypothesized that the repetitive melody (i.e. three verses sung to the same melody) between verses could facilitate text recall, while insufficiently-learned melody (i.e. three verses sung to three different melodies) might become additional burden in the learning process. The remaining experiment procedure was identical to experiment one except there was no delayed-recall task in experiment four. The result of the immediate learning task showed a significant effect of learning condition, indicating that participants in the sung-to-the-same-melody condition outperformed the other two groups (spoken/three verses sung to different melodies). The researcher hence concluded that the melody of a song can facilitate recall of text (i.e. make a text more memorable), especially when the melody is easily and sufficiently learned and hence is familiar to the learners. At the same time, familiar tunes can provide a melodic structure for the to-be-remembered materials, building connections between music and words, and offering solid retrieval cues for later retrieval. Therefore, the findings of the above-mentioned experiments have suggested that familiar melody can facilitate learning because the musical structure may provide another resource for text recall in the memory system, while unfamiliar tunes may become distraction or learning burden for text recall when the melody is insufficiently learned.. 16.

(25) Studies on connected texts and unconnected texts The aforementioned Wallace’s study (1994) has suggested that familiar tunes may provide an information-rich context when the learners are encoding the text, and this combination of the melody and the text can hence facilitate the text-recall process. However, several questions emerge if we take a closer look at the experiments; and it is noteworthy that the lyrics of the ballads used in these studies are all connected texts. That is, based on the findings of the reviewed studies, it is clear that the familiar melodic structure could facilitate text recall, especially connected texts. Then, what about unconnected texts (i.e., sentences/lexical items that are irrelevant or discrete to one another in the texts)? Could musical mnemonics equally facilitate the recall of unconnected texts? In order to shed light on this issue, this section will discuss studies that are set out to explore the effect of music as mnemonics on the recall of unconnected texts. Chazin and Neuschatz (1990) conducted a study to investigate the effect of song on recall of familiar and unfamiliar information. More specifically, the researchers made a list of mineral names (considered as unfamiliar information), describing their colors and characteristics (as familiar information), and recorded the list in two ways: spoken and sung to a familiar melody. The list of the mineral names and their characteristics was 20 unconnected sentences (e.g., “Cinnabar is bright scarlet and it’s very heavy,” or “Niccolite is shiny copper and it’s sort of brittle”). The researchers hypothesized that song could improve recall in memory because there exists association between the melody and words. They invited 26 eight-year-old primary school students and 20 adults (mean age 18 to 20) to join the experiment. These participants were randomly divided into four groups, older song group, older lecture group, younger song group, and younger lecture group. 17.

(26) The participants were asked to listen to the twenty sentences once, and then a handout was given to each participant. They then listened to the recording two times and repeated the information either in a spoken or sung way. Right after the practice was the immediate test. The participants freely wrote down any words or sentences they remembered. There was a delayed test a week later. The result indicated a main effect for type of learning (i.e. spoken versus sung), suggesting that recall of the names of the minerals (i.e. unfamiliar information) was significantly greater in the sung group than in the spoken group. However, recall of the characteristics of the minerals (i.e. familiar information) showed no difference among the four groups. As for the delayed test, the result showed no main effect on the information (i.e. familiar or unfamiliar) or type of learning (i.e. spoken or sung). That is, the superior effect of the sung condition ceased to emerge in the delayed test. The researchers thus suggested that there would be no sustaining effect of the musical mnemonics if the information and the melody were not over-learned in the learning process. Nevertheless, the findings of the immediate test showed that the familiar tunes could facilitate recall of unconnected texts by developing associative links with the to-belearned information and improve the retrieval process. Rainey and Larsen (2002) conducted two experiments to investigate whether familiar melodies could facilitate initial learning or long-term memory for unconnected texts (i.e. discrete lexical items). The researchers hypothesized that the learners could memorize the unconnected texts in fewer trials when presented in the sung mode than in the spoken mode. Moreover, they also assumed that the learners could retain the to-be-remembered materials for a longer time when learned in the sung mode than in the spoken mode. In experiment one, 79 adult participants (mean age 19.7) were randomly divided 18.

