Perceptions of Interaction Dynamics in Co-Creative AI: A Comparative Study of Interaction Modalities in Drawcto

摘要

This paper explores how different interaction modalities with AI agents affect human perception of the co-creative process. We utilize the Wizard of Oz methodology within Drawcto, a co-creative drawing system, to examine co-creativity across three scenarios: human-human, human-robot, and human-software interactions. Using a mixed-methods approach, we combined insights using the Observable Creative Sensemaking (OCSM) method with data from structured interviews. The study involved 20 participants engaging in a collaborative drawing task under each interaction scenario. Key findings reveal an average OCSM curve indicative of typical human-human interactions, varied themes in human-AI collaboration, and a notable influence of AI embodiment on participant perceptions, with the robot interface resembling human-human collaboration more closely than the software interface. Overall, this research offers valuable insights into how different interaction modalities influence the perceived role of AI in the co-creative process and provides design considerations for enhancing human-AI co-creative interactions.