Nikola Sekulovski, Meike Waaijers, Giuseppe Arena • PsyArXiv
Finding: This paper introduces a framework using large language models to automatically generate prior probabilities for Bayesian network modeling in psychology, specifically tested on PTSD symptom networks.
Why it matters: Demonstrates how AI can augment human decision-making in research methodology by automating the challenging task of specifying informed statistical priors.
Method: Uses LLMs to provide inclusion judgments for variable pairs, converts these to edge-specific probabilities, and validates against expert-derived networks with an R package implementation.
Jargon: Bayesian Graphical Modeling - statistical method for modeling relationships between variables as networks; Prior elicitation - process of specifying probability distributions that represent beliefs before seeing data.
Finding: The CIRCUS framework, a structured 6-stage gamification model for higher education, significantly improved student learning outcomes (from 68% to 82% performance) and engagement (89% earned all badges, 94% completed embedded games) when integrated into university courses.
Why it matters: This provides educators with a validated, systematic framework for implementing gamification in learning environments, addressing the common problem of fragmented gamification approaches in educational settings.
Method: Mixed-methods R&D approach with expert validation (Aiken's V = 0.87), practitioner assessment by 18 lecturers, and field trial with 62 undergraduate students across two technical courses.
Jargon: CIRCUS = Consider Needs, Inspect Content, Regulate Objective, Construct Prototype, Utilize Prototype, Summing-Up; Aiken's V = statistical measure of expert agreement on validity.
⚠️ Self-reported data only | No control group mentioned | Single study (not replicated) | Convenience sample (e.g., MTurk, students) | Limited statistical detail in abstract
Finding: This paper introduces a Bayesian framework that separates individual abilities from collaborative abilities in human-AI interactions, finding that users who can adapt to AI perspectives achieve better collaborative performance but not better individual performance.
Why it matters: This provides a principled method for measuring and understanding the distinct skills needed for effective human-AI collaboration, directly informing how to design AI systems and train users for better teamwork.
Method: Novel Bayesian Item Response Theory framework validated on human-AI benchmark data (n=667) that decomposes additive synergy from non-linear collaboration dynamics.
Jargon: Bayesian Item Response Theory - statistical method that models performance based on both person ability and task difficulty; additive synergy - benefits that simply add together vs. non-linear dynamics - complex interactive effects.
Finding: Cognitive reflection (measured by Cognitive Reflection Test) consistently predicts better performance on classic judgment bias tasks like anchoring, availability, and representativeness across two large studies, while self-reported thinking styles and trait self-control show weak associations with bias susceptibility.
Why it matters: This provides concrete evidence for which individual difference measures actually predict susceptibility to cognitive biases, directly informing behavioral science education and research on decision-making processes.
Method: Two complementary large-scale studies (N=1,071 and N=480) examined multiple measures of cognitive reflection, thinking styles, and self-control across classic heuristics-and-biases tasks, with Study 1 integrating nearly a decade of data and Study 2 using preregistered methodology.
Jargon: CRT (Cognitive Reflection Test) - measures tendency to override intuitive responses with deliberative thinking; anchoring bias - over-relying on first piece of information; availability heuristic - judging probability by ease of recall; representativeness heuristic - judging probability by similarity to mental prototypes.
⭐ Pre-registered | ⭐ Includes replication ⚠️ Self-reported data only | No control group mentioned
K. Atirbek, Roza Kadirbayeva, Ş. Dost et al. (4 authors) • Semantic Scholar
Finding: A structured two-rotation blended learning model that combines guided AI-supported activities with systematic analysis of AI outputs significantly improved pre-service teachers' digital pedagogy (large effect, d=0.93) and shifted them from uncritical acceptance to systematic verification of AI-generated content.
Why it matters: Demonstrates how to design educational interventions that promote critical thinking about AI tools rather than passive reliance, directly relevant to understanding AI's impact on learning and skill acquisition.
