Recursive Loops of Game-Based Learning: a Conceptual model.

Recursive Loops of Game-Based Learning: a Conceptual model. Paul R Kearney PhD Student School of Education Deakin University Melbourne Australia pkearney@unitec.ac.nz Maja Pivec Information Design, FH JOANNEUM, University of Applied Sciences, Graz, Austria Maja.Pivec@fh-joanneum.at Abstract: This paper addresses the issues surrounding knowledge and skill acquisition from Game-Based Learning. Using both the learning models of Garris, Ahlers, and Driskell (2002) and Klob (1984), an enhanced conceptual model of how and where game-based learning fosters learning outcomes has been presented. The new model contains a time dimension and shows the scaffolding of player abilities. The motivation behind this paper is to promote an understanding of the pedagogy delivered by computer games, both educational and recreational, to assist educators utilize these tools when including game-based learning into the teaching curriculum. Introduction Many academics compare video games to the act of teaching and do not embrace the learning potential that modern commercial computers games can offer. Brown (2002) suggests that students learn as a result of the framework or environment that fosters learning rather than as a result of the teaching. He maintains that today s students look upon technology as an integral part of learning. For many of them computer games have been part of their learning since early childhood. Brown suggests that there is a shift in the way that these students learn. The shifts include literacy, from text to multimedia; lectures, from teacher-centered to student-centered; reasoning, from deduction to transforming; and reading, from solitary to social exploration. Computer games are ideally situated to cater for these students. Oblinger (2004) suggests that educational environments involving computer games lead to deeper learning, and Buchanan (2000) states that the cognitive conflict from computer games enhances learning. However, much of the current research in this field is purely qualitative and many empirical studies are based on observation only. There has been little evidence to show how Game-Based Learning (GBL) works and this invites many skeptics to the discussion. Egenfeldt-Nielsen (2005), states that results in this overall research area are mixed, but do lend some support to computer games having a connection with spatial skills, problem-solving and to a very limited degree, eye-hand coordination (p.92). He continues with saying the transfer of improvements obtained in the computer game context to other areas of life have been notoriously hard to document (p.92). Egenfeldt- Nielsen s (2003) earlier study stated that eye-hand coordination was not improved when playing a popular arcade game title Super Monkey Ball, yet in a recent study by Rosser, Lynch, Cuddihy, Gentile, Klonsky, and Merrell (2007) using the same game, eye-hand co-ordination was not only found to be improved, but also transferred outside the context of computer game play. This suggests that research in the area of GBL is varied and results differ as widely as the methodologies employed.

Quality Learning from Computer Games Learning is defined in most dictionaries as the acquisition of knowledge or skills through experience, practice, or study, and learning outcomes are the knowledge, skills and abilities that the student will possess following the learning experience. The learning outcomes from computer games are often divided into skill based (technical, motor), knowledge based (declarative, procedural, strategic), and affective (confidence, attitudes, dispositions) (Garris, Ahlers, & Driskell, 2002). However, for quality learning to occur, the student must posses the necessary underlying cognitive abilities. For example, Bruner (1966) lists multitasking as one of the cognitive skills that improves the ability to learn, and Tolman (1948) states that spatial skills are necessary to create cognitive maps required for complex problem solving. Behaviour is learnt and decisions are made by evaluating the situation and considering all the options. Buchanan (2004) notes that game designers challenge players by using bots or non-player characters (NPCs) in their games that mimic human behavior. Bots do this by using what programmers call a decision tree. Human players do it intrinsically by monitoring the situation and manipulating it based on their own thoughts and perceived skill set. This is meta-cognition and Garris et al., (2002) suggest that it is part of the learning process within the game cycle. As shown in figure 1, from the feedback received, players make judgements based on evaluations and modify their behaviour and game play accordingly. Buchanan claims that experienced players consciously increase their mental space for visualisation and manipulation of problems. Their meta-cognition skills increase as they progress through the game. Buchanan (2004) suggests that game players possess an increased ability to multitask and mentally sort information. He states that computer games include all the underpinning characteristics for quality learning and Garris et al., (2002) list these characteristics as follows: Fantasy Imaginary or fantasy context, themes, or characters. Rules/Goals Clear rules, goals, and feedback on progress towards the goals. Sensory Stimuli Dramatic or novel visual and auditory stimuli. Challenge Optimal level of activity and uncertain goal attainment. Mystery Optimal level of informational complexity. Control Active learner control. Garris et al. also suggest that the learning outcomes occur outside of the game during reflection and debriefing (figure 1). INPUT PROCESS OUTCOME Instructional content Judgements System feedback Game Cycle Debriefing Learning outcomes Game characteristics Behaviour (Garris, Ahlers, & Driskell, 2002) Figure 1: Learning in GBL

