Development and evaluation of the STOU BrainLadder: a board-game intervention to enhance cognitive health among Thai elderly

The world in the 21st century is rapidly transitioning toward an aging society, generating significant social, economic and health challenges (United Nations, 2017; WHO, 2020). East Asian nations (e.g. Japan, Republic of Korea and Singapore) are in advanced stages, and Thailand is following this trajectory. Adults aged ≥60 represent 20% of the Thai population (National Statistical Office of Thailand [NSO], 2024); a threefold increase since 1994, projected to surpass 28% within a decade. This demographic shift underscores the urgent need to enhance the elder life quality and functional capacity. Government tools aiming at effectively dealing with this issue are therefore needed.

This rapid expansion impacts labor structures, healthcare systems and long-term care planning. Families and communities need to increasingly support elder independence and well-being. Cognitive health is vital, as age-related declines in processing speed, working memory, attention and executive function constrain daily activities and autonomy (NSO, 2024). Developing scalable, context-sensitive strategies to preserve cognitive health is therefore a national priority for sustainable aging in Thailand.

Cognitive health is the maintenance of mental capacities essential for independent living, life quality and social participation (WHO, 2020), encompassing functions like memory, attention, processing speed, reasoning, decision-making, planning and behavioral regulation (Chan et al., 2025). While short-term memory and processing speed decline with age (Jost and Kujach, 2025), evidence demonstrates that non-pharmacological approaches – particularly multidomain cognitive training (stimulating, social, physical activities) – can enhance function and slow decline. Promoting cognitive health is thus crucial during mild cognitive impairment (MCI), a transitional stage where intervention can mitigate deterioration (Huang et al., 2025).

Although dementia remains incurable, neurocognitive evidence indicates that targeted cognitive function can mitigate age-related decline. Recent studies show that single- and multidomain interventions (cognitive, physical and social) yield measurable gains in memory, executive function and cognitive performance in the elderly (Chan et al., 2025). Structured, demanding activities stimulate neural plasticity and delay deterioration in those at risk of mild cognitive impairment. Tools integrating cognitive exercises with movement and social interaction are thus a strategic approach to promoting cognitive health and life quality in elderly Thai people.

Evidence indicates both digital and analog games can enhance elders' cognitive abilities (Chen et al., 2022; Hongthai et al., 2025). Digital games improve memory, processing speed and reasoning but are underutilized due to usability barriers, a lack of digital skills and technology-related apprehension among elderly users (Tipayalai et al., 2025; Moldagaliyeva et al., 2026). In contrast, tabletop and board games offer socially-engaging activities that may support cognitive functioning and psychological well-being among older adults (Chen et al., 2022; Guardabassi et al., 2024). Some designs enhance balance and visuomotor coordination, reducing fall risk (Bufano et al., 2024). Gameplay thus provides an accessible form of structured engagement.

In Thailand, initiatives like game-based exercises with speech recognition (Tantisatirapong et al., 2021) and technology-mediated practice (Phanasathit et al., 2022) have shown promise. However, these remain small-scale digital prototypes, lacking integration of tangible play and structured social interaction. Despite Thailand's Third National Plan for the elderly (2023–2037) emphasizing holistic capacity and lifelong learning, evidence-based, socially-interactive tools remain scarce.

Most existing games emphasize isolated cognitive domain development rather than integrated cognitive enhancement. The proposed “to-be-developed game” is designed to promote the elders' holistic cognitive functions.

Given these empirical and policy gaps, a board-game intervention integrating digital media with tangible play and social interaction is needed. This study aims to design and develop a digitally-enabled board game to enhance cognitive functioning and quality of life for Thai elders via socially engaged learning. The innovation aligns with the UN's SDG 3 (United Nations, 2017) and Thailand's Third National Plan for Elder Development Strategy (2023–2037), which both emphasize active aging through health promotion and lifelong learning.

STOU BrainLadder development was guided by the design–play–experience (DPE) framework (Winn, 2009). DPE was suitable for this game-based learning intervention because the design elements, player interaction and user experience (UX) supported cognitive engagement. Specifically, design guided game mechanics, accessibility features, feedback systems and cognitive-task structures; play addressed interactions with the board-game loop, mini-games, peers and facilitators; and experience focused on usability, enjoyment, perceived challenge, social interaction and cognitive engagement.

Research objectives are: (1) to design and develop the cognitive health game, and (2) to evaluate the effectiveness of the developed game.

