Creative Sense-Making: A Cognitive Framework for Quantifying Interaction Dynamics in Co-Creation

Abstract

Collaboration is a powerful way to inspire and support creativity. The field of computational creativity is beginning to explore how co-creative agents might collaborate with humans during their creative process in co-creative systems. A cognitive theory or framework describing collaborative creativity would help design and evaluate co-creative systems. In particular, measuring collaboration outcomes involves being able to quantify interaction dynamics, e.g. the rhythm of interaction, style of turn taking, and manner in which participants are mutually making sense of a situation through time. However, there is a gap in the literature about quantifying and evaluating open-ended creative collaboration.

This dissertation extends the cognitive science theory of enaction and its conceptual framework, participatory sense-making, to the domain of open-ended creative collaboration to formalize this theory in computational models of creative collaboration. The proposed qualitative coding and analysis technique provide a means to rapidly and reliably quantify interaction dynamics continuously through time. This temporal data can be mathematically analyzed using continuous functions (e.g. moving averages, integrations) to classify different sense-making strategies and trends in collaboration.

The proposed creative sense-making framework is applied to empirical studies of human collaboration (in the domain of pretend play) and technical systems (in the domain of collaborative drawing) to establish its validity through cross-domain application and inter-rater reliability within each domain. A web-based co-creative drawing agent called the Drawing Apprentice was created as a technical probe to explore technical approaches and interaction designs to facilitate participatory sense-making in co-creative systems.