AI and Tech in Behavioral and Social Sciences

Abstract
The rapid growth of digital imaging technologies has made high-quality visual data increasingly accessible; however, the storage, transmission, and restoration of high-resolution images remain challenging in bandwidth-limited and resource-constrained environments. Although compression methods reduce file size, they may remove critical details required for scientific, medical, remote-sensing, and security applications. To address this limitation, this study proposes an attention-guided dynamic ensemble framework for Single Image Super-Resolution (SISR). The proposed method integrates several representative super-resolution models, including LapSRN, SRResNet, ResNeXt-based SR, SRCNN/FSRCNN, and ESPCN, and uses an attention-guided selection module to assign the most suitable model to different image regions based on local characteristics such as edges, textures, and smooth areas. The selected outputs are then fused by a convolutional integration network to generate the final high-resolution image. Experiments on DIV2K and BSDS300 show that the proposed method improves reconstruction quality, particularly in terms of structural similarity and texture preservation. On DIV2K, the proposed method achieved 33.40 dB PSNR and 0.9172 SSIM; on BSDS300, it achieved 28.13 dB PSNR and 0.8497 SSIM. These findings indicate that dynamic model selection can reduce the limitations of individual super-resolution models and improve detail recovery in feature-diverse images.

This quasi-experimental study examined the effectiveness of computerized cognitive rehabilitation in improving academic engagement among lower-secondary students experiencing academic underachievement. The study used a pre-test/post-test design with an experimental group and a control group. The statistical population consisted of lower-secondary students with academic underachievement in Tehran during the first semester of the 2023-2024 academic year. Thirty students were selected through convenience sampling and randomly assigned to an experimental group (n = 15) and a control group (n = 15). Academic engagement was measured using the Zarang Academic Engagement Questionnaire, which assesses cognitive, motivational, and behavioral engagement. The experimental group completed a 12-session computerized cognitive rehabilitation program using Captain's Log version 14, delivered twice weekly and designed to train 22 foundational and higher-order cognitive skills, including focused and sustained attention, divided and selective attention, working memory, auditory and visual processing speed, response inhibition, problem solving, and logical reasoning. The control group did not receive the computerized intervention during the study period. Data were analyzed with analysis of covariance after checking the main statistical assumptions. The experimental group showed a marked increase in academic engagement from pre-test (M = 80.86, SD = 9.87) to post-test (M = 103.26, SD = 11.10), whereas the control group showed only a small descriptive increase from pre-test (M = 86.46, SD = 11.36) to post-test (M = 88.80, SD = 10.00). ANCOVA indicated a statistically significant group effect after controlling for pre-test scores, F(1, 27) = 7.11, p = .013, partial eta squared = .208. These findings suggest that computerized cognitive rehabilitation may be a useful school-based adjunct intervention for strengthening academic engagement among students with academic underachievement, although future studies with larger samples, active control conditions, and follow-up assessments are needed.

Aggression during the early school years interferes with peer relationships, classroom adjustment, and academic functioning. School-based social-emotional learning (SEL) programs can reduce behavioral problems, yet children may benefit most from formats that are developmentally engaging and expressive. This study examined whether a digital art-based SEL program could reduce aggression in second-grade students.A quasi-experimental pretest-posttest design with a control group was used. Fifty-two second-grade students from public elementary schools in Mazandaran Province, Iran, were selected through cluster sampling and assigned to an experimental group (n = 28) or a control group (n = 24). Aggression was assessed at baseline and post-intervention using the Persian version of the Buss-Perry Aggression Questionnaire. The experimental group participated in an eight-week intervention comprising 16 sessions (two 60-minute sessions per week) integrating CASEL-based SEL competencies with digital art activities, whereas the control group continued routine instruction. Descriptive statistics and mixed-design ANOVA were used for analysis. The groups were comparable at baseline on demographic characteristics and aggression scores. From pretest to posttest, the experimental group showed marked reductions in total aggression (47.85 ± 5.20 to 31.80 ± 4.50), physical aggression (12.45 ± 2.10 to 8.30 ± 1.85), verbal aggression (11.80 ± 2.40 to 7.50 ± 1.90), anger (13.10 ± 2.60 to 9.20 ± 2.10), and hostility (10.50 ± 2.30 to 6.80 ± 1.70), whereas changes in the control group were small. Mixed ANOVA showed a significant group × time interaction for total aggression, F(1, 50) = 26.07, p = .001, η² = .34, as well as for physical aggression, F(1, 50) = 22.15, p = .001, η² = .31; verbal aggression, F(1, 50) = 18.92, p = .001, η² = .27; anger, F(1, 50) = 24.56, p = .001, η² = .33; and hostility, F(1, 50) = 19.45, p = .001, η² = .28. A digital art-based SEL program appears to be an effective and engaging school-based approach for reducing aggression in elementary school students. Larger multi-site studies with follow-up assessments are recommended.

