Language Learning最新文献

As we work to advance the field of second language (L2) pronunciation teaching and learning, we look to other fields that have guided advancements in the teaching of other aspects of language learning. Fields such as corpus linguistics, pragmatics, instructed second language acquisition (SLA), psycholinguistics, technology, and assessment have served to inform and shape major changes in applied linguistics, and their impacts have been eminent in the field of L2 pronunciation teaching and learning as well. In fact, the current popularity of artificial intelligence (AI) and automated speech recognition (ASR) development have been particularly facilitative in this collaborative movement, as they offer vast opportunities in applications and practices in L2 pronunciation.

This cross-sectional study examines the sociopragmatic use of pitch and fluency features in requests among second language (L2) Chinese learners at two proficiency levels alongside native Chinese speakers. Twenty-eight L2 learners completed a 40-item oral discourse task with two types of request-making situations: (a) a high-imposition request to a higher status person and (b) a low-imposition request to an equal-status person. Similarly to native speakers, both higher proficiency and lower proficiency learners adjusted fluency features (e.g., speech rate, pauses) according to situational demands. However, only higher proficiency learners showed nativelike adjustment in pitch range, and neither group adjusted pitch value (F0) in different situations. These results suggest that the development of L2 learners’ sociopragmatic use of prosody is dependent on specific prosodic features.

A one-page Accessible Summary of this article in nontechnical language is freely available in the Supporting Information online and at https://oasis-database.org

Current L2 utterance fluency literature tends to operationalize disfluency as isolated, individual features. However, disfluency features often co-occur at one location or across multiple locations in one utterance. This study explores the co-occurrence of L2 disfluency features in a speech corpus from 71 L1 and L2 speakers of English across proficiency levels on an elicited imitation task and an oral listen-to-summarize task. We segmented each participant's speech into analysis of speech (AS) units (k = 2,704), extracted 2,972 individual disfluency chains based on 15 disfluency variables, and subjected them to principal components analysis and hierarchical-based K-means clustering analysis to identify disfluency co-occurrence patterns and speaker profiles across tasks. Results showed that different disfluency co-occurrences can be interpreted around various repair behaviors, and these repair behaviors also differ across tasks. Further analysis concerning the disfluency–proficiency relationship suggests that whereas some disfluency co-occurrences are meaningfully associated with proficiency, others might not be.

This paper examines outsourced call center interactions to illustrate how these contexts can enhance pronunciation analysis and training. Public opinion in the United States and the United Kingdom regarding the perceived “pronunciation problems” of agents based in call centers in Outer-Circle English-speaking countries is typically negative. However, it is often difficult for researchers to pinpoint the specific issues involved, as access to authentic calls is scarce. This paper reports an investigation into the role that the differing use of prosodic conventions can play in call center interactions recorded in the Philippines between Filipinos and North Americans. A microethnographic analysis of call center data focused on prosodic features of interaction suggests that, where conflict occurs, it is mirrored in the prosodic features of the interaction. This has important implications for modeling effective interaction and training for high-stakes contexts.

A one-page Accessible Summary of this article in nontechnical language is freely available in the online Supporting Information and at https://oasis-database.org

The rising popularity of audiobooks in language learning has highlighted the need to understand their potential benefits in enhancing comprehension and the mechanisms driving these effects. In this registered report, we explored the hypothesis that reading-while-listening can enhance lower-level decoding skills, in turn freeing up cognitive resources to support comprehension. In a within-participant design, eighty-six intermediate-to-advanced Chinese learners of English read, read and listened, and listened to different excerpts of a novel. Contrary to our preregistered hypotheses, participants comprehended the text less well when reading-while-listening than when reading it silently. Both reading conditions yielded better comprehension than listening-only. Individual differences in segmentation skills and orthographic decoding supported comprehension but did not interact with the input condition in the way that theory would predict. These results confirm the importance of orthographic decoding and speech segmentation for comprehension, but they also point to gaps in theoretical understanding of reading-while-listening's presumed pedagogical advantages for comprehension.

This study expands on the practical application of the critical role of auditory processing in the rate of naturalistic L2 speech acquisition. In Study 1, the prosodic production of English by 46 Chinese college students was tracked over a five-month study abroad program in the UK. Learners with extensive L2 input opportunities demonstrated improvements in prosodic accuracy; however, those with pitch acuity below a certain threshold showed regression, potentially reinforcing L1 interference. To determine what percentage of participants fell below the auditory processing threshold determined in Study 1, Study 2 administered pitch processing tests to 400 Chinese college students learning English, all with normal hearing, and developed a provisional corpus to assess pitch acuity variation within this cohort. The comparison of findings from Studies 1 and 2 suggests that insufficient auditory precision hampers naturalistic L2 learning. Approximately the bottom 1.5 quartiles of the population (35%) may fall below this threshold. These learners could benefit from remedial strategies (e.g., explicit phonetic instruction, auditory training) to fully capitalize on their naturalistic L2 learning opportunities.

This study investigated the use of Artificial Intelligence (AI) models and signal detection processes to generate meaningful visual and ChatGPT-like narrative feedback on second language (L2) English intelligibility. To test the effects and perceptions of such techniques, three groups of learners (N = 90) received visual and narrative feedback (n = 30), visual-only feedback (n = 29), and no feedback (n = 31) in an online self-paced intervention with explicit instruction on segmental and suprasegmental features of intelligibility. Pre/postspeaking tasks were evaluated by raters for intelligibility, comprehensibility, and accentedness, as well as segmental and suprasegmental accuracy, in scripted and spontaneous speech. The results indicate that visual feedback improves prominence production, but only those participants who also received the narrative (i.e., ChatGPT) feedback improved in two of the three prosodic features and in intelligibility. However, those who received narrative feedback had the lowest perceptions of the practice activity helpfulness. Implications for the use and improvement of AI-based pronunciation feedback are provided.

The present study investigates whether congruency facilitation and figurative interference—two counteractive effects—persist in L2 collocational processing when both congruency and figurativeness are present. A primed lexical decision task was administered to 44 L1-Chinese L2-English learners and 40 L1-English speakers to assess response times for figurative congruent collocations, along with their matched literal congruent and figurative incongruent collocations. Results showed that while collocational priming was absent, both congruency facilitation and figurative interference emerged, with their effects modulated by L2 proficiency. Specifically, in low-proficiency learners, congruency facilitation appeared to outweigh figurative interference, whereas in high-proficiency learners, figurative interference became more pronounced as L1-based facilitation was suppressed. These findings suggest that L2 learners initially rely on their activated L1 semantic network but gradually shift toward developing L2 collocational representations as proficiency increases, though these representations may remain weak and insufficient to facilitate collocate access.

This study explores how word frequency affects verb-mediated prediction in L1 and L2 speakers, using a visual-world eye-tracking task. By manipulating frequency of nouns within subjects (higher; lower) and type of verbs used as predictive cues (semantically restrictive; neutral) in sentences (e.g., The {doctor/surgeon} {opened/moved} the box), we investigated the impact of frequency of early-processed words on prediction, hypothesizing that higher-frequency words might free up cognitive resources, thus facilitating lexical retrieval, integration with a subsequent predictive cue, and ultimately prediction. Results showed that both L1 and L2 speakers predicted the target object upon hearing restrictive verbs. However, the L2 group showed such a predictive behavior only when the sentences contained higher-frequency subjects, whereas the L1 group did so in both conditions but faster with higher-frequency subjects. These results suggest L2 learners’ sound ability to generate predictions, and underscore the importance of word frequency in facilitating both L1 and L2 prediction.