Developing serious games specifically adapted to people suffering from alzheimer

VIDEO GAMES FOR COGNITIVE TRAINING OF MENTALLY-IMPARED PATIENTS

INTRODUCTION

In the past decade, the video game industry has been developing and pushing technology to constantly offer players new, breathtaking experiences. The industry has been driving technology to create new hardware, achieve better performance, offer never seen before graphics and defining new ways for players to evolve in these virtual worlds. With over 155 million Americans playing video games, over 22 billion USD spent on the game industry in the U.S. in 2014 and more than half of U.S. Households owning a dedicated game console (ESA Essential Facts, 2015), it’s clear that this industry is a huge success with games now being ubiquitous in our culture. But what exactly could explain this tremendous success that video games are being part of?

THE PSYCHOLOGY OF FUN IN VIDEO GAMES

Back in the mid-1970s, Minhály Csíkszentmihályi, a professor in psychology, came up with the concept of Flow in an effort to explain happiness (Csíkszentmihályi, 1990). This concept, which has since become fundamental to the field of positive psychology, has been originally described as a feeling of complete and energized focus in an activity, with a high level of enjoyment and fulfilment, also described as being “In the Zone”. In video games, Flow is a well-known concept characterized by the immersion of the player loosing track of time and external pressure, by the feeling of control over the game experience, the sense of disconnecting from everyday life, and the sense of disappearing (Chen, 2007). Jenova Chen explains that in order for a game to have broader appeal than others, their design must evoke positive feeling. He also explains that in order to maintain a user Flow’s experience, the activity must balance the inherent challenge of the activity and the player’s ability to overcome it, as shown in figure 1. If the challenge is too difficult for the player’s skills, he will be overwhelmed and is likely to experience anxiety. On the other hand, if the challenge is too simple for the player’s skills, he will experience boredom. In both situations, the player won’t be in his Flow zone and may lose interest in playing the game.
However, while playing a given video game, different players will present different skill levels and won’t require the same level of challenge to experience Flow. Many games have been trying to overcome this situation by offering the player a choice from different difficulties at the beginning of the experience. One way of doing this is by requesting the player to choose a difficulty from a defined set, such as easy, medium or hard. Those difficulty levels would be intimately bound to the skill level necessary to overcome the challenges that will be presented to the players. However, this often requires the player to guess their skill level before even trying the game. This can lead to frustration since the player can easily make a choice that won’t correctly reflect the level of challenge that would suit his skills (Chen, 2007). Another way of approaching the different skills level problem is to implement Dynamic Difficulty Adjustment (DDA) into the game. DDA is a method allowing the game to dynamically modulate the difficulty of the different challenges in order to more accurately respond to the player’s skill level through the course of a game session (Hunicke, 2005). Examples of DDA includes design-based approaches, such as letting the player control the difficulty of the challenges by their actions, as proposed by Chen (Chen, 2007). In that sense, Chen gives the example of real life surfers developing skills and choosing to engage particular waves. Other approaches such as the one proposed by Hunicke (Hunicke and Chapman, 2004) presents a computational way of determining the player’s expectation of the difficulty in order to adapt the game to it. Independent of the approach used, the idea of DDA is to adapt the game to the player’s skill level rather than asking the player to adapt himself. The result is more accurate challenges difficulty adapted to a wide range of players.

