Machine learning in digital games: a survey

Artificial intelligence for digital games constitutes the implementation of a set of algorithms and techniques from both traditional and modern artificial intelligence in order to provide solutions to a range of game dependent problems. However, the majority of current approaches lead to predefined, static and predictable game agent responses, with no ability to adjust during game-play to the behaviour or playing style of the player. Machine learning techniques provide a way to improve the behavioural dynamics of computer controlled game agents by facilitating the automated generation and selection of behaviours, thus enhancing the capabilities of digital game artificial intelligence and providing the opportunity to create more engaging and entertaining game-play experiences. This paper provides a survey of the current state of academic machine learning research for digital game environments, with respect to the use of techniques from neural networks, evolutionary computation and reinforcement learning for game agent control.     

Implicit learning as a design strategy for learning games: Alert Hockey

Concussion education and prevention for youth hockey players has been an issue of recent concern amongst sport medicine practitioners and hockey’s administrative bodies. This article details the assessment of a sports-action hockey video game that aims to reduce the aggressive and negligent behaviours that can lead to concussions. The game, termed Alert Hockey, was designed to modify game playing behaviour by embedding an implicit teaching mechanism within the gameplay. In Alert Hockey, participants were expected to learn by simply playing to win, in contrast to playing to learn. We studied learning in an experimental simulated environment where the possibility to win the game was exaggerated as a consequence of desirable safety behaviours (positive learning group) and effectively reduced as a consequence of undesirable (negative learning group) behaviour. The positive learning group significantly improved their mean score on a composite behavioural indicator compared with no significant change amongst control group participants. The results demonstrate that implicit learning embedded in a sports-action game can lead to changes in game-play behaviour.     

The influence of virtual presence: Effects on experienced cognitive load and learning outcomes in educational computer games

Does the immersive design of an educational gaming environment affect learners’ virtual presence and how much do they learn? Does virtual presence affect learning? This study tries to answer these questions by examining the differences in virtual presence and learning outcomes in two different computer-based multimedia environments: a gaming environment with high immersive design vs. hypertext learning environment with low immersive design. As the main focus, the effect of virtual presence on learning is also explained and tested. By identifying virtual presence as a variable that may determine learning, it is argued that computer gaming environments present a new challenge for researchers to investigate, particularly, the effects of virtual presence on the immersive design of games in order to help designers to predict which instructional configurations will maximize learning performance. In general, results revealed that the high-immersive gaming environment leads to the strongest form of virtual presence but also decreased learning. Although regression analyses indicate that virtual presence positively influences trivial- and non-trivial learning outcomes, learners who learned in a low-immersive environment outperformed the gaming group. A mediation analysis showed that the relation between virtual presence and non-trivial learning outcomes is partly mediated through increased cognitive load.     

Commitment to multiplayer online games: An investment model approach

Multiplayer online games have highly committed players who continue playing these games for years. In the present study, commitment to multiplayer online games was examined from an investment model perspective. The investment model was originally developed to explain persistence in romantic relationships, ultimately identifying the three predictors of commitment as satisfaction, quality of alternatives, and investment size. Furthermore, commitment has been linked to greater accommodation (i.e., constructive responses) in relationship conflicts. The present study extended the model to the context of multiplayer online games. Participants (N = 176) who regularly played a multiplayer online game completed modified versions of the investment model and accommodation scales in an online questionnaire. Structural equation modeling analyses showed that satisfaction and investment predicted greater game commitment, which in turn predicted greater accommodation toward the game's problems (i.e., lower endorsement of exit behaviors, greater endorsement of voice and loyalty behaviors). Findings showed that the investment model provides a valid framework for examining game commitment.     

Exploring different technological platforms for supporting co-located collaborative games in the classroom

Computer Supported Collaborative Learning is a pedagogical approach that can be used for deploying educational games in the classroom. However, there is no clear understanding as to which technological platforms are better suited for deploying co-located collaborative games, nor the general affordances that are required. In this work we explore two different technological platforms for developing collaborative games in the classroom: one based on augmented reality technology and the other based on multiple-mice technology. In both cases, the same game was introduced to teach electrostatics and the results were compared experimentally using a real class.     

