Predicting Video Game Players’ Fun from Physiological and Behavioural Data

Abstract

Finding a physiological signature of a player's fun is a goal yet to be achieved in the field of adaptive gaming. The research presented in this paper tackles this issue by gathering physiological, behavioural and self-report data from over 200 participants who played off-the-shelf video games from the Assassin's Creed series within a minimally invasive laboratory environment. By leveraging machine learning techniques the prediction of the player's fun from its physiological and behavioural markers becomes a possibility. They provide clues as to which signals are the most relevant in establishing a physiological signature of the fun factor by providing an importance score based on the predictive power of each signal. Identifying those markers and their impact will prove crucial in the development of adaptive video games. Adataptive games that tailor their gameplay to the affective state of a player in order to deliver the optimal gaming experience. Indeed, an adaptive video game needs a continuous reading of the fun level to be able to respond to these changing fun levels in real time. While the predictive power of the presented classifier remains limited with a gain in the F1 score of 15% against random chance, it brings insight as to which physiological features might be the most informative for further analyses and discuss means by which low accuracy classification could still improve gaming experience.

Objectivity and Subjectivity in Games: Understanding Engagement and Addiction Mechanism

“…Such measure offered greater objectivity but generally costly (temporally and financially) and challenging to interpret [39]. Furthermore, psychophysiological measures are primarily independent of bias and can be measured continuously without breaking flow [27]. Meanwhile, UX's subjective assessment revolves around interviews, focus groups, in-game probes, and questionnaires, which are low-cost alternatives with fewer challenges around interpretation and offer variable insights both in depths and focus.…”

“…Game and gaming can generally be defined as the activity and action of playing games, respectively. In recent years, video games had been increasingly linked with potential benefits from the perspective of social, cognitive, and motivational [27]. Although playing games for recreational purposes could promote relaxation, challenge, and socialization [89], unrestricted gaming may become counter-intuitive where vulnerable individuals could be exposed to pathological gaming behaviors [12] and ultimately be addicted [45].…”

Dynamical modelling of player engagement and skill in the Quantum Moves 2 citizen science game

“…For in-stance, the winning team of the IEEE CIG 2017 Game Data Mining Competition [16] used deep neural networks to predict the remaining lifetime of each player using real world data. In another example, machine learning models predict the churn rate, the expected expenses or the fun of the player to adapt the game experience [17,18].…”

PASS: A Multimodal Database of Physical Activity and Stress for Mobile Passive Body/ Brain-Computer Interface Research

“…Physiological measures and passive body/brain-computer interfaces offer tremendous possibilities for monitoring individual functional states. In recent years, several works have shown that physiological measures can be used to assess e.g., the operator functional state of workers (i.e., workload, stress, fatigue), videogame player fun level, or even health markers (Banaee et al, 2013 ; Gagnon et al, 2016 ; Harrivel et al, 2017 ; Fortin-Côté et al, 2018 ). Moreover, it has been demonstrated that such assessment can be leveraged to augment interactions with intelligent systems, such as adaptive videogames or adaptive workload management systems (Parnandi and Gutierrez-Osuna, 2015 ; Aricò et al, 2016 ).…”