The purpose of the present study is to investigate the impact of the study abroad (SA) experience in an English-speaking country on Japanese university students. This study focuses specifically on the interplay of affective variables and L2 proficiency, using the framework of the L2 Motivational Self System. The data obtained through the questionnaires were analysed by using a multiple structural equation modelling (SEM) approach. The proposed SEM model showed that the L2 Motivational Self System could be developed into an integrative framework that accounts for not only L2 motivation but also the interplay of various affective variables and L2 proficiency all in one system. More specifically, the study found that the influences of the ideal L2 self, self-efficacy, and L2 learning attitude on motivated learning behaviour became stronger after SA experience. Additionally, L2 anxiety decreased significantly after SA. The ought-to L2 self, which had had less influence on motivated learning behaviour, turned out to have a significant positive impact on it after SA. Based on these findings, the authors argue that SA experience consolidates the robustness of L2 learners' motivation, which in turn contributes to maintenance of their motivated learning behaviour and, consequently, improvement of their L2 proficiency.

View full abstract

This study examined the scale of ideal second language (L2) self (Dörnyei, 2009) for college English learners by applying the Graded Response Model (GRM) (Samejima, 1969), one of the major models within item response theory. While Likert-scale questionnaires continue to be a popular method for investigating ideal L2 self, there have not been enough discussion on the improvement of the scale and the types of ideal L2 self that are relevant for a particular population. This study reports on an application of the GRM and analyzes data from college English learners who may not be particularly motivated to learn English. The GRM analysis, together with an analysis on an open-ended question, demonstrated that some items were more discriminating and informative than others, and that certain types of ideal L2 self were more relevant and realistic for the participants, pointing to the importance of careful item construction and selection.

View full abstract

Spoken and written English-to-Japanese word translation experiments were performed. By investigating the correlation between spoken and written lexical processing and the correlation between individual differences in lexical processing performance and individual differences in listening and reading comprehension, the present study aims to compare listening and reading comprehension from a word-level study. A word translation task was chosen as the experimental paradigm because it enables us to compare the participants' performance on the same test material. Spoken and written word translation performances correlated significantly in terms of both latency and accuracy. Their correlation with the participants' TOEIC® scores showed an interesting pattern. Step-wise regression analyses revealed some important predictors of word translation, such as familiarity. Through the analyses of errors, an attempt was made to quantify and measure the quality of the participants' lexical knowledge. The estimated quality of lexical knowledge yielded higher correlations with the participants' TOEIC® scores than translation accuracy. Theoretical and pedagogical implications of the results were discussed in terms of English lexical knowledge and processing for Japanese learners.

View full abstract

The involvement of phonological short-term memory (PSTM) in second language (L2) acquisition has been widely documented. However, it has not been empirically clear what types of PSTM strongly predict L2 listening comprehension. This study investigates the contribution of speech-processing memory to L2 listening comprehension under two conditions: individual sentences or random word strings, and the first language (L1) or L2. Based on TOEIC listening scores, thirty Japanese EFL learners were sorted into groups of advanced and intermediate learners. To measure each participant's PSTM through oral repetition tasks, four subtest items were created: L2 sentences, L2 words, L1 sentences, and L1 words. A multiple regression analysis revealed that L2 sentence-based PSTM was the only significant predictor of L2 listening comprehension (R² =.54). In addition, t-tests between advanced and intermediate proficiency groups showed that the gap between sentence-based PSTM and word-based PSTM was significantly greater in the L2 than in the L1. Moreover, the L2-L1 gap was significantly less in sentence-based PSTM than word-based PSTM. These findings suggest that L2 sentential knowledge, rather than L1 memory capacity or L2 vocabulary enhancement, plays a crucial role in L2 listening development.