Co-reflection in online learning: Collaborative critical thinking as narrative

This article presents findings from a comparative case study of the learning experiences of two graduate students in an online action research course. The key roles played by reflection and co-reflection, an emerging concept, are identified through the use of narrative analysis. Co-reflection is a collaborative critical thinking process mediated by language, broadly construed to include all meaningful signs. Two types of co-reflection are proposed: tacit and active. Regardless of type, the evidence shows that co-reflection involves cognitive and affective interactions in synergy with relationship building. To the study of group cognition, this study contributes evidence of the potential of co-reflection as a core process. The simple, flexible software tools used in the course (wiki-style collaborative software and simple email and chat programs) effectively supported inquiry learning and co-reflection by allowing learners to freely and easily create their own web pages and to adapt the tools for their different communication and learning styles.     

A theory of online learning as online participation

In this paper, an initial theory of online learning as online participation is suggested. It is argued that online learner participation (1) is a complex process of taking part and maintaining relations with others, (2) is supported by physical and psychological tools, (3) is not synonymous with talking or writing, and (4) is supported by all kinds of engaging activities. Participation and learning are argued to be inseparable and jointly constituting. The implication of the theory is straightforward: If we want to enhance online learning, we need to enhance online learner participation.     

Computer games in pre-school settings: Didactical challenges when commercial educational computer games are implemented in kindergartens

This article focuses on the didactical implications when commercial educational computer games are used in Norwegian kindergartens by analysing the dramaturgy and the didactics of one particular game and the game in use in a pedagogical context. Our justification for analysing the game by using dramaturgic theory is that we consider the game to be a multimodal performance utilising text, graphics, pictures, sound and animation. Similarly we analyse the didactic situation by using dramaturgic theories and concepts because we consider the didactic meeting between the medium (the game), children (the player(s), and teacher to be a dramaturgic situation comprising different roles, actions in progress, time and space. Our data material shows that the pre-school teacher is more or less absent during the children's playing with the computer games, but when the pre-school teacher involves him/herself, she finds it difficult to realise her ideal socio-cultural didactical project in which dialogue is a central medium for exploration and learning. Through our analysis of the data material we find that there are two different dramaturgies at stake; the built-in interactive dramaturgy of the game materialised in the gaming situation and the dialogical dramaturgy that the pre-school teacher tries to create in the didactical situation. This implies that there is a didactical dissonance between the learning space which the game and the learning space the pre-school teacher wants to construct and orchestrate.     

Assessing online collaborative learning: process and product

The assessment of online collaborative study presents new opportunities and challenges, both in terms of separating the process and product of collaboration, and in the support of skills development. The purpose of this paper is to explore the role of assessment with respect to the processes and products of online collaborative study. It describes a qualitative case study of staff and students perspectives on two UK Open University courses which have used a variety of models of online collaborative assessment. The findings underline the importance of assessment in ensuring online participation, and in supporting the practice and development of online collaborative learning. They have led to a number of recommendations for the assessment of online collaborative learning.     

The learning kit project: Software tools for supporting and researching regulation of collaborative learning

Computer-supported collaborative learning (CSCL) is a dynamic and varied area of research. Ideally, tools for CSCL support and encourage solo and group learning processes and products. However, most CSCL research does not focus on supporting and sustaining the co-construction of knowledge. We identify four reasons for this situation and identify three critical resources every collaborator brings to collaborations that are underutilized in CSCL research: (a) prior knowledge, (b) information not yet transformed into knowledge that is judged relevant to the task(s) addressed in collaboration, and (c) cognitive processes used to construct these informational resources. Finally, we introduce gStudy, a software tool designed to advance research in the learning sciences. gStudy helps learners manage cognitive load so they can re-assign cognitive resources to self-, co-, and shared regulation; and it automatically and unobtrusively traces each user′s engagement with content and the means chosen for cognitively processing content, thus generating real-time performance data about processes of collaborative learning.     

