Kris Ramonda

Although the benefits of extensive reading are well known, very little research has investigated the validity of assessment tools to measure general comprehension of graded readers. To address this, quizzes were authored for 42 level 2 Penguin graded readers and administered to 166 students over a semester. Item facility for high-scorers and low-scorers was calculated for the 168 question items from the 42 graded readers, and the resulting item discrimination (ID) score was used to categorize and group quiz items with shared structural and content-based features. The results showed that certain question types tend to be more effective for measuring comprehension of graded readers than others.

This study examined the impact that metaphorical pictures and semantic transparency had on meaning recall of English idioms for L2 learners. Twenty-seven idioms of differing semantic transparency (low, mid, and high) were selected and two types of pictures were drawn for each idiom. One picture type was a literal only representation while the other was a literal+figurative representation. These idioms and picture conditions (no picture, literal only, and literal+figurative) were counterbalanced and presented to student participants (n=64) via a PowerPoint input treatment. The delayed posttests measured meaning recall of the idioms’ L1 paraphrased meaning. The data showed that in absolute terms, literal-figurative pictures promoted better retention of meaning at all levels of transparency, but this finding was most robust for high-transparency idioms. However, a number of pictures, especially those with metonymical elements, led to overspecification, which has wider implications for pictorial input in general.

This article reports on the role pragmatic inferencing plays in accounting for the ways in which native speakers perceive and interpret the semantic transparency of idioms. Although previous studies have suggested that semantic transparency intuitions of idioms are partly motivated by the conceptual metaphors that underlie them (Gibbs 1992; Gibbs et al. 1997), findings from other studies (Keysar & Bly 1995, 1999) have raised questions concerning the arbitrariness of such intuitions. This study seeks to further address the discussion on the nature of semantic transparency by examining the role of pragmatic inferencing and encyclopedic world knowledge for understanding how native speakers interpret the relationship between the literal parts and figurative meanings of metaphorical idioms. To this end, semantic transparency ratings were elicited among fifteen native speakers of English for 222 metaphorical English idioms. Furthermore, raters provided qualitative support by justifying their ratings for a smaller subset of 30 idioms. These initial results were then triangulated by a follow-up exploratory study surveying etymological notes from a number of idiom dictionaries. The findings suggest that pragmatic inferencing via encyclopedic world knowledge plays an important role for the non-arbitrary way in which native speakers perceive the semantic transparency of idioms.

The benefits of extensive reading (ER) to second language learners are well known. However, there has not yet been widespread adoption of ER in the classroom. This is partly because graded readers are costly and challenging to implement into curricula (Davis, 1995). One solution is to introduce an online ‘virtual library’ of graded readers so that students can have unlimited access to simplified reading material. To date, however, few studies have examined how students bene t from online extensive reading.

To fill this research gap, I introduced a virtual library of graded readers to students (N=107) over a semester. At the end of the semester, students evaluated the usefulness of online extensive reading. In this study, I triangulate findings from the data provided by the online extensive reading tools with self-reported gains perceived by students. I also discuss some of the broader implications of implementing online extensive reading for both research and pedagogy.

Teacher beliefs are an important area of inquiry because research has found that these beliefs are often diverse (Breen et al., 2001) and strongly impact classroom practices (Borg, 1998, 2003; Burns, 1992; Farrell & Bennis, 2013). Therefore, uninformed teacher beliefs could be to the detriment of the L2 learner. Despite the fact that knowledge of idioms is a sign of proficiency (Kjellmer, 1991) and many adult L2 learners want to know them (e.g. Liontas, 2002), the classroom tends to be generally devoid of figurative words and expressions (Danesi, 1995). It is possible, however, that teacher beliefs contribute to such overly literal L2 input, especially when they are responsible for curricular choices in the classroom. To further investigate this, 15 native speaker university level EFL teachers in Japan completed a survey that elicited both quantitative and qualitative data on their beliefs about the value of metaphorical idioms for L2 learners. The data showed a considerable divide between participant raters on a number of measures. Much of this divide was due to differences in the raters' emphasis on a top-down or bottom-up view of language learning, as well as the relative importance they placed on the perceived frequency of individual idioms.

This study examines the lexical representation and recall of abstract words by beginning L2 learners of Spanish in the light of the predictions of the dual coding theory (Paivio 1971; Paivio and Desrochers 1980). Ninety-seven learners (forty-four males and fifty-three females) were randomly placed in the picture or non-picture group and taught twelve concrete and twelve abstract words they did not previously know. Subjects performed a recall task on an immediate and a delayed posttest.  The results showed that associating abstract words with pictures had a significant effect on their recall on the immediate posttest, but no such effect was found on the delayed posttest. The results suggest that associating abstract lexical items with pictures has a significant effect on memory representation and recall in the short term.  The findings also support the predictions of the dual coding theory and show that a concreteness effect can be created for abstract words by associating them with visual images.

Diagnostic Yes/No tests are a recommended and much researched assessment tool (Read, 2007; Nation, 2008), yet there is little research into how to apply them to address the mismatch between preexisting course vocabulary lists from commercial textbooks for a particular level and learners’ actual vocabulary knowledge. This study looked at a vocabulary battery of 240 words adopted with a textbook for a pre-intermediate level English course at a Japanese university. During the 1st week of instruction,
a Yes/No test including nonwords (pseudo-words) was administered in three forms with 85 items each. Approximately 100 students took each form. On the average, test takers claimed they knew 75% of the items on the list. A low false alarm rate supports Shillaw’s (1996) findings that the use of nonwords could
be lessened significantly in the Japanese context.

According to the Dual-Coding Theory (Paivio & Desrochers, 1980), words that are associated with rich visual imagery are more easily learned than abstract words due to what is termed the concreteness effect (Altarriba & Bauer, 2004; de Groot, 1992, de Groot et al., 1994; ter Doest & Semin, 2005). The present study examined the effects of attaching visual imagery to abstract words through use of a meaning recall test. Eighty-seven American university students of first-year Spanish participated in the study. Participants were placed in either picture or non-picture groups and were given a treatment of 12 abstract and 12 concrete words. The treatment included three input phases lasting approximately 17 minutes. The posttest and delayed posttest involved a meaning recall test to measure receptive knowledge in which participants were supplied the L2 lexical item and asked to write the L1 translation. The results indicated that participants in the abstract picture group outperformed those in the abstract non-picture group on both posttest and delayed posttest; however, no such effect was found for concrete words. Findings suggest that meaning recall of abstract words can be facilitated by usage of metaphorical, emotive, or symbolic imagery.

Although most of the world’s head-initial languages have a noun + relative clause constituent order, head-final languages are inconsistent in the syntactical positioning of relative clauses. This inconsistency could be due, in part, to competing elements between the heavy constituent principle (Hawkins, 1994) and Lehmann’s (1974) basic constituent order predictions. This study examines possible causes of head-final relative clause syntactical inconsistencies, how these inconsistencies impact working memory, and what potential coping mechanisms compensate for the heavy cognitive load of relative clause + noun constituent order. A comparative analysis taken from samples of relative clauses translated by native speakers into nine different languages (five noun + relative clause languages and four relative clause + noun languages) served as the basis for the study. Results indicate that relative clause + noun constituent order seems to place more restrictions on the types of relative clauses permissible in the language.