Improving Decision Support Systems and Disruptive Technology Adoption With Analytical Serious Games

Introduction

Serious games (SGs) are games for which the primary purpose is not entertainment (Abt, 1987). They can serve several purposes and a widely adopted taxonomy (Djaouti et al., 2011) distinguishes between: (i) games for education and training, (ii) games for awareness raising and communication and (iii) games for data exchange. The latter are the overall category of games used to collect data regarding how participants interact between them and with the game environment for further analysis. Therefore, this category includes serious games used for analytical and experimentation purposes. The term analytical games will be used in the remainder of the chapter to identify both categories. There is a growing attention towards SGs focused on the improvement of the decision-making processes in many domains, not only by providing training opportunities, but also as a technique to observe the decision-making processes in controlled settings to better understand how decision are taken in uncertain and volatile environments. Applications include defence, security, business, emergency preparedness and emergency response. More recently analytical games have demonstrated utility not only to understand the decision-making process at strategical, operational and tactical level, but also as a knowledge engineering support. The term knowledge engineering refers to the set of processes that relate to the extraction, structuring and use of knowledge to create systems able to reason about a specific problem (e.g., artificial intelligence systems). The term knowledge acquisition (KA) identifies the processes to acquire such knowledge from humans. In the last decades researchers have started experimenting with serious games as a support to KA. Applications included the discovery in text narratives relations between entities (Kondreddi et al., 2013), ontology alignment tasks (Thaler et al., 2011), image annotation (Morrison et al., 2010) and semantic web (Markotschi & Völker, 2010). The Knowledge Acquisition Analytical Game (K2AG) approach has been introduced (de Rosa, 2020) to support the design of intelligent systems. Successful K2AGs, include the Reliability Game (de Rosa et al., 2018; de Rosa et al., 2019) and the MARISA Game (de Rosa & De Gloria, 2020), where data has been used to design multi-source information fusion algorithms to be employed in intelligent systems (e.g., behavioral analysis services) on the basis of cognitive mimetic principles. K2AGs have proven to be very efficient and effective in terms of knowledge acquisition (i.e., time reduction, experiment simplicity and ability to extract the required qualitative and quantitative knowledge). In this chapter the authors discuss the results obtained through the use of the Maritime Unmanned System Trust (MUST) Game (de Rosa & Strode, 2022a, 20022b), which is a K2AG exploring the use of uncrewed systems¹ in maritime operations. The MUST Game is an analytical game which captures beliefs, attitude and perspectives of the participants with respect to the employment of MUS. This game aims at better understanding the relation between trust factors and MUS in different missions. The game explores how players make decisions with respect to MUS deployments as the scenario threat level increases. This allows to capture the important information on the trade-offs related to their use and the impact on mission planning activities. The game design followed a model-driven engineering approach (Wagner, 2018; de Rosa & De Gloria, 2021) that circled through the following design phases: (i) conceptual design, (ii) model design and (iii) implementation design. An accurate conceptual game design, does not only improve the quality of the overall SGs, but becomes a key element to create bridges towards other scientific fields, enhancing the ability of the SGs to improve the design of future systems that support human decision-making in different operational environments and the adoption of new technologies, such as emergent and disruptive ones (Armstrong, 2017) for which a clear uptake pathway still has to be identified.