(27) into two groups to remember a list of twelve names of baseball players either in a spoken way or in a sung version (i.e. sung to the familiar tune of “Pop Goes the Weasel”). There were two sessions in this experiment. At the beginning of session one, the twelve names were presented on the computer screen accompanying the aural input (either spoken or sung mode) in order to familiarize the participants with the names of the baseball players. After that, the participants began to learn the list individually by listening to the recordings. The numbers of the trials required to thoroughly learn the list were recorded. When the participants believed that they had learned all the names, they recited the list to the researcher. If the participants correctly repeated all the names on the list, the researcher would record the total number of the times the participants listened to the recording. If the participants missed any names, they went back to hear the recording until they completed the task. A week later, the participants were tested on their retention of the twelve names at the beginning of session two. If they failed to recall any name of the list, they would follow the same procedure in session one and relearn the list until they completed the task. The result showed no difference between the two groups in initial learning. However, for the recall of names a week later, the participants in the sung condition relearned the list significantly faster than those in the spoken version. The researchers thus suggested that while there might be no difference between the two groups in short-term recall, the participants in the sung group did relearn the list more effectively a week later. The researchers considered that this result was possibly due to more mental rehearsal between the two sessions. They believed that the participants in the sung condition “experienced ‘hearing’ the melody during the week and rehearsed their list automatically when the melody came to mind.” (Rainey & Larsen, 19.

(28) 2002, p. 184) However, in a later non-official discussion, several participants reported that certain names of the baseball players in the sung version were “glued together” and led to confusion (i.e., different names were “perceived as connected by the participants”). The researchers, therefore, decided to replicate the first experiment and provide new stimulus lists to avoid this confusion. In experiment two, the researchers created a list of 14 nonsense names which had the same number of syllables and the same consonant and vowel structure to avoid the previous singing confusion. The list was recorded in two ways, spoken and sung to the melody of “Yankee Doodle.” There were three learning conditions in the experiment: the sung condition, the spoken condition, and a new visual condition. For the participants in the visual version, they learned the list by watching the names on the computer screen, with the same presentation rate as the other two groups and without any auditory input. One hundred and two participants were individually tested and the other procedures remained the same as those in experiment one. The result indicated a significant main effect for learning conditions, revealing that the participants in the visual group required significantly fewer trials to initially learn the list than those in the other two groups. There was no significant difference between the spoken and sung group in initial learning. However, there was a significant difference in the number of trials required to relearn the list a week later. The result showed that the participants in the sung group required significantly fewer trials to relearn the list than those in the spoken group. Meanwhile, it was worth noting that while only four participants could perfectly recall the list in either the spoken or the visual group at the beginning of session two, there were strikingly eleven participants in the sung group achieving such an accomplishment. Based on the two experiments, the researchers concluded that in initial learning, 20.

(29) the learners in the visual condition achieved the best performance of recalling the unconnected text. However, the learners in the sung group outperformed the participants in the other two groups when recalling and relearning the unconnected texts a week later. The researchers argued that the familiar tunes could significantly facilitate the recall process because the integration of the names and the melody might serve as an effective retrieval cue. To sum up, based on earlier studies (see for example, Wallace, 1994; McElhinney & Annett, 1996), music might become a beneficial mnemonic device in initial learning when the learners could familiarize themselves with the melody, and the to-be-remembered text should be connected in the content. That is, the lyrics of the songs in these studies are related to one another, and hence form a cohesive unit that may improve the retrieval process. Moreover, the familiar melody could not only facilitate the recall of unconnected texts but also enhance the retention and relearning of the remembered text (e.g. Rainey & Larsen, 2002). This memory retention is especially critical to language acquisition because when learners could naturally and mentally rehearse the newly-learned sentence patterns or vocabulary when listening to the music familiar to them, they seem to be able to review the target structures in almost every place.. Transfer Appropriate Processing (TAP), Language Learning, and Pop Music Teaching In the previous sections, studies on music as a mnemonic device are reviewed. Most of the reviewed studies seemed to present the result that musical mnemonic can elicit beneficial effects on text recall when the tunes are familiar to the participants. However, when taking a closer look at these experiments, there are a number of points 21.

