Students with Autism Spectrum Disorder (ASD) generally exhibit lower levels of learning motivation in mainstream classrooms than their peers; however, the relationship between the basic psychological needs of ASD students and their learning motivation remains under-researched. Based on Self-Determination Theory, this paper theoretically analyzes how autonomy, competence, and relatedness influence learning motivation in ASD students. The theoretical analysis indicates that teachers’ need-supporting behaviors are unevenly distributed in inclusive classrooms, where autonomy support is scarce yet effective. The three mediation pathways are as follows: ASD’s executive function difficulties lead to autonomy frustration, subsequently reducing learning motivation; ASD’s limited cognitive flexibility and executive function difficulties cause competence frustration, thereby diminishing learning motivation; and social communication difficulties in students with ASD lead to belonging frustration, thereby reducing learning motivation. A sense of belonging serves as the foundation for autonomy and competence to function, with potential compensatory effects among the three needs. This paper proposes a theoretical mediation model of transmission from ASD characteristics to need impediments and subsequent changes in learning motivation, thereby providing practical guidance for teachers in inclusive education.

With the development of data-driven sports and artificial intelligence, computer vision and deep learning technology are now being used in new ways for tennis performance analysis and athlete training. Hard courts, clay courts, and grass courts are all different in terms of technology, tactics and physical demands; thus, many-surface-capable players are required at present. However, the previous method of tennis performance analysis was based on manual observation and statistical collection; thus, it suffered from human error, missing data, and slow feedback. Review of existing research on court-surface effects, traditional tennis data collection, and the application of deep learning and computer vision in tennis analysis. Explore applications of artificial intelligence in supporting the collection of match data, cross-surface tactical analysis, and personalised player training. Deep learning can be used to collect all-round, unbiased performance data on the players through coaches, identify any deficiencies in their game, and set up customised training plans. Complex visual environments, data integration, equipment costs and the interpretation of non-visual performance factors are also problems. In the future, research will integrate computer vision and wearable/physiological data to build lightweight AI systems for more accessible tennis training applications.

Social Anxiety Disorder (SAD) is widely recognized as a common and debilitating mental health condition, particularly among young adults navigating complex social environments. A substantial body of prior research has thoroughly investigated how individuals with SAD process facial expressions, given that these non-verbal cues serve as a core channel for emotional exchange and social signaling in daily interactions. However, the vast majority of laboratory studies in this area rely on static and isolated facial pictures as experimental stimuli. These simplified materials, while controlled, cannot adequately simulate the dynamic, multi-modal, and context-rich nature of authentic social encounters. This significant methodological gap may inadvertently restrict our full understanding of the cognitive and behavioral symptoms of social anxiety, potentially introducing uncertainties into clinical intervention practices. Consequently, the field urgently requires more ecologically valid experimental paradigms, such as immersive virtual reality environments. By conducting research within realistic social contexts, researchers can more effectively reveal the nuanced cognitive characteristics of SAD and subsequently promote the development of more targeted.

As contemporary communication increasingly shapes how individuals imagine possible futures, mental simulation may influence emotional and behavioural responses across both environmental and digital media contexts. This study employs a comparative case study approach, analysing IPCC climate change communication as a case of threat-oriented mental simulation and TikTok's algorithmic recommendation environment as a case of reward-oriented mental simulation through a Future-oriented Mental Simulation Framework. The study aims to examine how different forms of imagined future outcomes influence emotional responses and behavioural tendencies. The findings suggest that threat-oriented simulation tends to activate concern, anxiety, and protective motivation but may also lead to avoidance, while reward-oriented simulation tends to generate anticipation and approach motivation that can promote continued engagement and potentially repetitive use. The study demonstrates that the valence of imagined futures plays an important role in shaping behavioural direction and provides a cross-contextual psychological perspective for understanding risk-oriented and reward-seeking behaviour.

This study explored the perspectives and experiences of nine early education practitioners in Wuhan, China, regarding the development of empathy in early childhood. This study aims to understand early childhood teachers' perceptions of empathy, including what empathy is, children's expression of empathy, their current strategies to enhance children's empathy, and what they believe should be done to develop children's empathy. This qualitative study used semi-structured interviews and thematic analysis to gain comprehensive insights into the instructive perspectives and practical experiences of the participants. The findings of the study indicated that although most of the participants have basic knowledge of empathy, their understanding of empathy was largely oriented towards cognitive empathy; while instructional time was often fragmented in practice. This was primarily due to three major practical obstacles: a rigid assessment system, limited curriculum autonomy, and insufficient collaboration between home and school.This study supplements domestic research gaps on preschool teachers' perceptions of child empathy and provides empirical evidence for preschool social-emotional curriculum optimization.

On the Internet, fake news has become a phenomenon that deserves attention because of its social influence. Past research on fake news focused on repetition and presentation, and a research gap of lack of Asian population and Asian circumstances occurred. Therefore, the present study addresses two main research questions. First, it tests whether exposure frequency affects the perceived reliability of news and participants' willingness to share it. Second, it examines the reasons behind this pattern. To assess participants' attitudes toward fake news, researchers measured their sharing preference, their willingness to share the news with friends, and the news' perceived reliability, which refers to whether they believed the news was true. 74 participants participated in this research, and 68 participants' results were measured. The result shows that exposure frequency significantly influenced people's perceptions of fake news. Two exposures made people more likely to believe fake news, whereas three exposures reduced their belief in it. Limitations and directions for further research are discussed.

While institutional policies on regulating student generative AI (GenAI) use in assessment have multiplied in higher education, their soaring number may not guarantee effective governance. Previous studies have highlighted persistent barriers including violation of academic integrity, student overreliance, and assessment invalidity, whereas more internal factors around how policy per se can undermine prospective clarity and exacerbate governance dilemma are still underexamined. Therefore, this study interrogates existing issues around institutional GenAI-regulatory materials themselves. Employing the concept of 'interpretive load', this study examines the governance dilemma through a specific lens of how students bear the burden to navigate and understand the institutional policy architecture and language before starting assessments. Through a policy document analysis of a purposive sample of materials published by the UK's G5 universities, this paper identifies three issues – policy proliferation, policy fragmentation, and policy ambiguity. These three findings represent crucial mechanisms that progressively enable institutions to transfer responsibility for clarification to students, consequently placing interpretive load on students. Crucially, this paper concentrates on a relatively new notion of interpretive load, which implies ample scope for theoretical investigation and offers directions for future institutional policy making and research orientations within the field.

The theory of "ideological surplus value" proposed by Venezuelan Marxist philosopher Ludovico Silva provides important theoretical resources for understanding the nature of neoliberal education. By rereading Silva's core texts, this paper systematically sorts out the theoretical construction of "ideological surplus value," analyzes the operating logic of education as an ideological production mechanism, and further decodes how neoliberal education systematically extracts ideological surplus value through three mechanisms: the myth of human capital, the standardized examination system, and digital educational technology. On this basis, this paper draws on Silva's critical reflection on the concept of "revolutionary ideology," engages in dialogue with critical pedagogues such as Freire and Giroux, and attempts to provide a theoretical foundation for the reconstruction of contemporary critical pedagogy, moving toward a "de-commodified" educational imagination.