Digital Teaching Day 2025

Digital tools, and artificial intelligence (AI) in particular, can be used in a variety of ways to address specific challenges in school and higher education, such as the diversity of learners and increasing support needs. Since the release of OpenAI’s ChatGPT chatbot, AI has become a hot topic, and the possibilities for using AI to support teaching and learning seem limitless. It is crucial that such use is approached with critical reflection and grounded in sound subject-specific pedagogy. At the same time, it is becoming increasingly difficult to keep track of evidence-based and pedagogically sound AI-based tools. This presentation therefore outlines a wide range of potential applications for AI tools in higher education and schools. The focus is on both traditional AI tools, such as intelligent tutoring systems, and advanced, generative AI tools such as large language models (e.g. ChatGPT) and their pedagogically grounded application in lesson planning and design, as well as in the learning process – in each case with a view to teaching quality in terms of evidence-based teaching characteristics. Last but not least, the presentation will also examine contemporary learning and assessment cultures in order to counteract emerging phenomena such as ‘skill skipping’ or ‘deskilling’.

In the lecture section, the (academic) fundamentals and terminology
of the various forms of AI will first be explained; academic and pedagogical quality criteria for the evaluation of AI tools in an educational context will be identified; and the possibilities and limitations of AI tools will be discussed. AI tools and their potential applications will then be presented; these can be tried out and discussed by the students during the workshop section. In this context, the current research project ‘WoLKE’ will also be presented; this project is developing courses for trainee teachers of languages and STEM subjects, focusing on the subject-specific and critically reflective use of AI tools in language and STEM teaching.

Assistant Professor Dr Luzia Leifheit is a tenure-track professor of digitalisation, specialising in the didactics of algorithms and data science, at the University of Education in Schwäbisch Gmünd. Her research examines the effectiveness of various learning approaches – often game-based and physical – in fostering motivation and skills in computer science education, from primary school through to university.

Contact: luzia.leifheit(at)ph-gmuend.de

Assistant Professor Dr Heiko Holz is a tenure-track professor of Computer Science and Computer Science Education, specialising in digitalisation in education, at the Ludwigsburg University of Education. His research focuses on identifying the conditions for successful digitalisation in education and promoting these within digital teaching and learning systems, including human-computer interaction, digital game-based learning, intelligent tutoring systems, and digital interventions and assessments.

Contact: heiko.holz@ph-ludwigsburg.de

Assistant Professor Dr Heiko Holz is a tenure-track professor of Computer Science and Computer Science Education, specialising in digitalisation in education, at the Ludwigsburg University of Education. His research focuses on identifying the conditions for successful digitalisation in education and promoting them in digital teaching and learning systems, including human–computer interaction, digital game-based learning, intelligent tutoring systems, and digital interventions and assessments.

Contact: heiko.holz@ph-ludwigsburg.de

Assistant Professor Dr Luzia Leifheit is a tenure-track professor of digitalisation, specialising in the didactics of algorithms and data science, at the University of Education in Schwäbisch Gmünd. Her research examines the effectiveness of various learning approaches – often game-based and physical – in fostering motivation and skills in computer science education, from primary school through to university.

Contact: luzia.leifheit(at)ph-gmuend.de

Prof. Dr Matthias Wölfel is Head of the Institute for Intelligent Interaction and Immersive Experience (IIIX) and a professor (in the Department of Media Informatics) at Karlsruhe University of Applied Sciences (HKA), as well as an associate professor at the University of Hohenheim (in the Department of Social Sciences). Wölfel has received numerous awards; e.g. the 2024 Teaching Award (HKA), the 2021 Digitalisation Award (HKA), second place in the ‘Professor of the Year’ award (UNICUM Foundation) in 2017, and the 2013 ICT Innovation Award (BMWi).

Contact: matthias.woelfel@ph-ka.de

This article briefly introduces the Pedagogical Conversational Tutor (PET) and the potential educational applications of this AI-supported system in higher education.