Method: Quasi-experimental pre-post design with 94 participants, comparing an 8-week structured intervention (guided AI use followed by systematic AI output analysis) against a control group, using both quantitative measures and qualitative analysis of reflections.
Jargon: Digital-pedagogical competence = teachers' ability to effectively integrate digital tools into their teaching practice.
⚠️ Small sample (n=47) | Single study (not replicated)
Lowie V. Libongcogon, Glen Mae A. Ferrer, Arlene Jane A. Dionaldo et al. (5 authors) • Semantic Scholar
Finding: The study presents a Hybrid Play-Game-Based Learning Framework that combines Play-Based Learning and Game-Based Learning principles with ADDIE instructional design to create an AI-enhanced mobile app for early English literacy, showing high usability among young learners.
Why it matters: This demonstrates how to systematically integrate gamification and play elements into educational technology design, offering a concrete framework for HCI researchers developing learning applications and understanding child-computer interaction.
Method: Used descriptive-developmental design following the ADDIE model (Analysis, Design, Development, Implementation, Evaluation) to guide the app creation and testing process.
Jargon: ADDIE model - systematic instructional design framework with five phases; PBL/GBL - Play-Based Learning emphasizes child-led exploration, Game-Based Learning uses structured gameplay mechanics.
Roland Erich Uriko, Myron Tsikandilakis, Jaan Valsiner • PsyArXiv
Finding: Measuring only final emotional outcomes (like valence ratings) misses crucial psychological processes - for instance, visual-emotional ambivalence produces neutral ratings identical to genuine indifference, despite being functionally different states.
Why it matters: This challenges fundamental assumptions in HCI user experience measurement and behavioral research, suggesting current evaluation methods may miss critical information about how users actually experience and process interfaces or make decisions.
Method: Presents a novel real-time methodology for capturing evaluative dynamics as they unfold, rather than just measuring endpoints.
Jargon: Visual-emotional ambivalence = experiencing a stimulus as simultaneously positive and negative; Evaluative dynamics = the process of how evaluations change and develop over time.
⚠️ No control group mentioned | Single study (not replicated) | Limited statistical detail in abstract
Silvia Seghezzi, Maia Juliet Armstrong, Patrick Haggard • PsyArXiv
Finding: When people plan actions, they form memory traces not only of what they actually did, but also of alternative actions they considered but didn't take, leading to false memories where they believe they previously saw configurations from unchosen solution paths.
Why it matters: This reveals how decision-making processes create persistent mental representations that can bias future judgments, which is crucial for understanding cognitive biases in design thinking, team decision-making, and how people retrospectively evaluate their choices.
Method: Used a modified Tower of London puzzle task where participants solved problems with two equivalent optimal paths, then tested recognition memory for configurations from chosen paths, unchosen paths, and novel configurations.
Jargon: Tower of London task - a classic problem-solving puzzle used to study planning and executive function; Signal detection analysis - statistical method to separate true recognition ability from response bias.
⭐ Includes replication ⚠️ No control group mentioned | Limited statistical detail in abstract
Fredi Ganda Putra, S. Sutiarso, Nurhanurawati Nurhanurawati et al. (6 authors) • Semantic Scholar
Finding: The DECADE learning model synthesizes five major learning theories (cognitivism, constructivism, information processing, dual coding, and cognitive load theory) into a unified six-phase instructional framework for mathematics education, achieving high expert validation scores (>90%) for theoretical coherence and structural validity.
Why it matters: This addresses a critical issue in learning sciences where educational theories are often applied in isolation, providing a validated framework that could inform instructional design across domains beyond mathematics.
Method: Mixed-method developmental design using qualitative thematic analysis to map theoretical foundations followed by quantitative expert judgment validation of the integrated framework.
Jargon: Dual coding theory - the idea that information is better processed when presented through both visual and verbal channels; Cognitive load theory - framework explaining how working memory limitations affect learning.
Han Guo, B. Alias, Jamalullail Abdul Wahab • Semantic Scholar
Finding: Clear goal setting, effective curriculum coordination, and promoting professional development opportunities significantly improve teacher job satisfaction, while high visibility and student progress monitoring do not significantly influence satisfaction among teachers in Northwest China.
