Aniket Nagle

Linking Recognition Accuracy and User Experience in an Affective Feedback Loop

In an affective feedback loop, the computer maps various measurements to affective variables such as enjoyment, then adapts its behavior based on the recognized affects. The affect recognition is never perfect, and its accuracy (percentage of times the correct affective state is recognized) depends on many factors. However, it is unclear how this accuracy relates to the overall user experience. As recognition accuracy is difficult to control in a real affective feedback loop, we describe a method of simulating recognition accuracy in a game where difficulty is increased or decreased after each round. The game was played by 261 participants at different simulated recognition accuracies. Participants reported their satisfaction with the recognition algorithm as well as their overall game experience. We observed that in such a game, the affective feedback loop must adapt game difficulty with an accuracy of at least 80 percent to be accepted by users. Furthermore, users who do not enjoy the game are likely to stop playing it rather than continue playing and report low enjoyment. However, the acceptable recognition accuracy may not generalize to other contexts, and studies of affect recognition accuracies in other applications are needed.

Towards a system of customized video game mechanics based on player personality: Relating the Big Five personality traits with difficulty adaptation in a first-person shooter game

While personality is known to moderate the kind of video games people like to play, the link between personality and game mechanics has not been explored. Finding such a link can enable customizing game mechanics based on personality, potentially making games more enjoyable and suited to a wide range of players. The present work investigated the relationship between the Big Five personality traits and four different implementations of difficulty adaptation, a popular game mechanic, in a first-person shooter game. In Study 1, a linear regression model was derived to relate the five personality traits with enjoyment (ENJ) and gameplay duration (DUR) in the four difficulty adaptations. In Study 2, this regression model was used to construct a predictor that chose a difficulty adaptation which maximized ENJ and DUR based on player personality. The predictor was tested against dynamic difficulty adjustment (DDA), which matched difficulty to user performance. ENJ and DUR were significantly higher in the prediction group than the DDA group. The present work highlights the importance of difficulty adaptation as a game mechanic, and suggests that personality could also be related to other game mechanics. Accordingly, a framework for personality-based game mechanics customization is proposed to foster future research.

The effect of different difficulty adaptation strategies on enjoyment and performance in a serious game for memory training

Among the approaches used to make serious games and learning environments enjoyable, two important ones are 1) adapting task difficulty to match user performance, which leverages the theories of zone of proximal development and flow and 2) providing a high degree of control and choice, which uses constructs from self-determination theory. However, it is not clear whether the two approaches lead to similar results or whether focusing on performance may lead to suboptimal enjoyment. This is relevant since, in addition to being performance-oriented, serious games and learning environments also need to be enjoyable to be effective. To address this issue, we evaluated two kinds of difficulty adaptation in a simple memory training game: one adaptation based solely on difficulty-performance matching and the other based solely on providing a high degree of control/choice. Our results suggest that neither approach is optimal: players enjoyed themselves the most in the control/choice mode, but performed the best in the difficulty-performance matching mode. Player behavior in the game suggests that a trade-off between maximizing performance and maximizing enjoyment could be achieved by a combination of the two approaches. However, long-term studies would be required to determine whether such a combination could indeed be effective.

Can two-player games increase motivation in rehabilitation robotics?

Rehabilitation robots have the potential to greatly improve motor rehabilitation. However, the patient must be properly motivated to actively participate in therapy. Several strategies have been suggested to improve patient motivation, but one element has not yet been explored: playing with other people. We designed a two-player rehabilitation game played by two people using two ARMin III robots. We tested three game modes: single-player (competing against a computer), competitive (competing against a human), and cooperative (cooperating with a human against a computer). All modes were played by 24 healthy subjects who filled out questionnaires about their personality and in-game motivation. Almost all subjects preferred playing the two-player game modes to the single-player one, as they enjoyed being able to talk and interact with another person. However, there were two distinct player groups. One group liked the competitive mode but not the cooperative mode while the other liked the cooperative but not the competitive mode. Subjects who liked the competitive mode also put more effort into it. Finally, subjects’ personalities partially predicted what mode they would like. This emphasizes that two-player rehabilitation games have advantages over single-player ones, but that the right game needs to be chosen for each subject. An extended patient study is planned for the near future. Categories and Subject Descriptors H.1.2 [User/Machine Systems]: Human factors. H.5.1 [Multimedia Information Systems]: Artificial, augmented and virtual realities. J.3 [Life and Medical Sciences]: Health. General Terms Design, Experimentation, Human Factors.

High User Control in Game Design Elements Increases Compliance and In-game Performance in a Memory Training Game

Computer games are increasingly being used for training cognitive functions like working memory and attention among the growing population of older adults. While cognitive training games often include elements like difficulty adaptation, rewards, and visual themes to make the games more enjoyable and effective, the effect of different degrees of afforded user control in manipulating these elements has not been systematically studied. To address this issue, two distinct implementations of the three aforementioned game elements were tested among healthy older adults (N = 21, 69.9 ± 6.4 years old) playing a game-like version of the n-back task on a tablet at home for 3 weeks. Two modes were considered, differentiated by the afforded degree of user control of the three elements: user control of difficulty vs. automatic difficulty adaptation, difficulty-dependent rewards vs. automatic feedback messages, and user choice of visual theme vs. no choice. The two modes (“USER-CONTROL” and “AUTO”) were compared for frequency of play, duration of play, and in-game performance. Participants were free to play the game whenever and for however long they wished. Participants in USER-CONTROL exhibited significantly higher frequency of playing, total play duration, and in-game performance than participants in AUTO. The results of the present study demonstrate the efficacy of providing user control in the three game elements, while validating a home-based study design in which participants were not bound by any training regimen, and could play the game whenever they wished. The results have implications for designing cognitive training games that elicit higher compliance and better in-game performance, with an emphasis on home-based training.

How would you like to be rewarded? Relating the Big-Five personality traits with reward contingency in a cognitive training puzzle game

Rewards in serious games have had a long and at times contentious history about their effect on intrinsic motivation, with contradictory findings about whether they are detrimental to intrinsic motivation or not. While recent research suggests that rewards can be beneficial, the effect of reward contingency on motivation is still unclear. The link between player personality and different motivations to play games, and the differing reward preferences arising from the different motivations suggests that personality could moderate the effect of a reward contingency on intrinsic motivation. Knowing this effect could help in better leveraging rewards to make motivating serious games. Two studies, therefore, were conducted to investigate the relationship between personality and reward contingency. In Study 1, 60 participants played a block-matching puzzle game that can train executive functions, with each participant randomly assigned to one of three conditions: performance-contingent rewards, which gave rewards after achieving a specific performance level, task-contingent rewards, which gave rewards simply for completing a task, and no reward condition. Big Five personality traits values of the participants were assessed and related to their intrinsic motivation and performance using a linear regression model. A different combination of the five traits significantly predicted motivation and performance in each condition, with the performance-contingent condition yielding significantly higher enjoyment and performance than the other conditions. The efficacy of the linear model was tested in a follow-up Study 2, where 30 participants played the same puzzle game, with half the participants given performance-contingent rewards, and the other half being assigned a reward contingency that would maximize their intrinsic motivation and performance based on the model obtained in Study 1. Enjoyment and competence subscales of intrinsic motivation, and performance, were significantly higher in the group that received a personality-based contingency as compared to the one that always received performance-contingent rewards. The present study highlights the role of personality in moderating the effect of different reward contingencies on motivation and performance, and could explain the contradictory findings of existing research. The results demonstrate that serious games can be made more motivating and performance-oriented by using appropriately chosen reward contingency.