Serious Games in Cognitive Training for Alzheimer’s Patients Frédérick Imbeault, Bruno Bouchard, and Abdenour Bouzouane LIARA laboratory Université du Québec à Chicoutimi Chicoutimi, Canada {Frederick.Imbeault, Bruno.Bouchard, Abdenour.Bouzouane}@uqac.ca Abstract—Research on progressive dementia increased significantly in the past years due to the urgency of the aging population. Patients suffering from such dementia, for instance Alzheimer's disease, lose efficiency in cognitive spheres such as memory, planning skills, initiative and perseverance. Some researchers tried to evaluate the potential of close-to-reality simulations and generic video games for brain training to stimulate the cognitive abilities of AD patients. Using recent advances in artificial intelligence such as learning, activity recognition and guidance to enhance this concept of training, we are proposing, in this paper, a detailed explanation of an adapted serious game we designed for this purpose. A prototype has been developed showing how to exploit AI techniques to create an affordable and accessible tool for cognitive training and allowing in-game estimation of the patient's cognitive performance. Keywords-algorithms; design; human factors; cognitive training; cognitive performance; serious game; alzheimer's disease; cognitive impairment; I. INTRODUCTION Over years, the game industry created new platforms, such as the Nintendo Wii and the Nintendo DS, in order to attract a larger community of gamers. Theses platforms present new ways of gaming, enabling the entire family to play together. Constant evolution in these technologies recently paved the way for what is called serious video games. This new type of digital games specializes in other purposes than just entertaining, such as educating [1], leading societal impact on specific subjects [2], enhance individual user's aptitudes [3] and, more recently, train cognitive faculties of silver-aged gamers [4]. Recently, research on progressive cognitive dementia like Alzheimer's disease (AD) increased significantly due to the aging of the population. Some researchers turned themselves towards serious video games, in order to measure their potential in cognitive training of silver-aged gamers and AD patients. To explore this potential, most of the studies either use a realistic approach using close-to-reality simulations [5], or serious video games made for this purpose [4]. Nevertheless, AD patients need specialized training that will focus on four cognitive spheres; memory, planning skills, initiative and perseverance [6]. In addition, AD patients need specifically adapted challenges, and also need help to complete them. In that sense, studies show that it is more beneficial for AD patients to be helped through completion of a given task than just see the task failed and be presented with a new challenge [7]. Thus, to be correctly suited to this type of patient, a serious game should remain non-intrusive and present assistance to the patient through completion of the challenges. Moreover, each patient presents a different profile and might not necessarily need assistance in the same contexts or even in the same manner [8]. For example, a patient suffering from auditory disorders can benefit from a visual feedback, but auditory feedbacks should be avoided. Consequently, trainings should dynamically adapt themselves to a given profile to be fully effective. Yet, most games on cognitive training such as the popular Nintendo DS game Dr. Kawashima's Brain Training does not offer either help to the player in the completion of challenges or dynamic adaptation to the player's profile. Therefore, not only these games are not suited to this kind of player, but they does not allow the players to think about their errors and try to correct them, which is an important point in education and re- education [9]. We also noticed that the tools used in previous research [2,4,5] does not make use of modern artificial intelligence technologies such as learning, activity recognition (AR) and guidance in order to offer a personalized and more effective experience. Considering these problems and in order to contribute to the researches on cognitive training of AD patients, we are proposing, in this paper, the design and prototype of a serious game toggled to these patients. The prototype show how to exploit the recent advances in AI techniques to create a game that will constitute a cheap and accessible tool for AD patient's cognitive training and evaluation. The tool will also enable researchers to compare standard cognitive games to such specialized serious games. As for a gym can be used to train our body 3 or 4 times a week using specific equipment, the game will provide cognitive tasks and challenges to train mental faculties and help to slow down the degenerative process of cognitive functions of AD patients. Previous contributions of our labs, such as an AR model for AD patients [10], trial data sets for smart home researchers [11], and adaptive game mechanics for learning purposes [9] will be used to design and develop the game. This paper is organized as follows. Section 2 introduces the relevant related works in serious video games and cognitive training in Alzheimer's disease. Section 3 brings forward our theoretical contribution, including the conceptual model of the game. Section 4 describes the implementation realized to test our concept and presents