Eliciting User Requirements for AI-Enhanced Learning Environments using a Participatory Approach
Bibeg Limbu, Irene Angelica-Chounta, Vilma Sukacke, Andromachi Filippidi, Chara Spyropoulou, Marianna Anagnostopoulou, Eleftheria Tsourlidaki, Nikos Karacapilidis · 2025
Read Eliciting User Requirements for AI-Enhanced Learning Environments using a Participatory Approach by Bibeg Limbu, Irene Angelica-Chounta, Vilma Sukacke, Andromachi Filippidi, Chara Spyropoulou, Marianna Anagnostopoulou, Eleftheria Tsourlidaki, Nikos Karacapilidis (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-21088.txt.
Introduction
Bibeg Limbu and seven co-authors draw out what users want from AI-enhanced learning settings. Their paper is from two thousand twenty-five. Teachers, learners, and other school parties may want unlike things from such tools. A design that misses those wants is apt to fail in real use.
They run two rounds of joint workshops. The first maps the parties' tech and teaching traits, read through activity theory. The second draws out what they hope AI will do in schools. By crossing needs with hopes, clashes show up, and those clashes turn into firm user needs. The paper ends with tips for the next wave of design work.
Catalog record
| CREATOR | Bibeg Limbu, Irene Angelica-Chounta, Vilma Sukacke, Andromachi Filippidi, Chara Spyropoulou, Marianna Anagnostopoulou, Eleftheria Tsourlidaki, Nikos Karacapilidis |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 29 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-21088
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Eliciting User Requirements for AI-Enhanced Learning Environments using a Participatory Approach Bibeg Limbu, Irene Angelica-Chounta, Vilma Sukacke, Andromachi Filippidi, Chara Spyropoulou, Marianna Anagnostopoulou, Eleftheria Tsourlidaki, Nikos Karacapilidis Abstract This paper explores the needs & expectations of educational stakeholders for AI (Artificial Intelligence)-enhanced learning environments. Data was collected following two-phased participatory workshops. The first workshop outlined stakeholders’ profiles in terms of technical and pedagogical characteristics. The qualitative data collected was analysed using deductive thematic analysis with Activity Theory, explicating the user needs. The second workshop articulated expectations related to the integration of AI in education. Inductive thematic analysis of the second workshop led to the elicitation of users’ expectations. We cross-examined the needs & expectations, identifying contradictions, to generate user requirements for emerging technologies. The paper provides suggestions for future design initiatives that incorporate AI in learning environments. 1 Introduction The integration of AI in Education (AIED) should follow human-centered design principles, thus prioritizing human needs and capabilities, rather than replacing human roles or undermining human agency [ 11 , 3 ] . This necessitates interdisciplinary collaboration [ 6 ] . To promote human-centered design of AI-enhanced learning environments, the involvement of educational stakeholders, such as teachers and learners, in the design process of these environments is necessary [ 18 ] . At the same time, the integration should be guided by established learning and pedagogical theories [ 3 , 13 ] . To that end, participatory design is an approach that involves the active and meaningful participation of stakeholders, including end-users, in the design process. It aims to ensure that the solutions meet the needs and expectations of the end users [ 12 ] . This methodology emphasizes collaboration, co-creation, and shared decision-making between designers and stakeholders. We explored two distinct learning settings drawn from the broader project context (see Section 3 ): a higher education course for pre-service teacher training, and a course on Leapfrogging Industry 4.0 technologies designed for civic society watchdogs and EU civilian missions. Stakeholders from these settings were involved in two workshops with the aim of eliciting requirements for the design of an AI-enhanced learning platform for 21st-century skills. Qualitative data collected during the participatory workshops were analyzed using thematic analysis. To guide our analysis, we formulated the following questions: RQ1: What are the needs of educators and learners for teaching and learning 21st-century skills? In particular, we want to explore a) the current state of 21st-century skills training in practice; b) the tools and pedagogical approaches used, and c) the challenges faced. RQ2: What are the expectations of educators and learners of emerging technologies for the development of 21st-century skills? We aim to specify a) the challenges in designing learning activities, and how emerging technologies can solve them, and b) how educators monitor learner performance, what evaluation methods they use, and what types of AI-powered recommendations would they find valuable? 