The development stages of TSTM are shown in Figure 1.

Information about the first stage of TSTM development can be found elsewhere [28]. The main characteristics of TSTM are summarized in Table 1. These characteristics were influenced by work that was conducted by others [29].

TSTM is the result of interdisciplinary collaboration among health care professionals, researchers, serious video game consultants, and a single educational and therapeutic game development company. This collaboration made it possible to integrate several theoretical models of ADHD with clinical experience, the use of chess and VR by patients with ADHD, and other playful elements of video games that increase players’ motivation for engaging in therapy [30].

TSTM is theoretically driven by (1) both the Barkley [2] and Brown [21] models of ADHD and executive dysfunction and (2) our personal clinical experience of using chess with patients with ADHD [10]. The original Barkley model was based on the work of other researchers, such as Bronowski and Fuster [31,32]. This model focuses mainly on the hyperactive-impulsive ADHD subtype (type 2). The primary premise of the Barkley ADHD model is a deficit in behavioral inhibition. This deficit in inhibition is related to problems in the following four executive neuropsychological functions: (1) working (nonverbal) memory; (2) the self-regulation of affect-motivation and arousal; (3) the internalization of speech (verbal working memory); and (4) reconstitution, which is understood as the ability to manipulate verbal and nonverbal mental representations (behavioral analysis and synthesis) [2].

The Brown ADHD model postulates that the core deficit depends on executive dysfunction. In the Brown model, ADHD is considered to involve 6 major deficits in activation, focus, effort (motivation), emotion, memory, and action. In the Brown model, similar to the Barkley model, executive dysfunction is core to ADHD. However, both models have some differences [21]. The Barkley model focuses on the combined type of ADHD and emphasizes the relevance of behavioral inhibition. In contrast, in the Brown model, behavioral inhibition is just 1 of the 6 defective executive functions [2,21].

The TSTM prototype that was used in this usability study had 5 minigames; each minigame targeted specific cognitive abilities (Table 2) and particularly focused on inhibitory control, selective attention, cognitive flexibility, or processing speed [2,33]. The minigames were embedded into a VR forest context in order to provide a more immersive and motivating experience [28]. Furthermore, as stressed by Rabipour and Raz [23], “regular interaction with nature appears to facilitate improvements in cognitive function and behavioral control.” A minigame was defined as “a small, isolated game within the larger game environment that integrates unique game elements offering tools to improve strategic behavior” [15].

After an initial design of TSTM was produced, we iteratively tested modified versions in order to detect and correct all bugs [34]. Furthermore, we integrated all suggestions that were made by different users who tested the initial version. Accordingly, we included several modifications that allowed us to create a redesigned video game (redesign stage) [34]. We gathered further information on usability criteria, such as effectiveness, efficiency, user satisfaction, and the adverse consequences of use, and fixed some extra bugs. Thanks to this, we constructed the first functional prototype.

In conclusion, TSTM is a serious video game that is defined as either a game that was designed for a primary purpose other than entertainment [35] or a “computer-based [game] designed for training purposes” [36]. We also used gamification [37,38] to introduce chess and to teach the basic rules of chess to participants who did not know how to play the game.

The objective of our market study was to determine whether brain training via a serious video game, such as TSTM, would be of interest to mental health and education professionals. A survey that included the characteristics of TSTM and general questions about the game was designed. We used the Google Forms platform to disseminate Spanish and English versions of the survey through professional networks [39]. A total of 57 people responded, but 1 person was excluded, as he was neither a health worker or an education worker (he was a programmer). The local Committee of Medical Ethics did not require official review for study 1.

We only analyzed the percentage of responses. No statistical test was needed.

The second study included the first 40 consecutive patients with ADHD who were offered inclusion into a randomized controlled clinical trial (RCT) for testing the effectiveness of either TSTM or web-based chess [41]. This RCT study was a prospective, unicentric, randomized nonequality trial for patients with ADHD. All participants underwent drug titration (up to the optimum drug dose) and were determined to be clinically stable before the baseline evaluation. Patients were randomized into the following three groups: TSTM group (cognitive training via TSTM), the therapeutic chess group (web-based cognitive training via chess), and the control group (patients were called every week, but no cognitive intervention was used) [42]. The allocation ratio was equal in all 3 groups (35 participants per branch).