Cognitive health refers to the capacity to sustain mental abilities that support independent living and social participation (WHO, 2020). Rather than a sum of isolated skills, it reflects dynamic neural processes – especially neuroplasticity and cognitive reserve – that allow adaptation to age-related change while maintaining performance. These mechanisms support experience-dependent adaptation in later life; consequently, interventions that nurture multiple cognitive domains and sustain socially meaningful engagement can promote cognitive functioning in the elderly (Liu et al., 2024).

Cognitive functions – working memory, attention, executive functions and processing speed – operate as interdependent subsystems sustaining cognitive health (Cohen et al., 2019). Evidence demonstrates age-related decline across these domains, with the steepest trajectories in complex, resource-intensive processes (Li et al., 2021; Zaninotto et al., 2018). However, cognitive decline is not inevitable; structured social engagement can stimulate neural function and enhance learning potential in later life (Bufano et al., 2024). Accordingly, interventions combining goal-directed tasks with social interaction may attenuate higher-order cognitive decline and support daily functioning.

As cognitively stimulating activities, board games may promote cognitive health by integrating structured play, problem-solving, decision-making and social participation. They merge enjoyment with experiential learning, fostering development consistent with constructivist theory (learners construct knowledge via active practice and collaboration) (Piaget, 1972; Vygotsky, 1978). As structured, goal-oriented systems, board games recruit attention, planning and working memory while maintaining intrinsic motivation.

From an instructional-design perspective, effective board games integrate four principles: (1) simple rules that minimize extraneous cognitive load (Sweller, 2011); (2) calibrated challenges that promote flow and strengthens cognitive reserve (Csikszentmihalyi, 1990; Liu et al., 2024); (3) structured social interaction that supports social cognition and affective bonds (Samtani et al., 2026; Kelly et al., 2017); and (4) motivational affordances grounded in gamification and self-determination theory (SDT) supporting competence, autonomy and relatedness (Deci and Ryan, 1985; Koivisto and Malik, 2021). Despite robust international evidence, few Thai studies exist. Culturally-tailored board games integrating cognitive training, social interaction and adaptive feedback could enhance cognitive health among Thai elders.

Designing a board game to enhance learning and cognitive health requires an integrative framework bridging learning sciences, psychology and design. Game-based learning aligns with constructivism, where learners build knowledge through experience and social interaction (Piaget, 1972; Vygotsky, 1978). Kolb's experiential model situates gameplay in a cycle of experience, reflection, abstraction and experimentation (Kolb, 2015). Situated learning emphasizes participation in authentic communities of practice (Lave and Wenger, 1991). Collectively, these perspectives highlight active engagement as the central mechanism for durable and meaningful learning (Chen et al., 2022; Zhang et al., 2017).

Sustained gameplay emerges when experiences satisfy competence, autonomy and relatedness needs (Deci and Ryan, 1985) and calibrate challenge to skill to elicit flow (Csikszentmihalyi, 1990). Designing to these motives enhances enjoyment, self-efficacy and social bonding (Kappen et al., 2020; Koivisto and Malik, 2021). Cognitive load theory (CLT) favors minimizing extraneous load and maximizing germane load; thus, games for elders should use clear rules, simple iconography and bounded complexity (Sweller, 2011). Three design models operationalize this: the mechanics–dynamics–aesthetics (MDA) framework (Hunicke et al., 2004), Octalysis (Chou, 2015) and serious-game principles aligning learning goals with enjoyable play (Connolly et al., 2012). Together, these perspectives incited the application of the DPE framework in the present game design (Winn, 2009).

Serious-game design is supported through clear learning and development objectives. Extending the MDA framework, DPE links design decisions, player interaction and intended experience through four layers: learning, storytelling, gameplay and user experience, with technology as an implementation factor (Winn, 2009). Designers define game components, while players gain their experiences through gameplay.

The DPE framework guided game development to promote cognitive health among elderly adults. The learning layer linked activities with memory, attention, executive function and processing speed; storytelling used familiar everyday contexts; gameplay translated cognitive objectives into content, rules, challenges, decisions, feedback and progression; and user experience emphasized clarity, ease of use, interface design and engagement. Technology informed platform selection and implementation constraints. The game was developed through design, prototyping, playtesting, evaluation and iterative refinement to ensure usability and alignment with cognitive-health objectives.