View full abstract

The purpose of the present study was to examine the quality of statistical testing in foreign language teaching research in Japan. We reviewed t-tests, X²-tests and ANOVAs reported in the articles published in Language Education & Technology (LET) from vol. 38 to vol. 49, and calculated the post-hoc statistical power of each test. The findings of the present study were summarized as follows: (a) the sample size of most of the studies in LET ranged from 20 to 60, (b) the median of the effect sizes (in the case of t-tests) showed middle to large levels (d = 0.40-0.80), but (c) the statistical powers of many studies signified severely low levels (almost the 80% of the two- sample t-tests failed to show the statistical power greater than .80). The tendencies were quite likely to have originated chiefly from the inappropriate designs of the experiments or surveys, especially, mismatches between the targeted effect size and the actual sample size. We assert the importance of setting proper sample sizes based on a priori power analysis and precision analysis.

View full abstract

The present study aimed to identify influence of foreign language learning in elementary school in Japan on children. The participants were fifth and sixth graders in elementary school in Tokyo. The data was collected by asking them about similarities and differences between Japanese and English, about English pronunciation, and what they paid attention to when communicating, and was qualitatively analyzed. First, according to their answers, they were aware of differences between Japanese and English in length and ways to use a mouth and a tongue in pronunciation, of similarities and differences between loanwords of Japanese and the original words of English, and of how to say in communication, in order for the interlocutor to understand easily: loudness of voice, speed, stress, and choice of the words. Second, the changes of their answers from the first questionnaire to the second were various: writing on different items in the 2nd (17.6%), expressing the same linguistic items more accurately (6.9%), adding more to their first answer (6.6%), and so on. As a result, it was found that not the levels of all the participants' awareness attained to abstractness, but foreign language learning influenced linguistic awareness and would gradually influence the underlying linguistic ability.

View full abstract

In the discussion of the effectiveness of the task of shadowing, Kadota (2007, 2012) argues that one of the potential effects of the L2 shadowing training is to enhance the speed and efficiency of the subvocal rehearsal in the phonological working memory, and to enhance implicit learning of lexical chunks such as formulaic sequences.
The experiment to be reported here is an attempt to investigate how shadowing and listening tasks differently affect the participants' conscious subvocal process in reading English text silently. Twenty-five participants learning English as a foreign language (FL) in Japan were instructed to subvocalize (i.e., utter internally) a total of ten English passages in pre- and post-tests while tracing the lines of the text they were reading, and their behaviors were all video-recorded. The main result was that there was a significant increase in conscious subvocalization speed (words-per-minute data, etc.) in silent reading after the shadowing session, whereas there was not after the listening session. With this finding, the present research suggests that shadowing may enhance the L2 learners' subvocal rehearsal efficiency in the phonological working memory.

View full abstract

A needs analysis was conducted to elicit attitudes and preferences related to English education at a university of students majoring in Economics and Management. The participants in this study were 505 freshmen majoring in Economics and Management at a private college.
The materials used in this study were three questionnaires on the preferred English learning styles of the students, the English skills which the students wanted to enhance, and the English lessons the students hoped to experience. Also, open-ended questions were asked.
Results revealed the participants' characteristics: (a) they want to learn English using movies; (b) they want to enhance various English skills, particularly in the higher TOEIC group; (c) they want to take lessons to learn business English and want places or lessons to communicate in English, especially in the higher TOEIC group; (d) the higher TOEIC group participants tend to learn English related to economics and management; (e) many participants want to enhance TOEIC scores, particularly in the higher TOEIC and positive-attitude groups; and (f) the lower TOEIC and negative-attitude group want to take lessons in which they can relearn simpler English grammar and experience enjoyable lessons.

View full abstract

This study investigates the procedures for reading poetry of advanced Japanese EFL learners. It replicates, in a different setting, Hanauer's (2001) groundbreaking article, which demonstrates empirically the process of poetry reading in second language acquisition for the first time. The linguistic form of poetry is said to direct learners' attention due to its deviant use of language. However, little empirical data has supported this theoretical claim. Hanauer (2001) showed that the process of reading poetry enables the learners to focus on form without interfering with focus on meaning. The findings of the present study showed a similar effect to the previous study and indicated that Japanese EFL learners were engaged in a process of constructing meaning while noticing the linguistic features of poetry. Through an analysis of protocol data, the present study also illustrated that the discussion of poetry in pairs could facilitate a meaningful interaction between the text and learner, raising awareness of the target language.

