The purpose of the present study is to investigate how Japanese learners of English store and process derived words in their mental lexicon. L1 research in word recognition has given rise to three hypotheses. The affix stripping hypothesis claims that derived words are accessed by their component morphemes, while the full listing hypothesis claims that they are accessed as full forms. Hybrid hypotheses such as the Meta Model claim that the lexical properties of suffixes determine how derived words are processed. Our data show that Japanese learners of English with larger vocabulary sizes tend to decompose highly productive, semantically and phonologically transparent Level 2 suffixes, but not less productive, semantically and phonologically opaque Level 1 suffixes. On the other hand, the learners with smaller vocabulary sizes tend to process both kinds of suffixes by the Decompose Route. It is argued that the learners may have not acquired the meanings and phonology of the derived words with the less semantically and phonologically transparent suffixes. These results support hybrid hypotheses, even for L2 learners.

View full abstract

Semantic analysis for the establishment of mental representation is not necessarily complete and people have enourmous difficulty in detecting the anomaly in sentences such as, "They were trying to decide where to bury the surviving dead." Two experiments were carried out to examine depth of reading processing among Japanese EFL readers. In Experiment 1, the language effects (L1 vs. L2) on anomaly detection were investigated. In Experiment 2, whether or not anomaly detection occurs online was examined. The results of the two experiments showed that many readers had a tendency to process text in a shallow or incomplete manner, frequently failing to detect anomalies. This tendency was more remarkable in the L2 condition. Furthermore, participants' reading time revealed that an anomaly can be detected online, although their reaction to the anomaly might be delayed. Failures to detect contradictions are accounted for by assuming that non-detectors believe too strongly in the global text interpretations they create.

View full abstract

This study aims to clarify (a) how noticing a hole during output encourages learners of English to incorporate linguistic forms into their IL system and (b) what cognitive activities are triggered by noticing a hole. A total of 102 undergraduate students, who were divided into 3 groups based on their English proficiency, took part in an experiment, in which they worked on guided summarizing, wrote down what they had thought when noticing a hole and were exposed to relevant input. The participants were asked to work on the same guided summarizing in the following week without previous notice to examine how they incorporated linguistic forms into their IL system from the relevant input. The results are: (a) noticing a hole during output encourages learners of English to incorporate linguistic forms into their IL system from the relevant input, (b) noticing a hole triggers various cognitive activities and they have different influences on the incorporation of linguistic forms and (c) learners with high proficiency incorporate more linguistic forms into their IL system from the relevant input.

View full abstract

Words usually have multiple meanings, and some meanings are based on the same concept and others come from different concepts. This study focused on the former and investigated to what degree Japanese EFL students can understand different meanings of words (different L1 translation words). The results suggested that even when the participants understood the core meaning, most of them could not understand the non-prototypical meaning that shared the same concept with the core meaning, regardless of the participants' vocabulary sizes. This fact indicated no relationship between vocabulary breadth and knowledge about meaning other than the core meaning, and this was supported by non-significant partial correlation coefficients between the two. Therefore, knowing more than one meaning of words within the same concepts is a different aspect of vocabulary knowledge from mapping an L2 word onto only one L1 translation word.

View full abstract

The purpose of this research is to investigate how Japanese learners translate Japanese into English, and to consider the basis for effective sentence-level writing instruction. The participants for this experiment were 84 university students, who were asked to complete two tasks: (1) a translation task and (2) a sentence partition task. A total of 320 English sentences translated by 40 students and the partitioned units of each Japanese sentence were analyzed. The results showed that 1) during the translation process, skilled writers produced an English sentence on the basis of larger partitioned units than unskilled writers, 2) learners produced significantly better translations by writing based on sentence or phrase-level units than by writing based on word-level units, and 3) although the strong relationship between partitioned units and the scores of each translation item was not observed, sentence or phrase-level partition had significant effects on translating the item including comparative forms and the item associated with the use of the infinitive. The analyses suggest that teachers should encourage their students to deal with larger units when they produce an English sentence with the Japanese language clue, and supplementary information on particular grammatical items is also necessary for producing correct sentences.