These games may support cognitive function among elderly adults by combining decision-making, strategic thinking and social interaction. The board games may improve memory, attention and executive functions (Chen et al., 2022). Further, serious games with technologies like augmented reality may further intensify player experience (Guardabassi et al., 2024).

In Thailand, empirical work on cognitive enhancement has focused more on generic brain-training or analog games than digital formats. For example, Hongthai et al. (2025) observed memory and attention improvements; Kansopa et al. (2025) reported visuospatial benefits from digital Solitaire; and Makphin et al. (2024) found that therapeutic board games improved memory, problem-solving and social skills.

Studies substantiate that games stimulate neural functioning and social participation, yet initiatives lack system-level integration of cognitive mechanisms, learning and social interaction. Salient gaps include (1) no explicit design framework – such as Winn's DPE – to align mechanics with outcomes (Winn, 2009); (2) limited interactive technologies connecting cognitive training with social learning; and (3) underutilizing serious-game principles to link mechanics to learning, therapeutic aims and life quality outcomes.

In attempting to synthesize insights, this study developed a digital board game to strengthen cognitive health in the elderly under the DPE Framework (Winn, 2009), balancing enjoyment, meaning and cognitive value through integrated engagement. Research questions were: (1) What design features and learning mechanisms enhance elders' cognitive health? and (2) To what extent is the developed game effective?

Nine experts were purposively selected based on their educational backgrounds and experiences (gerontological nursing, media design, game development, elder education). All nine experts participated in both focus-groups sessions: the first session assessed conceptual feasibility, while the second session evaluated the prototype's quality and suitability for elder users. Feedback from both sessions was incorporated into the final design. A small technical team co-developed the physical board, user interface (UI), graphics and gameplay, and refined the user experience, while researchers served as principal developers to ensure alignment with study objectives.

Instead of using a pre–post control group design, this study employed an independent cohorts design, assigning participants to two independent groups (Group 1 for the initial prototype trial and Group 2 for the revised prototype evaluation) for three reasons: (1) to reduce practice and testing effects from repeated measures, which may inflate cognitive scores due to familiarity rather than true change (Duff et al., 2025); (2) to support iterative prototype development while minimizing carryover and contamination effects; and (3) to minimize unnecessary cognitive load among elders, which may influence data quality (Sweller, 2011). Both cohorts were recruited through voluntary participation.

Consistent with the formative user experience design and usability-testing focus of this phase, preliminary effectiveness was examined through short-term within-cohort changes in cognitive-performance indicators embedded in the STOU BrainLadder mini-games from Day 1 to Day 14. These indicators were mapped to working memory, attention, processing speed and executive function. Given the absence of randomization and blinded independent outcome assessment, the observed changes should be interpreted as preliminary performance indicators rather than definitive causal evidence of intervention efficacy.

Ethical approval was obtained from the Human Research Ethics Committee of Sukhothai Thammathirat Open University, Thailand (Certificate No. 022.68; Project Code: STOUIRB 2568/025.2404).

Group 1 – initial prototype trial (n = 30/cohorts): The inclusion criteria of the participants included community-dwelling elders (≥60 years) who residing in Nonthaburi Province, possess basic Android skills, basic literacy, the capability in providing informed consent and availability to participate in the study procedures. The exclusion criteria comprised individuals with scheduling conflicts or those who were unable to attend the trial sessions at the specified dates and locations. The participants were allowed to withdraw from the study at any point without reason.

The consent process was as follows: (1) permission to conduct the digital board game trial with eligible elders was obtained from the Nonthaburi Municipality. Nonthaburi Municipality is one among the 76 provincial municipalities in Thailand and was willing to participate in the study; (2) the study objectives, procedures and benefits were explained to responsible officers, and recruitment materials were distributed to voluntary participants; (3) prospective participants received a verbal explanation of the study activities together with the participant information form; (4) individuals were given one week to consider their participation and were provided with the information sheet and informed consent form; (5) the researchers explained details of the study activities; and (6) participants attended the sessions as scheduled, retaining the right to decline participation in any activity without penalty or impact on their rights or status in the study.

