Approaches to Embed Bio-inspired Computational Algorithms in Educational and Serious Games  Michela Ponticorvo 1*, Angelo Rega 2, Andrea Di Ferdinando1, Davide Marocco1 and Orazio Miglino1 1 Department of Humanistic Studies, University of Naples “Federico II” , Naples, Italy 2 IRFID, NeapoliSanit, Ottaviano, Italy *michela.ponticorvo@unina.it Abstract micro-worlds [Rieber, 1996] that can work as a lab where to experiment something: behaviours, emotions, strategies in a Bio-inspired computational algorithms can be effectively somewhat protected environment. Games can also start from employed to develop games for learning. In the present reality and go beyond, this is the case for hyper-realistic paper we will introduce different approaches to embed this games as war simulators or surrealist games. kind of models in Serious and Educational Games. For this paper purpose, we will now focus on digi- According to a multi-level description of game design tal/electronic games that, in the last years, have assumed an process, bio-inspired computational algorithms can be important role in the game market with an ever-increasing visible to the user, residing at an external, shell level; can be diffusion. Also their application in education has been invisible to the user, residing at the core, internal level; can massive for many reasons. Digital entertainment games be employed in the evaluation and tutoring level pertaining have some specific features that are very useful in an to user profiling and supporting learning and teaching educational context: games engender motivation [Malone, processes. This different approaches are explained by 1981], are engaging [Gee, 2003] and exploit learning by introducing some educational games example: BreedBot in doing [Aldrich, 2005]. which bio-inspired computational algorithms are used to To fully exploit game potentials, design plays a crucial role develop the player-game interaction and are explicitly [Chandrasekaran, 1990], as it must consider the interaction visible by the user; Learn2lead, where these techniques are between the player and the digital game, in the more general used to model the game mechanics and are invisible to the frame of human-computer interaction [Lieto and Radicioni, user; and Infanzia Digi.tales project in which these 2016]., keeping the lessons derived from neurosciences, techniques are functional to develop smart educational cognitive science and psychology, related to attention, materials by implementing adaptive tutoring systems. executive functions and spatial cognition [Bhatt and Freksa, 2010]. To implement these aspects, computational models can be 1 Introduction taken into account, in particular bio-inspired computational In recent years an epochal turn has been observed in models can be effectively embedded in games. education coming from a twofold pathway. On one side, a Bio-inspired computational models, at the edge between growing effort has been devoted to the use of new natural and artificial, are extremely fit for educational goals technologies, in particular ICT technologies, as educational [Ponticorvo et al., 2016] if the goal is to teach biological, tools. Technology-Enhanced learning (TEL) has intercepted psychological and social matters, as it will be evident later, this tendency by promoting new educational practices, new because they allow to convey knowledge about dynamic and communities and new ways of communication [Balacheff et complex system, emergence, evolution and development. al., 2009]. On the other side, a lot of interest has arisen about the use of game for learning. This interest is witnessed 2 Serious Games design process by the numerous research branches that emerged, game- The SG design process can run according to a multi-level based learning [Tobias and Fletcher, 2011], edutainment framework, with two concentric levels, the shell and core [Charsky, 2010], gamification of learning [Kapp, 2012], just level and a ubiquitous one, the evaluation and tutoring level to cite some. In particular many games have been developed [Dell'Aquila et al., 2016], represented in Fig.1. The shell under the label Educational Games and Serious games and the core level are present in every game of game, and, which include card, board and videogames. Serious Games more in general in almost every cultural product. The shell (SG) are games that educate, train, and inform [Michael and level represents the visible content that is immediately Chen, 2005], sharing the same educational mission. accessible to players. It frames the game engine, the game Why games are so appealing as educational tools? Games dynamics that are hold in the core level. The third level, the are often models of reality that simplify what happens in evaluation and tutoring level, is characterizing for real world including some relevant aspects of it. They are Educational and Serious games, as it allows, on the teachers In the case of educational games, the