This may be true for declarative knowledge, but to succeed in the fast paced action games available today, players must increase their procedural and strategic knowledge within the game itself. Shaffer (2006) calls this reflection-in-action, as opposed to reflection-on-action as would be the debriefing in figure 1. Shaffer suggests that the virtual worlds created by such games allow students to take action within the game and then reflect on this action, both during and after play. Kolb (1984) suggests that learning follows a cyclic pattern, and the reflection on experience is part of the learning cycle itself (figure 2), similar to Shaffer s reflection-in-action. However, Paras & Bizzocchi (2005) state that when play is broken up with reflection, the learning is reduced. However, if the reflection is dispersed within the game itself, the learner/player takes responsibility for the learning outcomes. Concrete Experience Feeling Active Experimentation Doing Reflective Observation Watching Abstract Conceptualisation Thinking Figure 2: Kolb s learning styles The reflection can occur during periods between the levels of the game, or while waiting for the game to complete a simulation, or even be part of the game itself. For example, Kearney (2005) compared the commercial game Counter-Strike with Quake III, both first-person shooter multiplayer computer games. In the game Counter-Strike, if players are shot, they are required to wait between missions until the remainder of their team wins or loses the level. This provides time to reflect on their game strategy, their decisions and subsequent actions, while they are passive observers of the game being played. In Quake III, players can re-enter the game immediately and no time for reflection is provided. The results of Kearney s study showed that players of the game Counter-Strike improved their multi-tasking ability by up to 2.5 times more than that of Quake III players; the time used for reflecting before reentering the game may have contributed to this improvement. Many publications that include learning models differ in their suggestion of where and how learning takes place. However, they all agree that learning outcomes are enhanced through the immersive characteristics of computer games where the attention of the player is focus on the goal of the game. They also state that when this immersion occurs, the game motivates the player to repeatedly engage in play. This type of motivation has been described as flow (Csikszentmihalyi, 1990). The concept of flow can be used to identify which computer games foster the persistent re-engagement of the player, by analyzing computer games with a game-flow analysis model (Sweetser & Wyeth, 2005; Kearney & Pivec, 2007).

