HI Data Forum @ LeaF – How could AI (help us to) understand emotions and motivation in learning?

Discover how Artificial Intelligence (AI) can deepen our understanding of human emotions, motivation, and learning. Emotions and motivation are essential for engagement and learning outcomes, yet they are complex, dynamic, and context-dependent. Through AI-powered approaches and multimodal methods, combining behavioural, physiological, and self-report data, researchers are developing new ways to capture and interpret these human processes.

The event is part of the Hybrid Intelligence: Human-AI Co-evolution and Learning in Multi-realities (HI) research program, which examines the role of data in advancing research and innovation in human-centred AI.

HI Data Forum @ LeaF brings together experts who are pioneering these advances:

• Haoyu Chen, Assistant Professor at the University of Oulu, explores how fine-grained bodily behaviours such as micro-gestures and spontaneous actions reveal hidden emotional cues beyond language and self-report.
• Olivia Metzner, PhD student at the University of Potsdam, presents research on using Large Language Models (LLMs) to automatically identify teachers’ motivational messages from real classroom data.
• Tiina Törmänen, Postdoctoral Researcher in the HI program, discusses methodological innovations for studying emotions and motivation in collaborative learning through integrated, process-oriented approaches.

A panel discussion will explore how AI can (help us to) understand emotions and motivation in learning, considering both opportunities and ethical challenges.

The event concludes with a networking session featuring refreshments and short presentations on ongoing initiatives such as the Profi7 program and LeaF services.

The hybrid event is free and open to all!
Secure your spot and register: onsite participation closes November 21, and online registration by November 26 at 12:00 (noon).

Register here.

 Program

13:00 Opening Tiina Törmänen

13:05-13:35 Human emotion understanding via bodily behaviors and goes beyond Haoyu Chen

13:35-14:05 From Words to Data: Classifying Teachers’ Motivational Messages in the Classroom With a Large Language Model Olivia Metzner

14:05-14:15 Break

14:15-14:45 Methodological advancements in studying emotions and motivation in collaborative learning - where do we go next? Tiina Törmänen

14:45-15:15 Panel discussion: How could AI (help us to) understand emotions and motivation in learning? - Haoyu Chen, Olivia Metzner, Hanna Järvenoja

15:15 - 16:00 Closing Session & Networking with Snacks

Presenters

Haoyu Chen is a tenure-track Assistant Professor at CMVS unit, in the University of Oulu, Finland, and Research Fellow in the Research Council of Finland. His research explores multimodal human behavior understanding, affective computing, and hybrid intelligence. Dr. Chen’s work focuses on decoding human emotions and cognition via fine-grained behavioral signals such as micro-gestures, social motion patterns, and embodied actions. He has published in top-tier venues such as CVPR, ICCV, ECCV, NeurIPS, IJCV, and ICML, and his recent projects involve spontaneous behavior analysis, emotion AI, and hybrid AI. Outside academia, he is a co-founder of two AI startups based in Helsinki, Finland, focusing on implementing the latest AI technologies to real world scenarios.

Olivia Metzner is a PhD student at the University of Potsdam, working in Rebecca Lazarides’ research group School and Instructional Research, and a fellow of the International Max Planck Research School on Learning, Institutions, and Future Evolution (LIFE). Her academic journey began with a Bachelor of Arts degree in Educational Science and European Ethnology from Humboldt-Universität zu Berlin, followed by a Master of Arts in Educational Science with a specialization in consulting and organizational development from TU Berlin. Her research focuses on teachers’ motivational messages and how LLMs can be applied to reliably detect these types of messages.

Tiina Törmänen is a postdoctoral researcher in the Hybrid Intelligence program, working under the theme Understanding humans in AI interaction, and in the Learning and Educational Technology (LET) Research Lab. Her interests cover emotions, motivation, and socially shared regulation in collaborative learning, and how to study these processes with multimodal multichannel data (e.g., video observations, physiological data, situational self-reports). She is also interested in motivational and emotional processes in learning with Artificial Intelligence (AI), and how AI could be used to both investigate and support motivation and emotions in learning.

Hanna Järvenoja's research interest is in the field of self-regulated learning, particularly motivation and emotion and their regulation in individual and social levels, socially shared regulation processes in collaborative learning, technology enhanced learning and supporting regulation of learning with technological tools. Previously, Järvenoja was leading, for example, a research project that studied elementary school students’ emotion regulation in collaborative learning situations (EmReg). Currently, she is continuing this line of research in a four-year long MoTor research project. In MoTor, that stand for "Not motivated - How regulation of learning builds up students' will to learn?", Järvenoja and her research group are investigating the complex reciprocal relationship between learners' regulation of learning and their learning motivation.

Register here.