Assessing HapHop-Physio: An Exer-Learning Game to Support Therapies for Children with Specific Learning Disorders

Abstract

The HapHop-Physio exer-learning game is a cognitive rehabilitation solution for children with learning disorders to train memory and attention functions. Therefore, HapHop-Physio must be assessed as an appropriate tool developed to support the enhancement of essential cognitive skills. In this paper, we aim to establish the validity of HapHop-Physio in a healthcare setting with children receiving training with this tool. HapHop-Physio was developed through interaction between clinical experts, one graphic designer, and game developers. Following an assessment framework for serious games in healthcare, the game’s rationale, functionality, validity, and data safety are described. Particularly, the validity was assessed with experts through a case study conducted in three phases with 12 children diagnosed with Specific Learning Disorders. The baseline of the cognitive profile of the children was obtained and trained with HapHop-Physio. General trends and findings were obtained through an exploratory analysis of the gathered results from the phases. The validity aspects were achieved through continuous feedback from the experts, allowing to improve the game in five features: game structure, scoring system, avatars, login system, and clinicians’ dashboard. The application of the assessment framework in HapHop-Physio guarantees end users that the game is safe and effective enough to be used for supporting memory and attention training. One relevant finding from the case study was that cognitive performance improved in 11 out of 12 children at the end of the training. The positive outcomes of this assessment indicate the game’s appropriateness for a healthcare setting. However, the improvement in cognitive performance cannot be associated with HapHop-Physio in this case study. Consequently, it needs to be evaluated in a controlled experiment (including a control group) to ensure that it leads to cognitive rehabilitation in reality.

1. Introduction

Cognitive skills such as memory and attention are essential for carrying out independent daily activities. Cognitive disorders usually appear in childhood, affecting school performance and social interaction. These disorders affect up to 10% of children worldwide [1]. Particularly, Specific Learning Disorders (SLD) embrace conditions in which learning and cognitive skills are significantly lower than expected according to a child’s age [2].

The treatment of disorders associated with cognitive and behavioral development is focused on training or enhancing these necessary cognitive skills for learning [3]. Conventionally, most therapies use training activities in which efficacy is evaluated using standardized batteries (scales). The training usually involves helping the children to learn skills and making a learning plan based on their strengths and the cognitive domains that need intervention [4]. In addition, the training targets a specific cognitive ability through repetitive practice on graded activities to improve the cognitive function [5]. These therapies are conducted by a specialist guiding the child through pencil and paper activities [6,7]. This circumstance is close to what children experience in school or at home, which could generate fatigue and lack of motivation, affecting the progress and results of the therapeutic session and even causing non-adherence to the treatment [8].

In recent years, the use of video games, and specifically serious games, has become an ideal format for training specific cognitive abilities for children with SLD. In [9], the authors found that using adaptive difficulty levels in serious games enables customized learning environments for children with learning difficulties. Depending on the user’s performance, this helps in maintaining optimal levels of engagement and cognitive improvement. Treatments using this approach are mainly used to train dyslexia [10,11]. However, other elements can improve the serious game approach.

Physical exercise significantly improves memory and attentional processes in clinical and healthy populations [12]. Technology has combined physical activity with cognitive tasks through ‘exergames’. Exergames are interactive video games that stimulate the players to make physical movements aimed at completing actions within the game [13]. In this way, exergames enable the gamification of conventional therapies to make them more attractive, motivating, and engaging for children [14].

The HapHop-Physio system is an exergame developed to support memory and attention therapies based on the Hopscotch product [15]. The game trains the auditory and visual components of memory and attention, along with the training of reading and writing processes in five difficulty levels. The system has an electronic mat as the input device for promoting physical activity in players. Theoretically, children with SLD could improve their learning through movement by maintaining the level of attention needed to perform cognitive activities [16,17].

Attention is an essential requirement for adequate cognitive functioning. Children with attention problems also have learning difficulties with a reduced ability to store information. A good level of attention requires the inhibition of irrelevant information and a focus on relevant information for prolonged periods [3].

Regarding SLDs such as dyslexia, there are other cognitive functions related to learning to read, such as the ability to decode visual stimuli, naming speed, vocabulary size, working memory capacity, and the ability to maintain attention and concentration. Grapheme–phoneme–grapheme associations represent a central aspect in learning to read and write. In the HapHop-Physio mini-games, these aspects are approached from memory [3].