What makes the PET unique is that content uploaded by lecturers – such as PowerPoint slides, videos and PDFs – is fed into the knowledge base of a local chatbot. This enables students to ask specific questions about this content and receive personalised learning support. The system also features a ‘hallucination control’ mechanism that verifies the chatbot’s statements. Furthermore, the article discusses general applications of AI in higher education.

Andreas Reich, M.Sc., is a research assistant at the University of Hohenheim in the Department of Communication Studies, specialising in media and usage research. His doctoral thesis focuses on machine learning, NLP and chatbots. In his part-time freelance work, he designs websites and programmes games, amongst other things.

Contact: andreas.reich@uni-hohenheim.de

Prof. Dr Matthias Wölfel is Head of the Institute for Intelligent Interaction and Immersive Experience (IIIX) and a professor (in the Department of Media Informatics) at Karlsruhe University of Applied Sciences (HKA), as well as an associate professor at the University of Hohenheim (in the Department of Social Sciences). Wölfel has received numerous awards; e.g. the 2024 Teaching Award (HKA), the 2021 Digitalisation Award (HKA), 2nd place in the ‘Professor of the Year’ award (UNICUM Foundation) in 2017, and the 2013 ICT Innovation Award (BMWi).

Contact: matthias.woelfel@ph-ka.de

In our presentation, we would like to explore how interaction with AI – for example, via ChatGPT – through targeted prompt engineering, understood as an iterative revision process (cf. Hayes & Flower 1980), help to build learners’ knowledge of text and text types and expand their writing skills (cf. Göpferich 2015).  We draw on data collected in collaboration with trainee teachers of German and English for primary and secondary education. The focus of our analyses is on how the students adapt their prompts to generate texts which they judge to be (situationally) appropriate in terms of the linguistic requirements of a specific text type, i.e. also with regard to the linguistic routines employed.

References:
Göpferich, S. (2015). Text Competence and Academic Multiliteracy: From Text Linguistics to Literacy Development. Narr Francke Attempt. http://nbn-resolving.org/urn:nbn:de:bsz:24-epflicht-1428851

Hayes, J. R. & Flower, L. S. (1980). Identifying the organisation of writing processes. In L. Gregg & E. Steinberg (Eds.), Cognitive processes in writing (pp. 3–30). Routledge.

The rapid development of artificial intelligence is having an increasing impact on university teaching in the field of programming. In the interdisciplinary supplementary programme ‘Digital Skills’ at OTH Regensburg, students from all disciplines acquire, amongst other things, basic programming skills. This article presents the findings of a qualitative study in which participants on the ‘Digital Skills’ supplementary programme (N = 6) solved a programming task using the AI tool ChatGPT. The results highlight clear differences between the participants in terms of their use of ChatGPT and their personal learning success in solving the task. This points to a clear need for action in higher education practice regarding the didactic design of programming courses, taking AI into account.

Student (Master’s in Business Informatics) at OTH Regensburg

Former tutor on the Digital Skills supplementary programme at the Regensburg School of Digital Sciences (RSDS)

Contact: dominic.hauser@st.oth-regensburg.de

Chair of Media Informatics

Coordinator for the Regensburg School of Digital Sciences at OTH Regensburg (RSDS)

Contact: markus.heckner@oth-regensburg.de

Chair in Business Informatics and Digital Transformation

Professor at the Regensburg School of Digital Sciences

Contact: ulrike.plach@oth-regensburg.de

Staff member on the BeDisc project (Build digital competence and explore Digital Sciences – supplementary course in digital skills at OTH Regensburg)

Contact: lisa.holzerschulz@oth-regensburg.de

AI enhances learning and takes on tedious routine tasks – but it can also encourage laziness and stifle initiative. This talk will demonstrate how AI can personalise learning processes, provide creative food for thought and bring dry theory to life. There are smart ways to use it – AI as a tutor, a source of ideas or a tool for structuring learning. However, there is also a risk that students will let the machines do the thinking for them and, in doing so, lose the joy of discovery. Conclusion: use AI sensibly, switch on your brain and always remain curious. (This abstract was formulated with the help of ChatGPT.)

