Conference Agenda

Visual outputs of generative Artificial Intelligence (AI), including AI-Images, have become an integral part of everyday media practices. Following the viral Pope AI-Image in February 2023 (Vilentchik, 2023), such images have proliferated on social media, for instance through entertaining trends like “italian brain rot” in 2025 (Gupta, 2025). While often used for entertainment, AI-Images also go hand in hand with significant risks, as they can be employed to manipulate, misinform and discredit individuals or groups (Bray et al., 2023). This becomes particularly evident in the case of deepfakes, which refer to “photorealistic images [...] that have been algorithmically generated or manipulated” (Jacobsen & Simpson, 2024, p. 1) and which can further intensify the spread of fake news (ibid.). Existing research indicates that recipients often struggle to reliably distinguish AI-Images from photographs (e.g. Bozkir et al., 2025), thereby amplifying the challenges AI-generated imaginary pose to information credibility. With the continuous advancement of technology, AI-Images are becoming increasingly sophisticated and harder to identify (Künne, 2024). Accordingly, media literacy that goes beyond simple recognition of AI-Images is essential for a responsible engagement with social media.

Young people are more likely to be exposed to AI-Images in their everyday media practices: In Germany, they spend an average of 231 minutes per day online, with social networks such as Instagram and TikTok among the most frequently used platforms (mpfs, 2025). As images are a central component of social media platforms, it can be assumed that adolescents come (unnoticed) into contact with them. In this context, teachers bear a special responsibility, as they are expected to foster media literacy in schools and raise students’ awareness of responsible and critical media use (KMK, 2016). Therefore, this study seeks to examine the extent to which Pre-Service Teachers exhibit media literacy in the context of AI-Images, enabling them to impart it in their future professional practice. This target group is particularly relevant, as they occupy a dual role as both active media users and future educators.

Media critique “involve[s] a comprehensive examination of (mass) media in their complex societal entanglements. In this sense, critical recipients […] are not only able to question media. They can also reflect on their own relationship to a media offering and subject the media to evaluative judgement.” (Schmitt et al., 2020, p. 332). These various facets of media criticism are integrated into Baacke’s concept of media literacy (Medienkompetenz), which he structures into four dimensions: media knowledge, media use, media production and media critique (Baacke, 1996). Within the dimension of media critique, he further distinguishes three sub-dimensions: analytical, reflective and ethicaldimension (ibid.). In the analytical sub-dimension, “problematic societal processes […] are captured in an analytically appropriate manner” (Hugger, 2022, p. 22). In the reflective sub-dimension, „this knowledge […] is applied to oneself and one’s own actions” (ibid.). In the ethical sub-dimension “analytical thinking and reflective consideration are coordinated and defined in a socially responsible manner” (ibid.).

The overarching question of the study, “How do Pre-Service Teachers act critically toward media in the context of AI-Images?”, is addressed through the research questions guided by Baacke’s concept of media critique:

RQ1: To what extent do Pre-Service Teachers grasp AI-Images from an analytical perspective? (analytical sub-dimension)

RQ2: How do Pre-Service Teachers evaluate different AI-Images ethically? (ethical sub-dimension)

RQ3: To what extent do AI-Images affect Pre-Service teachers’ own media reception and practices? (reflective sub-dimension)

This paper examines how early childhood teachers in two national contexts, the United Kingdom and China, understand, adopt, and negotiate digital practices in their everyday teaching. Digital technologies, platforms, and AI-enabled tools have become increasingly embedded in early learning environments, yet teachers’ perspectives on their usefulness, constraints, and pedagogical value remain under-examined. Early childhood education (ECE) is also positioned within shifting policy landscapes where expectations for digital competence, online safety, and data accountability are intensifying. In this context, the study asks: How do early childhood teachers in the UK and China perceive digital practices in their classrooms, and what contextual, cultural, and policy factors shape their adoption and use of digital tools?

The paper has three aims: (1) to map teachers’ attitudes towards digital practices, including perceived usefulness, ease of use, and behavioural intentions; (2) to examine how teachers understand the place of digital technologies in play-based and early years pedagogical environments; and (3) to analyse how national and local ECE policies influence teachers’ professional decision-making and classroom practices. Through these aims, the study contributes to ongoing European debates on digital childhoods, teacher professionalism in changing sociotechnical environments, and the conditions under which educational knowledge is produced and enacted in early years settings.

The study is informed by the Technology Acceptance Model (TAM), which is used to conceptualise teachers’ perceptions of usefulness, ease of use, attitude towards use, and intention to integrate digital tools. TAM constructs provide a lens for understanding why teachers choose to adopt or resist digital practices, especially in resource-unequal or culturally distinct environments. The work is further situated within sociocultural perspectives on digital learning environments, recognising that digital practice is shaped by institutional norms, policy frameworks, and culturally embedded values regarding childhood, learning, and technology.