“SERIOUSLY” FUN GAMES

Over the past decades, the always increasing popularity of video games as an entertainment medium had many researchers wondering about its potential as a tool in many fields, such as promoting healthy behavior in children (Majumdar et al., 2013), supporting rehabilitation in disabled patients (Prange et al., 2015), training medical personnel (Graafland et al., 2012, Graafland et al. 2014), increasing social awareness (Rebolledo-Mendez et al., 2009) and using Game-Based Learning as an education tool (Egenfeldt-Nielsen et al., 2011). Still using the potential of high enjoyment, these games are built for a primary purpose other than pure entertainment. The concept of Serious Games was first described by Abt (Abt, 1970), as games that “have an explicit and carefully though-out educational purpose and are not intended to be played primarily for amusement”. Early examples of serious video games includes The Oregon Trail (Rawitsch, 1978), produced in 1974 and designed to teach players about the life of 19th century American pioneers. In 1981, Atari developed The Bradley Trainer (Arcade-History, 2015), a game for the training of American army recruits in how to operate a Bradley tank. While the early era of serious games was mainly focused on military training (Laamarti et al., 2014), today’s application are a lot broader. Ranging from education, health care, training and many other fields, the research on serious games has been rapidly growing in the past decade.

THE AGING OF WESTERN POPULATION

In the past years, research on neurodegenerative disease such as Alzheimer’s disease (AD) increased significantly due to the urgency of the aging population. One notable area of such research is increasing life quality and autonomy of cognitively-impaired patients using Smart Home (Bouchard et al., 2012). For instance, the LIARA has been driving researches on activity recognition (AR) of AD patients in smart homes, using recent advances in artificial intelligence (AI) (Maitre et al. 2015, Tremblay et al. 2015, Fortin-Simard, 2015) and effective way to assists those patients in completing daily tasks by making efficient use of prompts to provide customized guidance as needed (VanTassel et al., 2011). In fact, studies have shown that it is more beneficial for AD patients to be provided with help in the completion of a given task than to simply see the task fail (Pigot et al., 2003). Further researches also shows that not only providing help in the completion of tasks will increase autonomy of AD patients, but specifically adapted use of different prompts, depending on the patient’s profile and nature of task, will increase efficiency of guidance (Lapointe et al., 2013). For example, auditory feedback can be adapted to certain profiles, but should be avoided with patients suffering from auditory disorder.
Another area that researchers have been exploring to overcome the aging of western population is to develop ways to cognitively train elders. To that end, some researchers have been driving efforts to develop tools such as close-to-reality simulation (Hofmann et al., 2003) or to make use of commercial games (Nacke et al., 2009) in order to train AD patient’s cognitive abilities. However, none of these approaches takes into account researches in cognitive performance for AD patients, showing that specialized training should focus on four cognitive spheres: memory, planning skills, initiative and perseverance (Baum and Edwards, 1993) Furthermore, those approaches either does not make use of any assistance or does not take into account recent researches in Smart Homes in terms of adapted assistance depending on the patient’s profile. Tools aimed at training AD patients cognitive skills should take into account all of these specificities in order to maximize its potential.

SERIOUS GAMES IN HELP OF AD PATIENTS

Seeing the enormous popularity of video games and the recent successful use of serious games in various spheres of application, we have been interested in accessing their potential for cognitive training of mentally-impaired patients such as AD. In that sense, our first paper (Imbeault et al., 2011) focused on using state of the art AI developed in Smart Homes for AR and assistance to develop a serious game prototype aimed at stimulating cognitive spheres of AD patients for cognitive training. The paper first introduces related work in the field of serious games and cognitive training of AD patients. It then presents our theoretical contribution, the in-depth description of a serious game prototype aimed at the cognitive training of AD patients. We describe the reasons behind design choices of the game settings and mechanics and also detail our different systems such as our DDA algorithm, the player’s profile adaptation for adapted assistance during challenges, and the in-game cognitive evaluation. Finally, it describes our implementation realized to test our concept and exposes our efforts to conduct experimentations, with a description of our experimental protocol to be used.