In the case of dysgraphia, memory allows to maintain the topic of the text. Consequently, memory plays an important role in increasing writing proficiency. Meanwhile, visuospatial skills are related to an adequate spatial organization that facilitates the subsequent reading of the text. Likewise, the proper management of attentional resources is a skill that facilitates learning to read and write [3].

Correcting errors seen in children with dyscalculia involve abilities such as spatial organization, visual attention, motor skills, judgment, reasoning, and memory. Working memory associated with sustained attention plays a central role in performing arithmetic operations [3]. All these cognitive aspects related here to SLD were taken into account in the functionality of HapHop-Physio.

The literature contains multiple studies relating the use of exergames for different motor and/or cognitive rehabilitation therapies. These studies are mainly focused on the elderly or adults with different mental disorders [18,19]. Patients and healthcare professionals generally view game-based interventions as fun and challenging [20].

Few games target children with cognitive disabilities. Some of these games captured physical data from wearables or sensors such as Kinect [21,22,23], offering playful experiences to reinforce attention, memory, and visual-spatial skills [24,25]. Nevertheless, these games do not report a comprehensive assessment of the technology to determine validity in terms of applying the solution to the target population in a healthcare environment.

In this sense, HapHop-Physio is a game specially designed and developed for targeting an underdiagnosed population: children with SLD. It aims to train basic cognitive skills and promotes learning through physical activity, hence its final characterization as an exer-learning game.

The HapHop-Physio project was a technological transfer initiative developed at the Universidad del Cauca in collaboration with the Fraunhofer Institute for Digital Media Technology in Germany. The ENI (Evaluación Neuropsicológica Infantil), a test battery developed in Spanish for Latin American children with similar language backgrounds [26], was used as a reference for the development and system scoring of HapHop-Physio. The ENI evaluates a wide spectrum of functions included in neuro-psychological domains, distributed in three sections. The cognitive functions section consists of 58 grouped tasks, including memory and attention. Three tasks and fourteen measures are included to assess executive functions. The last section evaluates the soft neurological signs. This clinical instrument is flexible and in accordance with the developmental stage in which the child is. It is standardized for different age groups. A previous version of the game was graded as usable and functional through testing and evaluation of two initial prototypes [27].

In this paper, an assessment framework was applied to evaluate whether the game is appropriate and secure as a support tool for the therapies. For HapHop-Physio to become a “qualified game”, it is necessary to disclose comprehensive information on its components and the claims for its purpose and final side effects. This framework provides consistent, transparent, and reliable assessment to eventually integrate HapHop-Physio into a real-life healthcare scenario. Indeed, it allowed us to improve the status of the current version of the game, which is now applicable for the next experimentation stage.

The remaining of the paper is distributed as follows: in the Methods section, the framework is described in terms of its items, categories, and themes along with the evaluation strategy for each item. In the Results section, each theme is addressed. In particular, the results of a case study for the validity category are presented. The Discussion session indicates the implications of the applied framework. Finally, conclusions and future work are stated.

2. Materials and Methods

A consensus-based framework for the assessment of serious healthcare games published by Graafland et al. [28] was applied to evaluate HapHop-Physio and identify those aspects that did not meet the items within the framework, so as to improve them. This section explains the assessment framework first, stating the most relevant aspects for the evaluation of HapHop-Physio and how they were measured. Then, it introduces the case study that allowed to measure the validity of the game.

2.1. Assessment Framework

This framework was developed by the Dutch Society for Simulation in Healthcare (DSSH) to assess the technologies (not devices) created for healthcare to check on the safety and validity of their use. It cannot validate the effectiveness or engagement of serious games.

The framework describes 62 items gathered in 5 main themes. Figure 1 displays the information about themes and categories. Due to the nature of HapHop-Physio (exer-learning game), not all items were evaluated. The first theme is “Game Description”, which has 13 items that include descriptions regarding the information of the intellectual and physical property of the game. There were 12 evaluated items in HapHop-Physio: operating system, version, web link, project type, access, adjunct devices, funding, advertisement policy, sources of income, affiliations, conflicts of interest, and disclosure.