core modeling process side, to understand if and how the player/learner has is addressed not only by the chosen kind of game, but also acquired the concepts conveyed by the Educational game. on what content we want to convey. If our goal is to build educational tools and materials which are related to biology, psychology and sociology, exploiting concepts such as emergence, complex and dynamic systems, evolution and development, we can glean from a wide class of bio-inspired algorithms. Bio-inspired computing [Pintea, 2014] is a field of study that exploits the study of natural phenomena to apply it to machine learning: from evolution to genetic algorithms [Goldberg, 2006], from natural complex systems to cellular automata [Chopard and Droz, 1998], from the nervous systems to artificial neural networks [Patterson, 1998]. A particular class of bio-inspired algorithms, Agent Based models (ABM), is, in our opinion, particularly well-suited for game design. ABM [Helbing, 2012] is a class of Figure 1 A multi-level game design process with shell, core computational models used to simulate phenomena and evaluation/tutoring level. Shell and core levels can be belonging to various domains ranging from biology to found in every kind of game whereas the evaluation/tutoring psychology and sociology starting from the action and level characterizes educational games interaction of simple agents. These agents are autonomous and can represent individual or collective entities such as groups. We adopt a wide definition of ABM: in SG, ABM is 2.1 The shell (game narrative) and core (game not used to understand collective behaviors starting from mechanics) levels simple rules, but it aims at representing in detail agents interaction and the agent itself. In other words, a great effort The shell level represents what the player sees, what we call is devoted to modelling agents too, in this respect the game narrative. Digital games, as many other cultural resembling multi-agents systems approach [Van der Hoek products, are expressed trough a narrative metaphor that and Wooldridge, 2008] where agents can be very complex. carries out the crucial role to give sense to the game. Let us If we adopt ABM, in the core level, every agent is defined consider for example, the Monopoly game. Throwing the in function of its sensory features, what it sees, hears, dice and moving on the boxes has the meaning to represent smells, touches in the setting and about the core, and action real estate commerce and this strongly helps to engage the endowment, what it can do to affect the core state. These player. actions must follow game rules that are defined both by In designing the shell level we have to define the context: setting constraints and by agent actions chances residing in who are the agents, what actions they can display, what the core level. interactions are possible between them. The shell level, As we are in the domain of digital SG, agents can also be based on narrative, holds an hidden level with a specific artificial agents: in this case the agent is not human, but a operation, the game engine, the core level. bot whose artificial intelligence can rely on bio- The game engine allows to implement core functionalities computational models as well. related to game dynamics, for example related to physics, animation, artificial intelligence, etc. The core level defines precisely the characters with sensory-motor endowment, the 2.3. The evaluation and tutoring level environment with its features and every possible interaction between characters and characters/environment. In a SG, a relevant role is played by the evaluation and These levels are in dynamic interaction and have strong tutoring level. A SG has an explicit educational goal that is effects one on the other: the narrative provides a frame to allow the player to accomplish specified educational where the hidden content resides. objectives. The evaluation and tutoring layer complements The shell level is necessary in providing a semantic context the core and shell layers. This level analyze player’s game to educational activities whereas the core level defines the performances relatively to the specified training objectives, skills or the abilities to be transferred. and provides the players and the trainer, whose role is In digital games, the core can host the game engine based on indeed relevant in Educational and Serious games, with computational algorithms. important information and data about the learning process. At this level we find learning analytics [Siemens, and Baker, 2.2 What the core level holds 2012], which are the measurement, collection, analysis and reporting of data about learners to improve the whole The core level holds the game mechanics, the engine. It can learning process. This level is also crucial from the teachers’ be conceptualized differently depending on the kind of point of view, as it provides specific tools and function to digital