Game Metrics Research Articles

Digital game-based learning in architecture education: Consolidating visual design principles in freshmen

Using games as educational tools has been a captivating subject in the academic domain. There is an increasing number of digital games designed to support architectural education. This paper introduces a serious game aimed at enhancing basic design knowledge for first-year architecture students. The game focuses on teaching and testing visual design principles such as emphasis, balance, and rhythm. Based on these principles, it allows students to create 2D compositions on a grid pattern by placing and manipulating simple shapes in terms of color, shape, and size. The final composition is evaluated by an artificial intelligence (AI) tool integrated into the game. This AI tool predicts the design principles present in the composition, providing three possible outcomes with associated percentages. The game, currently in the testing phase, has been played by 126 first-year students, and user experience has been assessed through questionnaires, surveys, and basic game metrics. The use of this game to teach visual design principles has proven to be an effective method for engaging students in active learning and enhancing their understanding and application of design concepts. The innovative use of AI to provide real-time feedback and the interactive nature of the game have fostered a deeper, experiential learning process. Additionally, students have proposed various innovative ideas to improve the gaming experience, suggesting potential enhancements that could lead to a more refined and enjoyable gameplay. These insights highlight the potential of digital game-based learning (DGBL) and AI-enhanced tools in creating an engaging and effective educational environment.

Validity of a motor-cognitive dual-task agility test in elite youth football players.

Agility, as the ability to react rapidly to unforeseen events, is an essential component of football performance. However, existing agility diagnostics often do not reflect the complex motor-cognitive interaction required on the field. Therefore, this study evaluates the criterion and ecological validity of a newly developed motor-cognitive dual-task agility approach in elite youth football players and compare it to a traditional reactive agility test. Twenty-one male youth elite football players (age:17.4±0.6; BMI:23.2±1.8) performed two agility tests (reactive agility, reactive agility with integrated multiple-object-tracking (Dual-Task Agility)) on the SKILLCOURT system. Performance was correlated to motor (sprint, jump), cognitive (executive functions, attention, reaction speed) and football specific tests (Loughborough soccer passing test (LSPT)) as well as indirect game metrics (coaches' rating, playing time). Reactive agility performance showed moderate correlations to attention and choice reaction times (r=0.48-0.63), as well as to the LSPT (r=0.51). The dual-task agility test revealed moderate relationships with attention and reaction speed (r=0.47-0.58), executive functions (r=0.45-0.63), as well as the game metrics (r=0.51-0.61). Finally, the dual-task agility test significantly differentiated players based on their coaches' rating and playing time using a median split (p<0.05; d=0.8-1.28). Motor-cognitive agility performance in elite youth football players seems to be primarily determined by cognitive functions. The integration of multiple object tracking into reactive agility testing seems to be an ecologically valid approach for performance diagnostics in youth football.

Game provenance graph-based representation learning vs metrics-based machine learning: An empirical comparison on predictive game analytics tasks

Predicting events and behavior in games is a crucial task of game analytics, both during gameplay, such as predicting wins, and after, like forecasting player churn. Game metrics have been widely adopted due to their human-readable and task-oriented nature. However, these metrics vary between games and require expert knowledge. Game analytics has evolved with techniques that primarily handle non-tabular and sequential data, like graphs and time series. However, existing methods still depend on traditional game metrics from extensive feature engineering. In parallel, graph representation learning has been applied to game provenance graphs — structures that log game events as nodes and link them through causal relationships. This approach may allow analysts to bypass traditional game metrics and learn relevant downstream features during training. This paper compares this metric-less graph-based representation learning with traditional metric-based machine learning in two predictive game analytics tasks: win prediction and hit prediction. Our results show that the same graph-based representation learning architecture achieves competitive results in win prediction and outperforms metric-based approaches in hit prediction, suggesting its suitability for predictive game analytics tasks. This method could significantly benefit as it does not require game-specific metric expertise, and its learned features are versatile across multiple tasks.

