The Martian Game: Exploring Open-Ended Question-Asking via an Online Gaming Environment

Abstract: Question asking is a crucial skill in human communication, influencing social cognition, creative problem solving, and information seeking. Yet, its cognitive mechanisms remain poorly understood due to challenges in studying it naturally. We developed The Martian Game, an open-ended online question-asking game that simulates open-ended problem solving in realistic contexts. Players are challenged to design a solar energy system for a Martian city through two stages: (1) a problem finding phase, where they ask an AI chatbot (“Mark”) questions to gather information; and (2) a solution-planning phase, where they write their proposed solution, which is also visually generated to them. Questions are coded for complexity, originality, and relevance; solutions are rated for originality and appropriateness, and compared to different solution templates. This game offers an ecologically valid, interdisciplinary tool to study question asking and supports the hypothesis that complex questions promote effective problem solving. A pilot online study (N = 105) validated the game, as well as related game performance to relevant cognitive capacities by measuring participants individual differences in creativity, intelligence, curiosity, and question asking capacities. We find that better performance in the problem findings stage led to more creative and novel solutions. Furthermore, individual differences in creativity, intelligence, curiosity, and question asking capacities facilitated gameplay. Overall, our game allows investigating open-ended problem solving in natural settings, thus highlighting its potential in examining complex behavior.