Issue Information

This study examines how primary school teachers evaluate AI-supported managerial decisions through the lenses of justice, trust, transparency, and acceptance. Rather than treating artificial intelligence solely as a technological innovation, the study focuses on how algorithmic support interacts with institutional power relations, professional judgment, and the legitimacy of school-level decisions. A qualitative case study design was employed with six participants selected through criterion sampling. Data were generated in a 49-minute scenario-based focus group interview in which four hypothetical cases enabled comparison between traditional human-centered and AI-supported decisions. The transcripts were analyzed through qualitative content analysis in MAXQDA 2026, considering manifest and latent meanings together.

The findings reveal what this study conceptualizes as a legitimacy paradox. Teachers were attracted to AI-supported decisions because data-driven procedures appeared capable of reducing seniority pressure, favoritism, personal discretion, and other human sources of inconsistency. Yet the same decisions lost legitimacy when the data, criteria, or reasoning behind algorithmic outputs became opaque, when classifications risked stigmatizing students, or when professional responsibility appeared to shift from educators to the system. Justice and trust therefore operated as central evaluative criteria across both traditional and AI-supported scenarios. Participants favored neither autonomous algorithmic authority nor unstructured human discretion, but a hybrid decision-support model in which AI contributes consistency and analytical capacity while contextual, pedagogical, and ethical responsibility remains human. The study argues that legitimate AI-supported school management depends not only on technical accuracy but on explainability, contestability, equitable criteria, accountability, and meaningful human oversight.

Online psychological counseling has expanded rapidly over the past decade, particularly following the COVID-19 pandemic, yet the intellectual structure and dynamics of the field remain insufficiently mapped. This study provides a bibliometric and science mapping analysis of research on online psychological counseling indexed in the Web of Science Core Collection between 2010 and 2026. A total of 2,733 articles and review articles published in English were retrieved from the Social Sciences Citation Index, the Science Citation Index Expanded, and the Emerging Sources Citation Index, and were analyzed through performance analysis and science mapping using VOSviewer and the bibliometrix package in R. Annual output increased approximately tenfold, from 27 documents in 2010 to a peak of 280 in 2021, followed by a phase of consolidation characterized by stable annual production. The corpus accumulated 75,378 citations, with a mean of 27.6 citations per document. The Journal of Medical Internet Research was the leading source; Andersson, G. was the most productive and the most cited author; Karolinska Institutet was the leading institution; and the United States, Sweden, and Australia were the most productive countries. Approximately 29.5% of the documents were internationally co-authored, and international collaboration was associated with higher citation impact. In the thematic map, author keywords formed six clusters organized around two motor themes—internet-based cognitive behavioral therapy for depression and anxiety and its randomized evidence base—alongside a basic and transversal theme on telehealth service delivery, a niche theme on disorder-specific treatments, and peripheral themes addressing child and adolescent applications and evidence synthesis. COVID-19, digital mental health, teletherapy, digital health, artificial intelligence, and machine learning emerged as new themes after 2020. The findings delineate the knowledge base of the field, identify its most influential actors, and highlight directions for future research, counselor education, and digital mental health policy.

The increasing integration of artificial intelligence (AI) into global education raises two fundamental questions: “Whose Intelligence?” and “Whose Responsibility?” This article examines the epistemic, cultural, and pedagogical implications of AI-mediated education, focusing on whose knowledge and values are represented, authorized, and potentially privileged through AI systems. It argues that epistemic authority provides a necessary starting point for understanding accountability, since responsibility depends on decision-making authority and the capacity to intervene. To address this relationship, the research proposes the Distributed Educational AI Responsibility (DEAIR) Framework, a conceptual framework that maps five stakeholder levels—policymakers and regulators, developers and technology providers, educational institutions, educators and mediators, and students and learners—across six educational AI risk domains: algorithmic bias, data privacy, academic integrity, cultural exclusion, cognitive overreliance, and inequality of access. The framework distinguishes between primary operational and mediating or oversight roles, emphasizing that responsibility is distributed but differentiated rather than uniform or exclusive. DEAIR further positions human epistemic authority, pedagogical judgment, and institutional accountability as central to responsible educational AI governance. The study contributes a theoretically grounded framework for examining how responsibility can be allocated across multiple levels of intervention when AI becomes embedded in educational decision-making and practice.

This study examines the relationship between Turkish speaking and writing motivation among Syrian middle school students under temporary protection. Using a quantitative approach with a descriptive and correlational survey design, the study included 126 students selected through purposive sampling during the 2025–2026 academic year. Data were collected using a personal information form, the Speaking Motivation Scale for Learners of Turkish as a Second Language, and the Writing Motivation Scale. Data were analysed using descriptive statistics, independent-samples t-tests, one-way analysis of variance (ANOVA), and Pearson correlation analysis. The findings revealed no statistically significant differences in motivation scores based on gender or perceived Turkish language proficiency. However, extrinsic motivation differed significantly by age, while significant differences were observed across all motivation dimensions according to grade level. Fifth-grade students exhibited lower levels of motivation than students in certain higher grades. A strong positive correlation was found between intrinsic speaking motivation and writing motivation (r = .798). In addition, writing motivation was positively correlated with extrinsic motivation and negatively correlated with amotivation. Based on these findings, the study recommends integrating speaking and writing activities and developing supportive instructional practices specifically tailored to fifth-grade students.