Why it matters: Provides empirical evidence on which leadership practices most effectively support teacher motivation and satisfaction, directly informing goal-setting strategies and team dynamics in educational settings.
Method: Correlational study analyzing questionnaire responses from 372 randomly selected junior high school teachers in Ningxia, China.
Jargon: Instructional leadership - leadership focused on improving teaching and learning through curriculum coordination, goal-setting, and professional development rather than just administrative tasks.
Finding: A Problem-Based Learning e-module incorporating local ecological content showed high expert validation and user satisfaction but failed to produce statistically significant improvements in critical thinking or science literacy compared to traditional methods in a four-meeting implementation.
Why it matters: This demonstrates the importance of implementation duration and prior exposure when designing digital learning interventions, providing valuable insights for educational technology design and pedagogical effectiveness research.
Method: Quasi-experimental pre-test/post-test design with 89 students across three conditions, using ADDIE instructional design model and expert validation process.
Rocío Segura Nebot, Milagros Sáinz Ibáñez, Soledad de Lemus et al. (4 authors) • PsyArXiv
Finding: Children ages 5-6 show significant social distance toward counter-stereotypical peers (especially boys who play with "feminine" toys), with rejection driven by gender stereotypes, aggression perceptions, and preference differences.
Why it matters: This reveals how gender stereotypes create early barriers to inclusive play and collaboration, informing design of educational games and team activities that could reduce bias and promote diverse interaction patterns.
Method: Mixed-methods study with 258 Spanish primary school students examining toy preferences, peer selection, and reasons for rejection through both quantitative measures and qualitative analysis.
Jargon: Gender Backlash = social penalties for violating gender norms; Black Sheep Effect = rejection of group members who deviate from group standards; Social distance = degree of separation or reluctance to interact with others.
⚠️ No control group mentioned | Single study (not replicated) | Limited statistical detail in abstract
Titiworada Polyiem, P. Nuangchalerm, Pongthorn Asawaniwed • Semantic Scholar
Finding: A structured mentorship program with four components (professional learning teams, collaborative teaching design, mentored practice, and reflective exchange) significantly improved novice teachers' instructional capabilities and pedagogical competencies.
Why it matters: This provides evidence-based insights into effective mentorship design and collaborative learning structures that could inform team dynamics research and skill acquisition in educational contexts.
Method: Mixed-methods study with 97 teachers across 15 disciplines, combining quantitative assessment scores with qualitative analysis of mentorship experiences.
Lanqing Ye, Yaoping Liu, Wannaporn Siripala et al. (5 authors) • Semantic Scholar
Finding: The "Xindi Applied Piano Pedagogy" integrates technology-enabled instructional design with cultural elements and applies deschooling theory to encourage creative development in music education, using a three-step teacher training model and PDCA cycles for continuous improvement.
Why it matters: This demonstrates how technology can be integrated with pedagogical frameworks to foster creativity in educational settings, relevant to learning sciences and creative skill acquisition.
Method: Qualitative study using in-depth interviews with 12 students across three schools and grounded theory analysis with three-level coding.
Jargon: Deschooling theory - educational philosophy that questions traditional schooling methods and promotes alternative learning approaches; PDCA cycles - Plan-Do-Check-Act methodology for iterative quality improvement.
Jose G. Tan, Jr., Blessa Kay F. Caballero • Semantic Scholar
Finding: Teachers actively use stemming and lemmatization (breaking words down to root forms) as instructional tools in English classrooms, which effectively helps students understand and retain new vocabulary.
Why it matters: This demonstrates how linguistic decomposition strategies can enhance learning and skill acquisition, providing insights for designing educational interfaces and learning tools.
Method: Qualitative study using classroom observations and interviews with nine university-level English teachers.
Jargon: Stemming = reducing words to their root form (e.g., "running" → "run"); Lemmatization = reducing words to their dictionary base form considering context and meaning.