View full abstract

The study examined whether latency (i.e., the temporal gap between the aural input and the shadower's repetition) affected shadowing performance focusing especially on the error rates and the use of top-down processing in word recognition. The 81 high school students shadowed passages that contained a word that was syllabically and semantically deviant from the original one, and then were divided into three groups (i.e., close, middle, and distant shadowers) based on their mean latency. The statistical analyses showed that there was almost no association between the latency group and the error rates, but that the close shadowers were less likely to be affected by the abnormality of the target words and repeated the words exactly as they had heard them. Based on these results, the author proposed that delayed shadowing, with latency of longer than one second, may need to be redefined as shadowing with as long a latency as the learner can afford. Finally, two empirical implications were made for future research.

View full abstract

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

View full abstract

This study aims to investigate the effects of 10-minute Sustained Silent Reading (extensive reading in class; henceforth SSR) on reading fluency and motivation of Japanese junior high school EFL learners. A total of 122 3rd years participated in the study: 60 of them (experimental group) were involved in reading graded readers or leveled readers extensively for ten minutes at the outset of one English class in a week, and 62 of them (control group) were not. Extensive reading tests developed by Edinburgh Project on Extensive Reading (henceforth EPER) were conducted as a pre-and a post-test to measure reading comprehension and reading speed. A 31-item questionnaire was administered to examine their motivation at the end of the program. The results showed that 10-minute SSR improved the reading fluency (reading accurately and rapidly) of the experimental group more significantly, and developed their intrinsic motivation to read English. These findings suggest that large amounts of linguistically appropriate input seem to promote efficient word recognition and syntactic parsing and to achieve fluent reading, and that easy, interesting materials and enjoyable reading experiences seem to promote a positive attitude and the growth of motivation toward second language reading.

View full abstract

This study compares the effects of shadowing and those of repeating during six repetition trials in terms of two aspects: reproduction rate and type of reproduced words. 48 university students performed both shadowing and repeating six times each, and their recorded voices were analyzed to see if there were any differences between the two kinds of training methods in terms of the reproduction rate and the number of function and content words reproduced correctly on each repetition trial. The results indicated that four or five repetition trials were enough to improve the rate of reproduction to a level near the ceiling point. It was also found that the participants could reproduce more content words with shadowing than with repeating, whereas some reduced function words were reproduced from the first trail of repeating, all of which may reflect differences in the language processing between shadowing and repeating performances.

View full abstract

The purpose of the study is to investigate whether learning a two-word collocation is superior in retention and retrieval of meaning to learning a single word. The focus is on the effect of a two-word collocation consisting of a known word and a new word for learners, such as school janitor. This study employed 39 high school students. They were asked to remember the Japanese meanings of 20 low-frequency English words, half of which were turned into two-word collocations by attaching high-frequency cues. In short, they tried to remember the Japanese equivalents of 10 single words and 10 two-word collocations. Two types of recall test were conducted. Test 1 presented all the items as single words by removing the cues from the two-word collocations (without the cue condition). Test 2 presented all the items as they were on the list (with the cue condition). Both of the tests were given to the same participants just after the learning phase and one week later. The main results were: (a) two-word collocations showed better retention of meaning than single words and, (b) two-word collocations showed better results in the recall tests with cues in the retrieval phase, whereas the opposite results were obtained without cues in the retrieval phase. Some pedagogical implications were drawn from these findings.

