Title: Beyond Intrinsic Motivation: Motivational design for healthy, autonomous engagement 

Abstract: Whether our goals are profit, behaviour change, civic engagement, or user well-being, motivation is a central issue in modern technology design, UX research, and HCI. Yet current approaches to motivational design have problems and limitations. In the attention economy, we motivate engagement by optimising content based on clicks and swipes - but this fails to distinguish between momentary shock or outrage and material users actually care for and value. Even more well-meaning approaches have problems - one standard approach is to "gamify" education and health activities by transforming them into satisfying task-reward loops and activity streaks. Such approaches are great at driving engagement, but they only work for simple tasks. When things get tough, and activities get uncomfortable (e.g., rehabilitation), challenging (e.g. deeper learning), or boring (e.g. cybersecurity practices), gamification hits a wall. Moreover the virtue of such approaches - their ease of implementation - comes with a fundamental shallowness: like the attention economy, gamification still focuses on unreflective engagement and arbitrary reward loops, poorly connected to the user's goals and values.

Outside the digital realm, the best motivational interventions tend to involve some kind of "coaching", connecting the activity to the user's interests, goals, and motivational style. How can we bring these approaches into motivational design, and make them work at scale? I propose an approach grounded in motivational psychology and enabled by LLM technologies, adjusting content to users' values and goals and targeting motivational interventions based on a model of their developing "motivational profile" for the activity. I discuss the empirical work I've done to date to ground this approach; addressing motivation in general technology use, social media, and cybersecurity. I then discuss the next steps and challenges along the way

Bio Hello! I'm an assistant professor at Aalborg University, formerly a Lecturer at University of Bristol, UK, and a postdoc with Elisa Mekler at Aalto and ITU Copenhagen. My background is in Computer science and Philosophy of Cognitive Science. In my research I combine the two: I draw on both theory-building and empirical data-driven approaches to clarify messy concepts like "autonomy", "agency" and "readiness-to-hand". The goal is generally to make these things tractable for empirical study and design while retaining the complexity that matters. I also make occasional trips into multisensory experience and gaming (player experience).

Affiliation: Aalborg University

LinkedIn: https://www.linkedin.com/in/daniel-bennett1981/

Scholar: https://scholar.google.dk/citations?user=KxrABMIAAAAJ&hl=en&oi=ao

Email: dtbe@cs.aau.dk