View full abstract

The primary purpose of this study is to investigate the effect of introducing the Accent Chants Method (ACM) into English classes in an elementary school. The ACM was created by the author and was hypothesized to be effective in developing L2 English learners' accent placement ability. The results of a paired-comparison t-test showed a difference between the pretest and posttest at a .01 level of significance, indicating the effect size of 0.48. This revealed that the ACM is effective to some extent in helping students learn how to determine the primary accent placement on words. With regard to individual test words, the results of McNemar's test indicated that the ACM improved students' ability to correctly determine the accent placement of many English words; however, for certain words, the method was not very effective or revealed the opposite effect. Generally, if the accent position in the English word occurred in a similar position in the Japanese loanword counterpart, students tended to answer correctly. If not, the result was opposite. This paper discusses the reasons for the participants' correct or incorrect placement of the primary accent, therefore shedding light on some problematic aspects of the acquisition of English prosody.

View full abstract

This study attempted to develop a new scale to capture foreign language learners' attitudes toward Computer-Assisted Language Learning (CALL). We first conducted two large-scale questionnaire surveys, targeting Japanese university students who learned English (n = 841), and determined the constructs of a Computer Assisted Language Learning Attitude Scale (CALLAS) using both exploratory and confirmatory factor analyses. The results of our factor analyses yielded the five factors as follows: (a) self-perceived computer skills, (b) beliefs about social significance of computer skills, (c) beliefs about the effectiveness of CALL, (d) attitudes toward computer-mediated communication, and (e) attitudes toward multimedia. Then, we further tested the reliability and the factorial structures of this new scale among multiple groups (junior high school, high school, and university students). The multiple groups structure equation modeling with the data of our follow-up survey (n = 1,487) supported the factorial validity of the scale among the groups with high reliability for each factor, showing that the factorial structure and loadings were invariant over the three groups. Pedagogical implications, especially in light of Aptitude-Treatment Interaction (ATI), are also discussed.

View full abstract

The purpose of this study is to observe the change of brain activity during auditory sentence comprehension by NS (native speakers) and NNS (non-native speakers, namely, Japanese EFL learners) using NIRS (near infra-red spectroscopy) topography. Brain activity of the bilateral superior temporal sulcus, including the auditory and Wernicke regions, was measured in both NS and NNS while they listened to short English conversations at speech rates that were varied using computer time-compression of the speech signal to 100%, 50%, and 25%. Results indicated that in NS, activity in the left hemisphere (associated with language processing and intelligible speech) increased as the speech rate doubled in speed. However, activity of the right hemisphere increased instead of the left when listening to a speech rate at 25% compression, which is not intelligible. On the other hand, the right hemisphere of NNS was more activated than that of the left even at standard speed. These results indicate that when speech is more ambiguous, brain activity might shift to the right hemisphere, which does not process sounds as language but more like melody, and hence differs from the processing of automatized language such as the first language in the left hemisphere.

View full abstract

This paper attempts to demonstrate and compare what vocabulary ESP learners are exposed to and how in two types of teaching materials and their writing texts. It also discusses how reading activities with their materials should be designed to maximize the development of learner vocabulary, which is incidentally learned through reading. The analytical points are: 1) which words are used in all reading materials in the course to ascertain whether if the vocabulary introduced in each text supports reading further materials; 2) the difference in frequency distribution of the vocabulary items between a commercialized textbook featuring a collection of reading passages with discrete topics, and the materials used in the target course in this study; and 3) how the vocabulary items between the reading texts and their worksheets (input) in the course and the students' writing output are related. The results of the vocabulary analyses imply that neither type of text features many recurring words. Rather, the contents and contexts of the learned texts become background knowledge to support comprehension of the contents and eventually the vocabulary of a new reading text if the texts are sufficiently relevant to enrich the knowledge of a specific topic.