Group 2 – revised prototype trial (n = 30/cohorts): This cohort comprised participants distinct from Group 1. The inclusion and discontinuation criteria, as well as the informed consent process, were similar to those applied in Group 1. In addition, participants were screened for cognitive status using the trail making test (TMT) forms A and B and a memory-screening test; inclusion thresholds were TMT-A ≤ 78 s, TMT-B ≤ 273 s, and a memory-screening score ranging from 14 to 19. Participants were excluded if they were classified under any of the following criteria: (1) acute illness or severe health conditions during the study; (2) inability to attend scheduled sessions at the designated dates and locations; (3) withdrawal of participation with consent; (4) inability to communicate effectively throughout the study; or (5) inability to complete TMT forms A and B within the specified time.

Five experts were purposively sampled to evaluate the prototype. The experts were required to possess a master's or a doctoral degree in a relevant field, such as educational technology, information technology or serious-game design, and at least five years of relevant professional experience.

1. The evaluation of the board-game prototype comprised 10 items on a 5-point Likert scale, adapted from prior work on game evaluation (Karavidas et al., 2022). To guide qualitative judgments, evaluation content covered a 16-aspect checklist (e.g. design, visuals, controls, usability). The item–objective congruence (IOC) was used to validate content validity by three experts. IOC values ranged within 0.67–1.00, exceeding the ≥0.50 criterion – indicating all items were appropriate to the intended constructs (Rovinelli and Hambleton, 1977).

2. The “STOU BrainLadder,” developed by the researchers, is a modular digital board game that sequences exposure to three mini-games – Simon, Matching and RuleMaster.

3. Screening instruments for board-game application trials.

• -

  Digitally adapted TMT form A and form B (Reitan, 1958) were administered via a standardized digital interface, controlling stimulus presentation and time response to capture latencies and errors and to address paper-pencil limitations. Form A measured processing speed, attention, visual scanning/sequencing; form B covered cognitive flexibility and set-shifting/executive control.

• -

  A memory-screening questionnaire, which was a 14-item self-report instrument assessing everyday memory in the elderly using a 4-point scale. The questionnaire, translated and adapted from guidance by Chankanya Tantisiripakorn and Sopha Greekeekraikool (Institute of Geriatric Medicine, 2020), was administered via Google Forms.

1. A game-play satisfaction questionnaire with a 10-item, 5-point Likert scale adapted from GUESS-18 (Keebler et al., 2020) was used to measure post-session satisfaction.

All instruments showed good internal consistency based on George and Mallery's (2010) criteria: TMT-A (α = 0.83), TMT-B (α = 0.86), the memory-screening questionnaire (α = 0.88) and the game-play satisfaction questionnaire (α = 0.81).

Cognitive health was defined as multidimensional cognitive functions in the elderly across four domains: processing speed, attention, executive function and working memory. These domains were evaluated through performance indicators derived from three embedded mini-games.

Cognitive functions referred to brain-based abilities measured through behavioral responses during digital board game play and were categorized into four domains as follows:

Processing speed was measured through the mean response time (in seconds) between stimulus presentation and correct response in the Simon and Matching games. Lower mean response times indicated better performance.

Attention was assessed through accuracy rate (percentage of correct responses) and the errors during gameplay in the Simon and Matching games.

Executive function was measured by identification accuracy and the number of trials required to reach the criterion (six consecutive correct responses) in the RuleMaster game.

Working memory was assessed through the maximum sequence length correctly recalled in the Simon game and the number of correctly matched card pairs within the allotted time in the Matching game.

The researchers synthesized relevant theories, models and studies on board-game development, analyzed cognitive decline in the elders and used Focus Group 1 results to assess feasibility. This phase yielded the STOU BrainLadder which could enhance cognitive health of the elderly in Thailand. Insights and conceptualizations of the researchers led to the development of the prototype, which was subsequently validated by Focus Group 2 to confirm its appropriateness before refinement and digitization.

Procedures included: (1) a participant trial (including two 20-min rounds with a 30-min break) with automatic log-in, during which the researchers provided on-site troubleshooting support, followed by a 14-day home-based trial with gameplay scheduled on Days 1, 7 and 14, with one 20-min session per day. These three time gameplays were designed to (1) minimize short-term practice effects associated with immediate repeated testing, (2) observe learning progression and early retention and (3) reduce participant fatigue while maintaining engagement. This intermittent schedule enabled (1) evaluation of short-term adaptation and usability stability over time; (2) expert appraisal to evaluate construct alignment; and (3) a revised trial under the same conditions, followed by in-application satisfaction ratings and interviews. Findings were integrated to evaluate the game's feasibility, usability, enjoyment and preliminary cognitive-performance patterns.