Persistent Re-engagement of Learners Beazzant (1999) suggests that the characteristics of commercial computer games create an environment where players are compelled to play to the extent of forming addictions. Garris et al., (2002) state that this addiction, or persistent re-engagement by the player, is what instructional designers strive to create. This would suggest that computer games foster behaviourism and the learning is achieved from drill and practice. Yet de Castell and Jensen (2003) argue that many educational games are not successful because they fail to immerse the play the way commercial computer games do, and it is this immersion that fosters a deep learning, not the low level of learning from drill and practice. In the study by Kearney (2005), the game Counter-Strike was observed to immerse the player in the game. Two teams of eight players sat quietly and focused for over two hours in what appeared to be very serious game play. Yet other teams in the same study played a similar game called Quake III and no player immersion was observed. The difference between the two games was that Quake III did not create same level of challenge nor difficulty that Counter-Strike did. Counter-Strike also had more rules, consequences of failure were increased, and the goal of the game was detailed enough to inform the player of the relevance of the gameplay, as suggested earlier in the list of characteristics by Garris et al., (2002). The evaluation matrix used by Sweetser and Wyeth (2005) and later modified by Kearney and Pivec (2007), found that the higher rating of flow within the game, the deeper the player immersion and the more likelihood of persistent re-engagement by the player. Quinn (1997) suggests that computer games can be highly effective when used in an educational environment. He also cites the concept of flow from Czikszentmihalyi (1990), in conjunction with Malone s (1981) critical elements of fantasy, challenge, and curiosity; both concepts are used and extended by Garris et al., (2002) for their model of GBL. Quinn goes further to suggest that Malone s challenge element is what maintains player engagement and creates a zone of difficulty where learning occurs. This could also be compared to Vygotsky s (1978) Zone of Proximal Development (ZPD), where the scaffolding or level of cognitive challenge must be appropriate for the learner s current abilities or learning will not occur. Quinn argues that cognitive challenges within the game lead to the practice of skills for problem solving. He states computer games are practice opportunities for cognitive skills. This may appear to some as drill and practice, however we call it recursive learning. Recursive Loops of Learning Recursive learning is a term usually applied to algorithms used in computer programming. A recursive loop is where the task is performed repeatedly until a counter of some kind has been incremented. We can apply the same methodology to game-based learning and suggest that the player will repeat the level or task until the learning outcome or goal has been achieved. The player s ability is then incremented and the game moves to the next level. Knowledge based skills are defined as declarative, procedural, strategic knowledge. Declarative knowledge being facts and data that are required to complete a task or to perform well within the task and these would be provided by the game or system feedback. Procedural knowledge is required to know how to approach the task and subsequently complete it. This could be referred to as knowing how to apply the declarative knowledge to a given situation. Strategic knowledge is the reasoning behind the task and how the task could be achieved in a different or more creative way. Each of these skills are achieved through reflection, but with many fast action computer games, it is reflection-inaction and occurs throughout the game cycle and within each level. This is the macro cycle. As skills and abilities are attained, the player advances through the game and increments their knowledge. We further suggest that depending on a player s ability or experience, the learning will occur only if the player enters the game at the appropriate level as shown in figure 3. Vygotsky suggested that the distance between the actual developmental level as determined by independent problem solving and the level of potential development as determined through problem solving is the where learning occurs (Vygotsky, 1978, p. 86). However, he suggests that this is facilitated through peer collaboration or

teacher involvement, but the computer game itself can act as the teacher. With multiplayer games, peer collaboration occurs between players and has been observed to foster learning (Kasvi, 2000). We propose an expansion on the model from Garris et al., (2002) to include a time dimension. This dimension allows us to follow the game play and the progression throughout the game. Within the model we can observe the macro and micro game cycles (figure 3) and include player reflection within the game, during play and between levels, and suggest where the different types of learning occur; skill based, knowledge based, and affective. Commercial computer games are known for creating social environments and cult followings surrounding the gameplay, the character attributes, and player s abilities, and we suggest this is where affective learning occurs. Garris et al., (2002) describes affective learning as including feelings of confidence, self-efficacy, attitudes, preferences, and dispositions (p.457). The skill-based learning appears to comfortably fit within the micro game cycle or levels within the game. For example, Rosser et al., (2007) found that the playing of commercial action games improved the surgical skills of laparoscopic physicians and decreased their error rate. There was no documented debriefing session for Rosser s study and it is assumed that the development of technical or motor skills occurs within the game itself. Figure 3 also shows how player ability and experience affects the challenge element and the level of learning (ZPD), and how the level of cognitive challenge can be appropriate for the learner s current abilities. The model also shows the inclusion of instructional design and game characteristics as critical elements of a game to enable the achievement of the learning outcomes, as well as the additional factor of player abilities.

Macro Game Cycle Reflection-in-Action (Declarative, Procedural, Strategic Knowledge) System feedback Judgements Player Abilities Behaviour Level 99 Instructional Design 3 4 Debriefing Reflection-on-Action Learning Outcomes Game Characteristics Level 1 2 Levels Completed (Abilities incremented) Social Environment (Affective Learning) System feedback Behaviour Judgements Micro Game Cycle (Skill based Learning, Cognitive Abilities) Figure 3: Recursive loops of Game-Based Learning (Kearney & Pivec).

