Conference Agenda

Objectives

Interdisciplinary education in higher education plays a critical role in preparing students to address complex, real-world challenges by integrating knowledge, methods, and perspectives from multiple disciplines (Cai et al., 2025; Spelt et al., 2009). As global issues such as climate change and AI ethics demand cross-disciplinary problem-solving, interdisciplinary approaches have been recognized as essential for cultivating innovation, critical thinking, and future-ready competencies (Markauskaite et al., 2024).

At the same time, artificial intelligence (AI) is emerging as a transformative force in higher education, offering opportunities for personalized learning, resource access, and cross-domain knowledge integration (Bond et al., 2024; Author, 2024, 2025). However, leveraging AI in interdisciplinary settings also presents challenges, including technological complexity, cognitive overload, ethical concerns, and limited user expertise in selecting appropriate tools (Cai et al., 2025).

Despite increasing research at this intersection, key gaps persist: limited understanding of AI’s role across diverse interdisciplinary models (Heeg & Avraamidou, 2023), fragmented discussions on AI’s pedagogical functions (Caratozzolo et al., 2022; Schmitt & Beneder, 2024), and insufficient analysis of its stakeholder-specific impacts (Spelt et al., 2009).

To address these gaps, this study conducts a systematic review of 59 recent studies on AI-supported interdisciplinary education in higher education. Specifically, it aims to:

(1) identify the forms of interdisciplinary education emerging in the AI context;

(2) examine the main research focuses at this intersection;

(3) analyze the functional roles of AI—such as evaluator, agent, monitor, and assistant—in interdisciplinary learning; and

(4) investigate how AI impacts students, educators, and institutions across cognitive, behavioral, and emotional dimensions.

This review offers a timely synthesis to guide researchers, educators, and policymakers navigating the evolving relationship between AI and interdisciplinary education.

Perspective(s) or theoretical framework

This study is grounded in an integrative theoretical perspective that connects interdisciplinary education theory with emerging conceptualizations of AI in education. Interdisciplinary education is framed as a pedagogical approach that integrates knowledge, methodologies, and epistemologies across multiple disciplines to address complex societal challenges (Spelt et al., 2009; Cai et al., 2025). Within this paradigm, the study draws on educational frameworks that position AI not only as a technological tool but also as a pedagogical agent capable of reshaping teaching, learning, and assessment processes (Author, 2024, 2025; Bond et al., 2024).

To examine AI’s functional roles in interdisciplinary education, the review adopts a lens that considers AI as an evaluator, assistant, agent, and monitor—roles that reflect AI’s potential to personalize learning, facilitate cross-domain integration, and support complex cognitive tasks (Caratozzolo et al., 2022; Schmitt & Beneder, 2024). Additionally, the study is informed by socio-cognitive perspectives that consider how AI influences stakeholder engagement at cognitive, behavioral, and emotional levels, aligning with broader understandings of technology-mediated learning environments (Spelt et al., 2009; Heeg & Avraamidou, 2023).

By combining these perspectives, the study provides a conceptual foundation for understanding the evolving relationship between AI and interdisciplinary education in higher education contexts, offering a structured lens through which to categorize and interpret empirical findings across diverse studies.

Generative Artificial Intelligence (GenAI) represents an accessible and transformative frontier in educational technology. By extracting intricate patterns from expansive datasets, it has the qualities to synthesize novel content that ranges from text and imagery to complex code and music (Peñalvo & Ingelmo, 2023). These capabilities inherently offer lecturers innovative ways to support and improve their teaching methods (Author3 et al., 2025; Author1, 2025). Recent studies show that GenAI tools, like Large Language Models (LLMs), are increasingly used in daily academic tasks, including lesson planning, designing assessments, and providing student feedback (Basty et al., 2025; Castillo-Martínez et al., 2024). In this context, GenAI has evolved from a hidden technical system into a visible and personalized "partner" that helps both teachers and students produce resources and explanations (Author1 et al., 2025). Systematic reviews and empirical studies indicate that, when used deliberately, GenAI can scaffold academic writing, offer rapid formative feedback, personalize content, and give educators’ time to focus more on relational and higher order pedagogical work. However, this simultaneously raises new concerns around academic integrity, epistemic depth, bias, and professional identity (Castillo-Martínez et al., 2024; Prilop et al., 2025).