Virtual learning spaces in distance education: tools for the EVA project

The EVA (espacios virtuales de aprendizaje, or virtual learning spaces) project applies artificial intelligence tools to teleteaching methods, in a way that eliminates or mitigates the need for synchronous and in situ education. (A) A taxonomy of the space of knowledge (also called ‘Learning Space'; currently our prototype teaches M.Sc. courses in Computer Science) is formed and discretized. (B) EVA finds each student's initial knowledge state (through a computer examination) and final (desired) knowledge state, and from these, a particular learning trajectory is designed for that student. (C) Personalized books (called polybooks, because they are formed by modules (chapters) written in a variety of media) are armed by concatenating — along the learning trajectory — modules from a large pool, and sent to the student through the net in a store-and-forward fashion. (D) EVA searches the net for teaching material which has not been indexed in the discretized learning space, using a tool (Clasitex) inside an agent that finds the main themes or topics that an article (written in natural language) covers. (E) EVA also schedules for each student synchronous activities (lectures in TV, teleconferences, on-line question and answering sessions, chats). (F) EVA suggests for each student suitable ‘classmates' (students having similar learning trajectories) in her town, as well as possible advisers (students or alumni having knowledge that the student is acquiring). The present status, problems, models and tools of EVA are presented.     

The use of online digital resources and educational digital libraries in higher education

This paper summarizes results from a national survey of 4,678 respondents, representing 119 institutions of higher education in the United States regarding their use of digital resources for scholarly purposes. This paper presents the following results: (1) demographics commonly used in higher education to categorize populations such as institution type or level of teaching experience could not reliably predict use of online digital resources, (2) valuing online digital resources corresponds with only higher levels of use for certain types of digital resources, (3) lack of time was a significant barrier to use of materials while, paradoxically, respondents indicated that they used them because they save time, (4) respondents did not tend to intentionally look to the Internet as a trusted resource for learning about teaching.     

Exploring selfish reinforcement learning in repeated games with stochastic rewards

In this paper we introduce a new multi-agent reinforcement learning algorithm, called exploring selfish reinforcement learning (ESRL). ESRL allows agents to reach optimal solutions in repeated non-zero sum games with stochastic rewards, by using coordinated exploration. First, two ESRL algorithms for respectively common interest and conflicting interest games are presented. Both ESRL algorithms are based on the same idea, i.e. an agent explores by temporarily excluding some of the local actions from its private action space, to give the team of agents the opportunity to look for better solutions in a reduced joint action space. In a latter stage these two algorithms are transformed into one generic algorithm which does not assume that the type of the game is known in advance. ESRL is able to find the Pareto optimal solution in common interest games without communication. In conflicting interest games ESRL only needs limited communication to learn a fair periodical policy, resulting in a good overall policy. Important to know is that ESRL agents are independent in the sense that they only use their own action choices and rewards to base their decisions on, that ESRL agents are flexible in learning different solution concepts and they can handle both stochastic, possible delayed rewards and asynchronous action selection. A real-life experiment, i.e. adaptive load-balancing of parallel applications is added.     

Designing educational software with students through collaborative design games: The We!Design&Play framework

In this paper, we present a framework for the development of collaborative design games that can be employed in participatory design sessions with students for the design of educational applications. The framework is inspired by idea generation theory and the design games literature, and guides the development of board games which, through the use of adequate stimuli, rules and props, facilitate students in extracting and expressing their needs, desires and prospects regarding future educational software. To evaluate the proposed framework three studies were conducted. The first study aimed at the design of a web learning platform with the participation of 62 undergraduate higher education students in 13 design sessions; in the second study, a structured design approach was employed (12 sessions, 54 students) with the same design objective for comparison reasons; in the third study, the framework was deployed for the design of an electronic assessment application so as to examine its applicability in different learning domains (8 design sessions, 28 students). Students were very positive regarding both their participation and experience with the design games, and the needs elicited. The games favored a quick, broad exploration of the design space and facilitated the elicitation of numerous diverse needs and ideas, almost twice as many as produced by the structured approach. They also facilitated the creation of an informal atmosphere and limited the effects of common social influences on idea generation, such as social loafing, evaluation apprehension and production blocking. The three studies indicated that the proposed framework may simplify the development and employment of effective and efficient participatory design sessions in educational settings.     

Falling in love with online games: The uses and gratifications perspective

Playing online games is experience-oriented but few studies have explored the user’s initial (trial) reaction to game playing and how this further influences a player’s behavior. Drawing upon the Uses and Gratifications theory, we investigated players’ multiple gratifications for playing (i.e. achievement, enjoyment and social interaction) and their experience with the service mechanisms offered after they had played an online game. This study explores the important antecedents of players’ proactive “stickiness” to a specific online game and examines the relationships among these antecedents. The results show that both the gratifications and service mechanisms significantly affect a player’s continued motivation to play, which is crucial to a player’s proactive stickiness to an online game.     