game we aim at building. support teaching processes. BrainFarm http://150.146.65.191/BrainFarm 3 Bio-inspired computational models in the proposed multi-level framework Bio-inspired computational algorithms can enter the SG design process in many different ways and at different level thus producing a diversified game typology. If the designer’s goal is to build a SG that is explicitly addressed to biology-related matters, bio-inspired computational models can flow from the core level to the shell level, thus becoming visible to the user. On the contrary, the designer can leave a traditional appearance to the game whereas the bio-inspired computational models work in an invisible Figure 1 The Brainfarm interface with custom neural manner, staying in the core level. Moreover bio-inspired networks computational models can be employed on the evaluation/tutoring level, providing the artificial evaluator The breeding is implemented through a user-guided genetic with a guise of artificial intelligence thus supporting the algorithm. The software side of Breedbot shows users a teachers’ role. In the following sections we will present population of nine wheeled robots, with infrared sensors and some example of such usage of bio-inspired computational motors and controlled by a simple feedforward neural models. network, representing an artificial nervous system. The neural network parameters are encoded in a genetic string that will undergo an evolutionary process guided by either 3.1 Bio-inspired computational models in the shell the users (artificial selection) or the machine (automatic level: when game narrative and mechanics selection). In this latter case the player can anyway converge manipulate the relevant evolutionary variables. By manipulating directly the parameters related to the genetic algorithm the player can understand the The first case is the use of bio-computational algorithms underpinning dynamics and experience different starting from the core and arriving to the shell level: the evolutionary pathways in a controlled environment. game mechanics are directly visible to the user. The shell In this example various bio-inspired computational models becomes transparent and what happens in the core level can that flow from the core to the shell level in a pervasive be accessed by the player; this way the SG becomes a manner. First of all, the robots are conceived as agents in the virtual laboratory where the user can directly manipulate the wide conception of ABM. Each robot, in fact, is seen as an relevant variables involved in the game, thus determining embodied agent interacting with a physical environment and the game evolution in an immediate manner. This direct with other robotic agents. manipulation takes place in a protected environment where Their artificial intelligence is implement adopting a failures or error do not determine a menacing outcome. This connectionist framework with artificial neural networks and virtue is counterbalanced by the unavoidable complexity their evolution/development carries out adopting reduction. evolutionary algorithms. Moreover the player can affect These games use Bio-inspired computational models for an directly the evolutionary pathway acting as a breeder that explicit interaction mechanism. The user interacts with the selects the preferred agents. The breeder acts as an expert game using traditional bio-inspired computational methods, using knowledge and expertise to select the best solutions. for example by evolving a population, training an organism, setting up an ecological system, etc. An interesting example of this kind of games is Breedbot 3.2 Bio-inspired computational models in the core and its sequels Bestbot, and Brainfarm [Miglino et al., 2008; level: using bio-inspired algorithms to model Ponticorvo et al., 2006]. the game engine. These are integrated software/hardware platforms that allow players, even without any particular computer skill, to The second case is the use of bio-inspired computational breed, within customizable virtual worlds, artificial models in the core level. The game engine is invisible for organisms that can be downloaded onto real robots (Fig. 2). the player that interacts with the game in a traditional These games can be reached though the following links: fashion, without perceiving what happens in the game engine. Bio-inspired computational algorithm are used to BestBot http://eutopia.unina.it/bestbot model complex system but the user does not interact with the computational models directly. These models can be derived from scientific theories in The game is played across numerous levels in which the many different domains and, obviously, the choice is driven player will lead teams in different corporation departments, by the designer educational objective. from the catering department to the research and As hinted in the introduction, one model that can be fit for development one. game design is ABM. This kind