One of the most important shifts in the history of artificial intelligence has been the gradual movement away from isolated computation and toward interaction-centered models of cognition.
Traditional AI systems have often been designed as tools, generators, classifiers, or autonomous agents. In these paradigms, intelligence is typically understood as something occurring internally within the system itself.
However, co-creative AI introduced a fundamentally different possibility: intelligence and meaning may emerge through interaction between humans and AI systems.
This idea lies at the heart of participatory sense-making.
Participatory sense-making originates from enactive cognitive science and describes how meaning emerges dynamically through social interaction between coupled agents. Rather than treating cognition as isolated internal processing alone, participatory sense-making proposes that interaction itself can become constitutive of cognition.
This perspective became especially important in the development of co-creative AI systems.
As humans and AI systems collaborate creatively, meaning does not simply originate from one participant and transfer to another. Instead, interpretation evolves dynamically, actions reshape future possibilities, collaborators adapt continuously, and creativity emerges through participation itself.
These ideas became central to the 2016 paper:
Empirically Studying Participatory Sense-Making in Abstract Drawing with a Co-Creative Cognitive Agent.
The paper represented one of the earliest empirical investigations of participatory sense-making within a human-AI creative system.
Participatory sense-making originally emerged within enactive theories of social cognition developed by Hanne De Jaegher and Ezequiel Di Paolo.
Their work proposed that social understanding is not merely constructed internally within isolated individuals. Instead, interaction dynamics themselves contribute directly to the construction of meaning.
This was a profound shift.
Rather than viewing communication as the exchange of pre-existing representations, participatory sense-making emphasized coordination, interaction rhythms, coupling, mutual adaptation, and shared activity.
The 2016 Drawing Apprentice research extended these ideas into human-AI co-creation.
The central question became:
Can humans and AI systems participate together in shared processes of sense-making?
Rather than merely producing outputs, could AI systems engage dynamically in the emergence of meaning itself?
To explore these questions, Nicholas Davis and collaborators developed:
The Drawing Apprentice was designed as a co-creative drawing partner capable of improvisational collaboration with human users in real time.
Unlike traditional drawing software, the system actively participated during the creative process itself.
The agent interpreted user sketches, generated visual responses, adapted dynamically to interaction context, and collaborated through reciprocal turn-taking. The interaction unfolded as an evolving dialogue between human and AI contributions.
Importantly, the system was not designed merely to complete drawings autonomously. Instead, the goal was to create a participatory creative interaction.
This distinction became foundational to later co-creative AI research.
One of the central insights of participatory sense-making is that meaning emerges through interaction itself.
In co-creative drawing sessions, users continuously interpreted AI responses, interaction rhythms, visual transformations, ambiguity, and evolving artistic trajectories. At the same time, the AI system responded dynamically to user actions.
This created a feedback loop in which human actions reshaped AI behavior, AI responses reorganized human interpretation, and new creative possibilities emerged recursively through participation.
Meaning therefore did not exist fully formed inside either collaborator alone.
Instead, meaning emerged through interaction between them.
The 2016 study empirically investigated these dynamics through collaborative abstract drawing sessions between users and the Drawing Apprentice.
The findings demonstrated that participants often interpreted the AI as improvisational, responsive, collaborative, and creatively engaged.
This suggested that participatory sense-making could emerge even within human-AI creative interaction.
The choice to study abstract drawing was especially important.
Abstract drawing creates a highly open-ended creative environment in which ambiguity remains high, interpretations shift continuously, and meaning must emerge dynamically through interaction.
This made abstract drawing an ideal environment for studying participatory sense-making.
Because the sketches were not rigidly predefined, participants continuously negotiated interpretation, intention, style, and direction collaboratively with the AI system.
The resulting creative process resembled improvisational dialogue more than conventional tool usage.
This was one of the earliest demonstrations that AI systems could participate in emergent collaborative meaning-making.
A major finding of the research was the importance of turn-taking dynamics.
Creativity did not emerge simply because the AI generated images. Instead, creativity emerged through reciprocal exchange, timing, adaptation, responsiveness, and evolving coordination between participants.
The interaction created moments of coupling, divergence, surprise, reinterpretation, and re-coordination.
These dynamics closely mirrored patterns found in human-human improvisational collaboration.
The research therefore suggested that co-creative AI systems should not be evaluated solely through final outputs. Instead, the dynamics of interaction itself become central to understanding creativity and collaboration.
This insight later became foundational to:
and the broader effort to quantify interaction dynamics in co-creative systems.
The broader significance of participatory sense-making extends far beyond drawing systems alone.
Traditional AI systems often treat humans as users, operators, or external input sources.
Participatory sense-making instead suggests that humans and AI systems may become dynamically coupled participants in shared cognitive processes.
This reframes artificial intelligence itself.
Rather than focusing solely on prediction, generation, optimization, or automation, participatory AI systems emphasize interaction, collaboration, adaptation, coordination, and shared emergence.
This interaction-centered perspective increasingly aligns with modern developments in human-centered AI, hybrid intelligence, interactive machine learning, embodied cognition, and co-creative systems research.
The early work on participatory sense-making helped establish many ideas that continue shaping co-creative AI research today, including interaction-centered evaluation, collaborative cognition, adaptive co-creation, turn-taking systems, creative trajectories, and dynamic coupling.
Importantly, the research demonstrated that the creative potential of AI systems may depend less on isolated intelligence and more on the quality of interaction they enable.
This insight remains increasingly relevant as modern AI systems become more conversational, adaptive, and integrated into human creative practice.
Participatory sense-making therefore represents more than a theory of collaboration.
It represents a fundamentally different way of understanding intelligence itself.
From this perspective, meaning emerges through participation, creativity emerges through interaction, and intelligence emerges through dynamically coupled systems engaged in shared processes of sense-making.