Association between game metrics in a simulation game and nursing students’ surgical nursing knowledge – a quasi-experimental study

BackgroundSimulation games are effective for acquiring surgical nursing knowledge during education by offering possibilities to learn theoretical knowledge through practical patient scenarios, thus preparing students for demanding surgical nursing care. Game metrics stored in the game system enable assessment of students’ behaviour while gameplaying. Combining game metrics with the assessment of a student’s surgical nursing knowledge allows versatile information to be obtained about the student’s learning outcomes. However, studies on game metrics stored in systems and their relationship with learning outcomes are scarce.MethodsThe aim here was to evaluate the association between game metrics in a simulation game and nursing students’ surgical nursing knowledge. Nursing students from three universities of applied sciences in Finland participated in a one-week simulation gameplaying intervention that included five surgical nursing scenarios. Students’ surgical nursing knowledge was investigated with a quasi-experimental, one-group, pre- and post-test design using a surgical nursing knowledge test. In total, 280 students filled in the knowledge tests. In addition, cross-sectional game data were collected at a single time point between pre- and post-tests. The data were analysed with descriptive statistics and multivariate analysis methods.ResultsStudents’ surgical nursing knowledge improved with the intervention. The total number of playthroughs was 3562. The mean maximum score was 126.2 (maximum score range 76–195). The mean playing time of all playthroughs by all players was 4.3 minutes (SD = 81.61). A statistically significant association was found between mean score and knowledge test total score (p < 0.0072), but no significant association emerged between mean playing time and knowledge test total score.ConclusionThe results indicated that the higher the mean score the better the students’ surgical nursing knowledge in the knowledge test. This study did not show that the time spent playing had an impact on students’ post-playing knowledge. Our findings support the idea that game metrics can be used in performance evaluation and the results can be used to improve nursing students’ readiness for challenging preoperative and postoperative clinical situations.

The Best Points to Score: An Econometric Approach to Competitive League of Legends

Professional League of Legends is incredibly complex. With 150+ playable characters, 100s of items to buy, and thousands of decisions to make, the possibilities are endless. So how, then, are people supposed to consistently win? Many analysts have approached this question, using the at-the-time best options of characters, items, and compositions. In this paper, we approach League of Legends from the constants; the things every game of League has and measure their value to game result. Through models, we find that tower destruction leads over every other factor. We also find that gold advantage plays a large role as well. Map objectives seem to impact game result less but have impactful results on tower destruction and gold advantage, which in turn impact game result. Additionally, we find that early game metrics, such as first blood or first tower, have significant impact on gold advantage, but are insignificant to game result.

Nursing Students' Scenario Performance: Game Metrics in a Simulation Game.

The purpose of this study was to examine nursing students' scenario performance in a simulation game by utilizing game metrics. A significant advantage of simulation games is that they can store large amounts of data. Although game metrics enable the objective evaluation and analysis of performance, their use in the evaluation of students' performance is limited. Nursing students ( N = 376) played a simulation game at home for 1 week. The resulting data consisted of game metrics stored in the game: number of playthroughs, mean scores, and mean playing times. The total number of playthroughs was 1,923. Statistically significant differences were found between different scenarios regarding the mean score ( p < .0001). Mean playing time was significantly associated with the mean score ( p < .05). Game metrics demonstrate nursing students' scenario performance in clinical reasoning skills in different scenarios in a simulation game.

Serious Game Analytics by Design: Feature Generation and Selection Using Game Telemetry and Game Metrics

The construction of prediction models reflecting players’ learning performance in serious games currently faces various challenges for learning analytics. In this study, we design, implement, and field test a learning analytics system for a serious game, advancing the field by explicitly showing which in-game features correspond to differences in learner performance. We then deploy and test a system that provides instructors with clear signals regarding student learning and progress in the game, which instructors could depend upon for interventions. Within the study, we examined, coded, and filtered a substantial gameplay corpus, determining expertise in the game. Mission HydroSci (MHS) is a serious game that teaches middle-school students water science. Using our logging system, designed and implemented along with game design and development, we captured around 60 in-game features from the gameplay of 373 students who completed Unit 3 of MHS in its first field test. We tested eight hypotheses during the field test and presented this paper’s results to participating teachers. Our findings reveal several features with statistical significance that will be critical for creating a validated prediction model. We discuss how this work will help future research establish a framework for designing analytics systems for serious games and advancing gaming design and analytics theory.

Proxies to the monthly active user number of Geo AR Mobile games - online search volume as a proposal.