The data were analyzed in two stages below.

Feasibility and appropriateness from Focus Groups 1–2 were analyzed via content analysis and yielded the “STOU BrainLadder”

1. Initial trial: Descriptive statistics of gameplay behavior, mini-game outcomes, aids/coins and technical errors; Wilcoxon signed-rank test; open-ended comments via content analysis were used.

2. Expert appraisal: Item/scale descriptive; dispersion/inter-rater consistency via coefficient of variation (CV); construct alignment via content validity index (CVI) were used.

3. Revised trial: Descriptive statistics of behavioral indicators per mini-game; Wilcoxon signed-rank test; Friedman's test for scores across Days 1, 7, 14 and Kendall's W (effect size) were used.

4. Satisfaction analysis: Descriptive statistics of questionnaire scores; interview remarks via content analysis were used.

Experts endorsed a Snakes-and-Ladders macro-loop for the game's familiarity and learnability, and urged curbing randomness and making skill progress more explicit. The experts recommended aligning tasks with four cognitive domains – processing speed, attention, executive function and working memory – and suggested using a fixed rotation of mini-games to balance practice opportunities and to strengthen logical validity and to reduce guessing. The experts advised giving participants brief work-examples, time limits where appropriate, graded on boarding hints and short rationales within RuleMaster to demonstrate rule application. With regard to accessibility, the experts recommended larger text, more generous touch targets and higher contrast for elderly users. Collectively, the feedback converged on a design in which the board governs pacing and the mini-games govern cognitive load, with concise performance feedback to support transparency and motivation.

Guided by expert appraisal and feedback from Focus Group 2, the researchers employed a deterministic loop – roll the dice → the avatar move according to the dice → the gameplayer may need to roll the dice again till the mini-game shows up for mini-game playing → play the mini-game → the feedback/score appears – so that board movement paces exposure while mini-games deliver targeted operations instantly. After ladder/slide effects are shown, the system randomly selects one mini-game out of the three mini-games (Simon, Matching and RuleMaster) at a time with equal probability (33% each) under a no-immediate-repeat rule: once a game is chosen, the next draw is from the remaining two (50% each) until all three mini-games have occurred and played; then the set reset button occurs. The rule screens provide brief work-examples and mnemonic cues. The cognitive mapping is as follows: Simon and Matching enhance processing speed, working memory and attention; RuleMaster enhances executive functions. After each mini-game, plain-language and domain-tagged feedback signal cognitive-health progress. Age-friendly designs safeguard for elder users are enforced (text ≥22 pt, buttons ≥1.5×1.5 cm, contrast ≥4.5:1, ≤5 top-level menu items), and a lobby and practice area are provided. The mini-games enhance cognitive practice and task-related performance changes.

The final-prototype is an Android two-part system: (1) the application comprising the securing and transmitting of client data and (2) the cloud platform for users which includes Firebase authentication, a Laravel REST API for data/admin, a MySQL analytics datastore, and Dockerised services on a Linode VPS for stability and scalability.

The interface prioritized clarity, accessibility and engagement for the elders and was organized into three stages.

1. Entry and setup, which included the Registration/Login screen, Lobby interface and Avatar selection, and which guided users in securely accessing the game and selecting personal avatars.

2. Gameplay interaction comprised practice and multiplayer modes. The game supports both independent and socially interactive cognitive activities.

3. Feedback and motivation include the post-game reward survey and real-time leaderboard, which enhanced reflection, sustained engagement and self-determined motivation.

1. Core game: Snake and Ladder

   • The player clicks the dice icon located at the left bottom of the screen to move the avatar according to the number shown on the dice. The avatar may sometimes fall down, move back or climb up a ladder. At the end, one mini-game (could be Simon, Matching or RuleMaster) from the right bottom will appear automatically as programmed for the individual or the group to play (Figure 1).

2. Mini-game play

   • Simon (Figure 2 (A)): Simon is designed to stimulate working memory and attention. The player needs to remember the sequence of piano sounds and the four colors – red, yellow, blue and green – that appear together in sequence with the sounds. The game allows a brief pause for the player to select the color. A correct selection is awarded a score, which is accumulated. Simon releases one, two or more colors sequentially, and the player needs to remember and select the colors in the order they appear. The more colors that appear, presumably the more difficult it is for the elder to remember and select the right colors and sequence. The researchers classified the difficulty level of the game into eight levels. An incorrect selection or answer leads the player back to the original starting point.