of models can be employed The game reproduces the day-by-day running of the to model both the interactions between agents and the agent department, including jobs with precise deadlines and itself. Moreover, if a careful description of agent is provided workloads. The player must assigning staff to work on those in terms of what it perceives, it knows about the external jobs. The basic challenge is to ensure that followers finish environment, and how it makes decisions upon its action, it all jobs in time and the leader must manage the assignment is possible to model the agent behaviours according to a of followers on jobs and their performance. With a smart specific psychological theory. management, the team can complete a respectable number Agent based modelling has been used to build Serious of job tasks within their deadlines. However, leadership Games in various contexts: for crowd simulation, economics involves more than management, and if the player uses a and artificial societies, just to cite some. strategy for developing followers, there will be an effective An interesting SG that exploits ABM to model and teach advantage in the game, in terms of completed tasks. team dynamics is LearnToLead (L2L) [Di Ferdinando et al., Indeed the followers are not all equals, as they are well- 2015]. L2L is a web-based game where the player covers characterized by ability level, motivation level, stress level the leader role and learns theories about leadership by and personalit. The player, acting as a leader must take these governing a team of artificial agents, the followers. The variables into account in assigning players to jobs and theoretical starting point is the Full-Range Leadership making action to affect the cited variables. It is, for Theory (FRL), a well-known and widely-employed theory example, possible to run workshops, organize team-building that explains leadership dynamics in small groups [Bass and events, perform one-to-one coaching, send memos, propose Avolio, 1994]. The game mechanics is developed by using training course, give lectures about performance, deliver two bio-inspired computational algorithms, namely ABM evocative speeches at staff meetings, etc. and artificial neural networks. On the core level, L2L is a logical structure where an In L2L game the shell level, the game narrative is clearly asynchronous interaction happen between the leader and the separated from the core level: it appears as a point and click followers. A turn-based structure to play is implemented, so game in a 2D environment which is played on the web. It that players always have an unlimited amount of time to takes place in a firms office, as it is evident from some carefully consider their actions, and consult reference decorative elements: desks, chairs, PCs and mobiles, stacks material about FRL if necessary before making an action. of paper, etc. (Figure 3). The player acts on the work environment and team dynamics by setting the working plan of each follower and influencing followers motivation, stress and their contribution to the team, with the action recalled before. More specifically, when the leader takes some decisions about one or more followers, these decisions affect the followers. These decisions become input for the follower's network and these inputs from the leader, together with the ones coming from the external environment, modify the agents’ internal states that, in turn, will change and influence the follower contribution to team job (Fig. 4). In brief, each agents’ behaviour is determined by external and internal variables: the first ones consist of leader’s behaviour (interact with followers), the working environment (total amount of workload, approaching of Figure 3 L2L office where the game takes place deadlines), and the social interaction with other followers. The internal ones are instead those relative to the psychological aspects of followers, between which This physical setting varies across all game levels looking motivation is the most important. In particular, three sub- nicer and nicer as the player advances in career. components of motivation are simulated: intrinsic, reward In L2L there is hierarchical interaction between the player and fear. The intrinsic component represents the internal acting as a leader and the followers, artificial agents. The form of motivation, driven by an interest or enjoyment in human player must manage the team of artificial agents, the activity. On the opposite, the reward and fear which stands for a team of workers in a bank, a post-office components model external forms of motivation, which rely branch or a local government office, for example. on external pressures like desire for reward, or fear for punishment. The peculiar features of these three components have been modelled using a different temporal In this game, the bio-inspired models are completely decay. In particular, the intrinsic component has a slower invisible to