Mobile game metrics have received attention since the emergence of big data technology and data-based decision-making. Among different metrics, the monthly active user number is usually significant because it shows the level of players' engagement and the profit of this game as a business. Therefore, the monthly active user number is valuable for researchers, analysts, and decision-makers interested in the mobile game industry. However, the actual monthly active user number data typically have the accuracy, accessibility, granularity, and cost problems. Therefore, a proxy to the monthly active user number would be helpful to facilitate the decision-making process. This paper proposes to capture user activity through the searches on the Internet from an information-seeking perspective. And the online search volume, wiki page view, social media posts and views are proposed as potential proxies. This paper proposes that the online search volume is an acceptable proxy for the monthly active user number in the context of Geo Augmented Reality (AR) mobile games through data analysis.

D3GF: A Study on Optimal Defense Performance Evaluation of Drone-Type Moving Target Defense Through Game Theory

Based on the paradigm shift in modern warfare, ground forces conduct pilot operations of wireless unmanned maneuvering systems, such as tactical drones, in the form of manned and unmanned cooperative tactics after deploying the relevant systems in the battlefield. However, security considerations for relevant systems are limited to the scope of using only the existing end-to-end encryption and public key authentication modules, and no defense strategy to actively respond to specialized cyber-electronic warfare threats has been officially established. To drastically reduce both the potential attack surface and security vulnerabilities of drones employed in network-centric-warfare, a proactive defense technology that expires the effectiveness of attacks by avoiding invasive action at the target is expected to be essential. Accordingly, this paper proposes the concept of active moving-target-defense (MTD), an element of cyber deception that minimizes the rate of success of cyber-attacks while conversely maximizing both defense predominance and attack complexity asymmetrically, exclusive according to partially observable Markov decision process (POMDP)-based threat modeling that considers both the internal and external operation sequences of target drones. To optimally design the proposed drone-type MTD based on the Pareto frontier, we additionally advanced and simulated a drone-based defensive deception game framework (D3GF), which represents a general-sum combat framework reflecting decision logics such as the perfect Bayesian Nash equilibrium, stochastic Stackelberg, and partial signal game. This study was conducted to compare and calculate the efficiencies of the drone-type MTD’s deceptive defense, which had not been considered in prior studies, by unique environmental features inside and outside the drone. Furthermore, we conducted a detailed performance evaluation considering game metrics based on sensitivity analysis. Hereafter, the drone-type MTD will be extended into an actual active drone protection technology combining cyber flare-type avoidance strategies and cyber camouflage-type disarrangement strategies by expanding its optimization domain as a hypergame, while integrating it with drone decoy elements.

Extended Energy-Expenditure Model in Soccer: Evaluating Player Performance in the Context of the Game.

Every soccer game influences each player's performance differently. Many studies have tried to explain the influence of different parameters on the game; however, none went deeper into the core and examined it minute-by-minute. The goal of this study is to use data derived from GPS wearable devices to present a new framework for performance analysis. A player's energy expenditure is analyzed using data analytics and K-means clustering of low-, middle-, and high-intensity periods distributed in 1 min segments. Our framework exhibits a higher explanatory power compared to usual game metrics (e.g., high-speed running and sprinting), explaining 45.91% of the coefficient of variation vs. 21.32% for high-, 30.66% vs. 16.82% for middle-, and 24.41% vs. 19.12% for low-intensity periods. The proposed methods enable deeper game analysis, which can help strength and conditioning coaches and managers in gaining better insights into the players' responses to various game situations.

Designing Serious Game Metrics for Family Caregivers of People with Dementia

The paper presents a conceptual framework and a metric instrument for assisting design and evaluating serious games for dementia family carers as the first step towards designing a game specifically designed to cater for their needs. The paper starts with a literature survey on dementia, dementia family carers and existing games for health issues. From this survey, three components of the carers’ needs (health, education and social), three platforms (social marketing, social media and games) and six game contents (game play, avatar portrayal, game world graphics, sound / music and storyline) were identified, and engagement as the metric of interaction was also identified. With these, a conceptual framework was constructed, identifying the relationship amongst these elements. Based on these, the metric instrument is devised using the Goal Question Metric (GQM) method. An exploratory experiment was conducted with six health-related games and five game professionals to assess the effectiveness of the instrument. It is found that the instrument can identify the successfulness of the games in terms of satisfying the three categories of needs of the carer, i.e., health, education and social, though it was suggested that a further experiment with more participants and focusing on one game would be needed to further verify the effectiveness of the instrument.

