Personalized Adaptive Gamification Systems to Improve Performance and Motivation

The objective of this dissertation is to empirically test the effects of personalized gamification on individuals' performance. In order to achieve this objective, multiple experiments and case studies are conducted, and methods to help in the design of personalized and adaptive gamified applications are presented. Gamification aims to implement game elements into non-game contexts with the goal of increasing the motivation and performance of individuals on a task or set of tasks (i.e., promote action or behavior). Studies support the potential for gamification to improve the motivation and performance of individuals in a wide range of applications; however, researchers caution that some caveats exist. Primarily, they caution that the current "one-size-fits-all" design approach, which assumes individuals are a homogenous group that reacts similarly to game elements, is not an optimal design approach. This is because research indicates that a game element that positively impacts an individual's motivation might not improve, or could even worsen, the motivation of another individual. Moreover, studies have shown that the positive effects of gamification on individuals' motivation may diminish over time. Due to these limitations, researchers are starting to explore how player type models (e.g., Hexad player type) and personality trait models (e.g., Five Factor model) can be implemented to advance gamification. Both personality traits and player type models aim to capture the differences between individuals, which can help explain the dissimilarities in their behaviors and attitudes. However, personality trait models can be understood as a high-level conceptualization of individual differences, not focused on any specific domain or behavior as opposed to player type models. Unfortunately, existing gamification methods do not personalize applications at an individual level or systematically capture data of an individual's interaction with an application to adapt accordingly. Researchers advise that in order to solve the diminishing effects of gamification and maximize its potential, more emphasis should be placed on designing methods for personalized and adaptive applications that are capable of motivating and improving individuals' performance. In light of these limitations, this dissertation first presents a study that explores the relationship between an individual's player type and their perception of game elements implemented in an educational gamified application. Subsequently, a controlled experiment is presented in which a gamified and a non-gamified physical-interactive application are used. In this experiment, the relationship that individuals' player type and their personality trait have with both their perception of and performance in the application are explored. The results of these studies show that individuals' performance and perception of the gamified applications used are associated with their player type. Moreover, the results reveal that the Hexad player type model was able to explain more of the variability in individuals' performance data than the Five Factor personality trait model. Combining these findings with previous research, a method to personalize gamified applications based on individuals' Hexad player type is presented. The method is intended to help designers in the systematic selection of game elements that are worth exploring. The method is based on a game elements recommendation algorithm that aims to minimize the design complexity of gamified applications while maximizing the potential to improve the motivation and performance of individuals. In addition, a machine learning method that analyses task and facial expression data to predict the performance of individuals in a gamified application is presented. The results of a case study reveal the capability and feasibility of the machine learning method to accurately predict the performance of individuals in gamified applications. This method could potentially be used to adapt the task difficulty and the game elements of an application. Finally, to explore the effects that personalized gamification has on individuals' performance, a randomized experiment is conducted. In this experiment, the performance of participants who interact with (i) a personalized gamified application, (ii) a non-personalized gamified application, (iii) a non-gamified application, and (iv) a counter-personalized gamified application is analyzed. The results reveal that individuals interacting with the personalized gamified application performed better than any other group. In contrast, those who interacted with the counter-personalized gamified application performed worse than any other group and did not show any performance improvement after interacting with the application for a second time. This dissertation provides empirical evidence of the benefits that personalized gamification has on individuals' performance. Due to the heterogeneity of individuals, the methods presented in this dissertation could be employed in a wide range of contexts to help personalize and adapt gamified applications with the objective to improve individuals' performance. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]