TOWARDS GUIDELINES TO SERIOUS GAMES FOR COGNITIVE ASSISTANCE

INTRODUCTION

Our work up to the publication of the article presented in the second chapter mainly focussed on going through current state of the art in different fields of research to build an effective tool for the cognitive training of patients suffering from Alzheimer’s disease (AD). Seeing the enormous success of video games in the past decades and the growing number of games built for another primary purpose than entertainment due to their effectiveness in different fields, it became clear to us that Serious Games would probably have great potential in cognitive training of AD patients. In order to effectively exploit this potential, we needed to understand how such a game would be built to respond to the needs of this particular audience. We studied the psychology of fun in video games and how the concept of Flow would help us keep the patients engaged in the experience, avoiding feelings such as anxiety and boredom (Chen, 2007). To do so, we explained how we adapted the well-know ELO raking system to create a DDA algorithm for a single player game, by having the player to score against the game itself. We also explain the importance of recreating a well-known environment to the patients and introduce them with challenges familiar to them. Furthermore, we expose our focus on assisting the patient throughout the different challenges using recent researches on using correct prompting for assistance based on the patient’s profile and nature of task for cognitively-impaired patients.
Following the implementation of our first serious game prototype, we had seen multiple approaches for cognitive training of mentally-impaired people. However, these researches either did not make use of specifically adapted tools for AD patients or did not use the potential of serious games, which had been proven effective in many fields. In order to contribute to the scientific community, we felt the need of creating guidelines for the development of serious games specifically adapted for the cognitive assistance of AD patients. Those guidelines could then be used to create different serious games targeted at AD patients or be adjusted to target other cognitively-impaired patients. The set of guidelines we created for this purpose cover four different aspects: (i) choosing right in-game challenges, (ii) designing appropriate interaction mechanisms for cognitively impaired people, (iii) implementing artificial intelligence for providing adequate assistive prompting and dynamic difficulty adjustment, (iv) producing effective visual and auditory assets to maximize cognitive training. Each of these aspects are introduced in this chapter and covered in-depth in the article presented in chapter 4.

CHOOSING RIGHT IN-GAME CHALLENGES

Previous researches show that, in order to avoid the need of learning complex mechanisms, we need to recreate a well-known environment to the patients and present them with adapted and familiar challenges (Laprise et al., 2010). In order to be able to measure the positive impact of the training we also need to conduct in-game cognitive evaluation during play sessions. In that sense, our first guideline suggest choosing
challenges that will allow evaluation, and keeping trace of the patient’s cognitive abilities from data collected during play sessions. In our game, we chose to implement activities from the well-known neuropsychological test called the Naturalistic Action Test (NAT) (Schwartz et al., 2002), which uses adapted activities based on routine actions of everyday life called Activities of Daily Living (ADL). This also allowed us to integrate proven score sheets for cognitive evaluation of the NAT directly in our game. We then suggest determining an appropriate number of steps for the chosen challenges, as too much steps could overload the patients and lower the benefits of the training. Since we use NAT-based evaluation in our game, we made sure to have compatible levels presenting from 8 to 12 steps. As a third guideline, we suggest keeping the player in his Flow zone to maximize engagement and positive impact of the training. In that sense, it is important to remember that it is more difficult for AD patients to learn new paradigms or complex mechanisms. In order to avoid confusion or frustration, and induce an enjoyable experience, the game must be straightforward and easy to learn. Finally, the article exposes our choice of using cooking activities in our serious game and explains the reasons behind this decision.

DESIGNING USER INTERFACE AND INTERACTION MECHANISMS

In order to have a healthy brain, it is known that physical activities can have beneficial effects on cognition (Kramer et al., 2003, Hillman et al. 2008). Thus, physical activity should be used in conjunction of cognitive training to increase its positive effects. In that sense, our guidelines suggest to promote ecological interactions in the serious game in order to reduce the learning time for the patient and lead to greater enthusiasm and engagement. However, it is important to do so with the targeted group in mind. Since elderly people can suffer from impaired motor skills, interactions based on whole body should be avoided. We also suggest having light interface for home-based exercises, like having portable interfaces that does not require any particular skills for configuration at home. Finally, we suggest taking advantage of the multimodal aspect of virtual reality (VR) technologies. For example, by using voice to indicate commands, or using vibro-tactile feedback as a memory aid (Kuznetsov et al., 2009). This also enables us to modulate the experience based on the patient’s profile, giving us multiple alternatives for people with auditory or visual disorder.

