The concept of healthy aging, the focus of the World Health Organization’s work on aging between 2015 and 2030, has sparked a growing number of cross-disciplinary research and creation initiatives to take action across multiple sectors to develop strategies for enhancing the autonomy and the well-being of older populations. Among the solutions are the assistive and rehabilitation technologies for the aging population promoted under digital health strategies for remote screening, preventive interventions, or dyadic or machine-coached rehabilitation [1]. Yet, the acceptability and accessibility of these technologies remains a challenge that needs to be addressed, with attention to the complexity of social, economic, and individual contexts of the living experience of older adults [2].

In this position paper, we focus on the gamification of digital training and rehab activities (known as serious games [SG]), which aims to deliver cognitive, emotional, and physical enhancement or rehabilitation routines for older adults [3,4]. For any behavioral intervention to become effective, it must first be acceptable and adaptable to fit the daily routines of the individuals. However, for many older adults, especially those with lower technological literacy or reduced cognitive and physical abilities, the barrier of technological access may be too stressful to overcome. We introduce Affective Game Planning for Health Applications (AGPHA) as an experimental framework based on the appraisal theory of stress and coping to inform SG design and evaluation processes.

Affective game planning builds on the premise that gamification offers a simulated challenge and a corresponding reward system that will motivate users to engage in enjoyable and meaningful health-related activities via information and communication technologies (ICTs). Here, we (1) propose a unifying framework based on a neurobehavioral model of stress and adaptation to address the need for a generalizable empirical approach that accounts for physical and psychological differences in appraisal and adoption of SGs, (2) explain how this model ties to existing theoretical and practical approaches to SG design (namely self-determination and flow theories), with particular emphasis on the necessity of designing enjoyable computerized health applications, and (3) propose a data collection and documentation approach that will enable the field to accelerate research and development of effective and adaptive SGs across multiple applications.

The motivation for this proposal comes from the field of behavioral neuroimaging and methods that it uses to evaluate and document the emotional and cognitive correlates of physiological and environmental phenomena. An influential theory that could explain the psychobiological variations in adaptation and learning, is the appraisal theory of stress by Folkman and Lazarus [5]. In this paper, we propose that the appraisal theory can incorporate 2 dominant game-design theories, self-determination [6] and flow [7,8] and propose a user-centered approach that emphasizes ludic and meaningful experiences in the mechanics-dynamics-aesthetics (MDA) design cycle [9]. We aim to address 2 of the major shortcomings in the field of SG studies: accounting for the complexity of a user’s game-playing experience and preferences [10]; and creating a standard empirical framework to assess the accessibility and efficacy of games that can benefit seniors’ health [11,12].

SG is not a new concept, and promoting the use of digital technologies for improving the quality of cognitive and emotional wellness of older adults has a long history. In their 1976 report of computer use for elder program, Jaycox and Hicks remarked the potential of game playing as a use case for building an intergenerational bridge to utilize the new informatics technology for generating easily sharable hypertexts that older generations can use to share experience and knowledge with younger generations [13]. In 1983, Weisman reported the applicability of computer playing in the daily routines of frail elderly (including those with Parkinson disease, dementia, and visual impairment and an average age of 85 years) and showed that 50 of the house residents agreed to repeat play slower adapted versions of 4 games with different skill requirements such as hand-eye coordination (Brick Out and Country Drive), audiovisual processing and reaction time (Ribbet), and memory (Hangman) [14]. They remarked the potential for games to increase attention and interest in the players on the one hand and to assist as diagnostic instruments on the other [14]. In 1984, McGuire [15] reported a study to suggest that computer-gaming broke the sedentary routine of residents of nursing homes after he measured elevated moods in 16 adults (compared with 12 nonplaying controls) 8 weeks after they were given access to Atari 2600 games (Bowling, Football, Breakout, Pac-Man, and Space Invaders). In 1986, Hollander and Plummer [16] introduced 10 games with various cognitive enhancement components (reaction, eye-hand coordination, and memory) in a community house of adults with an average age of 84 years and showed that, in principle, older adults were receptive to the idea of playing games for improving perceptual-motor skills, cognitive abilities, and attention spans from playing various games. The 17 participants who completed 3 weeks of training considered Trivia game (questions from popular topics in history, literature, and sports) and a computer-based Hangman (a multiplayer computer game to make words with given characters) to be the most interesting of all because they challenged them to think [16]. In 1987, Riddick et al brought 2 upright game arcades to a senior citizens retirement home and introduced them to 2 nonviolent and nonspatial games, Pac-Man and Donkey Kong, which were slow paced but allowed for progressive skill development—measurable after 19 sessions of playing. They also quantified a significant increase in arousal and recorded the accounts of players explaining why they found playing to be an uplifting experience—despite the fact that the game-play duration and pleasure declined [17].