The second theme is “Rationale”. For serious games (with a different purpose than entertainment), the goal or purpose of the game can be seen internally (e.g., completing the levels of the game) as well as externally (e.g., carrying out a therapeutic process). Thus, defining these purposes makes possible a better definition of who the end users of the game are, in addition to the technical requirements that must be taken into account for its correct operation. Eight out of the twelve items were evaluated for HapHop-Physio: goal or purpose, disclosure (of purpose), medical device, specific user groups, description (of user group), limits, disclosure (of user group), and patient care.

The “Functionality” theme helps to understand the process of interaction between the game and the player. For this, information about the content of the game, the way of communicating the goals within the game, and the evaluation of the performance are required. Above all, understanding the effect of the game on players’ learning is relevant for designers and developers. They should create gaming experiences where players develop knowledge, interests, and skills. In a game, a novice eventually becomes an expert by understanding and mastering the dynamics of the game [29]. The necessary items for this task were: learning or behavioral goals, relation learning and game-play, assessment of the progress in the game, assessment parameters, content management system, user uploaded content, content monitoring, restrictions and limits, potentially undesirable effects, disclosure (of effects), and measures taken.

The “Validity” theme determines, based on standards, whether a game such as HapHop-Physio adequately resembles the construct it intends to measure. According to the authors of the framework, the research for validity in these scenarios has consequent phases. Thus, the assessment of the items from the validity category was included up to the construct validity. The included items were: medical expert complicity, user group complicity, user testing, platform stability, face validity, content validity, and construct validity. In this paper, we mainly focused on evaluating the validity category with the help of experts and end users. Face validity refers to the appropriate representation of the game constructs. In particular, for HapHop-Physio the selected constructs were the scoring system, login system, and clinician’s dashboard. Content validity discusses whether the game content is legitimate for the experts. The assessed HapHop-Physio contents were the game structure and the set of avatars. Finally, the construct validity weighs the measurement capacity of the game. For HapHop-Physio, this was assessed through the case study detailed in Section 2.2.

Finally, in the “Data Protection” theme, user privacy must be a priority for developers of eHealth applications. The storage of data that are processed in a game such as HapHop-Physio must comply with the data protection laws of the region in which the game is located. The items assessed for the game were: data processing, patient privacy, data ownership, data storage period, data removal, data storage security, data transmission security, and disclosure.

The described framework was applied iteratively to the HapHop-Physio game to achieve an acceptable degree of completeness in most of the applicable items, in particular for the validity category. Several feedback sessions were held between neuro-specialists, graphic designers, web developers, and gaming experts to evaluate the proposed changes and missing aspects of game performance. The iterative assessment process stopped when it was possible to advance until the construct validity item (i.e., once the face and content validities were reached).

2.2. Case Study

One of the major activities to accomplish the assessment proposed by the framework (in the construct validity item) was to observe the interaction between the HapHop-Physio game and a small population sample. This activity helped in finding general trends from the collected data in the cognitive domains of auditory-verbal memory, visual memory, visual attention, and auditory attention.

A case study was proposed and conducted from March to June 2018 (12 weeks), in which 12 children participated. For this study, children were between 5 and 16 years old and diagnosed with SLD. The study took place in a local rehabilitation clinic, where children attended special programs. The ethical principles established in the Belmont Report of 1978, the Helsinki Declaration, and Resolution No. 8430 of 1993 from the Ministry of Health and Social Protection in Colombia were considered for the procedures. The autonomy and integrity of patients who decided to participate voluntarily with prior informed consent, approved by the Ethics Committee from the Universidad del Cauca, in session 6—1.38/6 from 29 April 2015, were respected, and the confidentiality and privacy of the information collected for the study were guaranteed.

This case study was approached in three phases. The first phase was the application of the ENI Battery to obtain the initial cognitive profile of the children in terms of attention and memory levels [30]. In the second phase, children recreated the normal rehabilitation process with the HapHop-Physio game. Finally, in the third phase, the cognitive profile of the children was re-tested. The progress and/or decrease in the cognitive difficulties were checked when comparing them with the initial test. For classifying data from the first and third phases of this study,

Table 1. Range and rankings of percentile values for defining the children’s profiles.

Percentile Values	Classification
>84	Superior
84	High average
26–75	Normal or average
16	Low average
5–9	Limit
<2	Very low

shows the rankings for the percentile values that helped the clinician in the definition of the cognitive execution profile of the children.

