This study examines the impact of Project-Based Learning (PjBL) on developing metacognitive skills among preservice science teachers (PSTs) in Northeast Thailand. A sample of 143 PSTs, including first-year students in General Biology 1, second-year students in General Physics 1, and third-year students in Basic Organic Chemistry, participated in an 18-week programme. The study aimed to assess changes in metacognitive skills before and after PjBL implementation, evaluate differences among academic years, and identify predictors of skill development. The methodology included a six-hour orientation workshop and a collaborative, project-based curriculum. Descriptive and inferential statistics were employed, with the item-objective congruence index (IOC) for content validity, paired samples t-tests for pre- and post-intervention comparisons, and Analysis of Variance (ANOVA) to examine differences across academic years. Multiple regression analysis was used to identify significant predictors of metacognitive skill development. Results showed significant improvements in metacognitive skills post-PjBL, with substantial enhancements across all subjects. ANOVA indicated significant differences among academic years, with third-year students demonstrating the highest metacognitive skill levels. Multiple regression analysis identified participation in PjBL and academic level as significant predictors of metacognitive skill development. These findings highlight the effectiveness of PjBL in enhancing metacognitive skills and underscore the importance of active learning and reflective practices in teacher education programmes. This study provides valuable insights into the impact of PjBL on PSTs' professional growth and instructional efficacy, preparing them for modern classroom challenges.

This study explores the impact of artificial intelligence (AI) integration on students' educational experiences. It investigates student perceptions of AI across various academic aspects, such as module outlines, learning outcomes, curriculum design, instructional activities, assessments, and feedback mechanisms. It evaluates the impact of AI on students' learning experiences, critical thinking, self-assessment, cognitive development, and academic integrity. This research used a structured survey distributed to 300 students through Microsoft Forms 365, yet the response rate was 29.67%. A structured survey and thematic analysis were employed to gather insights from 89 students. Thematic analysis is a qualitative method for identifying and analysing patterns or themes within data, providing insights into key ideas and trends. The limited response rate may be attributed to learners' cultural backgrounds, as not all students are interested in research or familiar with AI tools. The survey questions are about AI integration in different academic areas. Thematic analysis was used to identify patterns and themes within the data. Benefits such as enhanced critical thinking, timely feedback, and personalised learning experiences are prevalent. AI tools like Turnitin supported academic integrity, and platforms like ChatGPT and Grammarly were particularly valued for their utility in academic tasks. The study acknowledges limitations linked to the small sample size and a focus on undergraduate learners only. The findings suggest that AI can significantly improve educational experiences. AI provides tailored support and promotes ethical practices. This study recommends continued and expanded use of AI technologies in education while addressing potential implementation challenges.

Mindfulness, recognized as a protective factor against learning burnout in higher education, has garnered considerable attention, yet its underlying mechanisms remain underexplored. This study examined the relationship between mindfulness, regulatory emotional self-efficacy, and learning burnout. Data from 461 Chinese university students were collected using a correlational design and cluster sampling method, employing the Five Facet Mindfulness Questionnaire, University Student Learning Burnout Scale, and Regulatory Emotional Self-Efficacy Scale. Hypotheses were tested using partial least squares structural equation modeling. Results showed that Participants exhibited above-average mindfulness (M=3.090), learning burnout (M=3.278), and regulatory emotional self-efficacy (M=3.417). Results revealed that mindfulness is directly and negatively related to learning burnout (β=-0.679, t = 28.657, p < .001). Regulatory emotional self-efficacy (β = -0.357, t = 8.592, p < .001) was significantly and negatively related to learning burnout. Mindfulness was significantly and positively related to regulatory emotional self-efficacy (β = 0.638, t = 24.306, p < .001), and regulatory emotional self-efficacy (R2: from .461 to .537) partially mediated the relationship between mindfulness and learning burnout. Besides, the Importance-Performance Matrix Analysis revealed that managing negative emotions significantly contributes to learning burnout but performs poorly, whereas non-reacting demonstrates both the lowest contribution and performance. Findings suggest that mindfulness indirectly alleviates learning burnout through regulatory emotional self-efficacy, providing evidence-based insights for targeted mindfulness interventions in higher education.