The acquisition of unaccusativity in a second language (L2) is widely recognized as challenging. In L2 English, previous studies have reported non-target-like usage of unaccusative verbs in causative constructions (e.g., *Brian fell the rock), although such errors have received limited attention compared to overpassivization. This study investigates overcausativization errors from a first language (L1) perspective. Languages differ in whether certain verbs can appear in both intransitive and transitive constructions. For example, disappear is strictly intransitive in English but has a transitive counterpart in Japanese. This lexical discrepancy may lead Japanese learners to overgeneralize L1-based verb properties to L2 English. Specifically, they may judge transitive uses of non-alternating English unaccusatives as acceptable if the corresponding verbs alternate in Japanese. To test this hypothesis, an acceptability judgment experiment was conducted with 40 Japanese-speaking learners of English. The results showed that learners failed to reject English sentences with non-alternating unaccusatives that are alternating in Japanese, while they successfully rejected those with unergatives and non-alternating unaccusatives that are also non-alternating in Japanese. These findings suggest that learners are influenced by L1 alternation patterns when interpreting L2 verb constructions, leading to overgeneralization of transitive usage in English.

View full abstract

This study investigated the reliability of AI-based transcription for assessing reproduction rates in English shadowing tasks. Although shadowing accuracy has traditionally been scored manually, the process is time-intensive and prone to inter-rater discrepancies. Recent advances in speech-to-text systems, such as Whisper, suggest the potential for automatic scoring; however, validation against human judgment remains limited. In this study, shadowing and parallel reading performances by Japanese EFL learners were transcribed using Whisper and by two independent human raters. After both the Whisper-based and human transcriptions were completed, reproduction-rate scoring was conducted by human coders according to predefined scoring criteria. Reproduction rates and inter-rater reliability were evaluated using percent agreement, Cohen’s κ, Prevalence-Adjusted Bias-Adjusted Kappa (PABAK), Gwet’s AC1, and Intraclass Correlation Coefficient (ICC). Consequently, AI-based ratings closely approximated human judgments, particularly under the parallel condition, although κ values were consistently lower for AI–human comparisons. Overall, AI transcription provides a scalable and reliable tool for evaluating shadowing reproduction rates, reducing instructor workload, and offering immediate feedback to learners. Finally, the limitations and pedagogical implications of this study are discussed.

View full abstract

This study investigates the relationship between perception and production in the acquisition of English linking sounds by Japanese EFL learners. Specifically, the study examines whether learners who are more successful at perceiving linking boundaries also demonstrate more accurate production of linking sounds. A partial dictation task was employed to assess participants’ perceptual ability, while a reading-aloud task was used to measure productive performance. Data were analyzed at the participant level to evaluate the association between the two domains. The results showed that perceptual performance did not reliably predict production accuracy. Although perceptual accuracy was generally higher than production accuracy, learners displayed diverse perception–production patterns, including cases in which production performance was comparable to or exceeded perceptual performance. These findings indicate substantial individual variation in the perception–production relationship. Moreover, perceptual ability alone may not be sufficient to explain differences in the production of linking sounds. By focusing on a connected-speech phenomenon rather than isolated phonemic contrasts, this study highlights the complexity of the perception–production interface and underscores the need for a more nuanced view of how perceptual and productive abilities interact in second language phonological development.

View full abstract

This downstream case study tests whether Whisper transcripts can be used as inputs for construction-based proficiency research on L2 spontaneous speech using the Constructional Diversity Analyzer (CDA). We analyzed one-minute monologues from 107 Japanese learners in the International Corpus Network of Asian Learners of English (ICNALE) and their associated TOEIC scores. After AI-based pruning removed disfluencies from both Whisper transcripts and the corresponding human transcripts (Manual), CDA yielded a constructional diversity index and arcsine-transformed proportions for 11 constructions. Manual–Whisper agreement was strong for constructional diversity (r = .793; mean absolute difference = .0279) and analyzable construction-specific proportions (r = .648–.837), although absolute deviation varied by construction. Constructional diversity correlated positively with TOEIC in both conditions (Manual r = .370; Whisper r = .335), and an interaction model showed no evidence of a slope difference per 1 SD increase in constructional diversity (b = −0.758, 95% CI [−34.233, 32.718], p = .964). Conversely, AIC-based stepwise models using construction proportions showed modest fit (Manual R² = .098; Whisper R² = .071), with the passive proportion as the only consistent significant predictor. Overall, Whisper appears viable for scalable CDA-based profiling via constructional diversity, whereas construction-level proficiency claims warrant caution in this dataset.

View full abstract

This study examines how individual prosodic cues contribute to the prediction of syntactic structure during English listening comprehension by Japanese EFL learners. Spoken language comprehension relies on multiple sources of information, and prosody plays a crucial role when segmental cues are insufficient. While previous research has often treated prosody as a single construct, the present study focuses on the differential roles of pitch, duration, and pauses in online sentence processing. Using a gating paradigm, Japanese university students listened to temporarily ambiguous English sentences based on the late/early closure principle. Speech stimuli were presented incrementally under two conditions: natural speech with intact prosody and speech with flattened fundamental frequency (F0). Participants judged sentence structure based on shared acoustic input, and accuracy data were analyzed using generalized linear mixed-effects models. The results revealed a significant interaction between prosodic condition and sentence structure at a critical gate. In the natural speech condition, learners utilized durational cues such as phrase-final lengthening to predict syntactic structure. In contrast, under flattened F0 conditions, the use of durational information was disrupted, and learners relied primarily on pauses, and no significant effect of listening proficiency was observed. These findings suggest that pitch information supports the perception and interpretation of durational cues, indicating that prosodic elements function interactively rather than independently in L2 sentence processing.

View full abstract

This practitioner article describes weekly optional touchpoints in a non-credit, fully asynchronous English e-learning program designed to support study initiation and persistence. The touchpoints were not instructional. Instead, they offered shared time through brief chat prompts and minimal co-presence, providing start cues and light accountability with minimal instructor workload. Thirty-eight participants consented to data use. Program completion was defined as achieving at least 80% progress on the required course content within the program period. Descriptively, completion rates increased with higher attendance. Logistic regression analyses showed that attendance was significantly associated with completion. Because attendance and total study time were correlated and could not be modeled together without unstable estimation, a reduced model treated attendance as the sole explanatory variable. The attendance-completion association remained statistically significant and was similar after adjustment for self-reported baseline TOEIC^® Listening & Reading test scores in the TOEIC-available subset. These findings suggest that the observed association is not explained solely by initial proficiency, although causal inference is not warranted. Practical implications for low-burden support in asynchronous e-learning programs are discussed.

View full abstract