Current state of play: Children’s learning in the context of digital games

Digital games are an integral part of children’s lives, as 8- to 18-year-olds spend approximately 1.5 hours per day playing them (Rideout & Robb, 2019), and 76% of US children under the age of 18 play digital games (Entertainment Software Association [ESA], 2023). The appeal and potential benefits of digital gameplay have not been lost on educators who have increasingly incorporated games within their academic instruction even prior to the COVID-19 pandemic when digital devices were made more available for educational use (Dubé & Dubé, 2021).

One challenge for studying learning in digital games is that technology continues to evolve. Thus, researchers need to consider the implications of their findings for current applications, and how those applications will inform formal and informal learning. Both types of learning may include systems that take advantage of artificial intelligence (AI) to enhance personalization, or gameplay that involves extended reality (XR), augmented (AR), or virtual reality (VR). What remains unchanged, independent of technology used, are questions addressing how best to design efficacious games for learning and how to promote durable, transferable learning from gameplay to activities beyond the game screen. Our objective here is to briefly review work done to date addressing these questions and to suggest avenues for future research given the current and pending technological climate.

Next steps for research on games for learning

Both educational and recreational digital games result in learning and skill enhancement (Gentile & Gentile, 2021). Well-designed games incorporate developmentally appropriate practices that utilize cognitive, affective (including motivational), and social design features (Plass et al., 2015). These design features include incentives and rewards, feedback on the efficacy of game play, features intended to evoke emotional responses (e.g., visual design, music, and sound), and adaptive challenges that promote sustained game play (Plass et al., 2015).

Learning also takes place informally between players during digital games. For instance, Lane et al. (2022) discussed how Minecraft players collaborated on difficult “works of engineering” to build replicas of complex structures (such as the Taj Mahal) by planning, sharing, and coordinating online. This type of collaboration, that facilitates one’s learning and that of others, is also found on sites such as Twitch which is an interactive platform on which players live stream their game play (see Payne, et al., 2017).

How specific game features interact with individual-difference factors (e.g., cognitive abilities, prior game experience, affective states, time on task) warrants further examination to identify best ways to foster learning from digital games (Flynn et al., 2021). This examination necessitates further consideration of whether the game is researcher-designed, which typically emphasizes instruction, or one that is commercially available, whereby learning may be a secondary consideration or by-product (Flynn et al., 2021). The commercially available games that children regularly play tend to be more dynamic and include appealing features (e.g., engaging visual and audio stimuli). These features enhance players’ engagement and motivation but may derail learning of the intended educational content or skills. This issue may be most concerning for neurodiverse and younger players who may become overstimulated in highly dynamic game environments.

Games that sustain learning effects

Carefully-designed digital games have been shown to facilitate and improve children’s academic skills in diverse content areas such as math (Pan et al., 2022), science (Lane et al., 2022), language (Chen et al., 2020), reading (Ronimus et al., 2019), and to support the development of cognitive skills associated with learning, such as executive functions (i.e., memory, attention, inhibition; Eng et al., 2023; Homer et al., 2018). In fact, learning via games may be better or equivalent to learning via non-game-based tasks and activities (Ronimus et al., 2019), and may be highly effective among learners for whom traditional approaches are unsuccessful. For example, Ronimus et al. (2019) reported that struggling readers showed greater gains in reading skills when a literacy intervention included digital game-based learning tasks. This situation likely reflects the sustained engagement and motivation to persist in learning that game-based learning evokes (Deater-Deckard et al., 2014; Sailer & Homner, 2019). Games that provide players with practice that includes feedback, clear objectives, rewards for efficacious play, control over the difficulty of game play, and allow for mistakes and risks are most likely to promote durable learning (Gentile & Gentile, 2021). Other game features that enhance learning are competition and collaboration between players (Chen et al., 2020; Lane et al., 2022), variety in game dimensions (Raviv et al., 2022), the use of spoken text and conversational language, in-game advice, and opportunities for self-explanation or reflection (Mayer, 2019).

