Juan Zhang

Background: The pervasive integration of the internet into daily life has rendered Internet Use Disorders (IUDs) a particular threat to adolescents and young adults, whose vulnerability is amplified by a developmental stage characterized by ongoing neurodevelopment and self-exploration. This systematic review and meta-analysis aimed to evaluate the efficacy of mindfulness-based and mindfulness-informed interventions (MBIIs) for this population. Methods: Seven databases were searched from inception to April 2026 for randomized controlled trials (RCTs). Risk of bias was assessed using the Revised Cochrane Risk of Bias tool (RoB 2), data were synthesized via three-level random-effects models, and the certainty of evidence was evaluated using the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) framework. Results: From 28 included RCTs (26 for final analysis; 88 effect sizes), MBIIs yielded a significant overall reduction in IUDs ( d = −1.32, 95% CI [−1.73, −0.91], p < .001). Moderator analyses identified three significant factors influencing treatment efficacy: geographical region, measurement tool, and comparator condition. Discussion and conclusions: Our findings indicate that MBIIs represent an effective intervention for IUDs in youth, offering both therapeutic benefits and preventive potential. These results provide empirical evidence to inform clinicians and policymakers in developing targeted intervention strategies for IUDs.

Social-emotional development plays a key role in second language acquisition (SLA). With recent advancements in artificial intelligence (AI), there is increasing interest in how AI technologies can enhance social-emotional factors in SLA. This study aims to systematically synthesize empirical research on the role of AI in shaping social-emotional experiences in SLA, identify prevailing trends, summarize patterns in reported outcomes, examine proposed mechanisms, and highlight directions for future research. Following PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines, this scoping review encompasses 66 empirical studies published between 2022 and April 2025. All studies were independently reviewed by two researchers and appraised using the Mixed Methods Appraisal Tool (MMAT). Results: (1) a marked increase in scholarly attention to AI-supported social-emotional development in SLA; (2) AI’s reported potential in enhancing motivation and reducing anxiety, but evidence is limited by positivity bias, methodological inconsistencies, and concentration in East Asian contexts; and (3) five reported mechanisms through which AI may support social-emotional growth in SLA: dialogic interaction and perceived social presence; perceived competence, agency, and task value; emotion regulation and timely affective support; immersion, flow, and sustained engagement; and reduced evaluative pressure and psychological safety. At the same time, the review identifies insufficient critical attention to potentially negative or constraining aspects of AI-supported language learning, including learner over-reliance and reduced autonomy. This review offers a balanced synthesis of the current evidence and highlights implications for future AI–SLA research, pedagogy, and design.

Positive psychology (PP) has become a key perspective in second language acquisition (SLA), with growing attention to learners' emotions, motivation, engagement, well‐being, and self‐regulation. However, existing PP–SLA studies have not been sufficiently mapped in terms of how theoretical frameworks are selected, classified, and operationalized. This scoping review aims to map and classify publication characteristics (e.g., geographic distribution, journal rankings) and the theoretical frameworks and models that have guided the application of PP in SLA. Following PRISMA extension for Scoping Reviews (PRISMA‐ScR), the review was conducted in two stages. First, review, theoretical, and conceptual articles published between 2021 and 2025 were screened to identify higher‐level theoretical discussions in PP–SLA research. Second, empirical studies published in 2025 were examined to capture the most recent operational applications of these frameworks. A total of 38 studies met the inclusion criteria, comprising 6 review articles and 32 empirical studies. Findings indicate that PP functions as the broad conceptual umbrella of the field, while Broaden‐and‐Build Theory, Control‐Value Theory, Grit Theory, Flow Theory, Self‐Determination Theory, and Social Cognitive Theory represent the most frequently used specific theories/models. The review further classifies the seven major frameworks into four higher‐order categories: well‐being and flourishing frameworks, emotion‐centered theories, persistence and engagement theories, and motivational regulation theories. This scoping review contributes to PP–SLA research by moving beyond a list of individual theories and offering a hierarchical classification of theoretical framework use in the field. It also identifies key gaps concerning age‐sensitive theorization, digital and non‐formal learning contexts, cross‐cultural applicability, teacher‐related dimensions, and the need for stronger longitudinal and experimental designs.