AI FOR ASSISTIVE PROMPTING AND DYNAMIC DIFFICULTY ADJUSTMENT

The concept of Flow in video games is the ability to induce a high level of enjoyment and fulfilment, resulting in greater engagement from the player, by creating a balance between the player’s skill level and the difficulty of the challenges presented by the game (Chen, 2007). In order to maintain this balance, the game must adapt itself to the player by modulating the challenges to match the skills of the player (Hunicke, 2005). Since we are creating games for the silver-aged, it is important to keep in mind that they are not usually familiar with digital forms of games (Nacke et al., 2009). In that sense, we provide guidelines to use Dynamic Difficulty Adjustment (DDA) and assistive prompting to adapt serious games to AD patients. First, we propose to use Activity Recognition (AR) and the player’s profile to provide adapted assistance. Video games may not require complex mathematical model such as POMDP (Mihailidis et al., 2007) or HMM (Wilson and Philipose, 2005) used in smart homes, but the game should be able to provide adapted assistance on the task the player chose to undertake. It is also important to make use of the player’s profile in order to choose adequate assistance prompt (Lapointe et al., 2013). We then propose to make use of DDA to keep the player in his flow zone. In our game, we implemented an algorithm based on the well-known ELO ranking system (Coulom, 2010), adapted for a single-player game. To do this, we assign a rank to each level representing the expected difficulty, and assign a rank to the player that evolves depending if he completes the tasks easily or repeats errors and need assistance. We then make use of a normal distribution function FN(R1-R2) to give us a difficulty ratio, and adapt the challenges difficulty and assistance accordingly. This allows us to expect the level of assistance that will be required for a given level and smooth the passage from general to specific assistance and vice-versa.

PRODUCING EFFECTIVE VISUAL AND AUDITORY ASSETS TO MAXIMIZE COGNITIVE TRAINING

The last aspect of our guidelines concerns the production of assets specifically for AD patients. This is crucial since we must consider the visual and auditory troubles associated the effects of aging, since it represents the main risk factor for AD (Alzheimer’s Association, 2001). First, we suggest creating simple scenes as we know that aging can cause difficulties to find objects in visually complex scenes (Ally et al., 2009). We also explain that speed of movements of the cursor or objects in the scenes should be carefully designed. We then suggest using warm and bright colors with simple textures since they are the best seen by elderly persons (Jones and van der Eerden, 2008). We follow with the importance of creating good luminosity but avoid dazzling when creating 3D scenes, and use light to help the patients focus their attention at important objects in the scene. We then propose defining contrasts clearly to improve depth perception, since AD patients often lack the capacity discern figures from backgrounds and will have more trouble perceiving depth when looking at darker areas. One way of doing this is to exaggerate object’s outline thickness by using a well-known technique called toon-shading (DeCarlo and Rusinkiewicz, 2007). Finally, we suggest producing multiple assets to make use of different prompts when providing assistance, since specific profiles might not necessarily need assistance in the same context or in the same manner (Lapointe et al., 2013).

GUIDELINES FOR BUILDING SERIOUS GAMES AIMED AT COGNITIVE TRAINING OF AD PATIENTS

This chapter explained our journey from going through the literature of different fields of study in order to build an effective tool for the cognitive training of AD patients, to the creation of guidelines aimed at helping the research community to build similar tools or to be adapted for developing serious games targeted at other cognitively-impaired patients. The chapter introduced the different guidelines that we published to that end. The following chapter presents the article (Bouchard, Imbeault et al. 2012), that explains the origin of these guidelines and covers each of them in-depth. The result is a comprehensive and easy to use list that covers different aspects that should be considered when developing serious games aimed at AD patients.

Choosing the right in-game challenges for the patient

Previous researches show that AD patients need specifically adapted challenges [25], and also need help to complete them [16]. Consequently, trainings
should dynamically adapt themselves to a given profile in order be fully effective. This aspect may also impact positively the player’s engagement since it sustains its interest. In this section, we analyze how the in-game features that should be designed in order to fit the patient’s profile.