the users. In fact, they serve as effective decay than reward and fear, but can be activated only by technique to implement the FRLT, whereas in Breedbot (see appropriate leader behaviours (typically related to previous section), the bio-inspired models are relevant for leadership style). the interactive process. Another important variable to consider is the followers' This kind of games allows to observe dynamics that can personality, which has been modelled taking into account emerge form agents interaction and this is an important the McClelland [1978] theory. In particular, three different positive feature. It permits, in fact, to experience directly, personalities (or motivational drivers, according to even if in a controlled situation and a safe environment what McClelland theory) were considered: happens in a group context, thus complementing more a. Achievement: followers pursue excellence in traditional and theoretical learning methods. performance, a continuing drive for doing better all the The negative point is represented by the possibility that times. Excellence can be achieved through individual the emerging complex dynamics can slip away and generate efforts; unforeseeble outcomes. b. Affiliation: followers are interested in establishing, keeping, and restoring close personal relationships with others; 3.3 Bio-inspired computational models in the c. Power: followers pursue a status with impact on others. evaluation and tutoring level: when Bio- High power motivation induces highly competitive inspired algorithms model human trainer behaviour. expertise The stress level must be kept under control during the game The last case is about bio-inspired computational model in as well, because it affects the effort and the contribution of the evaluation/tutoring level. This level, that is followers to the team work. characterizing for educational games, foresees a smart Moreover, the followers ability has been simulated, as there interaction with the user/player. This smartness resides in are followers smarter (faster) than others doing their jobs. adapting, inferring, profiling and anticipation, functions that Thus, followers' performance is linked to their ability. mimic human teachers’ actions. All these variables interact among each other, with the In other words, at this level, it is necessary to foresee tools external stimuli and with the leader's behaviour, as depicted that extract two kinds of information: on one side, data in Figure 4. about the learner such as learning style, preferences, Personality and ability try to capture what the FRL theory weakness and strengths and, on the other side, about says about individual consideration, so that the same leader teachers behaviour in order to reproduce artificially human action may have a different impact on followers with trainer expertise. different personalities or abilities. On the other side, the The evaluation/tutoring level implements what an human leader who aims at raising the team motivation as high as expert in education would do while representing in an possible needs to perform some individualised effective and concise way what the learner does. considerations. Leaders should also pay attention when For example, this level provides an appropriate and timely assigning followers to the same workgroup, as conflicts may feedback to player action, it adapts to player special needs emerge depending on followers’ personality (Fig. 4). according to actual performance and the desired educational In this example, Bio-inspired computational models are goals, it tracks player performance in terms of achievements used to model the core of the game using an ABM and improvements. This smart interaction can mediated by approach, moreover each agent artificial intelligence is the use of Intelligent Tutoring Systems (ITS) [Carbonell, modelled using an artificial neural network, whose input and 1970]. output represent the already described external and internal Many examples can be found about this issue, as it has variables. arisen a strong interest since research about ITS was born in the seventies (for a recent review, see Wenger [2014]). One key feature in ITSs is the presence of a student model. To address the educational process it is crucial to pay attention to a particular student’s cognitive and affective states in order to tailor the whole teaching and learning process on the individual. To achieve this goal, it is necessary to build a student profile and a fruitful way to do it is to employ specific data analysis methodologies. Figure 4. Agent model in L2L Learning analytics rely on huge amount of data that can be used to improve learning. Educational data mining, for example, is a research branch devoted to processes designed for the analysis of data from educational settings to better as well as bio-inspired artificial learners. This can be done understand learners and the settings which they learn in. starting from the