Speech sound disorders (SSDs) can significantly impede a child’s communication skills, highlighting the need for early and engaging interventions. This study introduces Parak, a serious game designed for children aged 5 to 8 with SSDs, offering an interactive and motivating approach to speech therapy. In Parak, players use their voices to guide a pigeon through 23 levels, each targeting a specific Persian consonant, supported by vibrant visuals and interactive tasks. The game incorporates in-game rewards to enhance engagement and provides real-time feedback to support articulation improvement. Unlike traditional speech therapy, which can be repetitive and less engaging, Parak integrates therapeutic objectives with enjoyable gameplay, creating a more immersive learning experience. A controlled study with 20 children was conducted to evaluate the system, and the results demonstrate statistically significant improvements with small-to-moderate effect sizes in articulation accuracy, engagement, and motivation, suggesting promising preliminary evidence within a pilot-scale intervention. These findings suggest that gamified speech therapy can support improvements in clinical outcomes and user motivation in short-term, controlled settings. By combining entertainment with structured educational goals, Parak not only addresses the clinical needs of children with SSDs but also fosters a more positive attitude toward therapy, positioning it as a promising supplementary tool for pediatric speech and language rehabilitation.
The growing cultural and societal relevance of digital games has intensified ethical discussions surrounding game design and the values embedded in games. While prior research has predominantly focused on player experiences and outcomes, comparatively little attention has been paid to how ethical considerations are addressed upstream within everyday game design practices. This study addresses this gap by exploring how professional game designers perceive, interpret, and negotiate ethical issues that may shape the societal impact of digital games. Using a Grounded Theory approach, semi-structured interviews were conducted with 11 game designers. The analysis generated an emergent theory, Discussing Ethics in Video Games, comprising two interrelated categories: Ethical Issues, including concerns such as violence, representation, and inclusivity, and Ethical Responsibilities, reflecting designers' perceived roles in addressing these concerns within real-world design constraints. The findings indicate that ethical reasoning is embedded in design decision-making and is continuously negotiated alongside creative ambitions, technical limitations, and market pressures. By foregrounding designers' perspectives, this study highlights how ethical considerations at the design stage contribute to shaping the broader social and cultural dimensions of digital games. The proposed conceptual framework may support future academic research on ethical game design and provide industry practitioners with a reflective tool for identifying ethical issues and discussing responsibility during design processes. However, given the limited sample size and the regional concentration of participants, the findings should be interpreted as exploratory rather than generalisable.
Online user reviews provide rich data resources for exploring user experience in virtual reality (VR) entertainment devices. However, existing research on VR entertainment devices remains limited, and a systematic investigation of user satisfaction is still needed to support market-oriented product optimization. This paper develops an extended Kano-based framework for attribute classification and prioritization, grounded in satisfaction modeling. Specifically, LLM and BERT are employed for automated attribute extraction and attribute performance evaluation, respectively. At the same time, a LightGBM model is constructed to capture the nonlinear relationship between attribute sentiment and satisfaction. The Kano-based attribute classification mechanism is extended by jointly considering asymmetric contribution characteristics derived from SHAP values and the response structure revealed by correlation analysis. On this basis, a decision-making method based on nonlinear performance-gap mapping is proposed, which transforms qualitative attribute classification into quantitative improvement priorities and mitigates the allocation bias of traditional linear ranking under resource constraints. An empirical analysis of VR headsets demonstrates the applicability and practical utility of the proposed framework, providing scientific decision support for user-oriented product optimization and market development.