The rapid proliferation of artificial intelligence (AI)-driven educational interventions has generated substantial interest in their effectiveness for vocabulary acquisition; however, the existing literature presents inconsistent findings. This meta-analysis, drawing on 28 studies and integrating 55 effect sizes via a three-level random effects model, demonstrates a significant overall positive effect of AI agents on vocabulary learning (effect size = 0.904, 95 % CI: 0.724, 1.084, p < 0.0001). Furthermore, moderator analysis identified significant moderating effects of both AI embodiment ( F (1, 53) = 7.446, p < 0.01) and intervention duration ( F (20, 24) = 5.563, p < 0.001). Grounded in the student–AI interaction framework, this meta-analysis provides a theory-informed synthesis of how AI-supported instructional designs relate to vocabulary learning outcomes.

Although a phonology-based coding system (i.e. Pinyin) is universally taught to beginning readers in mainland China, in Macau no such system is taught to children learning Cantonese. To examine whether providing such a system to Cantonese-speaking children is beneficial for reading development in both first (Cantonese) and second (English) language, the present study first attempted to implement a Cantonese phonology-based coding system (i.e. Jyutping) intervention with Cantonese-English bilingual children in Macau. Participants were 67 K3 children studying in a local kindergarten. Compared with the control group (N = 33, mean age 5.76 years), after five sessions of training, the children with Jyutping training (N = 34, mean age 5.85 years) showed a significant increase in Chinese and English phonological awareness at both syllable and phoneme levels. These results highlight the effectiveness of phonology-based coding systems in early literacy development and underscore the educational value of incorporating Jyutping instruction in kindergarten settings.

Autistic individuals exhibit differences in Theory of Mind (ToM) compared to neurotypical (NT) individuals. The aim of this study was to meta‐analyse the neural correlates that contributed to the manifestation of the expression differences in ToM between autistic individuals and the NT population. A total of 328 autistic participants and 314 NT participants from 18 studies were included. We adopted Activation Network Mapping, which is a novel neuroimaging meta‐analysis method based on activation seeds and functional connectivity to identify brain networks, to investigate how the ToM network of the autistic group differed from that of the NT group. The thalamus and precuneus robustly participated in the ToM network of the autistic group. Moreover, the temporoparietal junction and the right hemisphere of the limbic system, especially the thalamus, caudate, and cingulum, were less involved in the autistic group's ToM network, compared to the NT group. Our findings provide the first quantitative evidence supportive of the distinct patterns in the ToM brain network in the autistic population. The current findings indicate that the primary difference in ToM task performance in autistic individuals may stem from altered information processing mechanisms rather than deficits in core ToM abilities.

Given that existing academic buoyancy measures do not capture learners’ everyday capacity to cope with setbacks in the L2 learning, an L2-specific scale is needed to assess second language (L2) buoyancy. This study aimed to develop and validate the Second Language Buoyancy Scale (L2BS). Using convenience sampling, data were collected from 554 university students at two mainland Chinese institutions and randomly split into two equal subsets (n = 277 per subset). Content validity was established via qualitative item generation (17 interviews) and expert review (ICC = .83). For structural validity, EFA on Subset 1 (KMO = .826; Bartlett’s χ²(6) = 616.99, p < .001) supported a single-factor, four-item solution with loadings > .65; CFA on Subset 2 showed good fit (RMSEA = .096, CFI = .992, TLI = .977). Internal consistency was strong (Cronbach’s α = .898; McDonald’s ω = .898). Construct validity was supported by AVE = .689 and small-to-moderate correlations with academic buoyancy, growth mindset, grit, and conscientiousness. Criterion-related validity was evidenced by hierarchical regressions (incremental variance: ΔR² = .173 for L2 engagement; ΔR² = .160 for L2 enjoyment) and machine-learning models (Random Forest/XGBoost/LightGBM), in which L2BS consistently outperformed academic buoyancy (best accuracies: 73.21% for engagement; 64.29% for enjoyment). Overall, L2BS provides a brief, reliable, and valid measure of L2 buoyancy with clear utility for predicting key L2 outcomes such as L2 engagement and L2 enjoyment.

Educational practitioners increasingly adopt mindfulness-based interventions (MBIs) to alleviate language learners’ stress when acquiring a second language (L2). Nevertheless, the influence of such interventions on L2 learning apprehension has not been consistent across previous research. To address this, the current study employed a three-level random-effect meta-analysis to synthesize prior research and explore the efficacy and mechanisms of MBIs in reducing L2 anxiety. This meta-analysis incorporated 21 effect sizes from 11 studies, revealing a significant decrease in L2 anxiety among students receiving MBIs compared with those who did not (Cohen's d : 0.425, 95% CI [0.219, 0.631], p < .0001). No significant moderators were identified. The findings indicated that MBIs played a vital role in reducing anxiety among L2 learners.