These historical citations foreground a growing body of similar contemporary work trying to acquire more refined empirical data to support the benefits of more refined digital playing for mental and physical health in older adults.

Since a decade, quantitative evaluation of game benefits has returned to the forefront. Basak et al trained 20 seniors (mean age of 69 years) on the personal computer game Rise of Nations for a total of 23 hours over the course of 8 weeks and showed improvements in cognitive tests (eg, n-back or mental rotation) compared with the untrained controls [18], even showing evidence for game-related changes in neuroplasticity [19]. Peretz et al performed a randomized trial in 155 seniors and compared a personalized cognitive training protocol versus various computer games assumed to involve cognitive functions (eg, Tetris, Target practicing games, and memory games) and showed game-related improvements (albeit weaker than personalized training) in overall cognitive scores [20]. In 2013, an influential publication in the Nature, a study of 64 participants (age 65 years, SD 5) provided evidence that training with a multitasking custom game (Neuroracer) significantly improved the cognitive control, with effects remaining up to 6 months after the training, concomitant with an increase in the power of theta-band electroencephalogram signals in the medial prefrontal region, an area important for memory and executive control [21]. Toril et al’s meta-analysis of all the literature between 1986 and 2013 revealed that video game training has a positive impact on cognition, albeit the effect size was moderated by factors such as age, personality, and experimental design [22]. However, in a recent (2017) randomized controlled study of Lumosity (nonaction cognitive training) versus The Sims (active control condition), they found that the control group (Sims) became less distracted and faster than the experimental group in performing cognitive tasks [23]. In an independent earlier study (2015), Rose et al demonstrated that simulation of 1-week life planning in the form of a game produced measurable improvements in real-life planning [24]. Overall, it seems that some specific cognitive benefits may be gained from specific elements designed in digital cognition-targeting games [25].

Beside cognitive domains, exercise games (exergames) have been proposed and empirically evaluated for their cognitive benefits, for instance, in older adults with neurological conditions [26], or physical benefits—in terms of improving balance and fall prevention [27-31]. Computer games have also been proposed for cardiac [32,33] and stroke rehabilitation [34]—subject of 3 Cochrane reviews, concluding on the insufficiency of statistical significance because of sampling heterogeneity and inconsistent reporting [35]. Beside quantifiable measures (cognition and physical fitness), which are still debatable [11,36], exercise games provide enjoyment and social interaction [37-44].

A recurring theme in studies that target older adults is that enjoyment trumps many challenges that would keep them from playing. A text analysis of the content of the literature concerning the older adult’s game experience revealed that many items were related to the potential health benefits of playing digital games, with the top 10 of most recurring terms including training, balance, physical, cognitive, exercise, and social [9]. Despite the general assumptions about technological gaps, older adults, at least in Canada, are onboard for digital play [3]. A recent survey of over 880 older adults in Canada indicates that 73% of the respondents enjoy playing games because they provide social and cognitive stimulation [45]. In a massive online multiplayer game study (World of Warcraft), Zhang and Kaufman [46] demonstrated that interindividual variations in enjoyment of the social game experience remain an important determinant of benefiting from the game play experience. Ruvio et al have shown that the fun factor in video exercise games predicts adherence to exercise routine compared with the no-game condition [32]. Of course, enjoyment is not the only factor in determining the meaningfulness of a media-related experience [47,48], but in the context of older adults playing, it seems to be important [49]. Many argue that to provide pleasurable and enjoyable experience is the sufficient reason why computer gaming is good for older adults [3,48-51].

In the Gerontoludic Manifesto (2015), De Schutter and Vanden Abeele have argued that to advance the field of gaming for older adults, we must move away from the biomedical model of solving the problem of age-related decline but instead focus on how different types of games provide a positive experience for their players [52]. Subsequently, in Gerontodulic Design (2017), it was proposed that in designing games for older players, the initial step must account for game aesthetics (ie, the emotional response to the games) which reflect the personal essence of gameplay experience and that designers must start by creating enjoyable and playful interfaces that provide connectedness and meaning through implementation of the game mechanics (ie, rules, challenges, and rewards). In this framework, game planning involves gathering iterative player-centered empirical data from the game dynamics (ie, the phenomenology of experiencing the gameplay), as its mechanics evolve through aesthetic improvements [9].

A purely biomedical approach to the problem of design is too reductionist to account for the richness of diverse factors that shape the experience of an SG player, yet designers employ various quantitative approaches and rely on aggregating data through meta-analyses to create a formulaic recipe for enjoyable games that have a higher likelihood of uptake and retention by the players [9,50,53-60].