The information obtained in phase II was stored in the game’s database, checked by the clinician, and analyzed through a simple exploratory data analysis to look for insights in the data collected after each session. The analysis was complemented with univariable and multivariable analysis. The considered variables are the classification results, the per-centiles of the results, as well as the gender and the age groups of participants. The final aim of this analysis is to obtain a set of hypotheses that should be tested in a more formal experiment. It is essential to consider the variables of the sample used in the normalization of the results. Variables such as age and gender are highly relevant in the normality criteria of cognitive functions within the characteristics of the patients [31].

3. Results

3.1. Game Description

The HapHop-Physio exer-learning game runs on the Windows operating system and it is currently on version 44. This version still does not have a web version available to download, and access to the game is only from local machines. The game was designed for academic and non-commercial purposes. Access to content and functionalities of HapHop-Physio are restricted only (and for now) to research projects in the Universidad del Cauca. The input device for interacting with the game is an electronic mat with nine positions to indicate different directions of movement; the mat has an additional key for special functions. The game was funded by different stakeholders. The Sistema General de Regalías (SGR) of Colombia supported two projects from the Universidad del Cauca that funded the research for the development of HapHop-Physio: Núcleos de Innovación and InnovAcción Cauca. This game also received financial support from the MinCiencias, Colombia.

The game is totally free of any kind of advertisement. Currently, there are no sources of income because of the academic nature of the game. One of the authors is affiliated with a rehabilitation clinic. However, this does not represent a risk since the clinic signed an intellectual cooperation agreement that forbids the commercial exploitation of the game and the economic revenue that might come from it. For the rest of the authors, their affiliations do not represent a source of bias or threat to the content of HapHop-Physio due to their academic nature. The authors of the paper share the intellectual property of the HapHop-Physio software along with the Universidad del Cauca. The conflicts of interest have been disclosed to the interested parties.

3.2. Rationale

The purpose of the game (external) is to support the practice of memory and attention therapies with auditory and visual components, along with the training of reading and writing processes. The goal of the game (internal) is to help the main character (Kaylo Rana) to get his memories back through giving correct answers to pass each one of the mini-games, which will challenge children’s memory and attention skills. HapHop-Physio has four modules containing mini-games, two modules for attention tasks, and two for memory tasks. Figure 2 displays the main menu of the game where the child has a representation through an avatar selected from a set.

The challenges in the mini-games are matched to therapeutic objectives such as increasing the child’s memory span in relation to what he/she hears or observes, as well as improving sustained and selective visual/auditory attention when receiving instructions through images, words, or sounds. Since the type of game is non-commercial and for academic purposes, the details of the goal of the game (both internal and external) are revealed through the informed consent given to each user. In the same way, the internal goal of the game is explicitly described at the beginning of the game.

The HapHop-Physio exer-learning game is not a medical device and does not replace a therapeutic intervention. Therefore, two items (class and approval by legal bodies) of the medical device category in this theme were not evaluated for the game. The cognitive rehabilitation solution offered by HapHop-Physio is targeted to children aged between 5 and 16 years old. Children diagnosed with SLD, or any diagnosis related to learning processes being affected, can train memory and attention skills with HapHop-Physio. As mentioned before, this game is a support tool for clinicians in the cognitive intervention/rehabilitation of children. HapHop-Physio is geared toward all genders. Children without any disorders diagnosed can also play the game.

Since the input device of the game is a mat, it requires children to move in different directions. Therefore, children with physical disabilities that prevent them from walking or moving around cannot play. In addition, children with severe or deep visual or hearing loss cannot interact with the game. The intended user group description was disclosed to the stakeholders. The HapHop-Physio game is not intended to be used in patient care. Therefore, the items ‘training courses’ and ‘SCORM compliance’ from the setting category within the rationale theme were not considered for the evaluation of the game.

3.3. Functionality

The purpose of the game is not focused on learning specific content. Instead, it is intended for children to practice their cognitive skills of memory and attention in their visual, verbal, and auditory components accompanied by the practice of reading and writing activities. To put these cognitive skills into practice, there are five types of mini-games in HapHop-Physio according to the cognitive challenges: selection, writing, sequencing, reaction, and numbering (new feature added). Each mini-game contains a text introduction that the child must read before starting to play; in the same way, the clinician can give their own instruction. The evaluation of the child’s progress in the game is given by the levels of difficulty that the game presents. Each mini-game has five levels of difficulty. HapHop-Physio has four game evaluation parameters: the score of each mini-game, the child’s performance in each mini-game, the level of each mini-game, and the response time. Depending on the level, there is a standard number of correct answers. Performance is graded by ranking the score obtained into low, medium, and high ranges. This ranking is determined by the ENI battery based on its evaluation instructions according to the level. For example, in level 2, children achieve a maximum of five correct answers in the memory mini-games; for this level, one correct answer means low range, two to three correct answers mean medium range, and four to five correct answers mean high range.