(30) worth noting. First, when initially learning the text, parts of the participants can benefit from the musical mnemonic (i.e., familiar tunes) and perform significantly better than the other groups (e.g., Gfeller, 1982; Wallace, 1994). Yet, the familiar tunes seem to create null effects or fail to facilitate some of the participants to have better recall performance (see for example, Rainey & Larsen, 2002). Second, while in Chazin and Neuschatz’s study (1990), the positive effect of the musical mnemonic ceased to continue on the delayed tests, the results of Wallace (1994) and Rainey and Larsen (2002) provided persuasive evidence for the beneficial effect on the musical mnemonic. In brief, the above reviews appear to have a facilitating effect in most cases; however, it is still an unanswerable question about how music can facilitate learners’ text recall. In addition, when it comes to the effect of music as a mnemonic device on unconnected texts, the results of relevant studies also leave a difficult question to answer. More specifically, the results showed conflicting evidence. On one hand, the musical mnemonic in Gfeller’s (1983) and Rainey and Larsen’s studies (2002) had negative or negligible effects (Experiment 1) on text recall during the initial learning process. On the other hand, when the participants were learning the unconnected texts in Chazin and Neuschatz’s (1990) study and Rainey and Larsen’s second experiment (2002), the familiar tunes significantly facilitated the recall process. Then, what may cause these discrepancies among the previous studies? Maybe one possible explanation of such inconsistency can be transfer appropriate processing theory (TAP). The reason for this inference is that almost all of the aforementioned studies have an obvious difference between how the participants learn the information and how they are assessed. To put it more specifically, during the learning process, the participants are asked to memorize the texts either in listening or visual modality; 22.

(31) however, when being assessed during the tests, the participants are requested to recall the texts in either written or spoken modality. This divergence apparently violates the basic tenets of transfer appropriate processing. In the following sections, the principles of transfer appropriate processing and its relevant theories on language learning will be reviewed to further explore the previous unanswered questions.. Transfer Appropriate Processing and Language Learning Recent decades have seen significant increases in the numbers of studies in cognitive psychology on second language acquisition (SLA). This psychological research is critical to SLA because it provides insights into how language is learned, processed, remembered, and retrieved (Lightbown, 2008). With the outcome of these studies on both cognitive psychology and SLA, language teachers are provided with a good opportunity to look over their teaching and readjust their classroom instructions. In this section, one perspective of cognitive psychology about how information is remembered and retrieved—transfer appropriate processing (TAP) will be examined and discussed to investigate its relationship with SLA.. Transfer Appropriate Processing (TAP) Transfer appropriate processing (TAP) can be traced back to its opposite standpoint to the concept of levels of processing. In 1972, Craik and Lockhart proposed the levels of processing framework, in which they claimed that in order to make learners have better memory or longer retention, the target learning materials should be deeply processed. That is, for example, when the inputs being processed are more superficial or shallower (i.e. involved no semantic analysis), the memory is less retainable than those processed more deeply (i.e. involved more semantic analysis). 23.

(32) Quite a few researchers conducted relative studies and supported this claim (see for example, Till & Jenkins, 1973; Walsh & Jenkins, 1973). However, Morris et al. (1977) questioned this contention and argued that not all the deeply-processed information could be retained longer or better—there should be some other possibilities. The researchers then conducted a series of experiments in which they tried to investigate the possible relationships between acquisition mode (i.e., input setting) and type of test (i.e., language use setting). In their experiments, the acquisition tasks included a semantic orienting task, which is considered as deep levels of processing, and a rhyme orienting task, considered as superficial levels of processing. During acquisition, the participants who learned the target words in sentences like “The ________ had a silver engine. (TRAIN)” were thought to process semantically (or more deeply), and for those who learned the sentence such as “_______ rhymes with legal. (EAGLE) (p. 522)” were considered to process nonsemantically (or more shallowly). After the acquisition phase, the participants were assessed on two types of test: standard recognition test and rhyming recognition test. The results indicated that “Semantic acquisition mode is superior to a Rhyme acquisition mode when subjects were given a standard word recognition test in which semantic information of a word needed to be compiled/assessed (Morris et al, 1977).” However, when the participants were given the rhyming recognition test, the memory performance was significantly better in the Rhyme acquisition mode than the Semantic acquisition mode, both in the immediate and the 24-hour-later delayed tests. The results thus do challenge the assumption of the levels of processing by revealing that not all the better memory traces are from “deeper processing,” but rather, sometimes the consistency in the way of the acquisition mode and the type of test can be more crucial during the learning and retrieving process. Therefore, the researchers 24.