   • Matching (Figure 2 (B)): The game promotes attention and short-term memory. The player is asked to pair the cards from the decks of 4, 6, 8, 10 and 12 cards. If the pairing is correct, the paired cards will remain “face up”. The player is supposed to find all pairs among the 12 cards and once the card matching at each level is correct and complete, the player will be awarded points.

   • RuleMaster (Figure 2 (C)): The game is designed to promote analytical thinking and logical reasoning for the elderly. The player needs to relate or link the revealed picture to the “descriptor” or fact or rule hiding behind the picture. If three answers are correctly and consecutively answered, the player will be awarded points. With the six correct and consecutive answers, the player will be also awarded points.

The player needs to play the three mini-games continuously within time limit.

The second phase examined the performance, usability and perceived quality of the STOU BrainLadder through participant testing and expert review. The findings were as follows.

1. Technical/usage issues – most frequently occurred problems were unclear UI (25.00%), followed by difficult-to-use/error-prone buttons (21.88%) and premature game exit rate (21.88%); the least frequently occurred problem was incomplete data recording (6.25%).

2. Gameplay behavior outcomes.

   • Simon – Mean memory sequences increased from 3.50 (SD = 1.20) to 4.10 (SD = 1.00), while errors decreased from 25.30 (SD = 8.10) to 20.10 (SD = 7.50), indicating improved sustained attention.

   • Matching – Correct pairs rose from 5.80 (SD = 2.50) to 7.20 (SD = 2.30), consistent with faster perceptual processing under time limits.

   • RuleMaster – Correctly inferred rules increased from 1.20 (SD = 0.80) to 1.60 (SD = 0.70), indicating a positive change in abstract reasoning performance.

   • Core game – Finish of dice rolls decreased from 15.20 (SD = 3.10) to 14.50 (SD = 2.80), and use of coin fell from 1.80 (SD = 0.90) to 1.10 (SD = 0.60), indicating more efficient planning and risk management.

Simon – Median of memory sequences were higher in Round 2 than in Round 1 (Z = −2.85, p = 0.004); median error rate was lower in Round 2 (Z = −2.20, p = 0.028), indicating improved working memory and attention.

Matching – Median of matched pairs increased in Round 2 (Z = −3.15, p = 0.002), indicating improved processing speed.

RuleMaster – Median of inferred rules were higher in Round 2 (Z = −1.98, p = 0.048), reflecting a positive change in executive-function-related performance.

Most participants reported the game as enjoyable and suitably challenging, though some required more time to learn initial rules. Matching was rated as easy and engaging, while RuleMaster was considered cognitively demanding and Simon was thought to have advanced too quickly. For RuleMaster, participants recommended enhancing real-life relevance, enlarging fonts and buttons and improving server connectivity.

Overall expert ratings were moderate to high (M = 3.40–4.40). The highest scores were given for mini-game quality/alignment and controls/buttons (M = 4.40), while the lowest were text/dialog clarity and limited dialog issues during play (M = 3.40). Agreement of what was good among the experts was little different (SD = 0.40–0.80; CV = 10.53–18.18%, all <20%). Content validity (S-CVI/Ave) was below the acceptable level 0.75, ≥0.80 benchmark (Polit and Beck, 2006).

Four integrated improvement priorities were identified, particularly for RuleMaster. First, content should link scenarios to daily-life contexts (e.g. home safety, nutrition), diversify activities across cognitive domains and use precise language with supportive icons. Second, narrative and feedback need pre-play guidance, in-game cues and post-round summaries to enhance usability. Third, validity and measurement should curb guessing through rule-based justifications, appropriate time limits and increase the number of pictures and descriptors. In addition, pictures and the descriptors need to be made relevant to daily living, routines and habits of the elders. Finally, accessibility and the user interface need to feature clear typography, high contrast, large buttons, simplified screens and adjustable visual/audio options tailored to the elderly's sensory needs.

Working memory-related performance in the Simon game increased steadily over the 14-day trial. The median of sequences recalled increased from 5.00 (Day 1) to 6.00 (Day 14). The Friedman test indicated significant differences across sessions, χ² (2) = 41.00, p < 0.001, with Kendall's W = 0.68 – which exceeded the 0.50 large-effect threshold (Fritz et al., 2012) and was consistent with favorable working-memory-related performance change.

