Title: Students’ Conceptions of Machine Learning

Duration: 2023 – 2025

Research Area: Responsible AI

The developments in the field of Artificial Intelligence (AI), in particular Machine Learning, play an increasing role in our everyday and professional life. Research and industry interest in the field of Machine Learning is steadily increasing. Thus, not only many professions will be affected by AI methods but also these methods will play an increasing role in everyday interaction of humans with technology.

It is, therefore, of crucial importance to understand how AI content and competencies can already be taught in schools, especially with regards to students’ everyday life in which they interact with these technologies daily. The chair of didactics of computer science concentrates on the questions of what students think about AI and how core ideas of Machine Learning can be taught in schools. For this purpose, pupils’ mental models of Machine Learning and central Machine Learning concepts are identified and Machine Learning workshops for school students as well as innovative learning materials are developed and evaluated.

Aims

Our primary objective is to delve into pupils’ conceptions of the mechanics behind machine learning and its associated technologies, aiming to illuminate their mental models. Through this exploration, we aim to identify key conceptual frameworks, facilitating the design of tailored learning interventions that resonate with students’ cognitive structures.

Problem

As AI and Machine Learning become integral to daily life [1], [2], pupils’ understanding of these concepts remains limited. Despite encountering these technologies in everyday applications and media, they often lack a foundational comprehension [3], [4], [5], [6], [7], [8]. Investigating their mental models not only sheds light on the learning process but also reveals common misconceptions, informing the development of effective teaching strategies to inspire and engage students [5], [9], [10].

Practical example

Findings on pupils’ conceptions about Machine Learning inform the design of learning materials and workshops available in German and can be viewed here: https://tu-dresden.de/inf/eduinf/ki-materialien. These resources include workshop concepts tailored for and tested with students of various ages, offering individual learning materials to teach specific Machine Learning concepts. Our goal is to equip teachers and educators with student-oriented, practical materials for teaching ML effectively.

Technology

The project utilizes a blend of quantitative and qualitative research methodologies, including surveys and interviews, to delve into students’ mental models of machine learning.

Outlook

Looking ahead, this project holds promise for enriching educational practices in the realm of Machine Learning. By refining our understanding of students’ mental models, we can tailor teaching approaches to foster deeper comprehension and engagement. This research has the potential to enhance not only individual learning experiences but also broader educational strategies for addressing the challenges posed by complex technological concepts. Further investigations in the area of conceptual change research could explore the efficacy of targeted interventions in addressing specific misconceptions about ML among students. Additionally, research could delve into the impact of cultural and contextual factors on the formation and evolution of mental models about Machine Learning in diverse student populations.

Publications

• JMarx, E., & Bergner, N. (2023a). Maschinelles Lernen in der Sekundarstufe I erlebbar machen: Workshop-Konzept zur Entwicklung einer intelligenten Museumsapp. 429–430. https://doi.org/10.18420/infos2023-053
• Marx, E., & Bergner, N. (2023b). Seminarkonzept zur fachlichen und fachdidaktischen Qualifizierung von Informatiklehramtsstudierenden zum Maschinellen Lernen. 10. Fachtagung Hochschuldidaktik Informatik (HDI) 2023, 10, 65–74.
• Marx, E., Leonhardt, T., Baberowski, D., & Bergner, N. (2022). Using Matchboxes to Teach the Basics of Machine Learning: An Analysis of (Possible) Misconceptions. Proceedings of the Second Teaching Machine Learning and Artificial Intelligence Workshop, 170, 25–29. https://proceedings.mlr.press/v170/marx22a
• Marx, E., Leonhardt, T., & Bergner, N. (2022). Brief Summary of Existing Research on Students’ Conceptions of AI. Proceedings of the 17th Workshop in Primary and Secondary Computing Education, 1–2. https://doi.org/10/gq2p2n
• Marx, E., Leonhardt, T., & Bergner, N. (2023). Secondary school students’ mental models and attitudes regarding artificial intelligence—A scoping review. Computers and Education: Artificial Intelligence, 5, 100169. https://doi.org/10.1016/j.caeai.2023.100169
• Marx, E., Leonhardt, T., Bergner, N., & Witt, C. (2023). Exploring students’ preinstructional mental models of machine learning: Preliminary findings. In J.-P. Pellet & G. Parriaux (Eds.), 16th international conference on informatics in schools: Situation, evolution, and perspectives, ISSEP 2023, local proceedings (pp. 233–236). Zenodo. https://doi.org/10.5281/zenodo.10015799

Literature

• [1] E. Brynjolfsson and T. Mitchell, “What can machine learning do? Workforce implications,” Science, vol. 358, no. 6370, pp. 1530–1534, Dec. 2017, doi: 10.1126/science.aap8062.
• [2] B. Zhang and A. Dafoe, “Artificial Intelligence: American Attitudes and Trends,” Center for the Governance ofAI, Future of Humanity Institute, University of Oxford, Oxford UK, 2019. doi: 10.2139/ssrn.3312874.
• [3] J. Szczuka, C. Strathmann, N. Szymczyk, L. Mavrina, and N. Krämer, “How do children acquire knowledge about voice assistants? A longitudinal field study on children’s knowledge about how voice assistants store and process data,” International Journal of Child-Computer Interaction, vol. 33, p. 100460, Jan. 2022, doi: 10.1016/j.ijcci.2022.100460.
• [4] H. Vartiainen, T. Toivonen, I. Jormanainen, J. Kahila, M. Tedre, and T. Valtonen, “Machine learning for middle schoolers: Learning through data-driven design,” International Journal of Child-Computer Interaction, vol. 29, p. 100281, Sep. 2021, doi: 10.1016/j.ijcci.2021.100281.
• [5] A. Bewersdorff, X. Zhai, J. Roberts, and C. Nerdel, “Myths, mis- and preconceptions of artificial intelligence: A review of the literature,” Computers and Education: Artificial Intelligence, vol. 4, p. 100143, Jan. 2023, doi: 10.1016/j.caeai.2023.100143.
• [6]A. Mühling and G. Große-Bölting, “Novices’ conceptions of machine learning,” Computers and Education: Artificial Intelligence, vol. 4, no. 100142, 2023, doi: 10.1016/j.caeai.2023.100142.
• [7] I. Evangelista, G. Blesio, and E. Benatti, “Why Are We Not Teaching Machine Learning at High School? A Proposal,” in 2018 World Engineering Education Forum – Global Engineering Deans Council (WEEF-GEDC), Institute of Electrical and Electronics Engineers (IEEE), Ed., Albuquerque, New Mexico, USA: Institute of Electrical and Electronics Engineers (IEEE), Nov. 2018, pp. 1–6. doi: 10.1109/WEEF-GEDC.2018.8629750.
• [8] P. Mertala and J. Fagerlund, “Finnish 5th and 6th graders’ misconceptions about artificial intelligence,” International Journal of Child-Computer Interaction, vol. 39, p. 100630, Mar. 2024, doi: 10.1016/j.ijcci.2023.100630.
• [9] D. Gentner, “Mental Models, Psychology of,” in International Encyclopedia of the Social & Behavioral Sciences, Elsevier, 2001, pp. 9683–9687. doi: 10.1016/B0-08-043076-7/01487-X.
• [10] I. M. Greca and M. A. Moreira, “Mental models, conceptual models, and modelling,” International Journal of Science Education, vol. 22, no. 1, pp. 1–11, Jan. 2000, doi: 10/btgzg6.

Team

Lead

•   Dr. Gregor Damnik, chair of didactics of computer science, TU Dresden

Team Members

• Erik Marx