(33) suggested that “superficial as well as semantic information played a role in remembering (p. 529),” and at the same time, processing should be transferappropriate to achieve the optimal outcomes. In short, the learning mode should be appropriate for the test type; that is to say, when the way the learners are acquiring the information can match the way they are tested, there will appear the best memory performance (Robinson-Riegler, 2008).. The Relationship Between TAP and the Studies on Music as a Mnemonic According to Lightbown (2008, p. 27), “[t]he fundamental tenet of TAP is that we can better remember what we have learned if the cognitive processes that are active during learning are similar to those that are active during retrieval.” This statement suggests that when language teachers are giving instructions, it is critical for them to ponder over the connections between the learning conditions and the ways of how they assess their students. Take a reading lesson for example. During the instructions, the teacher may ask the pupils to read aloud the materials, explain the meanings of the unfamiliar words, and make them infer the word meanings from the context to achieve the teaching goal—comprehend the whole reading material. Then, at the assessment phase, the teacher has the students write down a summary or freely recall the texts without any further instructions or writing practice. The results of the students’ performance can possibly result in null or negative effects because of the mismatches between the conditions of learning and assessment (e.g., reading vs. writing). Based on the previous discussion and basic concepts of TAP, it is necessary to reconsider the procedures of the previously-mentioned studies on music as a mnemonic. First, for the studies on familiar tunes, Gfeller (1983) made the 25.

(34) participants listen to the spoken or sung version of the target learning text and practice it in a spoken modality; however, she tested the participants in a written modality. In addition, in Wallace’s (1994) experiments, the participants were asked to learn the ballads in an auditory modality, while at the assessment phase, the researcher gave the test in a written modality by having the participants verbatim recall the texts. Secondly, for the studies related to the connected and unconnected texts, this kind of mismatch between the learning modality and type of test also appears. In Chazin and Neuschatz’s research (1990), the participants learned the list of the mineral names and their features in auditory and spoken/sung modalities, yet, they were tested in a written modality, in which the participants were asked to do a free recall of the text. Lastly, in Rainey and Larsen’s experiments (2002), although they provided the participants with auditory and visual learning modalities, the participants were tested in the spoken modality. Despite the fact that these studies all arrive at the conclusion that music as a mnemonic can facilitate text recall, there still exhibits several weaknesses and can be criticized on the following counts. First, in most of the cases, the participants had to learn the target materials in an auditory modality, while they were asked to take the tests in a written modality. It was very possible that the cognitive processes between the learning and testing phases were transfer-inappropriate. More specifically, after the participants learned the materials through the aural input, it is possible that they do receive and understand the information, remember it, but during the retrieval process, they just fail to produce it by spelling or writing it down. Similarly, even though in some cases, the participants were asked to recall the texts in a spoken way, there was still a flaw because they learned the texts in either auditory or visual modality. To conclude, with the previous discussion and exploration, it appears to be 26.

(35) acceptable that music as a beneficial mnemonic can be sustained by the existing empirical evidence. Moreover, the inconsistency among the findings of all these empirical studies could possibly be explained by the basic tenets of TAP. However, there is no research evidence to support the above-mentioned inference. That is to say, if researchers could establish an experiment that makes the mode of learning process match the type of test and proves the effectiveness of the musical mnemonic, then the empirical support for music as a facilitative mnemonic can be more valid.. Pop Music Teaching and Language Learning Based on the positive findings of the foregoing studies, music as a mnemonic can be of enormous value to language learners, especially for those in the EFL context. The reason for this argument is that EFL learners have no adequate and appropriate occasions to practically use English in daily lives. However, with the speed and efficiency of modern technology, English pop music can be easily obtained through the Internet. If EFL teachers can carefully choose English pop music for their pupils, with proper instructions and guidance, English pop music may serve as an effective mnemonic in foreign language classrooms. Following the previous contention, however, there has been little empirical research on the effects of songs as mnemonic devices. Nevertheless, a number of educators and language teachers have argued that integrating songs into language classrooms can significantly improve students’ learning in several ways. First, Mora (2000) and Gonzalez (2006) believe that songs can help to create a pleasant classroom atmosphere, lower students’ anxiety, and have a positive effect on language acquisition. Second, Abbott (2002), Diamond and Municz (1994) suggest that songs can provide formulaic chunks of language for the learners to accumulate their 27.