Impact of a game‐based intervention on fraction learning for fifth‐grade students: A pre‐registered randomized controlled study

AbstractBackgroundDigital game‐based learning is gaining increased attention from both researchers and educators for improving mathematics instruction. However, the evidence for game‐based learning is mixed and research with rigorous research design and analyses are limited.ObjectiveHere, in a pre‐registered randomized controlled study, we investigated whether a fraction game designed collaboratively by educational experts and professional game developers may serve as a useful tool to enhance students' fraction knowledge.MethodsWe assigned French fifth graders to either an experimental group who used the game (n = 110) or a control group (n = 78) who received traditional instruction on fractions. Fraction knowledge was assessed pre‐ and post‐intervention.ResultsResults show that students in the active control group had superior overall fraction performance than students in the experimental group at the end of intervention. However, the game had a positive effect on decimal learning. We also found a positive relation between game performance and overall fraction knowledge scores at post‐test.ContributionThe study highlights the importance of game metrics as indicators of personalized assessment tools. Given the increased usage of games in learning mathematics and the equivocal results on the effectiveness of these games, our study also highlights the importance of pre‐registration and randomized controlled studies.

Assessing the mental state of attention using a neurofeedback system and serious game tool

Attention Deficit Hyperactivity Disorder (ADHD) is a neuropsychiatric disorder, which is predominantly treated with the administration of psychostimulants. A complementary alternative for ADHD treatment is the neurofeedback (NFB), a method in which patients are trained to perceive and control their physiology, regulate their own brainwaves. As the NFB has a repetitive aspect and can be long lasting, “serious games” may be used to maintain interest throughout the process. This paper presents the development of a NFB system to assess the mental state of attention and analyze the improvement of concentrated attention using the Neurofeedback Space, a Serious Game tool. We present a preliminary validation to choose the brain region and the signal processing algorithm to be used. Training accuracy of 98.12% and test accuracy of 93.88% were achieved. To test the complete NFB system, an experiment with 5 volunteers, and 5 sessions was carried out. Three evaluation metrics were used to analyze the performance of the proposed system: the game metrics, the ratios alpha/beta and theta/beta throughout the sessions, defined here as brain data, and the Concentrated Attention Test. Almost all volunteers improved their performance in the Concentrated Attention Test. Regarding the game metrics, it was possible to observe improvements in the Attention and Score between the most sessions. The brain data results were inconclusive, as there was no common behavior pattern for the volunteers, but it was possible to notice a similarity between the behavior of the game metrics and the brain data in some volunteers in T3 and C3 electrodes. The attention test and game metrics improvement show that the volunteers were able to perceive their own attention and control it to advance in the game.

The Effects of Computer-Based Simulation Game and Virtual Reality Simulation in Nursing Students' Self-evaluated Clinical Reasoning Skills.

Research has shown that nursing students can learn clinical reasoning skills by engaging in simulation games. However, there has been no research regarding the effects of simulation games on clinical reasoning skills when nursing students also engage in virtual reality simulations. Furthermore, by engaging in simulation games, neither game metrics nor their impact on students' self-evaluated clinical reasoning skills has been studied. The purpose of this study was to evaluate the effects of these two kinds of simulation games using a one-group pretest-posttest design. Forty nursing students self-evaluated their clinical reasoning skills in three phases using the Clinical Reasoning Skills scale. Furthermore, the game metrics of the simulation games were analyzed, and the results clearly showed that students' self-evaluated clinical reasoning skills were systematically improved. There was a systematic association between better playing scores and better self-evaluated clinical reasoning skills in playing both kinds of simulation games. Moreover, students engaged more time in the virtual reality simulation than the computer-based simulation game. Overall, the results suggest that the use of a combination of both kinds of simulation games is an effective way for nursing students to learn clinical reasoning skills.

Improving motor function after chronic stroke by interactive gaming with a redesigned MR-compatible hand training device.