Dr Jörn Loviscach is a professor of engineering mathematics and technical computer science at Bielefeld University of Applied Sciences. Originally a physicist, he was previously a professor of computer graphics at Bremen University of Applied Sciences and, before that, a journalist, most recently serving as deputy editor-in-chief of the computer magazine c’t.

Website with blog: https://j3L7h.de/

The 2LIKE project develops AI-supported adaptive learning resources and personalised feedback for academic writing. The 2LIKE feedback tool goes beyond existing systems and offers targeted support at a micro-level to ensure compliance with academic standards, without generating ready-made text sections. Students receive detailed feedback on, for example, grammar, style, citations and the structure of their work. At the macro level, adaptive 2LIKE learning pathways enable flexible access to learning content at various proficiency levels and in preferred formats. The aim is to promote personalised learning within an increasingly diverse student body. As part of our presentation, we will also demonstrate how the concepts and tools developed are being transferred to the field of academic continuing education.

Contact: melina.klepsch@uni-ulm.de

Ansgar Scherp is a professor of Data Science and Big Data Analytics at the University of Ulm. Previously, he worked, amongst other roles, as a professor of Natural Language Processing and Data Analysis and was a member of the interdisciplinary research group “Language and Computation” at the University of Essex in England. Ansgar enjoys an excellent reputation in research in the fields of text and graph mining, particularly in the combination of symbolic and subsymbolic methods for data analysis. He combines methods from the fields of information retrieval, data mining and machine learning, as well as the Semantic Web. He applies his innovative data analysis methods to very large, distributed knowledge graphs on the web containing billions of edges, and to document corpora in the life sciences/medicine, social sciences, economics and the web.

Contact: ansgar.scherp@uni-ulm.de

Assessments play a central role in teaching – this refers not only to summative assessments used to determine grades, but also to formative assessments, for example in formats such as just-in-time teaching or peer instruction. As teachers, we should still determine the core content of the assessments ourselves, but we can enlist the help of generative AI when it comes to formulating and formatting the tasks. Precise tasks can be formulated using brief, bullet-point-style instructions. These tasks, or existing ones from various sources, can be formatted by the AI as required, for example for LaTeX, HTML or for direct import into an LMS such as Moodle.

Prof. Dr Robert Kellner has been a professor of physics at Rosenheim University of Applied Sciences since 2016. He runs workshops on active learning methods such as peer instruction and just-in-time teaching, as well as on the use of new digital opportunities in teaching. Through various projects, he is further developing methods such as HyFlex and SCALE-UP and utilising hybrid formats and AI in teaching. He has received several awards for his modern teaching methods.

Contact: robert.kellner@th-rosenheim.de

To date, linguistic and cultural diversity in higher education and schools has received little attention and, above all, has not been utilised constructively within multilingual learning settings (inclusive, functional, subject-specific) [1, 2]. The aim of the workshop is to present and jointly develop various artificial intelligence (AI)-based methods that promote linguistic diversity within learning groups in cooperative learning formats [3, 4], and to critically reflect on their potential. Particularly with regard to the training of future teachers, fostering a nuanced, multi-perspective understanding of complex phenomena through the use of a wide variety of sources in different languages represents a fruitful approach to expanding competences in the field of subject-specific (and subject-didactic) knowledge. The now widely available AI-based text generators enable teachers to create, with minimal effort, material that is precisely tailored to the capabilities of their learning group.

References:
[1] Repplinger, N.; Budke, A. ‘Is pupils’ multilingual life practice a potential focus for Geography lessons?’, European Journal of Geography 2018, Vol. 9 (3), 165–180.

[2] Heuzeroth, J. Multilingualism in geography lessons: Background, potential applications and difficulties in implementing multilingualism approaches within the context of subject-specific (and language-based) learning. k:ON – Cologne Online Journal for Teacher Education, 2023, Vol. 2, 116–133. https://doi.org/10.18716/ojs/kON/2023.s.7

[3] Brüning, L. (2010). Competence-oriented teaching through cooperative learning. An introduction to the repertoire of methods and examples of application. In: Praxis Geographie (42) Issue 12, pp. 6–14.