Guidelines

Keep trace of the patient’s cognitive abilities. One of the important features we were interested in is that the game would be capable of producing an in-game estimation of the patient’s cognitive abilities, using the data collected from the different activities. This will allow us to measure the positive impact of the game on the patient’s cognitive performance through the training sessions and keep a history of the estimations through time to fully evaluate the game potential. For testing real-life patient’s cognitive abilities in smart homes, our lab is using a well-established neuropsychological test called the Naturalistic Action Test (NAT) [13]. The test uses adapted activities based on routines actions of everyday life called Activities of Daily Living (ADL), in order to assess the patient’s errors using predefined score sheets. To answer our need of in-game cognitive evaluation, we decided to develop a game concept based on the activities used in this test, and to integrate the score sheets used for the evaluations in the game, in order to provide a fast estimation of the patient’s cognitive abilities during the play sessions.
Determine an appropriate number of steps for the challenges. Each challenges presented in the game should be completed in a correct number of steps. A high enough number of steps would correctly train the cognitive abilities of the patients. However, too much steps could overload them and lower the benefits of the game. As we decided to use an in-game NAT-based test as explained in the previous subsection, we determined that the game levels should be made of 8 to 12 steps, in order to assure they would be compatible with the NAT.
Keep the player in his “flow zone”. Keeping the player in is flow zone is important. Flow is a well-known concept in the video games community, representing the feeling of complete and energized focus in an activity, with a high level of enjoyment and fulfillment [28]. Maintaining the flow will make the game more fun for elderly gamers, improving their learning experience [14]. It is important to remember that it is more difficult for AD patients to learn new paradigms. Hence, the game must be straightforward and easy to learn in order to avoid confusion or frustration of the player. A good way to achieve this is to recreate a well-known environment and choose challenges that reflect the patient’s everyday life, as it will prevent the need of understanding complex mechanisms [15]. Also, placing them in a familiar context will smooth the learning curve and assure a less frustrating, and more enjoyable experience.

Our choice for the serious game environment

Considering the constraints we defined, and the fact that using ADL would allow the use of a NAT-based test in our game, we decided to base our serious game concept on cooking activities for multiple reasons. First, theses activities respond to the need of recreating a well-known environment for all patients [15]. Secondly, the NAT tests we are using at our lab with real patients are mostly done in a kitchen environment, so the integration in game will be easier since we have plenty of accessible data on the subject. Besides this, the importance of food in everyday life is quite crucial. Thus, not only making the patients prepare meals will train their cognitive faculties, but it will also make them able to repeat such tasks outside the training, i.e., at home [15]. Finally, cooking is a well-established subject in assistive technologies for elders [16], which means information from various researches are easily accessible and will allow us to effectively evaluate the patient through the training process.

Designing user interfaces and interaction mechanisms

It is known that physical activities can have benefic effects on brain and cognition [17], [18] as a result, several computer-based rehabilitation systems are based around functional activities. With studies reported in [17] the authors found that exercises effects on cognition were greater for exercice training interventions that exceed 30 min per sessions. Besides many effective results, when implementing these recomendations in traditional therapy approaches, people do often complain about repetive aspects of the exercices and healthcare cost are usually high [19]. In this section we do analyze how interaction aspects of serious games may serve for targetting these aspects.