Student engagement is crucial for the learning process. During the social isolation of the Covid-19 pandemic in Brazil, educators resorted to educational digital games to stimulate engagement in a remote context. To investigate aspects of teaching and learning, this research utilized data from a randomized controlled experiment involving 492 first-grade elementary school students in Rio de Janeiro, using mathematical games from a digital games platform. The aim was to identify engagement profiles and assess how the educational intervention affected them. Using Game Learning Analytics, data clustering (hierarchical method) separated the students into two groups: engaged and non-engaged. The engaged group exhibited unusual results: they showed no significant difference compared to the other groups and demonstrated no evolution between the tests, being characterized by a high need for mediation. In contrast, the non-engaged group achieved a large effect size (η2 > 0.14) across most mathematical skills when compared to the control group. This study demonstrates that the pedagogical intervention with digital games, conducted asynchronously and without mediation, is beneficial for mathematical skill development but is not effective for students with difficulties (cognitive or technological), who require educator mediation to overcome challenges.
Existing visitor experience frameworks are largely grounded in museum contexts and do not adequately account for the characteristics of large-scale expos, including high visitor flow, short visitor stay, and non-fixed visitor routes. This study extends Roppola's (2013) visitor experience framework to an expo environment and examines how resonance, expansion, and guidance relate to visitor experience within a multimodal narrative exhibition. Using the 2024 Asia-Pacific Sustainable Expo as the empirical setting, an exhibition integrating glove puppetry, AI-assisted script development, and RFID sensing technology was designed and implemented as a design intervention. Data were collected through on-site questionnaires (n = 120) and analyzed using PLS-SEM. The results indicate that Resonance (β = 0.361, p = 0.002) and Expansion (β = 0.366, p < 0.001) significantly predicted visitor experience, whereas Guidance was not a significant predictor (β = 0.177, p = 0.114); the model explained 72.1% of the variance in visitor experience (R2 = 0.721). These findings suggest that expo visitors may establish emotional resonance through familiar cultural symbols and use interactive technologies for self-directed knowledge expansion, rather than relying on systematic guidance, indicating that visitor experience processes are context-dependent. The study offers practical recommendations for multimodal narrative design in expo environments.
Live Operations (LiveOps) transform mobile games into always-on services whose millisecond feedback loops and revolving scarcity cues compress decision windows and intensify in-app purchases. While prior studies often model FoMO as an antecedent of flow, the affective route to spending in hyper-dynamic contexts is not well integrated. Building on the DeLone and McLean IS Success Model, flow theory, and dual-dimensional FoMO, we propose a Quality-Flow-FoMO-Impulse mechanism and test it using PLS-SEM on survey data from 557 Taiwanese players of Garena Realm of Valor (RoV) and Brawl Stars (BS). System and information quality significantly enhance flow, whereas service quality shows no meaningful effect. Flow subsequently amplifies External FoMO (status/competition) and Internal FoMO (belonging/continuity), which fully mediate flow's influence on impulsive microtransaction intention (R2 = 0.352). Multi-group analysis indicates stability across the two games. These results (i) explain when service quality loses salience in real-time LiveOps, (ii) reverse the conventional FoMO-before-flow sequence by showing flow as a trigger of FoMO, and (iii) translate into design guidance that prioritizes low-latency, transparent patch information, and calibrated scarcity cues to support conversion without undue pressure, thereby aligning monetization with ethical and regulatory expectations.
With the rapid development of information technology, virtual reality (VR), augmented reality (AR), entertainment computing and multimodal perception technologies are increasingly integrated into music education. As an important approach to cultivating artistic literacy, music education needs not only immersive presentation but also affective engagement, playful interaction and data-driven feedback. This study designs and implements a music education interactive platform that integrates VR/AR with multimodal perception. The platform collects and analyzes audio, visual, motion and interaction-log information, and then transforms learners' pitch accuracy, rhythm stability, gesture operation, attention state and task participation into real-time feedback and entertainment-based learning tasks. First, the study analyzes the limitations of current music education, such as single teaching methods, insufficient interaction, weak emotional stimulation and delayed feedback. On this basis, a platform architecture is proposed, including three-dimensional scene construction, music-symbol recognition, multimodal perception fusion, entertainment task design and real-time feedback mechanisms. The experimental results show that students using the platform significantly improved their mastery of music-theoretical knowledge and musical-instrument performance skills. The theoretical knowledge test score increased by an average of 29.6% across all experimental groups, and the accuracy rate of musical-instrument performance increased by 30%. Learners showed higher emotional engagement and willingness to continue using the system. In addition, students' overall satisfaction with the platform reached 92%, indicating that the combination of immersive VR/AR, entertainment computing and multimodal perception can effectively enhance learning motivation, interaction depth and teaching effectiveness. This study enriches the teaching methods of music education and provides a reference for the deep integration of entertainment computing and multimodal intelligent perception in educational scenarios.
