小坂 拓海

Abstract The Chunk-and-Pass model suggests that language acquisition involves learning to appropriately chunk language input into multiword sequences and to form more abstract linguistic representations. While the theoretical model has gained widespread attention in the language sciences, there are limited studies that adopt this model, particularly in the context of second language (L2) learning and teaching research. This study examines the effects of Chunk Reading Strategy Training (CRST), which was developed based on the model, on multiword processing in low-proficiency Japanese learners of English. A treatment group received CRST, while a control group underwent standard block-format reading training. A phrasal decision test assessed online multiword processing at pretest, posttest, and delayed posttest stages. The results indicated improvement in response times post-intervention for the treatment group, and only in the delayed posttest for the control group. This study, therefore, discusses the theoretical implications and limitations of CRST.

Abstract This study examines context effects on lexical processing by low-proficiency Japanese learners of English during sentence comprehension, and the role of individual differences in verbal working memory (WM). Thirty Japanese learners of English as a second language (L2) and 27 speakers of English as a first language (L1) were recruited for a self-paced reading task with high and low context sentences to assess lexical processing abilities, and a reading span task to measure WM capacities. Bayesian modeling indicated that, at the group level, L2 learners had slower lexical processing times for low context sentences compared to high context ones, especially in the spillover region, while L1 speakers showed no difference. At the individual level, higher WM capacities among L2 learners appeared to mitigate these context effects, supporting the interactive-compensatory model (Stanovich in Read Res Q, 10.2307/747348, 1980; Remedial Spec Educ 5(3):11–19, 10.1177/074193258400500306, 1984). The findings highlight context effects on L2 lexical processing and the crucial role of WM in managing these effects among L2 learners.

Abstract Recent psycholinguistic research underscores the significance of multiword units in language processing and acquisition, aligning with the Chunk‐and‐Pass framework. In this study, 55 low‐proficiency Japanese English learners and 27 native English speakers undertook a phrasal decision task featuring two trigram types: syntactically and semantically complete (send me letters) and fragmental (send me two) across four construction types (phrasal verb, simple transitive verb, caused‐motion verb, and ditransitive verb constructions). Results revealed that while native speakers distinguished between the trigram types, Japanese learners did less so. Additionally, Japanese learners struggled more with the ditransitive construction compared to other types, a challenge not faced by native speakers. This indicates that Japanese English learners may lack native‐like multiword chunking skills and that the learners’ processing efficiency differs by construction.