At the empirical level, beneficial game studies typically rely on well-established psychological motivational theories to understand who plays what, why, and how. In game studies, self-determination theory (SDT) of motivation [6] and flow theory [7,8] are dominant.

SDT posits that humans have an inherent need for autonomy and for developing competencies and relationships that will help them fulfill their goals for growth and self-actualization [6]. This theory is one of the most commonly applied in the field of game studies to explain relations between gaming and psychological need satisfaction [61-63]. The theory is also used as a framework to design gamified strategies for promotion of wellness [64]. Age-related variations in intrinsic and extrinsic motivations that underline SDT are important. In a meta-analysis of 11 independent studies, Birk et al aggregated 3041 samples of postgame player experience of need satisfaction (PENS) and intrinsic motivation inventory data and reported complex and age-dependent correlations between competence, relatedness, and intrinsic motivation [60]. Loos applied the SDT in examining the applicability of exercise games in older adults and found that although the exercise game satisfied the needs for autonomy and competence, they would not recommend the game to others; among the objections were the request for improved user interface or environmental background [65]. In fact, the issue of user interface and cognitive conditions in older adults is pertinent, especially because natural or pathological age-related decline in sensory, motor, and cognitive domains are expected to influence competence and need satisfaction [56,66].

The issues of interface and game accessibility are often addressed with the flow theory [7,8], according to which successful games must counterbalance their features along the axes of challenge and difficulty: if the level of challenge is too high and the player abilities are not commensurate, it leads to anxiety and discontinuation; if the level of challenge is lower than the abilities of the player, it leads to boredom and discontinuation. Nacke et al [67] examined the concept of flow in 2 groups of young and older healthy adults engaging in a pen-and-paper game versus a Nintendo puzzle game and tested the hypothesis that higher challenge associated with the Nintendo task would increase the flow and enjoyment of accomplishing the task. Using this method, they reported that although flow for older adults was correlated with challenge, positive affect, and arousal, for younger players, it was only correlated with challenge and that no affective correlations occurred in the younger sample. Belchior et al used the flow concept to compare the engagement of 45 older adults while training on a laboratory cognitive training stimulus versus training on commercial video games and reported significant increase in flow scores after 6 weeks of playing Tetris, as opposed to initially high but diminishing flow after 6 weeks of training on the laboratory game [68]. Marston et al showed different levels of flow experience in older adults trying different exergames, depending on which country the participants were recruited from [69].

Although SDT underlines the reflective component of playing motivation (competence, relatedness, and autonomy), the flow theory underlines the reflexive aspect (arousal, pleasure, frustration, and success). Our proposed theoretical framework unites these 2 by drawing on Selye’s General Adaptation Syndrome (GAS) [70] and Lazarus’ theory of coping, which incorporate motivational, relational, and cognitive components that give rise to individual differences in perception of, and coping with, stress, which is a physiological phenomenon linked to individual differences in biological factors such as metabolism, immune system, and adaptive learning (for an ontology by Lazarus, see [71]). This theoretical framework is supported by a wealth of evidence that enables us to investigate the interaction between biological and psychological factors that are known to impact not only the momentary experience of the game but also its learning and its long-term impact on broad range of health factors.

To provide empirical evidence for potential cognitive or health-related benefits appears to be a source of motivation for older adults to adopt computer games in their lives [52,92-95]. Empirical data are also necessary to make recommendations about which serious game to play, and how often [96]. Empirical methods to evaluate game benefits are often focused on evaluated on the expected outcome. In a 2012 review of SGs, Wiemeyer and Kliem [12] have listed a range of physiological functions that can be targeted with SGs, such as cardiovascular and cardiorespiratory system (eg, endurance, cardiovascular fitness, and prevention of cardiovascular diseases), energy metabolism (eg, weight control and prevention of obesity and diabetes mellitus), strength and flexibility (eg, posture and range of motion), bone structure (eg, prevention of osteoporosis), immune system (eg, prevention of cancer), and sensory –motor coordination (eg, reaction, balance, and fall prevention). Regardless of the outcome, the game interface mediates the perceptual experiences of activities that target a biological outcome; therefore, gaming is a primarily neuropsychological phenomenon. In other words, the game interface sits between the user’s perception of and interactions with the game and its transferable health outcomes. Although many scientists acknowledge the potential for health SGs, they often neglect the primacy of emotional variables (such as stress, flow or aesthetic experience) in game-related studies of older adults. These issues are often tackled by human-computer-interface (HCI) designers, who emphasize consideration of user’s cognitive and physical abilities that determine their experience and help them overcome challenges related to motivation and preferences [10,97], cultural contexts [53,98], affordance, control and self-determination [54,55,99-101], and accessibility [56,66]. We briefly propose two models that can be tested using a mixed methods approach.