Although the durability of learning induced by video games over time has been examined in some domains (e.g., perceptual/cognitive; Bediou et al., 2023), that durability remains understudied, particularly with respect to game mechanics that might facilitate transfer of knowledge. Transfer refers to the application of knowledge or skills acquired in one context to another similar (reflecting near transfer) or less similar situations (reflecting far transfer). Findings show that far transfer is the more difficult of the two to demonstrate (Barnett & Ceci, 2002). However, far transfer may be found in prosocial games whereby those who play them show gains in empathy and cooperative behaviors (e.g., Greitemayer, 2022) and among frequent players of violent games who show more aggressive behaviors than less frequent players (e.g., Anderson et al., 2007). A goal for future research may entail consistently distinguishing between multiple routes to far generalization - either far transfer (i.e., where learning in one environment allows for the immediate ability to solve new problems or show proficient performance in new environments) versus “learning to learn” (i.e., where learning in one environment doesn’t necessarily predict immediate enhancement in new environments, yet allows for faster learning in the new environment; Pasqualotto, et al., 2023). There also may be lessons learned from how entertainment games foster engagement in different contexts and promote far transfer. This issue also highlights the need for more long-term research evaluations to clarify how and under what circumstances children’s learning from digital games are impacted over time.

Leveraging AI and XR technology

Most modern educational games’ AI build on state machine technology (i.e., changing states depending on conditions) that emerged in the 1970s with Pac-Man, in which autonomous non-player characters (agents) took information from both the game and the player to determine what actions should then take place within the game (Gallagher & Ryan, 2003). Adding AI capabilities to educational games enables the various game elements to be calibrated to a learner’s individual needs, such as through AI bots, assessment, and rewards to scaffold learning (De Freitas, 2018). AI is also transforming how users interact with the system, for example, by using natural language processing to decode players’ in-game dialogue or to allow players to give commands and interact with the system through normal speech. These capabilities, however, also raise concerns about how data is created and used to inform how the AI-powered algorithms choose to engage children in games (De Freitas, 2018). Thus, researchers need to determine whether AI can engage children in games and if children can acquire knowledge and skills through these games. This effort entails identifying the cognitive, emotional, and social skills needed to use AI based games such that players’ learning benefits are enhanced and risks (e.g., distracting environments) are limited.

XR are currently trending as gaming technologies, including for children (Rideout & Robb, 2019). The immersive environments provided by XR may facilitate increased and sustained learning as players can practice skills in 3D settings that approximate real life settings and demonstrate cognitive and affective responses that one might emit in these settings (Bailey et al., 2019). XR games may be promising educational tools for teaching, assessment, interventions, and for promoting real-time physical activity and social interaction (Mado et al., 2022). Findings already show that VR can be used as an intervention for children with developmental disabilities (Mesa-Gresa et al., 2018) as demonstrated in motor improvements for children with Cerebral Palsy that exceeded traditional interventions (Chen et al., 2018). Future research needs to examine how physical and social activity within an immersive and dynamic environment may benefit or hinder children’s learning, particularly in classrooms, where immersive technology is increasingly found. Continuing efforts to examine the issues of transfer of learning from digital gameplay are warranted given the physical and social activities allowed within XR, especially those environments that closely match real-world situations.

Ethics of digital games for learning

Ethical concerns increasingly arise with respect to digital games. These concerns include children’s privacy, games that have in-game purchases, features of games that might increase addiction and exposure to negative content (e.g., inappropriate language, bias, harmful comments) from games that have collaborative features that are unmonitored or checked. Arguably, high-quality age-appropriate games are designed with these aspects in mind. However, many parents (and educators) fail to monitor details and features of games that seem appropriate and enjoyable to children.

Digital game ratings also warrant reevaluation and standardization as they remain largely unregulated (Blumberg et al., 2019) and many digital games labeled as educational lack research supporting that designation (Hirsh-Pasek et al., 2015). Organizations such as Common Sense Media provide educational ratings and reviews of digital games for educators and parents, but not at the point of purchase. The Entertainment Software Rating Board, a nonprofit organization for video game developers and publishers, also provides content and age-based ratings for video games and apps. These ratings flag content that is violent or for mature audiences given the presence of, for example, profanity or overt sexualized content. Similarly, these ratings help identify child-appropriate games (e.g., “E” for Everyone). However, designation of games that include potential learning or educational benefits remain unspecified. Such ratings, as substantiated by research and clear guidelines for use, and ratings of the age-appropriateness of game content, should be readily accessible to the interested public.