Conclusion Defining learning as the acquisition of knowledge or skills, suggests that Game-Based Learning is the vehicle that fosters the acquisition of the learning outcomes. The research for this paper suggests that Game-Based Learning occurs in a recursive loop and as such when the player skills are acquired, or incremented, the player moves on to the next level of the game. The model shown in figure 3 introduces a time element to allow the player to progress through the game increasing their knowledge and acquiring new levels of ability. This suggests that knowledge, declarative, procedural, and strategic is acquired over time and abilities or skills are incremented through experience. We have also shown detailed macro and micro cycles of learning within GBL and highlighted how player experience relates to the levels of the game itself. Where instructional design contributes to the challenge factor for the player, the characteristics of the game that promote player immersion contributes to the persistent re-engagement by the player. The level at which the player engages will affect the success of the learning outcomes. It is hoped that an understanding of the pedagogy delivered by Game-Based Learning will help educators utilize the design when including game-based learning into the teaching curriculum. This model is currently being used in a study of cognitive abilities achieved from recreation computer games, due for completion in 2008. References Beazzant, S. (1999). Dissertation: Children and video games: What's the fuss? Retrieved 12 April, 2003, from http://www.scottbezzant.btinternet.co.uk/downloads/dissertation.htm Brown, J. (2002). Learning in the digital age. Paper presented at the The Internet & the University: Forum 2001. Bruner, J. (1966). Toward a theory of instruction. Boston: Little, Brown. Buchanan, K. (2004). How an educator thinks about computer games. On the Horizon Retrieved 21st December, 2004, from http://www.msu.edu/~buchan56/games/educator_thinks_games.htm Csikszentmihalyi, M. (1990). Flow: The psychology of optimal experience. New York: Harper and Row. de Castell, S., & Jenson, J. (2003). Serious play. Journal of Curriculum Studies, 35(6), 649-665. Egenfeldt-Nielsen, S. (2003). Keep the monkey rolling: Eye-hand coordination in super monkey ball, Digra - Level up conference 2003. Utrecht University. Egenfeldt-Nielsen, S. (2005). Beyond edutainment: Exploring the educational potential of computer games. Unpublished phd thesis. IT-University of Copenhagen, Copenhagen. Garris, R., Ahlers, R., & Driskell, J. E. (2002). Games, motivation, and learning: A research and practice model. Simulation & Gaming, 33(4), 441-467. Kasvi, J. (2000). Not just fun and games - internet games as a training medium. In P. Kymäläinen & L. Seppänen (Eds.), Cosiga - learning with computerised simulation games. (pp. 23-34): HUT: Espoo. Kearney, P. (2005). Cognitive callisthenics: Do fps computer games enhance the player's cognitive abilities? Paper presented at the DiGRA 2005 Changing Views: Worlds in Play International Conference, Vancouver, Canada. Kearney, P., & Pivec, M. (2007). Immersed and how? That is the question. Paper presented at the Games in Action Conference, Gothenburg, Sweden. Kolb, D. A. (1984). Experiential learning: Experience as the source of learning and development. Englewood Cliffs, NJ: Prentice-Hall. Malone, T. W. (1981). What makes computer games fun? Byte, 6(12), 258-277. Oblinger, D. (2004). The next generation of educational engagement. Journal of Interactive Media in Education - Special Issue on the Educational Semantic Web, 2004(8), 1-18.

Paras, B., & Bizzocchi, J. (2005). Game, motivation, and effective learning: An integrated model for educational game design., Digital Games Research Association 2005 Conference: Changing views- worlds in play, Vancouver, 16-20 June 2005. Vancouver, British Columbia, Canada: Digital Games Research Association. Quinn, C. N. (1997). Engaging learning. Paper presented at the Instructional Technology Forum. Rosser, J. C., Lynch, P. J., Cuddihy, L., Gentile, D. A., Klonsky, J., & Merrell, R. (2007). The impact of video games on training surgeons in the 21st century. Archives of Surgery, 142(2), 181-186. Shaffer, D. W. (2006). How computer games help children learn. New York: Palgrave Macmillan. Sweetser, P., & Wyeth, P. (2005). Gameflow: A model for evaluating player enjoyment in games. Computers in Entertainment, 3(3). Tolman, E. C. (1948). Cognitive maps in rats and man. Psychological Review, 55, 189-208. Vygotsky, L. (1978). Mind in society: The development of higher psychological processes. Cambridge: Harvard University Press.