Generative artificial intelligence (AI) is reshaping creative work; however, its integration is neither uniform nor inevitable. We examine video game development, a multidisciplinary domain spanning art, programming, narrative, and design, as a critical site where workers actively negotiate the AI integration. Drawing on interviews with game creatives and informed by a political economy lens, we trace the structural conditions shaping AI adoption, the strategies creatives employ to selectively incorporate AI while maintaining critical distance, and the tensions produced by capitalist imperatives that provoke their resistance to AI. We articulate design tensions and the politics of AI integration in game development and adjacent creative domains. In doing so, our study intervenes in AI design discourse by challenging narratives of technological inevitability as well as labor and value extraction, towards envisioning possibilities for creative AI grounded in human agency.

Playtesting is a games user research (GUR) method used to evaluate design decisions based on feedback gathered from players with the goal to improve player experience. HCI games research has been actively working on and promoting best practices in GUR. However, these practices often require resources, knowledge and expertise, which are not readily available for indie video games developers. Thus, to better understand how GUR can support these developers, we conducted an interview study with 13 indie games professionals to learn about their practices and the challenges they face when doing playtesting. We report on the key findings from this study, including challenges with finding appropriate participants and handling the data from playtests. We provide a discussion of how existing GUR practices can be adapted and what HCI games research can do to help mitigate these challenges to make playtesting more accessible and impactful to indie video games developers.

Despite the advantages of using expert evaluation as a method within games user research (GUR) (i.e. provides stakeholders low cost, rapid feedback), it does not always accurately reflect the general player’s experience. Testing the game out with real users (also called playtesting) helps bridge this gap by giving game developers an in-depth look into the player experience. However, playtesting is resource intensive and time consuming, making it difficult to implement within the tight time frames of industry game development. AI can help to mitigate some of these issues by providing an automated way to simulate player behaviour and experience. In this paper, we introduce a tool called PathOS+—a playtesting interface which uses AI playtesting data to help enhance expert evaluation. Results from a study conducted with expert participants shows how PathOS+ could contribute to game design and assist developers and researchers in conducting expert evaluations. This is an important contribution as it provides game user researchers and designers with a fast, low-cost and effective game evaluation approach which has the potential to make game evaluation more accessible to indie and smaller game studios.

Game studios aim to develop titles that deliver a fun and engaging experience for players. Playtesting promises to help identify opportunities to improve player experience and assist developers in achieving their design intent. However, a lack of research on the added value of playtesting means that many studios are still uncertain about its commercial viability and impact on product success. This gap in understanding is further complicated by the vague definition of "success" afforded by sales figures and review scores. In this paper, we assess reported feature quality of three commercial titles by analyzing playtesting reports and game reviews. By comparing themes and design issues expressed in game reviews to the results of pre-release playtesting for each game, we aim to highlight the value of playtesting and propose a set of guidelines for selecting playtest methods based on the needs of a given game evaluation. Through the real-world case studies presented, this paper contributes to the growing domain of games user research and highlights the value of playtesting in game development.

Board games present accessibility barriers for players with visual impairment since they often employ visuals alone to communicate gameplay information. Our research focuses on board game accessibility for those with visual impairment. This paper describes a three-phase study conducted to develop board game accessibility adaptation guidelines. These guidelines were developed through a user-centered design approach that included in-depth interviews and a series of user studies using two adapted board games. Our findings indicate that participants with and without visual impairment were able to play the adapted games, exhibiting a balanced experience whereby participants had complete autonomy and were provided with equal chances of victory. Our paper also contributes to the game and accessibility communities through the development of adaptation guidelines that allow board games to become inclusive irrespective of a player's visual impairment.

Playtesting is a key component in the game development process aimed at improving the quality of games through the collection of gameplay data and identification of design issues. Visualization techniques are currently being employed to help integrate quantitative and qualitative data. Despite that, two existing challenges are to determine the level of detail to be presented to developers based on their needs and to effectively communicate the collected data so that informed design changes can be reached. In this paper, we first propose an aggregated visualization technique that makes use of clustering, territory tessellation, and trajectory aggregation to simultaneously display mixed playtesting data. Secondly, to assess the usefulness of our technique we evaluate it through interviews with professional game developers and compare it to a non-aggregated visualization. The results of this study also provide an important contribution towards identifying areas of improvement in the portrayal of gameplay data.

The influx of investment in virtual and augmented reality in recent years has brought with it a renewed interest in spatial audio. Spatial audio in virtual environments and games has been neglected previously due in part due to computationally expensive nature of the processing involved. Here we introduce G-SpAR, a GPU-based spatial audio renderer for the Unity3D game engine. The system develops a pathfinding method that shortens occlusion and obstruction, increasing performance in-game.

Exertion games, also referred to as exergames, have become popular because they combine physical activity (PA) with game mechanics, such as actions, challenges, and achievements. Exergames have been also used to encourage PA among older adults, as technological interventions to help achieve the latters’ health and wellness goals and as aids to rehabilitation. To the best of our knowledge, no systematic review of empirical studies on exergaming and older adults’ PA has been reported in the literature. Our review indicates that exergames make a measurable contribution to the improvement of health and wellness goals of older adults. Our systematic review identifies 9 categories and 19 themes of exergame applications in the domain of older adults’ PA. We aggregate these categories and themes into three broader exergaming clusters, of “training,” “rehabilitation,” and “wellness.” Additionally, we outline pathways for future empirical research into applying exergames as health and wellness interventions for older adults through physical activities.

Visualization techniques can facilitate the understanding and exploration of relationships in usertesting data. For example, data from players' in-game movement can be combined with interview data or questionnaire results. However, the process of amalgamation is not straightforward, because the underlying data often exists in different formats. Another challenge is making these visualizations simple enough to provide a quick overview for producers, but also detailed enough to be usable and practical for gameplay programmers. Although various visualization techniques have already been introduced in this domain, most of these techniques focus on displaying large amounts of quantitative telemetry data without integrating qualitative or contextual data on player experience. Moreover, most of the current visualizations are static representations of usertesting data, so they cannot dynamically adjust to users' (e.g. producers, programmers) needs. Hence, there is a need for an interactive visualization tool that can adjust data representation based on the nature and detail level of data required from different members of a development team. This paper reports our current development efforts on a tool that assists data collection and provides a dynamic and interactive representation of usertesting data. We also report two initial studies to evaluate the effectiveness of the tool with game developers to guide our future development.

Improving game design is a hard task. Few methods are available in games user research (GUR) to test formally how game designs work for players. In particular, the usefulness of user tests (UTs) for game designers has not been fully studied in the CHI community. We propose a novel GUR method called Biometric Storyboards (BioSt) and present a study demonstrating how a Classic UT and a BioSt UT both help designers create a better gameplay experience. In addition, we show that BioSt can help designers deliver significantly better visuals, more fun, and higher gameplay quality than designing without UTs and that classic UTs do not provide this significant advantage. Our interviews support the idea that BioSt provides more nuanced game design improvement. The design implication is that a game designed with the BioSt method will result in high gameplay quality.