Relatedly, we recommend greater investment in media literacy research that includes examination of digital game goals, to inform a national and international educational policy. Media literacy is defined as the ability to “access, analyze, evaluate and create messages” (Livingstone, 2004) and remains notably absent in US school curricula if not within other countries within Europe and Asia (Brodsky et al., 2021). Thus, media literacy education needs to become a standard accompanied by resources that address how to select affordable and effective age-appropriate games, and initiate discussions about the privacy and personal safety issues that are evoked through gameplay.

Conclusion

Much remains to be discovered about crafting age-appropriate, high-quality games for children (Blumberg et al., 2019). Thus, we urge investigating which specific game elements (e.g., rules, feedback loops, interactivity), affordances (e.g., avatar representation, embodiment), and strategies (e.g., storytelling, role-models, challenges) support learning during gameplay for different-aged neurotypical and neurodiverse children under what circumstances and over what length of time. Such investigations should be coupled with science-backed policies that provide evidence-based guidelines to help parents and educators choose games and apps to support learning. Our discussion here shares thoughts about attaining these goals.

Biographies

Fran C. Blumberg is a professor in the Division of Psychological and Educational Services within Fordham University’s Graduate School of Education. Her research concerns the study of children’s problem-solving skills in the context of formal and informal learning settings such as digital game play. She co-organized the first Ethical Games Conference with conference founder Celia Hodent and Sebastian Detterding.

Rachel M. Flynn is an assistant professor of child and adolescent development at San Francisco State University. Her primary research examines the impact of digital games on children’s cognitive development. She is interested in studying individual differences factors, such as age, prior exposure, enjoyment, and attention, and game features that differentially impact media effects.

Bruce D. Homer is a professor of educational psychology at the Graduate Center of CUNY, where he studies the role of digital technologies in children’s lives. His research focuses on social-cognitive development, including theory of mind and executive functions, learning, and children’s wellbeing. Of particular interest is how design features can support development and learning processes in digital contexts, particularly for children with greater needs.

Jakki O. Bailey is an assistant professor at the University of Texas at Austin in the School of Information. Dr. Bailey examines the psychological and social impact of immersive and interactive technology on human development. In addition to her academic research and service, Dr. Bailey has advised children’s media company executives and policy makers on the psychological, social, and ethical implications of VR in youth’s lives.

Cassondra M. Eng is an NIH-funded T32 sports neuroscience postdoctoral scholar at Stanford University School of Medicine. She investigates the neural mechanisms underlying changes in executive function from active VR interventions with neurodiverse youth. Her interdisciplinary research bridges cognitive neuroscience, learning sciences, developmental psychology, and human-computer interaction to optimize immersive contexts that foster brain and cognitive functioning crucial for academic achievement and over wellbeing.

C. Shawn Green is a professor of psychology at the University of Wisconsin-Madison and Editor-in-Chief for the Journal of Cognitive Enhancement and Associate Editor for the journal Technology, Mind, and Behavior. His research focuses on human learning broadly construed, with a particular interest in methods to improve individuals’ perceptual and cognitive skills via either purposefully designed training regimens or commercial off-the-shelf video games.

Michail Giannakos is a professor of interaction design and learning technologies in the Department of Computer Science at the Norwegian University of Science and Technology, and the head of the Learner-Computer Interaction Laboratory. His research interests include interaction design, child–computer interaction, learning analytics, and computing education. Giannakos serves as the Editor-in-Chief of the International Journal of Child-Computer Interaction. His work has been awarded multiple best paper awards and distinctions.

Stamatis Papadakis is an assistant professor at the Department of Preschool Education, University of Crete, Greece. His expertise lies in educational technology with a specific focus on mobile learning. His research primarily explores mobile learning, particularly leveraging smart mobile devices and associated applications in preschool and primary education.

Douglas A. Gentile is distinguished professor of psychology at Iowa State University. He has authored over 150 peer-reviewed scientific journal articles on the positive and negative effects of mass media on youth. He is a fellow of multiple associations, received the distinguished lifetime contributions award to media psychology by the American Psychological Association, and was named one of the Top 300 professors in the U.S. by the Princeton Review.