Understanding the success of Wii-like games

The arrival of the Wii-like games has promoted a strong integration of video games in centers for elderly. All over the world, it is common to see players who are their late eighties or in their early nineties. To understand this true success, it is necessary to analyze why is the Wii different from others. While interactions with traditional games were essentially based on the couple (keyboard/mouse), with Wii it is rather the natural that prevails. No more need for buttons or arrows in order to move, just do the right gesture and you are done. We refer this as ecological interactions. With such interactions, the learning time is greatly reduced if not absent. Such an aspect is specially important when dealing with elderly. Moreover, because of these realist interactions, the player is more likely to give credit in his task and hence to be engage in it. With more engagement, one can expect more fun and motivations that will let the user perform all the necessary exercises.
Although playing a major role in the Wii success, it is clear that the factors cited above can not, on their own, explain this console’s major success. Besides, one notes that these factors constitute the core of Virtual Reality (VR) technologies which aims to create a virtual environment where the user would felt being in reality. In fact what differentiates these games from VR applications is primarily their affordability in terms of development, maintenance and usage. This explain why games are particularly adapted to rehabilitation. In contrast to VR applications, games are so accessible, nowadays various games are dedicated to home-based rehabilitation.

CONCLUSION

The video game industry has been rapidly growing in the past decades and its potential has interested the scientific community in various different fields. The understanding of the psychology of fun and the concept of flow in games paved the way for the development of games which are built for a primary purpose than entertaining, called Serious Games. These games have gained interest in fields such as promoting healthy behavior in children (Majumdar et al., 2013), supporting rehabilitation in disabled patients (Prange et al., 2015), training medical personnel (Graafland et al., 2012, Graafland et al. 2014), increasing social awareness (Rebolledo-Mendez et al., 2009) and using Game-Based Learning as an education tool (Egenfeldt-Nielsen et al., 2011).
In the past years, research on neurodegenerative disease such as Alzheimer’s disease (AD) had significant increase due to the urgency of the aging population. Notable areas of research to address this issue include researches in Smart Homes to increase the quality of life and autonomy of cognitively-impaired patients. In this field, researchers have been interested in using technology deployed throughout the living environment (sensors, RFID, effectors, etc.) to help its resident by providing assistance in daily tasks (Bouchard et al., 2007, Mihailidis et al., 2007). Indeed, studies shows that it is more beneficial for AD to be helped through the completion of tasks, rather than simply see these tasks fail (Pigot et al. 2003). Research in this field also shows that assistance must be adapted depending on the patient’s profile and nature of task (Lapointe et al. 2013). On the other hand, some studies have also been exploring different methods and the benefits of cognitive training for the elders. To that end, researchers have been using tools such as close-to-reality simulations (Hofmann et al., 2003) or commercial games (Nacke et al., 2009) in order to train the cognitive abilities of AD patients. However, none of them has been at the same time presenting challenges focused on cognitive spheres that should be trained for AD patients (Baum and Edwards, 1993), and implementing state of the art Artificial Intelligence (AI) in order to provide specifically adapted assistance through the completion of different tasks (Lapointe et al. 2013).
The objective of the first paper (Imbeault et al., 2011), presented in chapter 2, was to exploit the potential of Serious Games and make use various researches in the field of assistance in Smart Homes and cognitive training, in order to build a tool specifically adapted for the cognitive training of AD patients. In that sense, the paper presents an in-depth description of our Serious Game prototype we developed specifically for the cognitive training of AD patients. We then explain our choice of using a Serious Game for that matter and describe the reasons behind design choices of the game settings and mechanics. We also detail our different systems such as our Dynamic Difficulty Adjustment (DDA) algorithm, the player’s profile adaptation for adapted assistance during challenges, and the in-game cognitive evaluation. Finally, we expose our experimental protocol and our ongoing efforts to conduct experimentations.
The objective of the second paper (Bouchard, Imbeault et al. 2012), presented in chapter 4, was to create guidelines for the development of Serious Games specifically adapted for the cognitive assistance of AD patients. The idea behind those guidelines was to create a source of information easily usable by the research community to either create similar tools targeted at AD patients, or to create tools targeting other types of cognitively-impaired patients by adapting those guidelines. The paper was organized to provide information in four different aspects: (i) choosing right in-game challenges, (ii) designing appropriate interaction mechanisms for cognitively impaired people, (iii) implementing artificial intelligence for providing adequate assistive prompting and dynamic difficulty adjustment, (iv) producing effective visual and auditory assets to maximize cognitive training.
In a larger perspective, we think our work will help further research on the development and usage of Serious Games in the field of cognitive training. We realize that this kind of game has proven efficiency in many fields and could have high potential in the re-education of mentally impaired patients, such as patients suffering from AD. Our research is part of a wider project conducted by the LIARA laboratory of the University of Quebec at Chicoutimi, that specialize in Smart Homes, activity recognition and assistance for AD patients. We hope that this research will generate enthusiasm in the scientific community for the usage of games for cognitive training.
Finally, I would love to conclude with a personal note on my experience. The journey from my initiation to the complex and fascinating world of scientific research to writing down those last lines has been excessively rich and rewarding. Even with modest contributions to the scientific community, the acceptation of our presented papers by peers is a token of acknowledgement for our hard work, which is of great value. Through this adventure, I have learned a lot and had the chance to work with a team of highly talented people working in different fields but driven by the same goal: to help patients suffering from AD. I am grateful for the opportunities I have been given and the generosity of the team working at the LIARA. I hope our contributions will help further research on the development of efficient tools that can be used to help AD patients.