New rehabilitation strategies enabled by technological developments are challenging the prevailing concept of there being a limited window for functional recovery after stroke. In this study, we examined the utility of a robot-assisted therapy used in combination with a serious game as a rehabilitation and motor assessment tool in patients with chronic stroke. We evaluated 928 game rounds from 386 training sessions of 8 patients who had suffered an ischemic stroke affecting middle cerebral artery territory that incurred at least 6 months prior. Motor function was assessed with clinical motor scales, including the Fugl-Meyer upper extremity (FM UE) scale, Action Research Arm Test, Modified Ashworth scale and the Box and Blocks test. Robotic device output measures (mean force, force-position correlation) and serious game score elements (collisions, rewards and total score) were calculated. A total of 2 patients exhibited a marginal improvement after a 10-week training protocol according to the FM UE scale and an additional patient exhibited a significant improvement according to Box and Blocks test. Motor scales showed strong associations of robotic device parameters and game metrics with clinical motor scale scores, with the strongest correlations observed for the mean force (0.677<Ρ<0.869), followed by the number of collisions (-0.670<Ρ<-0.585). Linear regression analysis showed that these indices were independent predictors of motor scale scores. In conclusion, a robotic device linked to a serious game can be used by patients with chronic stroke and induce at least some clinical improvements in motor performance. Robotic device output parameters and game score elements associate strongly with clinical motor scales and have the potential to be used as predictors in models of rehabilitation progress.

Motivational Profiling of League of Legends Players.

Player motivation is a key research area within games research, with the aim of understanding how the motivation of players is related to their experience and behavior in the game. We present the results of a cross-sectional study with data from 750 players of League of Legends, a popular Multiplayer Online Battle Arena game. Based on the motivational regulations posited by Self-Determination Theory and Latent Profile Analysis, we identify four distinct motivational profiles, which differ with regards to player experience and, to a lesser extent, in-game behavior. While the more self-determined profiles “Intrinsic” and “Autonomous” report mainly positive experience-related outcomes, a considerable part of the player base does not. Players of the “Amotivated” and “External” profile derive less enjoyment, experience more negative affect and tension, and score lower on vitality, indicating game engagement that is potentially detrimental to players' well-being. With regards to game metrics, minor differences in the rate of assists in unranked matches and performance indicators were observed between profiles. This strengthens the notion that differences in experiences are not necessarily reflected in differences in behavioral game metrics. Our findings provide insights into the interplay of player motivation, experience, and in-game behavior, contributing to a more nuanced understanding of player-computer interaction.

An analysis of in-game feedback provided by coaches in an Australian Football League competition

ABSTRACT Background: Coach observation studies have been a mainstay of coaching science research for decades, with a major focus on the use of instruction and feedback by coaches. However, relatively few studies have investigated feedback provided in a live competition setting, with a majority focussing on the training environment. As such, little is known about the way that feedback changes as a result of the intensity of a competitive setting. Additionally, feedback variables of interest are featured inconsistently between studies. Purpose: The purpose of the current study was to investigate the quantity and nature of feedback from coach to player in an Australian rules football context, specifically the elite-level Australian Football League (AFL) competition. An entire season's worth of coach-player feedback was recorded and coded against a wide range of feedback variables most commonly found in the literature. A secondary aim of the study was to examine changes in feedback against various game metrics (winning/losing quarter, margin). Method: Five coaches from one AFL team, all with several years of experience at the elite level, were recorded across 21 regular-season games from the 2018 season. Feedback quantity, along with several feedback variables of interest (valence, autonomy-support, descriptive/prescriptive, individual/group, etc.), were coded during each quarter of play. Game information (game outcome, quarter outcome, margin at the start of the quarter, point differential for the quarter) was also recorded. Results: Across 84 quarters of football, 1301 coach-player feedback instances were observed. Mean values across all quarters revealed more negative (20%) than positive (13%) feedback, more prescriptive (81%) than descriptive (19%) feedback, more controlling (58%) than autonomy-supportive (8%), and more individual (70%) than group (30%) feedback was provided. Task-related feedback (60%) was more common than process-related (37%), self-regulation feedback (3%) or self-level feedback i.e. praise (2%). More positive feedback was provided in winning quarters than in losing quarters, while more controlling feedback was present during losing quarters than during winning quarters. Coaches provided significantly more feedback overall when the game was in the balance than quarters where the team was 95% or more likely to win or lose. Conclusions: The paper presents novel findings about feedback use in an elite competitive team sport setting. Areas where the use of feedback does not align with current theory and evidence are highlighted, particularly regarding feedback valence and autonomy support. Feedback variables more commonly associated with improved learning and performance in the literature appear to generally be provided more frequently during winning quarters, but a causal relationship is yet to be determined.

How Could the Use of Game Elements Help Students' Affective and Cognitive Engagement During Game Play?

Researchers have posited different types of engagement, distinguishing between behavioral, cognitive, and affective engagement and theoretical frameworks have helped explain the psychological aspects of engagement. However, game researchers should examine all types of engagement using multiple methodologies as a means to understand what students are learning from educational games during game play. Conclusive results require psychological aspects and learning characteristics to be considered, but also require a deeper understanding of the intricate links between learning and game mechanics for engagement. This article presents the findings from a qualitative study with thirty participants that focuses on the importance of affective and cognitive engagement during game play with educational games. To do this, the researchers used Ferran Alsina, a game that would help to develop learning competences of primary education skills. Researchers obtained the experiences of students through a game play session, basic game metrics, think-aloud protocol, observation and focus groups. Results shows that the game provided participants an active participation associated with both affective and cognitive engagement. Without attention to cognition the atuhors risk losing valuable data that relate to a student's learning. Researchers should consider multiple qualitative methodologies and game play experience analysis as student experiences are qualitative.

Development of a Mobile Game to Influence Behavior Determinants of HIV Service Uptake Among Key Populations in the Philippines: User-Centered Design Process.

BackgroundOpportunities in digital distribution place mobile games as a promising platform for games for health. However, designing a game that can compete in the saturated mobile games market and deliver persuasive health messages can feel like an insurmountable challenge. Although user-centered design is widely advocated, factors such as the user’s subject domain expertise, budget constraints, and poor data collection methods can restrict the benefits of user involvement.ObjectiveThis study aimed to develop a playable and acceptable game for health, targeted at young key populations in the Philippines.MethodsAuthors identified a range of user-centered design methods to be used in tandem from published literature. The resulting design process involved a phased approach, with 40 primary and secondary users engaged during the initial ideation and prototype testing stages. Selected methods included participatory design workshops, playtests, playability heuristics, and focus group discussions. Subject domain experts were allocated roles in the development team. Data were analyzed using a framework approach. Conceptual frameworks in health intervention acceptability and game design guided the analysis. In-game events were captured through the Unity Analytics service to monitor uptake and game use over a 12-month period.ResultsEarly user involvement revealed a strong desire for online multiplayer gameplay, yet most reported that access to this type of game was restricted because of technical and economic constraints. A role-playing game (RPG) with combat elements was identified as a very appealing gameplay style. Findings guided us to a game that could be played offline and that blended RPG elements, such as narrative and turn-based combat, with match-3 puzzles. Although the game received a positive response during playtests, gameplay was at times perceived as repetitive and predicted to only appeal to casual gamers. Knowledge transfer was predominantly achieved through interpretation of the game’s narrative, highlighting this as an important design element. Uptake of the game was positive; between December 1, 2017, and December 1, 2018, 3325 unique device installs were reported globally. Game metrics provided evidence of adoption by young key populations in the Philippines. Game uptake and use were substantially higher in regions where direct engagement with target users took place.ConclusionsUser-centered design activities supported the identification of important contextual requirements. Multiple data collection methods enabled triangulation of findings to mediate the inherent biases of the different techniques. Game acceptance is dependent on the ability of the development team to implement design solutions that address the needs and desires of target users. If target users are expected to develop design solutions, they must have adequate expertise and a significant role within the development team. Facilitating meaningful partnerships between health professionals, the games industry, and end users will support the games for health industry as it matures.

Roles, Rules, and Keys: How Different Play Configurations Shape Collaborative Science Inquiry

Taking place within the Science Through Technology-Enhanced Play (STEP) design-based research project, this paper describes how first- and second-grade students use their bodies to imaginatively become microscopic particles inside of a collaborative simulation of state change (e.g., liquid to gas) and within two distinct play spaces: a modeling activity and a game activity. As participants at once bring play to life in the classroom and learn about microscopic particles, they introduce and enact distinct configurations of roles, rules, and keys that shape their moment-to-moment activity and their pathways through science inquiry. Participants’ integration of distinct roles, rules, and keys during stretches of play foster on the one hand cycles of open-ended creation, observation, and destruction, and on the other hand generation of pragmatic strategies for winning assessed through game metrics. Overall, this paper introduces a framework for tracking the collaborative instantiation of roles, rules, and keys, provides preliminary documentation of how these characteristics shape learning, and invites reflection on how to move beyond general claims about games benefiting learning.