The rise of platform-based gaming services has provided the opportunity for gaming to be not just a recreational activity, but also a form of digital labor. Within this context, the emergence of game companions, individuals who assist clients in gameplay and provide social interaction for a fee, represents a significant development. This study examines the ways in which gendered discourses shape and structure labor within the game companion industry. Utilizing Feminist Poststructuralist Discourse Analysis (FPDA), the research analyzes interviews with game companions to explore how they navigate power relations and construct professional identities. The analysis reveals three primary patterns: technical skill serves as a key source of legitimacy; emotional labor is unequally distributed (with women doing more) and undervalued; and platform structures, along with prevailing social discourses, reinforce gendered divisions of labor through specific mechanisms. Concurrently, workers actively negotiate their professional identities by adapting to, resisting, and reshaping dominant gender norms through daily discursive practices. This study explores game companionship as a new form of entertainment work facilitated by platforms, thereby expanding the research on gaming labor, platform labor, and gender. It shows how gendered disparities are both reinforced and challenged through daily discursive practices.
Despite extensive evidence on the effectiveness of games as learning spaces, players’ perceptions of learning transferability remain underexplored. Transferability refers to the ability to apply knowledge or skills beyond the context in which they were acquired. How players perceive and narrate this process is relevant to understanding the adoption and potential effectiveness of games for learning. This article reports a study exploring players’ perceptions of learning through video games, the transferability of game-related skills, and their views on the future of gaming and online spaces. Framed within a larger multi-method study, the Reflexive Thematic Analysis was based on two open-ended questions answered by a non-probabilistic sample of 627 Portuguese-speaking players (M=27.09; SD=11.84years). Participants identified several skills and ideas that they perceived as acquired through video games and transferable to other areas of life, in cognitive, social, cultural, emotional, and creative dimensions. They also described a recurring tension between perceived benefits and risks, alternating between educational and economic opportunities, as well as concerns about addiction, isolation, and commercial exploitation. Although situated within the Portuguese context, these findings encourage broader discussion of gaming’s role in everyday life and support the need for further qualitative research on perceived transferability.
Multi-agent formation control is the key technology of immersive experience in an entertainment system. Differences in kinematics and perception in Heterogeneous Multi-Agent Systems (HMAS) make it difficult to couple cooperative formation and real-time interaction. Multi-Agent Proximal Policy Optimization (MAPPO) Decentralized execution with centralized training addresses partially observable policy learning, but does not model user interaction dynamics. This paper proposes three innovations: heterogeneous attention formation control mechanism uses the relative state coding between agents to generate dynamic weights to realize adaptive role assignment; The interactive reward shaping function integrates the user operation trajectory into the MAPPO global value function to improve the response consistency; The hierarchical policy distillation framework reduces computational complexity by sharing the underlying policy network for homogeneous subgroups and retaining the heterogeneous top-level decision headers. In the simulation experiment containing 10 heterogeneous agents, the average position error of formation decreased to 0.11 m, and the formation holding time accounted for 96.3%. The median user interaction delay was 0.28 s, and the task completion rate was 92.7%. Compared to the original MAPPO, the number of steps required for convergence is reduced by 41%. The three mechanisms described above collaboratively address the triangular contradiction in heterogeneous multi-agent formations from three dimensions. The heterogeneous graph attention control mechanism handles the dynamic heterogeneity among agents, ensuring the formation maintains geometric accuracy even under heterogeneous conditions. The interaction-driven reward shaping mechanism embeds user intent into the policy gradient, guaranteeing consistency between formation responses and user actions. The hierarchical policy distillation mechanism reduces the dimensionality of the policy space through parameter sharing, enabling joint training to converge within an acceptable timeframe. Together, these three mechanisms achieve joint optimization of heterogeneity adaptation, interactive response, and real-time convergence.