View full abstract

Pronounceable words are divided into two types; one is known words and the other is unknown words or pseudowords. According to L1 research, many theorists have found that phonological decoding based on word pronunciation has correlation with reading comprehension skills, whether the words are known or pseudowords. In Japan, one L2 study dealt with known words and showed that there is a relationship between the efficiency of phonological decoding and reading comprehension. However, there is hardly any study concerned with pseudowords. Because Japanese people tend to recognize a word as a whole form, not as assembled letter-strings based on letters-to-sound translation rules, their lexical route efficiency might be activated, but it can not be presumed that their non-lexical route efficiency is the same. The following study was initiated in order to clarify how Japanese learners' reliance on lexical route in learning words would relate with reading comprehension. This study showed that there is a more significant relationship between the efficiency of phonological decoding and reading comprehension with pseudowords than there is with known words. This result suggests that word learning be done with the tactic of segmenting words with the knowledge of letters-to-sound translation, rather than with holistic memorization that Japanese learners of English are more likely to use.

View full abstract

Group oral test formats have recently been introduced to the range of oral performance tests because the assessment of L2 learners' authentic conversational competence is considered important in the current era of globalization. However, these tests are not widely administered in Japan and only a few studies dealing with group oral interaction have been carried out to date. This study explored three facets - the severity of raters, the difficulty of rating categories, and participants' speaking ability - when Japanese learners of English encountered oral interaction in a small group. The Common European Framework of Reference (CEFR; Council of Europe, 2001) was utilized for the rating in this study. The analysis was executed by means of multi-faceted Rasch measurement, which analyzed sources of variation and estimated "measures" from the raw scores given by raters. The raters were considered as self-consistent but showed significant differences among themselves as well as with other research. Utilizing multi-faceted analysis and building a consensus among raters by having them identify certain attributes in the speakers are recommended approaches to control such variances. The CEFR rating categories were proved to be valid; however, more detailed rating scales should be created for the situation in which all participants are Japanese, as the speaking ability of most of the participants was judged as low.

View full abstract

This research was conducted in order to investigate the effect of foreign accents of English on the degree of listening comprehension demonstrated by Japanese learners. To this end, we administered a test in which recordings of English passages made by a speaker of Received Pronunciation (RP) and another speaker whose native language is Hindi were played back to two groups of Japanese university students, with two varying levels of English ability, the "Upper-Level (UL) Group" (n=30) and the "Lower-Level (LL) Group" (n=30). The results indicated that (1) The LL Group demonstrated a low degree of listening comprehension of both the RP and Hindi accent speakers due to a lack of linguistic knowledge, whereas the UL Group demonstrated a gradual increase in comprehension over time through phonetic and linguistic mediation, particularly towards the accent of the Hindi speaker; and (2) Both the LL and UL Groups heavily favored a standard accent of English over the Hindi accent in terms of degree of comprehension. Our results found that when the listener has a relatively low proficiency in English, a mismatched interlanguage speech intelligibility disadvantage can be produced.

View full abstract

This study addresses the interpretation of the results of vocabulary size tests with reading proficiency estimation in mind. Despite the number of vocabulary size tests developed so far, how the raw scores should be interpreted, or put into a formula, to obtain estimated vocabulary size of L2 learners seems to leave room for improvement. In this study, three different interpretive approaches, namely the 'level,' 'size,' and 'confidence' interpretations were compared in terms of resulting estimated vocabulary sizes and their correlative strength with reading proficiency. According to test results for 162 Japanese university students, the study found that the size-approach yielded the largest estimated vocabulary size of 5442.77 words in average. It was also found that the vocabulary size calculated using a confidence-based approach showed the strongest correlation with L2 reading proficiency, followed by a size-based approach with a slightly lower correlation coefficient. The results indicated that although comparatively new, the confidence-based approach holds potential for estimating vocabulary size as well as reading proficiency, and the traditional 'size' interpretation was deemed as the most practical choice for the time being.