(36) prefabricated database for future production. Third, because of the repetitive feature in the chorus of songs, lyrics can increase the possibility of automatization for language learners’ performance (Gonzalez, 2006). Finally, Murphey (1992) has highlighted the importance of pop music in language learning because it conveys meanings, allows for different interpretations, enriches teaching materials, and increases students’ motivation and interest. In the above sections, the application of songs as a mnemonic has shown its advantages in language learning. However, few empirical studies have been conducted to prove this contention. As an EFL junior high school teacher, I suppose there are some possible reasons. First, songs have long been considered as a supplementary language teaching material due to the immense pressures of a teacher’s tight teaching schedule. Teachers can hardly squeeze time for songs, not to mention creating a series of experiments to verify this assumption. Second, songs seem to be considered as an enjoyment in language classrooms, rather than a serious theme or a pedagogical device; therefore, its research and pedagogical value has more or less been ignored. Finally, due to the few existing studies related to this subject, language teachers may have difficulty in establishing associated studies. Nevertheless, it is worth noting that certain characteristics of pop music may provide beneficial language learning resources. First, the lyrics of pop music can be considered as “connected texts” in which a message is usually conveyed, or a story is told to the listeners. Second, the basic structure of a pop song—the verse and chorus—are usually repeated and employ the same familiar melody that can reinforce the memory of the lyrical phrase. This repetitive feature can perfectly fit in with Ashcraft’s memory framework (1994) in which during acquisition, the learning materials should be “practiced repeatedly.” Moreover, according to Campbell (2006, p. 28.

(37) 15), pop music “appeals to a large percentage of the population” and “is familiar and widely heard.” This feature supports the idea that pop music is popular with many people, consists of familiar tunes that are easy to remember, and therefore can consolidate the memory structure of the melody and the texts.. Major Findings and Limitations The major findings and limitations in the studies reviewed above are summarized as follows. 1. Music as a mnemonic device could facilitate text recall but fail to achieve a consistent result. 2. Familiar tunes could better facilitate text recall than unfamiliar ones could. Meanwhile, with the help of the musical mnemonic, connected texts could be learned more easily than unconnected ones. 3. Most studies reviewed violate the tenets of transfer appropriate processing, that is, the learning conditions mismatch the conditions of the assessments in the experiments. 4. Educators and language teachers support the contention of adopting songs in language classrooms, but few empirical experiments exist to verify it. 5. The studies reviewed only include primary school students or adults; there are little research focusing on middle school students (e.g., junior high school students).. Research Questions Based on the literature reviewed above, the current study aims to adopt English pop music as a mnemonic in language classrooms and examine its effect on EFL junior high school students’ text recall performance. The research questions in the 29.

(38) study are proposed as follows. 1. Will English pop music enhance EFL junior high school students’ text recall performance in the sung group better than the spoken group or the control group? 2. Will English pop music facilitate EFL junior high school students’ learning of the past subjunctive mood in the sung group better than the spoken group or the control group?. 30.

(39) CHAPTER THREE METHODOLOGY In this chapter, the design of the current study will be described. Detailed information of the entire research method will include the background of the participants, the design of the study, the materials, the procedures of data collection, assessments used in this study, and data analysis of this study.. Participants The participants in the current study were 76 ninth graders studying in a junior high school in Taipei City, recruited from three classes (Class A, Class B, and Class C). There were 28 students in Group A, 24 in Group B, and 24 in Group C. These students were all invited to the current study. They had five English classes a week, 45 minutes each class. Their English teacher was the researcher in this study. All of them had learned English for more than six years since they were third-graders. Among all the students, 31 were male; 45 were female. Each class has an equal distribution of the boys and girls. Their average age was fifteen years old. During the learning sessions, the three classes were randomly assigned to three groups: the Experimental Group A (the sung group), the Experimental Group B (the spoken group), and the Control Group. There were certain reasons for inviting the students to join the current study. First, the three groups were of similar English proficiency level based on their average English scores of the previous five semesters (前五學期英文總平均). Second, all the participants coming from these three classes were those who had the admissions from senior high schools or vocational ones and needed not take the Basic Competence Test (BC Test). Because of the relaxed atmosphere, it was expected that 31.