Gamification is an active teaching methodology with many applications, ranging from education to corporate environments. Education is one of the most prominent fields for gamification. English language courses can benefit greatly from their implementation, as they often struggle to retain students. Institutions must address issues such as constant demotivation, high dropout rates, and frequent turnover of teaching professionals. This is exacerbated by the non-compulsory nature of foreign language proficiency in Brazil, where learning English is often considered an extracurricular activity. This article presents the results of the development and application of the Expressa game, which is a gamified approach designed to improve engagement in English language teaching and learning. A Systematic Literature Review was initially carried out to identify games used in English language teaching and learning. Following this study, the Expressa game was developed and tested using the case study method in two different classrooms with 26 students to validate its mechanics and dynamics. The results enabled the effectiveness of Expressa as a support tool, as well as its applicability in a teaching and learning environment.
Although the word ‘fun’ is very prevalently used to express positive experiences, researchers have been cautious about applying fun as a research concept due to its all-encompassing and ubiquitous nature. In this paper, we aim to clarify a specific area in the phenomenon of fun by scrutinizing it through the lens of motivation to play video games. In particular, the concept of fun is positioned within a model of motivational development incorporating hedonic (situated enjoyment) and eudaimonic (long-term motivation) factors. We present results from our exploratory survey study (N=903), broadly targeted to video game players in the USA. By using cluster analysis (k-means), we identified six player types according to their gameplay motivations. The analysis of these differently motivated groups revealed that ’fun’ remained the strongest reason to play games for all groups (i.e., Self-Actualized Fun, Bored Fun, Competitive Fun, Instrumental Fun, Compelling Fun, and Nostalgic Fun), even though Affective Engagement, a motive factor incorporating fun and enjoyment, was not a significant determinant in the clustering process. The results are in line with a conceptual distinction between the deeper, needs-based and the more superficial and hedonically driven orientations towards enjoyment. Practical and theoretical implications of these findings are discussed.
Science fiction serves as a powerful medium for exploring future technologies and their societal impact, offering rich opportunities for entertainment experiences that engage audiences with speculative futures. However, existing AI-based story generation approaches often lack systematic grounding in social-cultural analysis, limiting their potential for creating meaningful interactive entertainment content. This paper presented a novel multi-agent system that combines the Archaeological Prototyping (AP) methodology with large language model collaboration to generate science fiction stories with enhanced diversity while maintaining quality. Our approach operated in two phases: first, we constructed an AP model through a three-stage pipeline using evidence-based modeling and multi-agent competitive prediction; second, we employed a hierarchical multi-agent architecture for narrative generation, where an Overseer Agent coordinated Setting and Outline Agents while maintaining consistency with the AP model. Experimental results demonstrated that compared with the baseline method, our multi-agent approach improved story diversity while maintaining quality—a trade-off that temperature-based approaches fail to achieve. The generated narratives can serve as foundations for various entertainment applications, including interactive fiction, games, and trans-media storytelling experiences. Lastly, this architectural pattern, combining deterministic structural grounding with multi-agent ideation, is transferable beyond the AP model to other domain-specific ontologies and structured knowledge frameworks.