The patients included in this early usability stage tested the initial version of TSTM between December 6, 2019, and February 22, 2020, at preinclusion. In order to provide all available information before entering into the RCT, patients were offered the opportunity to test TSTM at preinclusion. Of the first 40 patients, 3 were excluded from the statistical analyses either because they did not complete the test or because the data provided were incomplete. The remaining 37 patients tested TSTM and eventually entered the study. However, not all patients were allocated to TSTM group, as randomization took place during the inclusion visit (day 0), and the data presented in this paper were recorded at preinclusion.

The inclusion criteria were (1) an age of 12-22 years, (2) competency in the Spanish language, and (3) written informed consent. The exclusion criterion was epilepsy, since the use of VR is not recommended for individuals with epilepsy, as per the official recommendations of PlayStation VR [43]. Ethical approval was obtained from the local Committee of Medical Ethics at Puerta de Hierro University Hospital – Majadahonda (Madrid, Spain; research project code: PI 187/19; approved on December 1, 2019). Written informed consent was obtained from participants and at least 1 parent.

The materials used for this study included the video game itself (TSTM), VR glasses, PlayStation 4 controllers (Sony Group Corporation), test kit consoles, monitoring screens, and headphones. The VR software runs on a PlayStation 4 (Sony Group Corporation) test device, which allows for the tracking of movement through the camera (eg, position, head movement, speed, etc).

Demographics, previous VR experience, and vision problems were noted by using an ad hoc questionnaire. The testing time was always monitored by a professional looking through a screen (Figure 2).

Initially, participants attended a single session in which they tested the TSTM experience. Patients with ADHD were given the opportunity to test a single minigame. After passing the tutorial of the minigame to be tested, a higher level of difficulty was tested. Patients could test each minigame for as long as they liked (ie, only for the duration of that session). After testing a single minigame, patients rated their opinion about the minigame by using an ad hoc questionnaire, which included (1) 5 general questions about the TSTM experience and (2) 2 questions about possible adverse effects. Based on previous studies [44-46], we included information about perceived dizziness and motion sickness, which were rated by using a 3-point Likert scale and 2 questions about patients’ feelings before and after testing TSTM. Standard questionnaires were used as a reference [47] by adapting them to the characteristics of our video game and using language adapted for children. We arbitrarily determined that a user satisfaction rate of 75% would indicate a good level of acceptance by users.

Participants were divided into the ADHD combined and inattentive subtypes because some authors consider these subtypes to be different disorders and believe that they should be treated separately. Thus, we feared that the different subtypes of ADHD would impact our usability measures. A descriptive analysis was performed by using the mean proportions (%) of participants’ ratings for categorical variables and means and SDs for numerical variables. Chi-square tests were performed to compare patients with ADHD (ADHD combined subtype vs inattentive subtype) and to compare frequent and infrequent video game players. We used SPSS version 20 for Macintosh (IBM Corporation).

Table 4 displays the basic demographic and clinical characteristics of participants. The testing time for the video game lasted between 10 and 40 minutes (mean 21.31 minutes, SD 6.77 minutes). Most participants were male (25/37, 68%), were right-handed (29/37, 78%), never repeated a school year (26/37, 70%), and were diagnosed with at least 1 comorbid mental disorder (28/37, 76%). Further, 38% (14/37) of participants wore glasses, 59% (22/37) had previously used VR, and 70% (26/37) played video games regularly. We found no statistically significant differences between the combined and inattentive subtypes of ADHD.

Figure 3 displays the proportion of children and adolescents who liked each minigame. All minigames were based on the opinions of the participants, and percentages ranged from 80% (chess: 4/5) to 100% (Kuburi: 22/22; Teka Teki: 23/23). With regard to the results for comprehensibility, the ease of play, and the ease of control of the PlayStation 4 controller, all proportions surpassed 60% (22/36). All minigames were easy to play, and only Teka Teki had an easy-to-play percentage of below 80% (22/36, 61%). Teka Teki was the only minigame in which the participants pointed out that it was difficult to achieve improvement after repeating the minigame. We found no statistically significant differences after comparing the ADHD combined subtype to the inattentive subtype and frequent video game players to infrequent video game players.

Table 5 displays the percentages of children and adolescents who had positive opinions, which were based on the two highest scores (“much” or “very much”). With regard to the discomfort of VR glasses and motion sickness, we considered any level of discomfort to be a negative opinion, even if it was just “a little bit” of discomfort. Most children and adolescents provided very positive responses to all questions. Furthermore, 5 out of 36 (13.1%) children and adolescents reported the discomfort of VR glasses and motion sickness, and just 1 (3%) reported this clearly.