(40) the participants would accept the researcher’s music instructions and the corresponding instructions more easily. Third, the participants were all ninth graders, who were assumed to have less difficulty receiving the instructions of the song lyrics than the seventh or eighth ones. Therefore, the participants should have no problem in understanding the chosen song lyrics.. Design of the Study The aims of this study were to investigate the effects of musical mnemonic on text recall and the participants’ learning of past subjunctive mood. The duration of the current study lasted for four weeks. Four English pop songs were employed in this study. Each song was learned for one week. The participants were divided into three groups: two experimental groups (the sung group and the spoken group) and one control group. There was certain common ground among these three groups. First, the control group learned the same texts (i.e. pop song lyrics) as the experimental groups did. Second, the participants in each group received identical instructions (e.g. learning the vocabulary and the meaning of the songs). That is, they had the same activities during the learning sessions. However, the participants in different groups learned the song lyrics in different ways: those in the sung group had the opportunity to listen to the song several times each class and sang to it; those in the spoken group listened to the song only once and then learned it by listening to the spoken recording and read it out loud; and those in the control group listened to the song only once, but had no other auditory input. In order to examine whether the musical mnemonic could facilitate the participants’ text recall performance and their learning of the past subjunctive mood, two types of tests were conducted in the current study. First, the pretests and posttests 32.

(41) of the text recall task were conducted to see whether the participants’ text recall performance would be different under different learning conditions. More specifically, the text recall tests were conducted to examine whether the participants in the sung group could better maintain the target texts than the other two groups. The second type of test, the target structure test, was administered to investigate whether the participants could subconsciously learn the past subjunctive mood without any explicit grammar teaching during the one-month experiment. More details about these two types of tests will be elaborated in the section of assessments.. Materials Pop Songs Selected for the Learning Sessions The teaching materials utilized in the current study were comprised of four English pop songs. In order to improve the students’ English learning, the songs were carefully selected for the EFL classrooms. The principles of choosing the appropriate songs are described as follows. Based on Lems (1996) and Poppleton’s (2001) suggestions, songs that are used in the EFL classroom should contain the following features: 1. The vocal performance is clear and the speed of the song lyrics is not too fast; therefore, the listeners can follow the flow of the music with no difficulty. 2. The vocabulary weight is appropriate to the learners’ proficiency level. Based on such a criterion, the song lyrics in this study only consisted of the vocabulary that was included in the 1200-word list of the Basic Competence Test. That is, it was assumed that the average students should be able to recognize more than 90 to 95 percent of the words in the lyrics. 3. The pop songs are suitable for the students in terms of content: the song lyrics 33.

(42) should contain no violent or mature issues or other inappropriate themes. After careful consideration, the four chosen pop songs were: Because I Love You by Shakin’ Steven, Tears in Heaven by Eric Clapton, One of Us by Joan Osborne, and If I Were a Boy by Beyonce Knowles (see appendixes A, B, C, and D).. Target Structure Besides the investigation of the students’ text recall performance, the current study also intended to examine whether the musical mnemonic could unobtrusively and imperceptibly allow the participants to learn the target structure. That is, all four of the pop songs contained a specific and the same target structure (i.e. the past subjunctive mood). There were two reasons for choosing this structure. First, the participants should not formally study it in the textbooks until the first year in senior high school, therefore, it could be considered as a novel knowledge to the participants. Second, the past subjunctive mood presented in if-clauses should be suitable for the participants to learn by themselves because they were equipped with certain basic knowledge about if-clauses in the previous semester. Therefore, the sentence pattern may not be too difficult for them in the one-month study. An example of this target structure is: “If I were you, I would ask my parents for help.”. Procedures for Data Collection The current study was conducted at the second semester of the participants’ third school year. As noted earlier, the duration of the experiment lasted for one month. The participants in each group had five learning sessions each week. They learned one song for one week. The first and fifth sessions were used for tests (i.e. the text recall tests and the target structure tests), while the second, third, and fourth sessions were 34.