regularities extracted by Educational data Data mining, for specific learning goals too, can be run mining and by modeling learner/teacher and their interaction adopting bio-inspired methods as neural networks [Lu et al., exploiting, once again bio-inspired computational 1996]. algorithms. It can be also useful to run data clustering analysis and Bio- inspired methods can be used for clustering data, thus 4. Discussion illustrating another way these methods can be embedded in The design process that leads to Educational and Serious the evaluation/tutoring level. Games can derive useful hints and borrow models from bio- Data clustering consists in finding homogeneous groups in a inspired models, meaning that artificial intelligence has a dataset and Bio-inspired algorithms can be employed to find deep impact on how cognitive elements are embedded in new methods for clustering that include the human expert people-centred design for games [Vanden Abeele and Van role. In particular Interactive Evolutionary Computation Rompaey, 2006]. (IEC) techniques [Bintrup et al., 2006] can be used. In this The multi-level framework proposed in this paper goes in case, a human breeder selects cluster configurations on the this direction and allow to explicitate relevant issue on the basis of their graphical visualizations. future direction for the contribution of artificial intelligence Data clustering is based on the analysis of explicit and cognitive issue to game design. Indeed bio-inspired information and quantitative variables (dimensions) that computational methods can be applied effectively in describe a given phenomenon and on latent and implicit designing Serious and Educational games as they are information captured by human cognitive mechanisms: this isomorphic to teaching subject in the case of biology, implicit analysis is what characterize human experts, also in psychology, sociology. education domain. The human experts are usually trained Moreover teaching and learning with digital games can lead for many years to recognize (categorize) natural phenomena to neglect some relevant aspect that are, on the contrary, on both explicit and latent information even if they cannot fundamental in other educational contexts, such as physical explain how they do it. embodiment, autonomy, social interaction, evolution and IEC allow to embed this feature in Evolutionary development. These aspects allow biological organisms to computation with the intervention of a human operator that successfully adapt to unknown and changing environments interacts with the artificial evolution process. and widen artificial intelligence to embodied artificial intelligence [Pfeifer and Iida, 2004]. The positive feature of this kind of models is that, applied to teaching and learning processes, they can capture interesting regularities that help profiling the student/player/user. This Acknowledgments process supports teaching and improves learning, but it The INF@NZIA DIGI.tales has been funded by Italian doesn’t foresee a complete teachers substitution. This Ministry for Education, University and Research under compensates the dark side these methods display, that is the PON-Smart Cities for Social Inclusion programme. temptation to image the educational process with a learner totally immersed in a digital, automatic, artificial Authors would like to thank Onofrio Gigliotta for Breedbot environment without any human contact. materials. It is our opinion, on the contrary, that, especially in some life periods, such as infancy, the social dimension of References education cannot be neglected: it should be rather supported [Aldrich, 2005] Aldrich, C. (2005). Learning by doing: A by these methods and algorithms that try to replicate comprehensive guide to simulations, computer games, teacher/student interaction but not cancelled. This is the and pedagogy in e-learning and other educational rationale behind Infanzia Digi.tales, an on-going research experiences. John Wiley & Sons. project whose goal is to provide smart digital objects to be used in learning and teaching process in children. [Balacheff et al., 2009] Balacheff, N., Ludvigsen, S., De Moreover it is worth underlining that this example is doubly Jong, T., Lazonder, A., Barnes, S. A., & Montandon, L. interesting as it shows how to build learners profile using (2009). Technology-enhanced learning. Berlin: Springer. bio-inspired computational models and indicating a new [Bass & Avolio, 1994] Bass, B. M., & Avolio, B. J. (1994). way of implementing a smart interaction. Improving organizational effectiveness through Introducing an human expert in the evolutionary process, it transformational leadership. Sage. shows how to go beyond ITS and propose a new framework [Bhatt and Freksa, 2015] Bhatt, M., & Freksa, C. (2015). with educational agents (EA), working in dynamic Spatial computing for design—an artificial intelligence interaction. If we conceive both learners and teachers as