[4] Katja Adl-Amini, Vanessa Völlinger (2021). Cooperative learning in the classroom. In: Institute for Educational Analysis Baden-Württemberg (ed.) Effective Teaching, Vol. 4. Self-published, Stuttgart.

Teacher of Geography, Social Sciences, Philosophy and German as a Foreign Language (DaF) / German as a Second Language (DaZ) at a comprehensive school, and lecturer at the Institute for Geography Education at the University of Cologne. Author for various journals on geography education and lecturer for the Goethe-Institut. His main areas of research include the relationship between systemic thinking, causality and language; language awareness and multilingualism; and the impact of metacognition in geography lessons.

Contact: johannes.heuzeroth@uni-koeln.de

Teacher, musician and translator. He currently works as a teacher of German as a foreign language in Tbilisi, Georgia, as well as an author and lecturer for, amongst others, the Goethe-Institut. As a linguist and educator, he was involved for several years in research projects in the fields of German as a second language and educational psychology (Friedrich Schiller University, Jena). He also coordinated numerous international school projects.

Contact: wolf.zippel@gmail.com

How can AI tools help with the production of learning materials? As well as assisting with the design of learning content, generative AI technologies also support teachers in creating audiovisual materials. Through practical scenarios drawn from course planning, participants will discover where AI offers significant time savings and new pedagogical potential – and which tasks cannot (yet) be automated. The workshop focuses, on the one hand, on the generation and, on the other, on the optimisation of audio and video content – always with a view to its use in participants’ own teaching.

Markus Tischner is a filmmaker, learning expert and trainer. He has been working as a filmmaker since 2008 and specialises in the production of educational media and science communication. He also heads the media production department at the Institute for Learning Innovation (ILI) at Friedrich-Alexander University Erlangen-Nuremberg (FAU). There, he not only produces
educational and science films, but also runs seminars on media production as part of the ‘Learning Design’ Master’s programme. He also trains researchers in the fields of science communication and the production of educational media. Since 2022, Markus has placed an additional focus on AI-assisted production of educational media.

Contact: markus.tischner@ili.fau.de

The workshop will introduce the fundamentals of Artificial Intelligence (AI) and provide participants with examples of how AI can be applied to selected tasks in the teaching of science (in this case, physics). Participants will then have the opportunity to try out AI applications for themselves. With regard to the use of AI in research, we will demonstrate how AI methods can be used to analyse complex datasets. With regard to the use of AI in teaching, we will demonstrate how interactive learning environments can be created and utilised with the help of prompt engineering and simple web-based tools. The workshop is aimed at those who have little previous experience of using AI for research and teaching in the didactics of the natural sciences.

Peter Wulff is an assistant professor of physics and physics education at Heidelberg University of Education. His research focuses, amongst other things, on the use of AI to support physics-related teaching and learning processes. He is currently developing a web application for AI-supported physics problem-solving processes as part of a collaborative project – WasP – (funded by the Klaus Tschira Foundation).

Contact: peter.wulff@ph-heidelberg.de

Marcus Kubsch is an assistant professor of physics education at Freie Universität Berlin. His research explores, amongst other things, the new possibilities that AI methods offer for generating insights in the field of science education. In addition, his research addresses ethical aspects and bias in the use of AI in digitally enhanced teaching and learning environments.

Contact: m.kubsch@fu-berlin.de

Fabian Kieser is a PhD student in the Department of Physics Education at the Heidelberg University of Education. His research focuses on the use of AI methods, particularly those from the field of computer-based language processing, to specifically enhance problem-solving skills in physics.

Contact: kieser@ph-heidelberg.de

Paul Tschisgale is a PhD student in the didactics of physics at the IPN in Kiel. His research focuses, amongst other things, on the analysis and recording of problem-solving processes in physics through the use of AI and data science methods, as well as the generation of automated, adaptive feedback based on these findings. A particular focus is on supporting talented and potentially high-achieving pupils, especially those taking part in the Physics Olympiad.

Contact: tschisgale@leibniz-ipn.de