The HapHop-Physio content management system is restricted to the person in charge of the research project designated by the Universidad del Cauca, who can continue to develop the concept around the game as they see fit. Likewise, it is under the supervision and evaluation of the clinician in use of the game. The content of the game can only be modified by the person in charge of the content management system at the exclusive request of the clinician. Content monitoring is performed by the specialist. The lack of automatic programming of the game flow depending on the child’s cognitive profile and its current state of progress is a current limitation of the game that is not yet covered. Currently, the therapist in charge of guiding the session will decide which mini-games are going to be played in the sessions and which ones are restricted, depending on the cognitive profile of the children.

Falls or physical accidents due to movement on the mat represent the potential undesirable effects of the game. These potential side effects are disclosed to stakeholders through informed consent. To avoid this, children are asked to play without shoes or socks on the mat. As long as there is large friction between the foot and the mat, accidental slippage can be avoided.

3.4. Validity

Neuro specialists were involved in the design process from the start of the project. In addition, children were part of this process. The information about the complicity can be found in the publication of the first version of the game. Technology experts and digital health professionals were also part of the multidisciplinary team that developed HapHop-Physio. Nevertheless, the item ‘educationalist complicity’ from the design process category in this theme was not evaluated. User testing took place, and a usability evaluation was performed and reported with results regarding changes in the aspect of the game, such as the palette color. Although HapHop-Physio was developed as a multiplatform game, the drivers for the input device only worked on the Windows operating system.

3.4.1. Face and Content Validity

As described, HapHop-Physio complies with the applicable items of the assessment framework (47 out of 55 items). The fulfillment of the previous themes (and their categories) depended on the continuous understanding (in discussion sessions) of the concepts around the support that the game can provide to clinicians and children. This process is reflected in the improved features of the game that help in ensuring validity by face and content.

For face validity, end users judge the game’s similarity to the represented constructs of pencil and paper therapies guided by clinicians. An example of such constructs is in

Table 2. Mini-game descriptions for face validity.

Mini-Game	Description	Instruction	Rating
Words list from auditory-verbal memory	A list of words is told to the children. They must choose the correct answer among distractors. The answers could be presented as images or written words.	On the screen, the instruction is shown: “Listen carefully the next words”. Regardless of the order, the game presents the correct answers among three distractors each time for every dictated word.	One point is awarded for each correct word/figure. The scores are added to have the total number of correct answers for each level. The memory span is defined by the number of correct answers.
Figure list from visual memory	A list of figures is shown to the children. They must choose the correct answer among distractors. The answers could be presented as images or written words.	On the screen, the instruction is shown: “Observe carefully the next figures”. Regardless of the order, the game presents the correct answers among three distractors each time for every shown word.
Number of objects in visual attention	A figure is presented to the children. They must count elements within this figure that have a particular characteristic among distractors to obtain the correct answer.	On the screen, the instruction is shown: “Observe the figure and count how many elements are”. The game presents a figure(s) with several distractors among the correct elements that need to be counted.	One point is awarded for obtaining the correct number/reacting to the correct element. The scores are added to have the total number of correct answers for each level. The attentional volume as well as the concentration and sequencing ability is defined by the number of correct answers. One point is awarded for every correct reaction.
Reaction to sounds in auditory attention	A particular sound is presented to the children. They must react to the correct answer among distractors.	On the screen, the instruction is shown: “Press the button when you listen a particular word”. The game presents the correct answer among several distractors until completing a defined number for the correct answer.

: the description, the instruction, and the rating of some mini-games are detailed.

On the other hand, for content validity, experts analyze the game’s content to determine how appropriate it is. In

Table 3. Improved features of HapHop-Physio.