Simple reinforcement learning agents: Pareto beats Nash in an algorithmic game theory study

Repeated play in games by simple adaptive agents is investigated. The agents use Q-learning, a special form of reinforcement learning, to direct learning of behavioral strategies in a number of 2×2 games. The agents are able effectively to maximize the total wealth extracted. This often leads to Pareto optimal outcomes. When the rewards signals are sufficiently clear, Pareto optimal outcomes will largely be achieved. The effect can select Pareto outcomes that are not Nash equilibria and it can select Pareto optimal outcomes among Nash equilibria.Acknowledgement This material is based upon work supported by, or in part by, NSF grant number SES-9709548. We wish to thank an anonymous referee for a number of very helpful suggestions.     

Student interactions in online discussion forums: their perception on learning with business simulation games

Digital technology offers new teaching methods with controversial results over learning. They allow students to develop a more active participation in their learning process although it does not always drive to unequivocal better learning outcomes. This study aims to offer additional evidence on the contribution of business simulation games to students’ learning outcomes, considering student interactions in online discussion forums. We conducted a qualitative research with the online discussion forums of 5 different courses at bachelor and master levels, which involves 41 students’ teams. The final sample was composed of 3681 messages posted by the students. The results reveal that some generic and specific managerial skills exert a positive influence on learning outcomes. Students mostly highlighted teamwork, decision-making, information processing, reaching agreements, and dealing with uncertainty as the most relevant contributions of the game towards their learning. These results have instructional and pedagogical implications for determining the best way to enhance students’ motivation and learning outcomes when using digital technology methods, which involves recommendations that affect their design and monitoring.     

Micro‐adaptivity: protecting immersion in didactically adaptive digital educational games

The idea of utilizing the rich potential of today's computer games for educational purposes excites educators, scientists and technicians. Despite the significant hype over digital game‐based learning, the genre is currently at an early stage. One of the most significant challenges for research and development in this area is establishing intelligent mechanisms to support and guide the learner, and to realize a subtle balance between learning and gaming, and between challenge and ability on an individual basis. In contrast to traditional approaches of adaptive and intelligent tutoring, the key advantage of games is their immersive and motivational potential. Because of this, the psycho‐pedagogical and didactic measures must not compromise gaming experience, immersion and flow. In the present paper, we introduce the concept of micro‐adaptivity, an approach that enables an educational game to intelligently monitor and interpret the learner's behaviour in the game's virtual world in a non‐invasive manner. On this basis, micro‐adaptivity enables interventions, support, guidance or feedback in a meaningful, personalized way that is embedded in the game's flow. The presented approach was developed in the context of the European Enhanced Learning Experience and Knowledge TRAnsfer project. This project also realized a prototype game, demonstrating the capabilities, strengths and weaknesses of micro‐adaptivity.     

The effects of computer-based games and collaboration in large groups vs. collaboration in pairs or traditional methods

As new technologies have emerged in the last few years, the learning process has been changing. New and powerful e-learning systems are being developed and new teaching methods can be used in classrooms. In this paper, we present a computer-based game with an educational background that is played on a large-size tabletop display. The game can be used as reinforcement for educational content related to historical ages. The game uses natural interaction. A study to compare the traditional learning method with a collaborative learning method using the game was carried out. A group of up to 12 children could learn together using the game. The experience of children in large groups was also compared with the experience of children playing in pairs. One hundred children between 8 and 11 years old participated in the study; they were divided into three groups (LGroup, Pairs, TClass). When the pre-test and the post-test results were compared, it was shown that the children learned the contents in all three groups. The results also showed that there were statistically significant differences between the traditional method and the game played in a large group in favour of children who played the game in the large group. The knowledge acquired was independent from gender and age. There were no statistically significant differences between learning in large groups or learning in pairs. In both cases, the children expressed their satisfaction for the game and found it easy to use. Therefore, playing games of this type collaboratively in large groups or in pairs can be a valuable learning method that can be combined with traditional methods.