⚠️ No control group mentioned | Single study (not replicated) | Limited statistical detail in abstract
Prof. (Dr.) Shailaj Kumar Shrivastava, Chandan Shrivastava • Semantic Scholar
Finding: The paper examines ethical concerns around generative AI in content creation, including deepfakes, bias, privacy, copyright issues, and author replacement, while advocating for transparency, human verification, and responsible disclosure practices.
Why it matters: This directly addresses critical questions about AI's impact on creative work and human collaboration that are essential for understanding how AI tools are reshaping design, content creation, and professional practice.
Jargon: Deepfakes - AI-generated fake videos or images; Gen AI - Generative Artificial Intelligence systems that create content.
Arifin, I. Jampel, D. Sanjaya et al. (4 authors) • Semantic Scholar
Finding: School principals in remote areas adapt their instructional leadership through contextual strategies like strengthening internal solidarity, self-training, and focusing on educational service sustainability despite structural and resource limitations.
Why it matters: Demonstrates how leadership and collaboration strategies adapt under resource constraints, which is relevant for understanding team dynamics and adaptive behavior in challenging environments.
Method: Multi-case qualitative study across three schools using interviews, observations, and thematic analysis.
Jargon: Instructional leadership - educational leadership focused on improving teaching and learning processes rather than just administrative management.
Finding: A bibliometric analysis of visual search research found that papers sharing data had modestly higher citation impact, with the advantage concentrated in the first four years post-publication and not varying by data granularity.
Why it matters: Provides evidence-based insights for researchers about professional incentives for open science practices, relevant to understanding how behavioral factors influence academic decision-making and collaboration.
Method: Extended an existing systematic audit with bibliometric metadata to compare citation impacts between data-sharing and non-data-sharing articles.
Jargon: Field-weighted citation impact - a metric that normalizes citation counts by field and publication year for fair comparison across disciplines.
Njabulo Ndlovu, P. Sifolo, N. Tshipala • Semantic Scholar
Finding: This bibliometric review of 373 articles reveals that while AI research in tourism is growing rapidly (especially post-2019), very few studies specifically address small and medium tourism enterprises, with most focusing on larger companies or using generalized models that don't account for SMTE constraints.
Why it matters: The finding that AI tools are not being designed for resource-constrained small businesses highlights important gaps in how AI impacts different types of work environments and organizational contexts.
Method: Bibliometric analysis using Scopus database and VOS Viewer software for network mapping and thematic clustering of publication trends from 2014-2025.
Jargon: SMTEs = Small and Medium Tourism Enterprises; bibliometric analysis = statistical analysis of publications to identify research trends and patterns.
Finding: A 7-minute stay in airport quiet rooms reduced anxiety, while a seating area reduced negative affect more than quiet rooms, with some evidence that sensory processing sensitivity influences these affective responses.
Why it matters: This demonstrates how environmental design features impact user experience and emotional responses, providing empirical evidence for inclusive UX design principles in public spaces.
Method: Quasi-experimental within-subject design measuring real-time affective responses across three different airport environments with varying sensory characteristics.
Jargon: Sensory processing sensitivity - individual differences in how sensitively people process sensory information from their environment.
⚠️ No control group mentioned | Single study (not replicated) | Limited statistical detail in abstract
R. L. Jr., Norita E. Manly, J. Lobo et al. (13 authors) • Semantic Scholar
Finding: Physical education teachers in the Philippines face five major challenges implementing a new health-focused curriculum: inadequate instructional materials, insufficient time, lack of teacher training, missing equipment, and poor administrative support.
Why it matters: This illustrates how organizational barriers can undermine educational innovation implementation, relevant to understanding change management in learning environments.
Method: Qualitative study using focus groups and interviews with five PE teachers, with expert validation of instruments and triangulation of results.
Jargon: PATH-Fit = Physical Activities Towards Health and Fitness curriculum; triangulation = using multiple data sources to validate findings.
⚠️ Self-reported data only | No control group mentioned | Single study (not replicated)