Objective: Chronic depression accompanied by high self-criticism remains difficult to treat and is often associated with persistent functional impairment. This study examined whether Cognitive Analytic Therapy (CAT), delivered with digital support tools, reduced depressive symptoms and self-criticism in adults diagnosed with persistent depressive disorder (PDD). Method: The study used a randomized pretest-posttest-follow-up design. Sixty participants with PDD and elevated self-criticism were allocated to either CAT (16 weekly sessions) or a waitlist treatment-as-usual (TAU) condition. Outcomes were assessed with the Beck Depression Inventory-II (BDI-II) and the Levels of Self-Criticism Scale (LSCS) at baseline, post-treatment, and 2-month follow-up. Data were analyzed using mixed-design analysis of variance. Results: Significant Group x Time interaction effects were observed for depression, F(2, 116) = 45.32, p < .001, eta_p^2 = .438, and self-criticism, F(2, 116) = 52.18, p < .001, eta_p^2 = .473. In the CAT group, BDI-II scores declined from 32.45 (SD = 4.12) at baseline to 18.20 (SD = 3.85) at post-test and 17.10 (SD = 3.90) at follow-up, whereas the TAU group showed only modest change. LSCS scores in the CAT group declined from 85.60 (SD = 8.45) to 58.30 (SD = 7.20) and 55.40 (SD = 6.95), while the TAU group remained comparatively stable. Conclusion: Within the limits of the present design, CAT supplemented with digital monitoring and self-help tools was associated with substantial and sustained reductions in depressive symptoms and self-critical thinking. The findings support CAT as a promising intervention for chronic depressive presentations marked by harsh self-relating.

Eating disorders (EDs) and body image disturbance remain major mental health concerns among adult women, and the persistence of shame, self-criticism, and relapse after standard treatment highlights the need for compassion-based interventions. This study investigated the effectiveness of Compassion-Focused Therapy (CFT) integrated with digital tools in improving body appreciation and self-compassion while reducing eating disorder psychopathology in adult women with disordered eating symptoms. Using a semi-experimental pretest–posttest design with a two-month follow-up, 60 women from Tehran, Iran, were randomly assigned to either an intervention group receiving 12 weekly group CFT sessions with digital support or a waitlist control group; 53 participants completed the study (26 in the intervention group and 27 in the control group). Outcomes were assessed using the Body Appreciation Scale-2 (BAS-2), the Eating Disorder Examination Questionnaire (EDE-Q), and the Self-Compassion Scale–Short Form (SCS-SF), and the data were analyzed using mixed ANOVA. The findings showed significant Time × Group interaction effects for all three outcomes, indicating superior improvement in the intervention group relative to controls. Body appreciation increased from 2.85 ± 0.71 at pretest to 3.92 ± 0.64 at posttest and remained stable at 3.88 ± 0.68 at follow-up, with a significant interaction effect, F(1.68, 85.64) = 42.35, p < 0.001, η² = 0.45. Eating disorder psychopathology declined markedly from 3.65 ± 0.89 to 2.10 ± 0.75 and was maintained at 2.15 ± 0.78, F(1.72, 87.72) = 55.18, p < 0.001, η² = 0.52. Self-compassion improved from 2.45 ± 0.62 to 3.55 ± 0.58 and remained at 3.50 ± 0.60, F(1.85, 94.35) = 38.92, p < 0.001, η² = 0.43. In contrast, the waitlist group showed no meaningful change across time. Overall, digitally supported CFT appears to be an effective and durable intervention for improving body image and reducing eating disorder symptoms in adult women.