Table des matières

ACKNOWLEDGEMENTS 
TABLE OF CONTENT
LIST OF FIGURES 
CHAPTER 1 VIDEO GAMES FOR COGNITIVE TRAINING OF MENTALLY-IMPARED PATIENTS 
1.1 INTRODUCTION
1.2 THE PSYCHOLOGY OF FUN IN VIDEO GAMES
1.3 “SERIOUSLY” FUN GAMES
1.4 THE AGING OF WESTERN POPULATION
1.5 SERIOUS GAMES IN HELP OF AD PATIENTS
CHAPTER 2 SERIOUS GAMES IN COGNITIVE TRAINING FOR ALZHEIMER’S PATIENTS 
I. INTRODUCTION
II.RELATED WORKS
III.DESIGNING A GAME SUITED FOR AD PATIENT
A. Game design: Choosing the right challenges
B. Game software architecture
1)Patient’s profile adaptations.
2)Activity recognition for dynamic assistance and cognitive abilities estimation
3)Dynamic difficulty adjustment (DDA)
IV.IMPLEMENTATION OF A PROTOTYPE
A. Developing the game
B. Gameplay
V. UPCOMING EXPERIMENTATIONS
A. Phase 1: Experimenting with trial data sets
B. Phase 2: Experimenting with alzheimer’s patients
C. Analysing the experiments results
VI.CONCLUSION
CHAPTER 3 TOWARDS GUIDELINES TO SERIOUS GAMES FOR COGNITIVE ASSISTANCE 
3.1 INTRODUCTION
3.2 CHOOSING RIGHT IN-GAME CHALLENGES
3.3 DESIGNING USER INTERFACE AND INTERACTION MECHANISMS
3.4 AI FOR ASSISTIVE PROMPTING AND DYNAMIC DIFFICULTY ADJUSTMENT
3.5 PRODUCING EFFECTIVE VISUAL AND AUDITORY ASSETS TO MAXIMIZE COGNITIVE TRAINING
3.6 GUIDELINES FOR BUILDING SERIOUS GAMES AIMED AT COGNITIVE TRAINING OF AD PATIENTS
CHAPTER 4 DEVELOPING SERIOUS GAMES SPECIFICALLY ADAPTED TO PEOPLE SUFFERING FROM ALZHEIMER
1 Introduction
2 Choosing the right in-game challenges for the patient
2.1 Guidelines
2.2 Our choice for the serious game environment
3 Designing user interfaces and interaction mechanisms
3.1 Understanding the success of Wii-like games
3.2 Guidelines
4 Producing visual and auditory assets for cognitive training
4.1 Understanding guidance in smart homes
4.2 Guidelines
5.1 Preliminaries on “Dynamic Difficulty Adjustment” (DDA)
5.2 Guidelines
6 Prototype, implementation and upcoming experiments
Conclusion
Acknowledgment
References
CHAPTER 5 CONCLUSION
REFERENCES

Télécharger le rapport complet