(43) for the music instructions and classroom activities. The participants in different groups learned the English pop songs in different ways. No explicit grammar teaching was included in the instructions. It is also important to note that, due to some unexpected condition or uncontrollable factors (e.g. students’ physical conditions or absence), the actual numbers of participants that were present varied from time to time, ranging from twenty to twenty-five individuals. The procedures for conducting the study and collecting the data are illustrated in Figures 1 and 2 below.. Week 1. Week 2. Week 3. Week 4. (04/30~05/04). (05/07~05/11). (05/14~05/18). (0521~05/25). 04/30. 05/04. 05/07. pre. post pre. 05/11. 05/14. 05/18. post pre. 05/21. 05/25. post pre. post. Figure 1. Administration Timing of the Text Recall Tests in the Present Study. Week 1. Week 2. Week 3. Week 4. (04/30~05/04). (05/07~05/11). (05/14~05/18). (0521~05/25). 04/30. 05/04. pre. 05/07. post. 05/11. 05/14. post. 05/18. 05/21. 05/25. post. Figure 2. Administration Timing of the Target Structure Tests in the Present Study. As shown in Figure 1, the purple boxes represent the pre- and posttests of the text recall task. During each week, the first learning session was used to administer the pretest of the text recall, and the fifth session was to administer the posttest of the text recall. In addition, as shown in Figure 2, the orange boxes represented that the 35. 05/28. post.

(44) target structure tests were administered five times during the experiment. The first target structure test served as the pretest, and the following four tests served as achievement tests, through which it was expected that the participants’ gradual growth of the target structure could be observed.. Assessments The assessments of the participants’ text recall and target structure in the current study were elaborated in the following subsections.. Pretest 1: The Singing Task The goal of the singing test was to check whether any of the participants were familiar with the song lyrics before the learning sessions. At the beginning of each session, the researcher made the participants listen to the target song once. After that, the participants were asked whether they had heard the song or not by raising their hands. The researcher counted and recorded the numbers. During the singing part, the participants individually sang to the researcher. If the participants could not hum the melody or control the words (i.e., they could not complete a single sentence), they were considered to be the ideal candidates for this study. However, if the participants could sing the song and achieve a comprehensible level, their data on the posttest of the text recall would be excluded from the final analysis.. Pretest 2: The Target Structure Task The goal of the pretest of the target structure in the first week was to know whether before the experiment, any of the participants had known or learned the target structure (i.e. the past subjunctive mood). Then, the other pretests in the following 36.

(45) weeks also served as posttests, through which it was hoped that the participants’ overall growth could be observed. The pretest of the target structure (see appendixes E, F, G, and H) was a listening test, including twenty multiple-choice items, with three options in each test item. In addition, in the twenty multiple-choice items, only five items focused on the target structure, and the other fifteen ones (e.g. grammatical multiple choices the participants had learned in the previous five semesters) were served as distractors. The listening transcripts were made by the researcher herself and recorded by native speakers. More specifically, if the song was performed by a male singer, the listening test was recorded by a male native speaker, and vice versa. The options were spoken out only once in each test item, and the verbs in each option were stressed to emphasize the differences for the participants to notice. The example of the options containing the past subjunctive mood in the test item is listed below: (A) If I were you, I would ask my parents for help. (B) If I am you, I would ask my parents for help. (C) If I was you, I would ask my parents for help. During the test, the participants were required to listen to three sentences in each test item. After three sentences, the participants had three seconds to choose the correct one. Only one sentence of each test item is correct. If they got the correct answer, they would get one point. The total score of this test was twenty points, but only the scores of the five test items pertaining to the target structure were counted. The test of the target structure was given to the participants in the first session before the singing test each week. Prospective participants were those who fail to get the correct answers for the target structure test items. If the participants got all the correct answers for the target structure items in the first week, it was assumed that they had known the target structure in advance and hence their data on the following target 37.