
Deaf and Hard-of-Hearing (DHH) learners persistently face academic disparities due to communication barriers and insufficient pedagogical adaptations. While AI-based Tutoring Systems (AITS) offer promising adaptive and multimodal solutions, their efficacy remains quantitatively underexplored. This systematic review and meta-analysis synthesizes empirical evidence on the effectiveness of AITS for DHH learners compared to conventional instruction. Following Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines, we searched five databases (Web of Science, Scopus, ERIC, PsycINFO, IEEE Xplore) and included 18 studies with 1,245 participants. Effect sizes were pooled using random-effects models. AITS demonstrated a moderate, significant positive effect on academic outcomes (Hedges’ g = 0.48 , 95 g = 0.55 ) and ASL-based systems ( g = 0.61 ). High heterogeneity ( I^2 = 67% ) reflected diverse AI implementations. AI-powered tutoring systems significantly enhance learning for DHH students, particularly when integrating sign language recognition and adaptive scaffolding. Future development should prioritize mature, communication-first platforms to bridge the educational divide.
This study explores how a physics professor’s understanding of teaching for conceptual change evolves through sustained collaboration with a science education professor. Using a phenomenographic case study approach, we analyze shifts in the professor’s conceptions over time, revealing key transitions from viewing student misconceptions as barriers to recognizing conceptual frameworks as bridges for learning. Findings highlight the role of cognitive conflict, conceptual resources, and metacognitive engagement in shaping pedagogical change. The study contributes to understanding faculty development in STEM education, emphasizing the impact of interdisciplinary collaboration on instructional transformation. Based on these findings, future research and policy recommendations are proposed.
Divine forgiveness is central to many religious traditions and is often thought to inform interpersonal forgiveness. However, little is known about how daily experiences of divine forgiveness shape real-time conflict within romantic relationships. Using a 14-day dyadic daily diary design, this study examined whether perceptions of divine forgiveness regarding treatment of one's partner predict forgiveness, amend-making, and anger in different-gender, cohabiting romantic couples (N = 82 dyads; 87.8% married; 77.4% White; 83.5% Christian/Catholic). Multilevel models that parse between- and within-person variance in divine forgiveness revealed that on days when individuals reported greater divine forgiveness, they also reported more forgiveness and amend-making toward their partner, suggesting that experiences of divine forgiveness may support same-day relational repair. Between-person effects further revealed that women experiencing higher levels of God's forgiveness reported making more amends for their behavior, while men higher in divine forgiveness reported being more forgiving; these effects were corroborated by partner reports and extended to reports of these outcomes the following day. Interestingly, divine forgiveness was not associated with self-reported anger but showed mixed effects regarding perceptions of partner anger. Together, findings suggest that divine forgiveness operates both as a dispositional resource and as a dynamic process that supports relationship repair, particularly for minor, daily conflicts.
Co-rumination is a social emotion regulation strategy characterized by extensive and exhaustive discussions of stressors and negative emotions with another person. While research establishing the costs and benefits associated with co-rumination was formative, the narrow focus on explaining heightened internalizing symptoms and increased relationship quality (i.e., the “trade-off hypothesis”) in mostly adolescents has precluded important advancements. Through a social psychological lens, we pave a way forward by offering a new theoretical conceptualization with which to view co-rumination that emphasizes the need to 1) explicitly consider the dyadic or social nature of co-rumination, 2) examine the heterogeneous content of co-ruminative discussions that extends beyond individual-level stressors, 3) directly assess the goals of co-ruminative conversations, and 4) consider with whom people co-ruminate. We then connect this theory of co-rumination to influential theories and concepts across psychological subdisciplines to show how this conceptualization of co-rumination can be further advanced when studied through a multidisciplinary perspective. This novel theoretical conceptualization and multidisciplinary application advances beyond the trade-off hypothesis by calling on researchers to consider the nuances of co-rumination across diverse populations and contexts. Re-orienting co-rumination as such may prove fruitful to better understand difficult and upsetting conversations with close others, ultimately improving recommendations for how individuals seek and provide support through challenging times.
This study presents a real-time, bidirectional system for translating American Sign Language (ASL) to and from English using computer vision and transformer-based models to enhance accessibility for deaf and hard of hearing users. Leveraging publicly available sign language and text–to-gloss datasets, the system integrates MediaPipe-based holistic landmark extraction with CNN- and transformer-based architectures to support translation across video, text, and speech modalities within a web-based interface. In the ASL-to-English direction, the sign-to-gloss model achieves a 25.17% word error rate (WER) on the RWTH-PHOENIX-Weather 2014T benchmark, which is competitive with recent continuous sign language recognition systems, and the gloss-level translation attains a ROUGE-L score of 79.89, indicating strong preservation of sign content and ordering. In the reverse English-to-ASL direction, the English-to-Gloss transformer trained on ASLG-PC12 achieves a ROUGE-L score of 96.00, demonstrating high-fidelity gloss sequence generation suitable for landmark-based ASL animation. These results highlight a favorable accuracy-efficiency trade-off achieved through compact model architectures and low-latency decoding, supporting practical real-time deployment.