View full abstract

This study investigated the significance of the effects of word-recognition speed, accuracy, and automaticity on three different English-reading proficiency groups: two groups of English as a foreign language (EFL) learners and one group of native English speakers. In addition, whether the effects of word-recognition speed, accuracy, and automaticity varied according to word frequency levels was examined. Automaticity of word recognition was measured in terms of the coefficient of variation of reaction times (CV_<RT>), proposed by Segalowitz and Segalowitz (1993). Overall results showed that word-recognition accuracy and speed had significant effects in differentiating among the three reading proficiency groups, while moderate effects were observed for word-recognition automaticity on them. In other words, the more proficient a person becomes in reading, the more quickly, accurately, and automatically he or she can recognize words. Interestingly, the effects of word-recognition accuracy and speed grew larger when word frequency decreased, while such changes were not observed for word-recognition automaticity. These results suggest that learners need to recognize words with a wider range of frequency--not only accurately but also rapidly and automatically.

View full abstract

Words having multiple meanings cause difficulties in text comprehension because readers possibly interpret such words as the primary meaning even when the words are used as other meanings. The current study examined how Japanese EFL learners interpreted those words that were used as the secondary meanings in contexts when the participants knew the primary but not the secondary meanings. The results showed that there was no significant difference in the correct rate of interpreting homonyms between reading proficiency levels. However, the types of error they produced were different: the upper group wrote more answers that fitted the context than the lower group who stuck to the primary meanings even when they understood the target contexts. These results suggest that the better readers had greater flexibility in changing their interpretation to one that fitted the contexts, but the poor readers persisted in assuming the meanings they already knew and could not change their interpretation.

View full abstract

The present study investigates the role of vocabulary processing in incidental L2 vocabulary learning through reading within the framework of the type of processing-resource allocation (TOPRA) model (Barcroft, 2000). According to the TOPRA model, the allocation of L2 learners' cognitive resources determines what aspect of L2 vocabulary (e.g., meaning, form, and form-meaning connection or FMC) is learned. The model also predicts that one type of processing depletes remaining resources that otherwise could be used to process other aspects of vocabulary; this is considered to be a necessary consequence, as learners' cognitive resources are limited. Within this framework and using university-level Japanese EFL learners, the present research compares the effects of semantic processing and form processing with the lack of any particular vocabulary processing (control). The results show that, in general, semantic processing inhibits the learning of forms and FMCs of new L2 vocabulary, whereas form processing accelerates the learning of new L2 forms but inhibits the learning of FMCs. These results suggest that processing-resource allocation plays an important role in incidental new L2 vocabulary learning.

View full abstract

The main purpose of this study is to examine how valid peer assessment can be in a high school setting in Japan. One hundred fifty seven students and five teachers participated in this study. The observed scores of the peer and teacher assessments were analyzed using Multi-faceted Rasch Measurement. According to the Rasch analysis, 94% of students did not show misfitting behavior, which suggested acceptable validity of peer assessment (Myford & Wolfe, 2003). Peer assessment was also examined at a class level, and this also fitted into the model. In addition, characteristics of misfit raters were investigated in terms of unexpected responses and English proficiency level, but no specific characteristics in common were found. However, some overall tendencies for peer assessment were observed. The average scores of peer assessment were higher than those of teacher assessment, whereas the standard deviations of peer assessment were smaller than those of teacher assessment. These results indicate that peer assessment of Japanese high school students has reasonable validity in a certain context, although student raters tend to be more lenient compared to teacher raters, and they tend to avoid extreme scores.

View full abstract

Many studies of L2 reading have used the learner's native language as the measurement language to estimate the learner's text understanding. In order to examine the effect of the measurement language, the present study used the recognition task with both native and target language and compared EFL learners' responses to the explicit and implicit information of a text. The results revealed that recognition rates are different between the native language (Japanese) and the target language (English). In addition, the extent of these recognition rate differences between languages varies according to recognition sentence types. These results suggested that the difficulty in recognizing inference sentences may be caused by readers' dependence on literal English expressions, implying that learners tend to pay attention to the words and phrases that appeared in a text when they recognized the inference sentences, even if they understood the implicit information.

