Your Gameplay Says It All: Modelling Motivation in Tom Clancy’s The Division

Abstract

Is it possible to predict the motivation of players just by observing their gameplay data? Even if so, how should we measure motivation in the first place? To address the above questions, on the one end, we collect a large dataset of gameplay data from players of the popular game Tom Clancy's The Division. On the other end, we ask them to report their levels of competence, autonomy, relatedness and presence using the Ubisoft Perceived Experience Questionnaire. After processing the survey responses in an ordinal fashion we employ preference learning methods based on support vector machines to infer the mapping between gameplay and the reported four motivation factors. Our key findings suggest that gameplay features are strong predictors of player motivation as the best obtained models reach accuracies of near certainty, from 92% up to 94% on unseen players.

Motivational Profiling of League of Legends Players

“…These results indicate a slight linear relationships between certain behaviors and motivational regulation. This is not surprising, as previous research examining game metrics and self-reported experience measures also reported low to medium correlations (Canossa et al, 2013 ; Schaekermann et al, 2017 ; Melhart et al, 2019 ). Among the 14 metrics we studied, the four motivational profiles varied significantly in terms of their winrate and assists in unranked matches, and kill-death-assist ratios in ranked matches (see also Table 3 and Figure 4 ).…”

“…The aim of this study was to explore how players' underlying motivational regulations relate to their experience and in-game behavior in a MOBA game. In contrast to previous research on predicting motivation from in-game metrics (Melhart et al, 2019 ), we present a novel, theory-driven approach for detecting motivational profiles, and compare these in terms of player experience and in-game behavior.…”

“…Canossa et al ( 2013 ), for example, investigated bivariate correlations and applied multiple supervised learning methods to identify relationships between in-game behavior in Minecraft (Mojang, 2011 ) and motivational factors, as measured by the Reiss Motivation Profiler (Reiss and Havercamp, 1998 ). Melhart et al ( 2019 ), in contrast, employed support vector machines to predict motivation in Tom Clancy's: The Division (Massive Entertainment, 2016 ) based on game metrics. They found that both linear and non-linear models successfully predicted motivation with an average accuracy of 65.89 and 75.62% respectively.…”

“…In turn, people's motivation for playing games shapes their player experience and in-game behavior (e.g., Yee et al, 2012;Canossa et al, 2013;Schaekermann et al, 2017;Melhart et al, 2019), as well as their well-being (Przybylski et al, 2009;Vella et al, 2013;Perry et al, 2018). However, while concepts from motivational psychology, particularly Self-Determination Theory (SDT, Deci and Ryan, 2000), commonly inform research on player experience (Tyack and Mekler, 2020) and game analytics (e.g., Canossa et al, 2013;Melhart et al, 2019), the notion of motivational regulation (Deci and Ryan, 2000)has received limited attention in the context of games (Tyack and Mekler, 2020). This is an unfortunate gap in our understanding of the player-computer interaction, as motivational regulations have been found to determine to what extent people experience positive emotions and need satisfaction, as well as how persistently they engage in a behavior (Neys et al, 2014).…”

From Chess and Atari to StarCraft and Beyond: How Game AI is Driving the World of AI

“…Additionally, companies are more and more collaborating with research institutions, to bring the latest innovations out to the industry. For example, Massive Entertainment and the University of Malta collaborated to predict the motivations of players in the popular game Tom Clancys The Division [49]. Other companies, such as King, are investing heavily in deep learning methods to automatically learn models of players that can then be used for playtesting new levels quickly [25].…”