Abstract—Persuasive technological strategies directed towards changing users’ attitudes or behaviors have thus far been applied in commercial and health promotion contexts, but not for educational purposes. The main reason for this lack of research in educational settings can be attributed to an insufficient understanding of which attitudes or behaviors should be adopted through the evaluation of students’ effective learning outcomes. In this study, we present an association between interpersonal conflict and positive learning and depicted the speaker’s intention and the hearer’s interpretation for cooperative and competitive player roles in light of game theory. Index Terms—Computer-supported collaborative learning, flaming, insults, interpersonal conflict. I. INTRODUCTION During collaboration in computer-supported collaborative learning (CSCL), students engage in a wide variety of collaborative and learning behaviors, which impact each student‘s learning in a variety of ways [1]-[3]. Throughout CSCL tasks, interactions between students are recorded in protocols. The analysis can be time consuming due to the large amount of data that must be managed. Hence, an automatic coding procedure for coding dialogue acts is required. Reference [4] recognized at least four general activities for the analysis of the interactive processes in collaborative learning dialogues: social, cognitive, affective, and meta-cognitive. [5], however, argues that it is inconceivable to dissociate cognitive tasks from social tasks because all cognitive tasks have a social component. In addition to these activities, other variables can be identified, for example, dialogue structure and sequence [6] and didactic-pedagogic issues [7]. Some researchers [8]-[10], have encouraged collaborative knowledge sharing. Reference [11] and [12] argued that social interaction is crucial to productive collaborative learning. Cognitive processes in collaboration, such as problem solving [13], are as important as social factors, such as motivating environments with positive affective relationships [14], [15]. Studies from the 1970s have demonstrated that conflict Manuscript received July 17, 2014; revised October 8, 2014. This work was funded in part by NSF grant REC-043779 to ―IERI: Learning-Oriented Dialogs in Cognitive Tutors: Toward a Scalable Solution to Performance Orientation.‖ David N. Prata and Patrick Letouze are with Federal University of Tocantins, CO 77.001-009, Brazil (e-mail: ddnprata@uft.edu.br). Stefano Cerri is with Université Montpellier, CO 34000, France (e-mail: cerri@lirmm.fr). Evandro Costa is with the Federal University of Tocantins, Palmas, CO 77.001-019, Brazil (e-mail: ebc.academico@gmail.com). and interaction promote cognitive development [16]-[20], and describe distinctions between conflict and content [21], consistent with Piaget‘s [22] discussion of the equilibration process. Several studies have found that cognitive conflict and learning emerge from the process of collaboration, when students mutually engage to co-construct shared knowledge [1]-[3]. Additionally, other studies have provided evidence suggesting that cognitive conflict does not only occur in purely collaborative and consensus-based processes. For example, [23] and [24] provide evidence that conflicting ideas often lead to interpersonal conflict. This study aimed to evaluate how social behavior can influence student‘s learning. Previous studies (e.g., [25]) have found insults to be associated with positive learning. In this study, we describe complementary findings for the role of insults in collaborative learning, depicting the speaker‘s intention and the hearer‘s interpretation for cooperative and competitive player roles in light of game theory. II. PLAYER ROLES IN COLLABORATIVE LEARNING Collaborative learning can be seen as a social game in which agents (players) are able to cooperate or compete in order to solve a problem. For game theory, the classic prisoner dilemma, the goal is to win goodness; competition means win-lose, cooperation means win-win, but there is also the awkward ‗lose-lose‘ collaborative situation. The medium could be communication, negotiation, and/or mediation. We sought to investigate who was most likely to gain knowledge (the goal) in CSCL games by examining the roles of the person who cooperates or competes as well as the person who sends information or receives it, and the effects of communication, negotiation, and mediation on the knowledge-building process. Based on the findings of [25] and [26], we have developed Fig. 1. Fig.1. Speech acts for learning gains. The analysis of a set of chat protocols using this coding scheme allowed us to identify speech acts [27] that correlated with learning gains for cooperative and competitive student player roles. These categories were qualified in terms of David N. Prata, Patrick Letouze, Stefano Cerri, and Evandro Costa A Game Approach to Assessing Learning Outcomes International Journal of Information and Education Technology, Vol. 6, No. 2, February 2016 137 DOI: 10.7763/IJIET.2016.V6.673