View full abstract

Listening anxiety is known as an affective factor influencing the second/foreign language listening comprehension process. However, the construct has been examined and discussed as unique to each individual learner and more or less a static learner attribute, which makes it rather difficult to examine the real-time effects of listening anxiety on cognitive processing in second/foreign language listening. The present study introduces "listening stress" as an alternative to listening anxiety and examines its debilitating effects, focusing on the use of coping strategies. A small-scale stress-inducing experiment was conducted with 9 Japanese learners of English. The results show that increased listening stress hinders the use of coping strategies, debilitating listening comprehension, especially in higher cognitive processing, such as inferencing and generalization/application of information. For future research, both quantitative and qualitative analyses of data obtained from more subjects, more control of factors within subjects, and further refinement of the instrumentation were considered to be necessary.

View full abstract

A Story Retelling Speaking Test (SRST) was developed to provide a speaking assessment tool that has high practicality and positive impact on teaching and learning. The current study aims to investigate the test quality of the SRST by analyzing test-takers' actual test performance based upon four test components--story length, reading comprehension questions, keywords, and opinions. The results showed that the four SRST components function fully or partially as intended, which is positive evidence for the validity of the test. The effects of text length on production volume were shown to be not consistent, which suggests negligible impact of memory on the SRST performance. Additionally, most test-takers understood the story's content in the reading section and most keywords were used in the retelling section. A substantial number of test-takers stated opinions; further, revisions for further eliciting opinions have been proposed.

View full abstract

The purposes of this study are (1) to longitudinally measure the listening and reading proficiency of Japanese senior high school (SHS) students for about three years and to investigate what percentage of the students made (a) progress, (b) little progress, and (c) negative progress in English learning during specific year(s); (2) to describe the longitudinal developmental patterns of the listening and reading proficiency of each student; and (3) to investigate how strongly their initial listening and reading proficiency in the month of April of their first year in SHS affect their progress during the approximately three years of SHS. Two hundred and fifty-two Japanese SHS students took TOEIC Bridge tests four times during their three years in SHS. The results of the analyses show that (1) even in a top-ranking SHS, it may be a little difficult to observe students' progress in a one-year period, (2) there are many different types of longitudinal developmental patterns in both listening and reading scores, and (3) the progress in listening scores might be affected by the initial listening score and the progress in reading scores may be slightly affected by the initial reading scores--the effect on listening scores might be higher.

View full abstract

The paper mainly reports the results of the questionnaire study conducted in Finland in 2008 in order to capture perceptions Finnish primary school pupils have about English language learning. In all, 277 Finnish pupils (180 5th graders and 97 3rd graders) at two different primary schools participated in the study. The questionnaire was also administered to 284 Japanese pupils (145 9th graders and 139 7th graders) in order to situate the responses from the Finnish pupils in an international comparison. The questionnaire consisted of four sections with 32 questions in all. The two questions in Section A were sensitising questions to familiarize pupils with a five-point Likert scale, the 12 questions in Section B asked the pupils reasons for studying English, another 12 questions in Section C asked the pupils how they were enjoying their English lessons, and the six questions in Section D inquired the pupils' attitudes toward English language learning in general. The results obtained from the questionnaire have disclosed that teaching English as a school subject does not necessarily induce disinterest in English.

View full abstract

Ibaraki University started a new English education program in 2004. The new curriculum was developed based on proficiency levels. Each level course shared the same objectives, can-do statements, syllabus, textbooks and grading system. This paper reports the process of designing an evaluation system for this new English program. An evaluation system was designed from the perspectives of "language assessment" and "program evaluation". Students' proficiency, achievement and self-learning motivation were assessed and program effectiveness was evaluated in quantitative and qualitative designs. Achievement tests, can-do self-assessment worksheets, background questionnaires and syllabus evaluation surveys were developed. Teachers, students, administrators were involved in this evaluation system. The data from students' grades and surveys was collected and analyzed. The importance of program evaluation is discussed.

View full abstract