The Free-to-Play (F2P) business model relies on in-app purchases, and practitioners encourage early monetization under the assumption that faster conversion improves long-term retention. Whether first-purchase timing is actually protective against churn, however, remains empirically underexamined in longitudinal settings. This study addresses this question using survival analysis of service-long gameplay logs from the mobile idle RPG “Dealers Only Squad,” as a case study of a commercial F2P title. Contrary to the initial hypothesis that immediate converters would exhibit weaker protection due to limited product knowledge, Day-0 converters are associated with the lowest churn risk relative to non-payers (45% reduction over the follow-up window), while later-converting cohorts show smaller and more uniform protective effects—a pattern indicating that early conversion reflects pre-existing user–game fit rather than impulsive purchasing. Product type and price significantly moderate this relationship: immediate converters of mid-priced, low-complexity products show some of the strongest protective effects, whereas late converters of low-priced, low-complexity products show weaker protection than non-payers. These findings suggest that conversion strategies should be tailored to player tenure rather than simply pushing low-price products for rapid conversion.
Interactive systems increasingly use body gestures as input for entertainment, yet experiential value and real-time recognition constraints are often examined separately. This paper investigates finger snapping as an input modality for rhythm games and explores how game design can enhance its experiential value. Study 1 showed that finger snapping had strong experiential potential, with high enjoyment, immersion, flow, and challenge, while interviews suggested that combining snapping with other gestures could further emphasize its unique acoustic and tactile feedback. To support this design, Study 2 evaluated single-channel electromyography (EMG) recognition under short signal windows of 16–256ms, clarifying the trade-off between latency and accuracy. Based on these findings, we developed SnapTune, an integrated multi-gesture rhythm game using differentiated note types and explicit visual feedback. In Study 3, a 256ms window with 8-shot calibration achieved approximately 80% accuracy, improved enjoyment compared with the Study 1 baseline, and was perceived as acceptable by most users.
Physical activity (PA) data are collected and visualized to promote awareness and motivation, yet research on their public presentation remains fragmented across HCI subcommunities. We focus on the public context as a critical yet underexplored setting, where PA data are made visible beyond private interfaces in co-located environments, posing opportunities and challenges for interaction. To build a systematic understanding, we analyze a representative corpus of 30 evaluated samples of PA data public presentations in terms of data input, presentation output, display context, and interactivity. Synthesizing evaluation results across studies, we identify four impacts at the affective, cognitive, behavioral, and social levels. We examine privacy issues in public settings. This work provides the first structured overview of PA data public presentation research across HCI domains. We further discuss nine design considerations and three research agendas, bridging perspectives from PA data presentation, health promotion, and social interaction in public contexts to inform future design and research in this emerging domain.
Draw rates in computer chess tournaments and elite correspondence play have reached levels that diminish both the competitive and entertainment value of the game. The root cause is the near-perfect play enabled by modern neural-network-enhanced engines, which consistently find drawing continuations even from asymmetric positions, reducing the scope for creative and decisive play.We introduce nugget positions – formally defined via a Win–Draw–Loss probability model as positions satisfying a 75% win-probability criterion with near-zero loss probability – and the Delta WDL Indecision Index (ΔWDL), a new measure of strategic difficulty that identifies strategic singularities: positions where neural engines cannot statistically discriminate between candidate moves, exposing the limits of data-driven convergence.We present a systematic mining pipeline that extracts nugget positions from repositories of master-level games, using Stockfish and the Shashin-augmented engine ShashChess as evaluation tools. The pipeline combines engine-based WDL analysis with a set of structural heuristics — including material balance, tactical complexity, draw avoidance, and sensitivity to errors — to isolate positions that are genuinely resistant to draw-forcing tendencies rather than merely unbalanced in a trivial sense.The theoretical framework is grounded in Shashin’s model of chess as a complex system, which classifies positions by the drift of their evaluation over time and provides a natural characterization of the pawn center topologies most likely to generate strategic singularities. Empirical validation through engine tournaments confirms that playing tournaments using nugget positions dramatically reduce draw rates.