Validity	Feature	Previous Versions	Current Version
Face	Scoring system	The game’s scoring system did not rate the mini-games according to the score ranges designated by the ENI battery. The scoring was standardized according to a different criteria due to misunderstandings between experts and developers.	After two assessment and feedback sessions, the scoring system was corrected. Now, it follows the ENI battery guidelines considering the number of questions and correct answers for each mini-game and its levels.
	Login system	There was no login system, thus the game progress information was lost from the HapHop-Physio system at the end of each session.	The login system was developed to save user performance information during multiple sessions. The ultimate goal of saving this information is for the expert to see how the child is progressing.
	Clinician’s dashboard	Experts guiding the therapies did not have a tool to retrieve or visualize the patients’ information.	The clinician’s dashboard was developed to display summarized information about each patient and their progress on each game session.
Content	Game structure	The structure of the game presented three types of mini-games according to the type of skill they trained: selection, writing, and sequencing. Each module of the game presented different types.	An improvement of the game structure allowed to add two more mini-game types: reaction and numbering.
	Set of avatars	Children could only play as the main character, Kaylo Rana. This could reduce their engagement to the game.	A set of avatars representing the user in different ways (6 characters, 6 animals, and 6 customs divided to represent male or female gender) was created. The design of these characters is consistent with the natural environment of the game and its visual aesthetics.

, details about improvement of the considered features are shown.

From these results, face and content validity were fulfilled within the framework. In Figure 3, the game structure by modules and the added types is displayed. Therefore, construct validity is the following item that HapHop-Physio met as an exer-learning game.

3.4.2. Construct Validity

Following the application of the assessment framework for the construct validity, a case study was conducted to measure differences in the skills it intends to train. It is important to highlight that due to the non-experimental nature of this study, the findings detailed here are suggestive and not conclusive. The identified trends and insights from the case study only intend to fulfill the holistic evaluation of HapHop-Physio as exer-learning game (eHealth application).

Twelve children voluntarily participated in the study (four female participants). The age range was from 6 to 15 years old (mean: 10.93 ± 3). All patients completed a median of 69% of the proposed therapies (minimum completion of 50%). The initial memory volume of the children was established (memory span) with the ENI battery. Patients had an initial median memory span of four words (auditory-verbal memory) and four figures (visual memory) in phase I. For phase III (after interaction with HapHop-Physio), the median memory span improved to five words and five figures.

Figure 4 shows the distribution of the children’s cognitive profiles before (green) and after (yellow) the training. The cognitive profiles were classified according to each children’s evaluation results using the measurement in Table 1. For example, in the auditory-verbal memory, the results of phase I showed that two children presented a very low performance. In phase III, only one child was found to present a limited performance. This shows an upgrade in the ranking classification. A similar behavior was detected in other cognitive domains: visual memory, visual attention, and auditory attention.

To complement the previous information, Figure 5 presents the evolution of the minimum and maximum percentile values for the children’s cognitive classification using a boxplot. Specifically, it compares the distribution of the classifications reached in phases I (pre-training) and III (post-training) of this case study. For example, the post-training minimum values of the auditory-verbal memory and visual attention subdomains are greater than their corresponding pre-training maximum values.

Similarly, a multivariate analysis of the results of the cognitive profile percentile values (both pre-training and post-training) was performed in order to obtain trends in the gender and age group variables. Making a comparison by gender, Figure 6 shows that in the pre-training evaluation the male results were below those of females. On the other hand, all males improved in the post-training period. Although there were also better results in females, the group became sparser and maintained some minimums, especially in the auditory-verbal memory subdomain.

Age groups were determined considering the children’s development stages defined by the Centers for Disease Control and Prevention (CDC) [32]. The trends between groups are clearly distinguishable (Figure 7). The age group between 9 and 11 years old (five children in middle childhood) has a clear disadvantage compared to the other two groups: 6 to 8 years old (three children, also called middle childhood) and 12 to 15 years old (four young teens). In general, the scores improved from one evaluation to the next. However, new hypotheses need to be defined to search for the factor generating this trend in the most populated age group in this study.