The research adopts a sequential mixed-methods design:

Phase 1 consists of a cross-national online survey (n = 215) with early childhood teachers in the UK and China. The survey includes TAM-based Likert scales and open-ended questions. Quantitative analysis involved reliability testing, descriptive statistics, and regression modelling to examine relationships among constructs. Qualitative responses were coded thematically.

Phase 2 comprises a Critical Policy Discourse Analysis (CPDA) drawing on national and local early childhood policy documents from England and China (2001-2025). Using a thematic coding framework, the analysis focuses on eight themes: digital competence, online safety, curriculum expectations, assessment/data practices, teacher professionalism, resource allocation, governance, and screen-time guidance. The CPDA provides macro-level insights into how digital practices are framed, legitimised, and regulated.

Phase 3 consists of semi-structured interviews followed by walking interviews with early childhood teachers in Sheffield and Xi’an. The walking interviews, conducted in classroom environments after the main interview, elicit contextualised reflections on real digital practices, classroom decision-making, and pedagogical reasoning. Interview analysis will follow reflexive thematic analysis.

This paper offers an original contribution by integrating mixed-methods data within a cross-national comparative design. It responds directly to the ECER 2026 theme by examining how teachers navigate the changing conditions of knowledge and action within digitally mediated learning environments. The international comparison highlights how broader sociocultural, political, and policy ecosystems shape the possibilities and limits of digital practice. The findings will be relevant to European researchers, policymakers, and practitioners seeking to understand digital transformation in early childhood education and its implications for teacher development, equity, and children’s digital experiences.

This study examines how teachers in Kazakhstan develop and enact media literacy (ML) practices in their classrooms and how these practices are shaped by demographic, experiential, and institutional factors. Drawing on a mixed-methods design, the study investigates how professional development (PD), school conditions, and teacher characteristics interact to support or constrain ML implementation across diverse school contexts.

The pedagogy of ML builds on constructivist principles, emphasising problem-solving, active learning, and critical reflection. It encourages both teachers and students to analyse and interpret various forms of media, including news articles, social media posts, and visual narratives, to understand how meaning is constructed in contemporary society (Hobbs, 2018; Potter, 2022). While digital technologies are central, ML extends beyond digital skills to include critical thinking, ethical media engagement, and participatory communication (Schmitz et al., 2024).

Teachers’ capacity to work with sensitive curricular content and exercise professional judgement plays an important role in shaping classroom pedagogy, particularly in education systems characterised by strong policy control (Chong et al., 2021). Contemporary ML frameworks, including those addressing social media environments, shape how ML is understood and enacted in classrooms by providing conceptual guidance for teachers’ instructional decisions (Cho et al., 2024). In response to the spread of misinformation and persistent digital inequalities, ML is increasingly understood as a means of supporting democratic participation and the development of critical digital citizenship.

Despite an expanding ML literature, most studies remain prescriptive or are based in high-resource Western contexts. Few studies investigate how ML unfolds in under-resourced or centralised systems where structural inequities and hierarchical governance shape pedagogy. In Kazakhstan, where many educators operate under strict curricular policies and uneven resource distribution, teachers’ engagement with ML is often shaped by both constraint and innovation (Yeleussiz & Qanay, 2025).

While global ML frameworks emphasise core competencies such as analysing and evaluating media, research highlights the need to adapt these frameworks to local policy and educational conditions (Weinand, 2018). Teachers’ ability to implement ML depends not only on training and access to tools, but also on professional confidence and opportunities for context-sensitive development (Le et al., 2025).

Emerging research from the Asia-Pacific region highlights the need to understand ML as a contextually situated practice rather than a universally transferable skillset (Lin et al., 2013; Weninger, 2017). Lin et al. (2013) show that ML in Singapore has been shaped by national policy priorities and institutional arrangements rather than adopted wholesale from global frameworks, while Weninger (2017) demonstrates that teachers’ understandings and classroom enactment of ML are strongly influenced by curricular expectations and school-level conditions. In addition, research on digital poverty in teacher education underscores how unequal access to digital resources and limited institutional support can restrict teacher agency, indicating that structural conditions might be addressed alongside pedagogical expectations (Arantes et al., 2025).

Given the limited empirical research on teacher ML in post-Soviet and Central Asian contexts, this study explores how teacher and school-level factors shape ML implementation. It addresses the following questions:

• RQ1: How do teachers perceive their own ML proficiency and classroom practice, and to what extent do school conditions support their engagement with ML?
• RQ2: How do teachers’ demographic characteristics, experience levels, and institutional contexts influence their ML practice?

This study contributes to broader debates on educational equity, teacher agency, and curriculum innovation in transitional education systems like Kazakhstan’s. By situating ML within local institutional and policy contexts, the study offers insights into how global educational ideals can be adapted in practice and informs both international ML research and national teacher education policy.