View full abstract

This study aims to examine the relationship between L2 learners' real-time processes of essay writing and the ratings of their products. To investigate the relationship, keystroke logging data were collected and analyzed. Previous studies on L2 writing have claimed that good writers tend to show sub-processes of writing such as (a) planning, (b) formulation, and (c) revision more clearly than novice writers. However, these insights had been given only by the use of retrospective methods, and had not been tested well using quantitative methodologies such as keystroke logging. Thus, the present study recorded 35 Japanese EFL learners' keystrokes during their essay writing. The time-series transition of their writing, which was operationalized as the change in numbers of words over time, was fitted to the linear regression model and the Poisson distribution model. The results suggested that the time series transition of most of the writers exhibited better fitness to the linear regression model. Also, our multivariate correlation analyses among the indices representing the participants' tendencies of writing and the ratings of the products revealed that some of the indices predicted a part of the variance in the ratings. Possible applications of this approach in future research and some pedagogical implications are also discussed.

View full abstract

The purpose of this study is to analyze the use of the passive voice, focusing on the long passive, in academic writing, mainly in the field of engineering. Some previous studies indicate that the overuse of the passive voice has been criticized for a long time in academic writing and a number of journals have started to give guidelines that discourage the use of the passive voice. However, much passive usage can still be found in many fields of academic writing. In addition, variations of using the passive voice in research articles among different academic fields are required to be examined in detail. Therefore, the main focus of this research is on the use of passive voice in academic writing in the field of engineering. In this study, the characteristics of the use of the passive voice were analyzed in a small corpus of academic articles from four major engineering journals. It was found that some verbs frequently appeared with a by-phrase and the passive form of them was often used with a mathematical presentation as its agent. The reason is that mathematical presentations are usually complex, long expressions and that may be difficult to turn into the subject of a sentence.

View full abstract

There are two purposes in this study. The first is to make a readability formula for Japanese EFL learners based on their characteristic cognitive process of reading. The second is to discuss the validity of readability values measured by word length and sentence length in the theoretical framework of working memory. Our research here is divided into two phases: Phase I attempts to develop a new readability formula exclusively for American native English speakers, and subsequently with the validity of the RBF verified, we move on to Phase II, which is to develop another formula for Japanese EFL learners, since the former has so long a history and a greater number of research studies on practical formulae. Note that there are four steps of independent investigations in each Phase: First, we focus on deriving a formula of phonological coding time for English text estimated by the number of syllables and phonemes. Second, we discuss whether the distribution of the mean length of the chunk processed at one time is related to the hypothesized memory constraints for language processing. Third, we attempt to develop a readability formula based on the above discussions. Finally, we verify the validity and effectivity of the above formula. We found that higher readability value leads to faster reading speed suggesting that the readability formulas based on the estimated time of phonological coding and the working memory constraints are highly valid and feasible. Educational implications for EFL reading and oral reading drawn from these investigations are also provided in the last section.

View full abstract

Seleaf <http://www.mintap.com> is a cloud-based search engine for a tagged corpus of spoken English along with video pictures from movies. This free site stores 35 hours of 18 American premier movies, which are broken down into 44,536 scenes, 32,270 phrases, and 193,106 words. Seleaf has been developed primarily for language educational use. The authors discuss how to enrich language teaching using this newly developed search site in a variety of ways. First, language teachers can instantly show and repeat the meaning of a particular word or phrasal expressions in a visual context in class. Second, teachers can offer examples of various phonetic realizations of spoken English in different situations. Third, the Shadowing Drill function helps learners to read aloud or shadow-read any phrases repeatedly along with their rich visual context. Each phrase is realized as a unit of breath group most of which are realized within two seconds, which is hypothesized to be kind to human cognitive load in working memory. This article also reports our classroom treatment and learning outcome using Seleaf as well as the results of questionnaires asking the learners' awareness and data collected from the error logs.

View full abstract