Although research on GenAI in higher education has increased rapidly, the literature remains fragmented in at least three respects, as follows. First, while faculty perceptions are increasingly recognized as crucial, many studies prioritize students’ viewpoints or capture lecturers’ perspectives only superficially, offering limited insight into how educators construe GenAI’s pedagogical value, ethical implications, and impact on professional identity (Crompton & Burke, 2023). Second, demographic variables such as gender, age, and teaching experience are often collected as descriptive controls but are rarely theorized or examined systematically as moderators of educators’ attitudes and acceptance, leaving unclear whether, and how, these characteristics shape lecturers’ orientations toward GenAI (Fadlelmula & Qadhi, 2024). Third, emerging work frequently reports attitudes, intentions, or general openness toward GenAI, yet gives comparatively less attention to how these perceptions relate to actual stages of integration and frequency of use in teaching practice, as well as to the concrete opportunities and challenges lecturers encounter in their classrooms (Bergdahl & Sjöberg, 2025).

In response to these gaps, the present study has been directed towards examining lecturers’ perceptions of GenAI in higher education and its potential implementation in instructional practice. Particular attention is given to the role of demographic characteristics and patterns of use in shaping lecturers’ adoption of GenAI, as well as to the opportunities and challenges educators relate to and associate with when integrating GenAI into their teaching. Accordingly, the study addressed the following research questions:

1) Are there significant differences in lecturers’ perceptions towards the integration of GenAI based on demographic variables such as gender, age and teaching experience?

2) What is the relationship between lecturers’ perceptions, lecturers’ stages of GenAI integration and the frequency of GenAI use?

3) What opportunities and challenges do lecturers perceive in the integration of GenAI in education?

The emerge of generative AI tools is shaping future demands on higher education research and practices (Chan & Colloton 2024) and raising questions about teaching scientific working and writing (Buck & Limburg 2023). Those are relevant for the global context since AI tools are widely spread among students. This is especially relevant for the European context, since the EU AI Act is leading towards the approach to protect democratic values, equality and human decisions based on those values with regulation but also AI literacy (EU 2024). The growing use of AI applications such as ChatGPT, Claude, and others can be observed in teaching and studying (Laupichler et al. 2022; Zawacki-Richter et al. 2024). AI systems can be used to summarize seminar literature, analyze collected data, or explain complex theories in a “simple” way. It can be assumed that students acquire most AI skills outside of lectures. Higher education faces the question of what this development means in the context of good scientific practice (Bozkurt et al. 2024).

Emerging from a workshop, the AI Collective at Goethe University Frankfurt is a project of professors, students, and researchers who explore and reflect together on teaching and studying educational studies in times of artificial intelligence. The project is designed as an open format to transcend institutional boundaries, enabling an exchange on educational discourses.

Our activities circulate the question of how we want to study and teach in the age of AI applications. This project aims to integrate AI as a cross-cutting topic in research and teaching and to bridge the gap between students' acquisition of AI practices and the demands of an educational science degree program. Various formats of cooperation among status groups address our central question:

1. We (Nikolas Fexon) designed a pilot seminar “Shaping AI together” on teaching and learning goals and AI with concrete application reflection, which is in the evaluation process. The aim is to involve students not only as learners but also as researchers and to jointly explore with educational scientists new ways of using AI systems in higher education. Teachers and students engage in dialogue about educational teaching and learning objectives. It also allows for application-oriented reflection on AI systems. For the department, this seminar offers the following opportunities: new impetus for its own research, exchange on didactic possibilities for the use of AI, and contribution to an innovative teaching. The evaluation results of the seminar will guide the development of further formats.
2. We conducted a student survey on active studying with AI exploring how students imagine learning processes and scientific writing practice and reflect their use of AI. In the survey students could also articulate their imaginaries of exchange formats in teaching and learning of scientific research and practice. The survey was conducted in fall 2025. 165 surveys have been completed.
3. In regular meetings of the “AI Collective” we are preparing an interdisciplinary guide on AI use in scientific writing. The meetings of the collective are intended to bring together the perspectives of the institutes and status groups. In discussing the pilot seminar and the student survey, the aim is to develop a handbook on active studying and teaching scientific practices with AI. It enables an exchange about the demands of research and teaching in the age of AI and can be used in a modularized way if necessary. Furthermore, students from the respective degree programs are to be invited to the meetings and represented with their perspectives.

At the ECER conference we will present our concept of the AI collective and first results of the survey on students’ perspectives and the seminar evaluation.

The rapid emergence of generative artificial intelligence (GenAI) tools has fundamentally altered the landscape of academic writing instruction across European and international higher education contexts. While considerable attention has focused on institutional policy responses and detection mechanisms, comparatively little research has examined how students themselves perceive the integration of these technologies within their writing processes, and whether such perceptions differ across educational systems with distinct approaches to writing instruction.

This exploratory cross-national study examines undergraduate students' writing process conceptions and their perceptions of GenAI use in academic writing, comparing data from the University of Tartu (Estonia) and California State University Long Beach (United States). The comparison is noteworthy given the divergent writing instruction traditions in these contexts. European higher education, including Estonian universities, typically lacks the systematic writing support infrastructure characteristic of North American institutions, such as dedicated first-year composition courses, writing centres, and Writing Across the Curriculum (WAC) programmes (Björk et al., 2003; Donahue, 2009; Harbord, 2018). This structural difference raises important questions about whether students' underlying conceptions of writing, and their subsequent integration of AI tools, vary across these educational traditions.

The study addresses three research questions: 1) What factor structure emerges from an exploratory analysis of the AI-adapted Writing Process Questionnaire with US and Estonian undergraduate students? 2) Do students' AI-related writing beliefs constitute a structurally distinct construct from traditional writing conceptions, or are they embedded within existing patterns of writing difficulty and epistemic belief? 3) In what ways do US and Estonian undergraduate students differ in their writing conceptions, including their orientation toward AI in writing?

The theoretical framework draws on established research into writing process conceptions, particularly the distinction between knowledge-telling (surface-level) and knowledge-transforming (deep-level) approaches to writing (Bereiter et al., 2023; Scardamalia & Bereiter, 1987, 2005). Students who view writing as knowledge transformation engage in effortful, reflective processes that develop and create knowledge through the act of writing itself. Research with European doctoral students has demonstrated that such conceptions relate significantly to academic well-being, productivity, and perceptions of the learning environment (Lonka et al., 2018). Maladaptive conceptions, including beliefs in innate writing ability, perfectionism, and procrastination, have been associated with writing blocks and reduced academic engagement (Cerrato-Lara et al., 2017).

The emergence of GenAI introduces new dimensions to these established constructs. If writing is understood as a process of knowledge transformation, what happens when AI can generate content that bypasses this cognitive work? Do students who hold more surface-level conceptions of writing embrace AI differently than those who view writing as epistemically generative? These questions have particular relevance for European educational contexts, where explicit writing instruction may be less systematically embedded in curricula.

This study contributes to the European educational research community by providing empirical data on how students in a European context perceive and potentially integrate GenAI within their academic writing. The comparison with US data offers insights into whether the structural differences in writing support systems relate to different patterns of AI adoption and conceptualisation. Furthermore, by adapting and extending an instrument originally validated with Finnish doctoral students (Lonka, 2014) to include AI-related dimensions, this research advances methodological approaches for measuring evolving writing conceptions in the age of artificial intelligence.

The findings have implications for writing pedagogy across European higher education institutions, particularly regarding how educators might address students' underlying beliefs about writing when introducing or responding to GenAI tools.