Attention-related performance in the Simon game showed favorable change as median error rates decreased from 20.30% (Day 1) to 16.15% (Day 14). The Friedman test confirmed significant differences across sessions, χ² (2) = 58.07, p < 0.001, with Kendall's W = 0.97, which was consistent with favorable attention-control performance change.

Processing-speed-related performance in the Matching game showed an increase in the median number of correctly matched pairs from 9.00 (Day 1) to 10.00 (Day 14). The Friedman test yielded χ² (2) = 55.06, p < 0.001, with Kendall's W = 0.92, which was consistent with favorable processing-speed-related performance change.

Executive-function-related performance in the RuleMaster game showed an increase in the median number of correctly identified rules from 1.00 (Day 1) to 2.00 (Day 14). The Friedman test showed significant differences across sessions, χ² (2) = 38.08, p < 0.001, with Kendall's W = 0.64, and consistent with favorable executive-function-related performance change.

Overall participant satisfaction was high (M = 4.13, SD = 0.94). Item-wise analysis revealed consistently high ratings across all dimensions. The highest mean score was for game controls/buttons (M = 4.17, SD = 0.75), and the lowest was for understanding training objectives (M = 3.87, SD = 0.90).

The feedback highlighted four areas: first, enhancing accessibility for sensory changes (font size, contrast and sound); secondly, broadening game variety and challenge to stimulate higher-order thinking and sustain motivation. Thirdly, regular sessions reduced loneliness and strengthened social connection, and finally, facilitators' friendliness and support enhanced technology acceptance and engagement.

The game trial showed short-term positive changes in working memory and executive function. These results align with cognitive-training principles underpinning executive control (Cohen et al., 2019). The findings suggest short-term patterns in trained domains, particularly working memory and inhibitory control, rather than the broad or durable changes across multiple untrained domains. The results also aligned with the WHO cognitive-health framework, which emphasizes the importance of maintaining functional ability and cognitive well-being in later life (WHO, 2022). Improvements in Simon task performance are also consistent with CLT and executive-control principles, whereby reduced memory load and improved attentional control may enhance executive efficiency (Cohen et al., 2019; Sweller, 2011). Repeated practice may likewise enable participants to internalize task rules and streamline response selection, which may explain why positive changes were most evident in functions closely related to executive control and working memory.

Participants' perceptions of RuleMaster's complexity are consistent with CLT, as ambiguous rules and dense information can increase extraneous load and impede learning (Sweller, 2011). Progressive scaffolding, clearer sequencing and multimodal cues may reduce this burden, particularly for elderly adults, who may need to allocate additional cognitive resources to understanding task procedures before engaging the target cognitive processes (Hallewell Haslwanter et al., 2020). Educational background and digital literacy may also influence the training experience. Participants who were more familiar with structured learning or digital interfaces encountered fewer barriers to understanding instructions and navigating the platform, which in turn supported engagement and performance.

The game's social and board-based format may have contributed to participant engagement by encouraging planning, decision-making and structured problem-solving in a meaningful interactive context. Recent board-game research similarly suggests that group-based board games support social interaction, cognitive engagement and well-being, particularly when difficulty is appropriate for participants (Guardabassi et al., 2024). In addition, the digital STOU BrainLadder platform captures fine-grained behavioral traces, including response times, error patterns and decision sequences, which may help distinguish task familiarity from genuine cognitive-performance change. Such data support the interpretation of short-term performance changes while reducing conflation between practice effects and longer-term cognitive change. Digital learning research further suggests that gamification may strengthen motivation and self-directed learning in open and distance contexts (Barua and Bharali, 2023; Por and Muniandy, 2023).

However, participant feedback indicated that the user experience may have been constrained by usability and accessibility issues central to age-friendly digital environments (WHO, 2022; Hallewell Haslwanter et al., 2020). Small fonts, low contrast and small buttons reportedly caused input errors and may reflect age-related changes in vision and motor coordination. If unaddressed, these barriers may increase extraneous cognitive load and contribute to digital exclusion (Pan et al., 2025). Universal design principles may therefore support confidence, usability and sustained participation. Larger text, higher-contrast displays, adequately sized touch targets, clearer instructions, step-wise tutorials and brief demonstrations reduce input burden and improve accessibility. Familiar training content also increase relevance and supports transfer (Bufano et al., 2024; Kelly et al., 2017).