perspective. In Studying visual and spatial reasoning for agents, ABM allows us to model effectively this interaction design creativity (pp. 109-127). Springer, Dordrecht. and to try to build bio-inspired artificial experts in education [Brintrup et al., 2006] Brintrup, A. M., Ramsden, J., & [Michael & Chen, 2005] Michael, D. R., & Chen, S. L. Tiwari, A. (2006). A review on design optimisation and (2005). Serious games: Games that educate, train, and exploration with interactive evolutionary computation. inform. Muska & Lipman/Premier-Trade. In Applications of Soft Computing (pp. 111-120). [Miglino et al., 2008] Miglino, O., Gigliotta, O., Ponticorvo, Springer Berlin Heidelberg. M., & Nolfi, S. (2008). Breedbot: an evolutionary [Carbonell, 1970] Carbonell, J. R. (1970). AI in CAI: An robotics application in digital content. The Electronic artificial-intelligence approach to computer-assisted Library, 26(3), 363-373. instruction. IEEE transactions on man-machine systems, [Patterson, 1998] Patterson, D. W. (1998). Artificial neural 11(4), 190-202. networks: theory and applications. Prentice Hall PTR. [Chandrasekaran, 1990] Chandrasekaran, B. (1990). Design [Pfeifer and Iida, 2004] Pfeifer, R., Iida, F. (2004). problem solving: A task analysis. AI magazine, 11(4), Embodied artificial intelligence: Trends and challenges. 59. In Embodied artificial intelligence (pp. 1-26). Springer, [Charsky, 2010] Charsky, D. (2010). From edutainment to Berlin, Heidelberg. serious games: A change in the use of game [Pintea, 2014] Pintea, C. M. (2014). Advances in bio- characteristics. Games and culture, 5(2), 177-198. inspired computing for combinatorial optimization [Chopard & Droz, 1998] Chopard, B., & Droz, M. (1998). problems. Berlin: Springer. Cellular automata. Springer. [Ponticorvo et al., 2006] Ponticorvo, M., Walker, R., & [Dell’Aquila et al., 2016] Dell'Aquila, E., Marocco, D., Miglino, O. (2006). Evolutionary robotics as a tool to Ponticorvo, M., di Ferdinando, A., Schembri, M., & investigate spatial cognition in artificial and natural Miglino, O. (2016). Educational Games for Soft-Skills systems. Artificial Cognition Systems, 210-237. Training in Digital Environments: New Perspectives. [Ponticorvo et al., 2016] Ponticorvo, M., Di Ferdinando, A., Springer. Marocco, D., & Miglino, O. (2016). Bio-inspired [Di Ferdinando et al., 2015] Di Ferdinando, A., Schembri, computational algorithms in educational and serious M., Ponticorvo, M., & Miglino, O. (2015). Agent based games: some examples. In EC-TEL (pp. 636-639). modelling to build serious games: The learn to lead Springer International Publishing. game. In Inter. Work-Conference on the Interplay [Rieber, 1996] Rieber, L. P. (1996). Seriously considering Between Natural and Artificial Computation (pp. 349- play: Designing interactive learning environments based 358). Springer International Publishing. on the blending of microworlds, simulations, and games. [Gee, 2003] Gee, J. P. (2003). What video games have to Edu. Tech. research and development, 44(2), 43-58. teach us about learning and literacy. Computers in [Siemens & Baker, 2012] Siemens, G., & d Baker, R. S. Entertainment (CIE), 1(1), 20-20. (2012). Learning analytics and educational data mining: [Goldberg, 2006] Goldberg, D. E. (2006). Genetic towards communication and collaboration. In algorithms. Pearson Education India. Proceedings of the 2nd international conference on [Helbing, 2012] Helbing, D. (2012). Agent-based modeling. learning analytics and knowledge (pp. 252-254). ACM. In Social self-organization (pp. 25-70). Springer Berlin [Tobias & Fletcher, 2011] Tobias, S., & Fletcher, J. D. Heidelberg. (2011). Computer games and instruction. IAP. [Kapp, 2012] Kapp, K. M. (2012). The gamification of [Van Abeele & Van Rompaey, 2006] Vanden Abeele, V. learning and instruction: game-based methods and A., & Van Rompaey, V. (2006). Introducing human- strategies for training and education. John Wiley & centered research to game design: designing game Sons. concepts for and with senior citizens. In CHI'06 ext. ab. [Lieto and Radicioni, 2016] Lieto, A., & Radicioni, D. P. on Human factors in computing systems (pp. 1469- (2016). From human to artificial cognition and back: 1474). ACM. New perspectives on cognitively inspired AI systems. [Van der Hoek & Wooldridge, 2008] Van der Hoek, W., & [Lu et al., 1996] Lu, H., Setiono, R., & Liu, H. (1996). Wooldridge, M. (2008). Multi-agent systems. Effective data mining using neural networks. IEEE Foundations of Artificial Intelligence, 3, 887-928. transactions on knowledge and data engineering, 8(6), [Wenger, 2014] Wenger, E. (2014). Artificial intelligence 957-961. and tutoring systems: computational and cognitive [Malone, 1981] Malone, T. W. (1981). Toward a theory of approaches to the communication of knowledge. Morgan intrinsically motivating instruction. Cog. Sci., 5(4), 333- Kaufmann. 369. [McClelland, 1987] McClelland, D. C. (1987). Human motivation. CUP Archive.