2 Related work 2.1 Participatory design We organised two participatory workshops to collect data to elicit user requirements. Participatory design is an approach that prioritizes the active involvement of participants, particularly end-users, in the design and development process [ 16 ] . Participatory design also involves end-users, stakeholders, and other relevant parties in the decision-making processes of a product, service, or system. Participants, as active contributors, offer insights to guide the development process, an element that characterizes this methodology [ 17 ] . By fostering collaboration, co-creation, and shared ownership among stakeholders, participatory design aims to develop solutions that better address real-world challenges and meet the diverse needs of the community. It aims to create inclusive, user-centered, and contextually relevant solutions by actively engaging those affected by or benefiting from the design outcomes. Participatory design recognizes the expertise and lived experiences of the participants, enabling them to contribute their perspectives throughout the design process [ 7 ] . This approach can lead to greater user satisfaction, increased usability, and more socially responsible outcomes. 2.2 Activity Theory Figure 1: Uden & Ching’s [ 19 ] Ecosystem for AIED design Activity Theory (AcT) emphasises a holistic "ecosystem" perspective in the requirements analysis, instead of the excessive focus on technology put by other models like the TPACK model [ 19 ] . It prioritizes the importance of studying the real-life use of technology as part of the unfolding human interaction with the world over the traditional HCI methods, which focus on formal, abstract models of the user and the system as separate entities within the interaction. AcT facilitates a rich, systemic understanding of users’ needs embedded within their actual work, tools, rules, and communities [ 9 ] . Using AcT in the analysis of user requirements supports the design of solutions that address the actual underlying problems and constraints, thus improving the relevance and sustainability of the solutions [ 5 ] . AcT conceptualizes activity systems as the fundamental unit of analysis for examining complex, real-life human practices. An activity system is a dynamic, historically evolving, goal-directed, and tool-mediated structure of human practice [ 9 ] . Following Uden & Ching’s [ 19 ] interpretation of AcT in the context of AIED (see Figure 1 ), AcT presupposes that the relationship between the Subject (Teachers) and the Object (Learning Outcomes) is mediated by the Artifacts (Technology, including AI). This activity system is situated within a Community (Schools), which is governed by a set of Rules (Pedagogy & Curriculum) and organized through a Division of labour . 3 Methodology Figure 2: Workshops structure per phase of implementation. We adopted a participatory approach in line with the assumptions of AcT. In particular, we invited stakeholders from two pilots of a European research project (augMENTOR 1 1 1 https://augmentor-project.eu ) to participate in this study. These stakeholders consist of higher education teachers, teacher trainers, and policymakers with extensive experience in distance learning and the use of technologies in their daily teaching practice. The pilots were organized and led by Kaunas University of Technology (Pilot 1) and the University of Patras (Pilot 2). The structure of the study is presented in Figure 2 ; each pilot conducted one workshop per phase (two in total). The workshops’ structure was the same for all pilot partners, with minor adjustments to meet the needs of different groups of participants. Storyboards were used to scaffold and communicate the workshop structure between the pilots. Each pilot partner was responsible for facilitating their workshops. The participants from the first workshop were encouraged to participate in the second workshop. This is aimed at maximizing user engagement, facilitating a collaborative atmosphere, and ensuring continuity, enabling participants to build upon earlier discussions. Insights were collaboratively documented using an online whiteboard. In Workshop 1, there were 6 participants in Pilot 1, and 6 participants in Pilot 2. In Workshop 2, there were 8 participants in Pilot 1, and 5 participants in Pilot 2. No personal data was collected during the study. Participants confirmed their voluntary participation with signed consent forms, and the study was approved by the project’s Ethics Advisory Board and Data Protection Officers of the pilot partners. 4 Results We used thematic analysis to synthesize insights from the collected qualitative data. Thematic analysis is a method for identifying, analysing, and reporting patterns (themes) within data [ 2 ] .