The interpretation of the Day 1 to Day 14 cognitive-function findings requires caution. Although statistically-significant changes were observed in the game-embedded indicators of working memory, attention, processing speed and executive function, these changes were measured through repeated exposure to the mini-games themselves. Therefore, the observed patterns may reflect a combination of cognitive engagement, task familiarization, strategy acquisition, improved understanding of game rules, reduced interface burden and practice effects. Without an active-control group, the study cannot isolate intervention-specific cognitive effects. The findings should therefore be interpreted as preliminary within-cohort performance indicators reflecting short-term change, rather than as definitive evidence that STOU BrainLadder independently caused durable cognitive improvement.

Relatedly, the short intervention period and relatively small, homogeneous sample, particularly in terms of educational background and digital literacy, reduce confidence in the broader generalizability of the findings (Bufano et al., 2024). Participants with relatively similar educational and digital experiences may have encountered fewer barriers to using the platform, and the findings may therefore not fully reflect the experiences of elderly adults with lower educational attainment, lower digital literacy or less familiarity with technology.

Preliminary findings from STOU BrainLadder suggest that participation in a socially-interactive digital game-based learning program was associated with improved performance on measures related to working memory, attention, processing speed and executive function. Because these outcomes were derived from within-cohort repeated gameplay without a control group, they should be interpreted as preliminary effectiveness indicators rather than definitive evidence of causal cognitive improvement.

The digital platform generated fine-grained behavioral data, including response times, errors and decision sequences, which provide useful insight into learning processes and possible influences of task familiarity and strategy used.

The observed benefits appear to be limited to short-term outcomes. The short intervention period, small sample and absence of an active-control group limit causal inference and may not allow firm conclusions about long-term cognitive maintenance or transfer to everyday functioning. Further, longitudinal controlled studies with ecologically-valid outcome measures are therefore needed to determine whether sustained cognitive and functional benefits can be achieved in real-life contexts.

The STOU BrainLadder was proposed as a cognitively-stimulating, socially-interactive group activity for Thai elderly adults in non-formal education settings and community centers. At this stage, it should be positioned as a promising pilot approach rather than as a proven intervention for long-term cognitive maintenance.

At the community level, pilot implementation should be supported by facilitator guidance, usability testing and inclusive UI/UX refinement to improve accessibility for elderly users. Priority design features include larger text, higher contrast, clearer instructions, step-wise tutorials, brief demonstrations and adequately sized touch targets. Maintaining the game's socially-interactive format may also help support engagement, cooperation and continued participation among elderly adults.

At the policy level, relevant agencies may consider integrating age-friendly digital cognitive-training games into active aging and digital inclusion strategies in the ways that are consistent with the WHO Decade of Healthy Aging (2021–2030) and the development of age-friendly digital environments (WHO, 2020, 2022). Wider implementation should be supported by continued research and development investment in serious games for cognitive health, together with inclusive UI/UX standards that reduce barriers to access and participation.

The Android-based game prototype is appropriate for initial implementation and testing. Technology adoption among elderly adults requires consideration of device cost, availability, ease of use, digital skills and real-world support. This aligns with Pan et al. (2025) and Tipayalai et al. (2025), who indicated that technology use among Thai elderly adults is associated with device accessibility, affordability, digital skills, perceived usefulness and user-friendliness, rather than operating-system popularity. Similarly, Moldagaliyeva et al. (2026) found that the digital divide among elderly adults should be examined through device access, usage skills and benefits for daily living, rather than device ownership alone. Therefore, selecting Android was a practical decision under technical and budgetary constraints. Cross-platform development, including iOS, should be considered in the subsequent phase.

At the research level, future initiatives should emphasize rigorous evaluation through longitudinal active-control together with randomized controlled trials, together with adaptive difficulty and personalized feedback systems. Future studies should also recruit more heterogeneous samples, and examine whether training effects differ according to educational level, digital literacy, prior technology experience and cognitive risk. These design considerations may help clarify the generalizability, sustainability and practical applicability of the intervention across diverse elderly adults.

It is also recommended that long-term active-control trials, with randomized controlled trials (RCTs) with ecologically-valid outcomes, be carried out to determine whether there are long-term effects and sustained real-world benefits. More heterogeneous samples and multidimensional assessments are needed to clarify whether the intervention can support daily functioning across diverse groups of elderly adults.