3.5. Data Protection

The HapHop-Physio game collects data from patients through a form in the account creation. In addition, it stores the game progress and performance on each mini-game automatically from the interaction with the player. The game must store patient-specific data to allow clinicians to assess their progress in the therapies. The only personal information stored is: name, Colombian ID number, age, and parents’ contact information. The collected progress data are: session number and date, type of mini-games played in the session, difficulty levels of each mini-game, results of each mini-game, and mini-game runtime. This data collection and storage must be authorized by patients and their legal tutors, and it follows to the Colombian legislation (cf. Section 2.2). The collected data are of the sole ownership of the Universidad del Cauca for anonymized research only. Data are stored for a maximum period of 5 years after the finalization of the therapies for (anonymized) research purposes. According to the Colombian legislation, users can request the complete removal of their data from the system at any moment. Therefore, the HapHop-Physio system allows to remove the user’s personal data and game records after a user’s request.

Collected data are stored in a local database, accessed only by the person responsible for the research project designated by the Universidad del Cauca. This person is also responsible for the anonymization of patient data before sharing information with other authorized researchers. All databases and storage systems are not connected to or dependent on any cloud service. All data protection and use aspects are disclosed to end users and their legal tutors (when necessary) before the first use of the game. Patients (and their legal tutors) must agree with these terms to use the game.

4. Discussion

HapHop-Physio is an exer-learning game designed and developed for training memory and attention skills, addressed specially for children with SLD. As any eHealth application, it must be complete, secure, and clear for the patients and clinicians. Consequently, an assessment framework for serious games [28] was applied to evaluate HapHop-Physio in this paper. The five framework themes were described through every applicable item, with a particular focus on the validity category applied to validate the face, content, and construct of the game.

The main results from the assessment of the face and content validity included the improvement of HapHop-Physio in five features: scoring system, login system, clinician’s dashboard, game’s structure, and set of avatars. In addition, findings in the described case study constituted the evidence for the construct validity. Among these findings, the most promising one was the improvement in the cognitive profile of the children after a training period with HapHop-Physio. However, an experiment using a control group is necessary to observe the real effects of HapHop-Physio. Analyses by subdomains, number of subjects classified, gender, and age group were carried out to detect trends in the case study results.

Furthermore, the clinicians in charge of the therapies pointed out some unexpected observations and possible positive side effects of the game:

• The use of technology for rehabilitation therapies was a good motivation for children; their disposition to attend and to perform the therapy was better.
• Effort and dedication in reading and writing activities were evidenced in patients with difficulties in these types of tasks.
• Children improved their attention to the instructions of the game to give correct answers, pass the level, and obtain the rewards of the game.
• Memorizing images and sounds from musical instruments was pleasant to children and helped them to improve this ability.
• Keeping a daily and accurate record of the child’s responses to track progress or difficulties was essential to continue strengthening or advancing the rehabilitation process. It was also essential to show parents what was conducted in therapy and how the progress was over time.

The clinicians also noticed one negative side effect of the game:

• Writing activities at advanced levels such as 4 and 5 discouraged patients, since they took longer to answer or write all the words and only managed to play one or two mini-games per session.

Based on the obtained results, findings, and these observations from experts, the application of the framework allows for assessing HapHop-Physio as appropriate and safe to be used in a healthcare setting. Nevertheless, only 48 of the 55 items in the framework could be addressed. In particular, this paper did not address the concurrent and predictive validities, since they require experimental results with a significant number of patients and a control group.

Consequently, the next objective is to fulfill the framework assessment (i.e., provide the concurrent and predictive validities) to prove the efficacy of HapHop-Physio. For addressing this, two complementary and challenging tasks should be tackled. First, depending on the calculation of the sample size, recruiting the minimum number of children who could be involved in the experiment with HapHop-Physio could be difficult. SLD is an underdiagnosed condition in Colombia. The control group must be determined using this number to enable the generalization and causal inference of the results. The second task consists of measuring the impact of the physical activity in children through wearable technology [33] to determine the validity in terms of the comprehensive nature of HapHop-Physio (exer-learning game).

5. Conclusions

HapHop-Physio must clearly disclose comprehensive information on its claims, purposes, and data use to become a qualified game. To verify such disclosure, we applied the assessment framework described in this paper. The positive outcomes of the assessment of this exer-learning game indicate that it is secure and appropriate for a healthcare setting. Improvements found in cognitive profiles of 11 out of 12 children are merely suggestive to be associated with HapHop-Physio from the case study. More studies are necessary to improve the functionality and validity details of the game. In addition, it needs to be evaluated in a controlled experiment to ensure that it leads to cognitive rehabilitation in reality, fulfilling its external purpose.