From Instructor to Collaborator: What a 90-Participant Study Reveals about Human-Agent Collaboration in a Mobile Serious Game
arXiv:2605.27384v1 Announce Type: cross Abstract: This position paper reflects empirical data collected during my PhD from a large-scale within-subjects study (N = 90). The study compared a highly human-like, spoken embodied conversational agent (ECA) against a low human-like text base agent (no embodiment, text bubble only) within a mobile, Unity-developed game about pre-decimal UK currency. The game included two agents with different roles-an Instructor (Alex) and a Shopkeeper/Collaborator. Users interacted using voice and mouse input. The quantitative data I collected included a usability q
The proliferation of AI agents necessitates deeper empirical understanding of human-AI collaboration dynamics to optimize their integration into practical applications.
This study provides early empirical data on the efficacy of different AI agent embodiments in collaborative scenarios, which is crucial for the design and deployment of future AI systems.
Our understanding of preferred human-AI interaction modalities is refined, suggesting that human-like agents may not always be superior and that specific roles impact perceived utility.
- · AI agent developers
- · UX designers
- · Gaming industry
- · Human-computer interaction researchers
- · Developers solely focused on fully human-like AI agents without empirical valida
- · Legacy interaction design paradigms
Further research into specific design parameters for AI agents in collaborative settings will accelerate.
Optimized AI agent designs will lead to more effective enterprise applications and educational tools.
Widespread adoption of specialized AI agents could redefine workflows across various industries, enhancing productivity and necessitating new training paradigms for human workers.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI