Design methodology of analytical games for knowledge acquisition

Abstract

Analytical games explore a problem or a domain with a research purpose. Considerable research is ongoing to investigate improvements to analytical game design, execution and exploitation. Moreover, the fast-paced technological developments in many fields, such as artificial intelligence and virtual reality, make it even more compellingto account for the advantages and limitations of these new capabilities. In game design, the use of digital means is often regarded as a mere technical factor that relates to the platform selection, facilitator support and data recording processes. In this work a shift in perspective is proposed, to move from technology-oriented design selection criteria towards a broader assessment of the design choices. In fact, the introductionof technology (i.e., automation and autonomy) will not lead to a substitution of tasks, but will intrinsically change the game environment. This work introduces a framework to provide a structured guidance on the aspects to be factored in the different design phases of an analytical game, including the potential impact of the adoption of automation and autonomy. The proposed approach is based on previous research in the fieldof simulation-based serious gaming, model-driven engineering and human factors engineering. The framework is applied to Knowledge Acquisition Analytical Games as a case study.

The Value of Wargames and Tabletop Exercises as Naturalistic Tools

“…• • Participants in both events espoused various "soft benefits" (positive and often intangible outcomes that were not explicitly sought after by game developers), such as broadened perspectives and challenged assumptions. This finding aligned with results from other wargames (e.g., Dorton et al, 2017;Dorton et al, 2020;de Rosa & Strode, 2022b).…”

Foresight for ethical AI

“…Premortems do not guarantee an exhaustive set of risks will be identified (Bettin et al, 2022 ). Similarly, methods such as analytic games (e.g., de Rosa and De Gloria, 2021 ) and other checklists based on naturalistic inquiry (e.g., Dorton, 2022 ) are bounded in scope by not only the scenarios and injects examined, but also the expertise of the participants involved in their application. More generally, there is a long-standing challenge of developing tools for an envisioned world with technological change (e.g., Woods and Dekker, 2010 ).…”

“…While the premortem is a sound candidate for the challenge that we present, it should be noted that it is not the only NDM approach that could be employed to identify AI-driven harms. Recent work has shown that analytic wargames involving experienced players in naturalistic settings can be used to test disruptive technologies and uncover emergent behaviors (de Rosa and De Gloria, 2021 ), and to explore how the work system may stretch with the introduction of new technologies (Dorton et al, 2020 ). Other approaches have included the use of naturalistic methods such as the critical incident technique to develop evidence-based checklists for AI developers (e.g., Dorton, 2022 ).…”

Editorial Vol. 8, Issue 4

“…"Design methodology of analytical games for knowledge acquisition", by De Rosa and De Gloria [1], introduces a framework to provide a structured guidance on the aspects to be factored in the different design phases of an analytical game, including the potential impact of the adoption of automation and autonomy. The proposed approach is based on previous research in the field of simulation-based serious gaming, model-driven engineering and human factors engineering.…”