This study presents a model for designing intelligent learning ecosystems that enhance the quality of digital education through the integration of artificial intelligence and blended learning, with Islamic Azad University as the empirical context. The research addresses persistent challenges in e-learning, including limited interaction, unequal access, and the need to respond to diverse learner profiles through technology-enhanced educational design. A mixed-methods approach was employed. In the qualitative phase, semi-structured interviews were conducted with 15 experts in education and educational technology selected through purposive sampling. In the quantitative phase, data were collected from 384 faculty members and university staff using stratified random sampling across regions and academic fields. Qualitative data were analyzed through thematic analysis, while quantitative data were examined using Partial Least Squares Structural Equation Modeling (PLS-SEM), Artificial Neural Networks (ANN), and the MABAC multi-criteria decision-making method. Findings revealed that the proposed ecosystem is built around three core dimensions: blended learning, artificial intelligence capabilities, and digital education quality. Blended learning was defined through flexibility, interaction, personalization, and infrastructure, while AI capabilities included educational data analysis, intelligent recommendation, intelligent support, and automated assessment. The quality of digital education was reflected in learner satisfaction, learning effectiveness, and educational interaction. The model demonstrated strong explanatory power (R² = 0.712). ANN results identified learner satisfaction and learning effectiveness as the most influential indicators, and MABAC ranked intelligent support as the highest-priority AI capability. The study concludes that integrating AI-driven support into blended learning environments can provide a practical pathway for strengthening digital education quality and informing future policy and implementation in higher education.

The purpose of this study was to design a qualitative model of school principals’ performance in the Department of Education of Karaj County based on a meritocracy approach. The research method was qualitative and based on grounded theory. The participants included organizational and academic experts related to the research topic, who were selected through purposive sampling. A total of 15 participants were interviewed until theoretical saturation was achieved. Data were collected through semi-structured interviews and analyzed using open, axial, and selective coding. The findings indicated that the qualitative model of school principals’ performance in the Department of Education of Karaj County with a meritocracy approach consists of six dimensions: talent attraction and retention, sufficient job-related information, support for the meritocracy process, awareness and analysis of international educational systems, knowledge management, and facilitation of organizational learning. In addition, 20 components were identified, including creating a platform for knowledge production, improving the psychological climate and organizational atmosphere, increasing efficiency, optimizing processes, better resource management, achieving improved student outcomes, enhancing the learning environment, improving educational quality, increasing educational equity, enhancing the effectiveness and productivity of the educational system, reducing costs, moving toward strategic goals, viewing the school as a learning organization, preventing rent-seeking, promoting meritocracy, developing a culture of citizenship and social responsibility, strengthening trust in the educational system, fostering creativity and innovation, and increasing staff commitment. Overall, 168 indicators were extracted and categorized within the paradigmatic model of the research in the form of causal conditions, contextual conditions, intervening conditions, strategies, and consequences.

The present study aimed to examine scientific authority indicators in Iranian sport management journals and compare them with international standards. A mixed-method design was employed. In the qualitative phase, semi-structured interviews were conducted with 15 experts in scholarly publishing and sport management. Thematic analysis identified three core dimensions influencing scientific authority: content quality, structural quality, and developmental strategies. In the quantitative phase, seven bibliometric indicators—impact factor, H-index, total citations, cited half-life, acceptance rate, international collaboration, and indexing status—were analyzed across 13 Iranian and 11 international sport management journals. Independent t-tests were used to compare groups. Results revealed significant differences in impact factor (0.48 vs. 2.70, p < .01), H-index (8.5 vs. 56.8, p < .01), and acceptance rate (30.7% vs. 20.9%, p < .05), indicating a performance gap. Differences in cited half-life and international collaboration were not statistically significant. Findings suggest that while Iranian journals demonstrate growth in publication volume, structural and citation-based authority indicators remain comparatively lower. Policy recommendations include strengthening peer-review rigor, improving internationalization strategies, enhancing English-language publishing, and aligning evaluation practices with international frameworks such as DORA and the Leiden Manifesto. This study provides evidence-based guidance for improving the global visibility and scientific authority of sport management journals.