Consider the model in Figure 2 which illustrates a simplified model of the interrelations between the most basic elements of adaptation. Organisms, from the most basic to the most complex, evolve through an intricate chain of interactions that are tied to basic metabolic regulation of the internal milieu while surviving through a highly variable external milieu (the very bottom of Maslow’s hierarchy of needs) [102]. Brain development involves a chain of interactions with the external world. They begin with sensory processing of the outside information that enters the body through the primary senses: vision, hearing and touch; and movement. Senses and movement further evolve as we grow (or decline as we age) by learning skills that support this process [103]. The genetic and neurochemical signaling pathways that support this learning process are tied to reward and fear processing, and the brain learns the world through an iterative evaluation of the situation and stimuli that are rewarding and enhance its well-being and avoiding or fighting those that threaten it [104-107]. These mechanisms create a closed-loop control system, any of which if broken will put the body in the state of allostatic load (or stress) [108,109]. The physiological manifestation of a stress response reflects the end product of myriad adaptive changes resulting from the interactions between the brain and other organs aimed at restoring the allostasis (ie, the ability to maintain stability through change). Neuroscientists argue that higher human functions, such as communication, creativity, empathy, and playfulness, are all manifestations of the humans extending this exploratory and interconnected terrain—sensing, moving, being rewarded, and learning [110,111]. Whether a game-playing experience produces a sum total of rewarding experiences or not will impact all elements of the system. Individual differences in movement, sensory processing, skills, and learning abilities will determine to what extent an activity is rewarding or stressful as well. We argue that in designing any SG for assistive health care, this big biological picture must be accounted for creating optimal flow experience. Considering that GAS affects the metabolic substrates of behavioral adaptation, the theory of appraisal fits the big picture by bridging between behavioral and biological factors that interact during a game-playing experience. For example, one might ask whether appraisal of the game reward will modify behavioral factors such as speed of execution of the game, and metabolic factors such as heart rate, and whether that will transfer to change in higher functions such as hedonic experiences, learning and memory.

Now consider another experimental model in Figure 2, a testable model of how game appraisal will predict later game-playing experience and adoption. The primary point of encounter between a user and a conceptual game is appraisal of the value of the exercise. If the game is solely presented as a source of entertainment, then it will have no stressful component, and the choice to play or not will depend on the player’s personal and cultural preferences for pastime. However, if the game is presented in the context of mental or physical health or as a preventive lifestyle strategy, then it implies benefits. At the first encounter with this game, the user will either believe the promise of benefits or reject it as useless or impossible to do. If it is rejected, then they will not further engage in the process. However, if they do subscribe to the beneficial narrative, then the ability to play or not play will become attached to the notion of one’s self, and from there on, the interaction with the game will depend on the individual’s appraisal and coping processes with the physical and mental demands of learning and executing the game successfully. In this case, 2 scenarios can unfold that are described below.

AGPHA offers a scientific framework aimed at the development and evaluation of SGs on a wide range of effectiveness measures. A schematic diagram of the AGPHA is presented in Figure 3. AGPHA is a recursive mixed-methods evidence-based and user-centered process consisting of 4 elements: (1) defining the health-related problem that an SG is to address, (2) identifying or designing a ludic intervention (whether curated from existing games or new), (3) a standardized procedure for data collection during game evaluation or design, and (4) an archiving system that would allow tracking the evolution of the SG design and application over time.

The first step in the process is for scientists to identify the health domain that can be targeted using an interactive ICT, for example, telerehabilitation for stroke recovery, cognitive enhancement, emotional intervention, education, physical fitness, monitoring and data collection, diet and lifestyle, and social networking. As health-related interventions are inherently stressful, the primary aim is to identify or design game-like interfaces that motivate users to learn and sustain playing by minimizing the stressfulness of the activity and maximizing the enjoyment experienced from engaging in the playing.

For this reason, the SGs for health can benefit from a Gerontoludic extension to MDA, which consists of recursive evaluation of the following components:

Figure 3 illustrates the scope of data that need to be collected in the AGPHA framework. AGPHA is an iterative mixed-methods approach to user-centered, evidence-based game design. It relies on the evaluation of the reflective (interests) and reflexive (experience) response to games that are developed through the MDA cycle. AGPHA relies on iterative testing and continuous documentation of the process of design and SG evaluation, with a common data thread to enable future aggregate studies.