The development of co-creative AI and interaction-centered models of artificial intelligence did not emerge from a single publication or isolated technical breakthrough. Instead, the field evolved gradually through an interdisciplinary research trajectory spanning computational creativity, human-computer interaction, cognitive science, enaction, and adaptive systems research.
Across this body of work, a consistent theme emerged:
creativity and intelligence are not merely properties of isolated agents.
They emerge through interaction.
The following timeline traces the development of these ideas from early investigations into distributed cognition and creativity support tools to later frameworks for co-creative AI, participatory sense-making, quantified interaction dynamics, and enactive AI systems.
The earliest foundations of this research trajectory emerged through investigations into distributed cognition, collaborative creativity, and perceptual interaction systems.
One major early contribution was:
presented at ACM Creativity & Cognition 2011, which won the conference’s Best Student Paper Award.
The paper explored how creativity emerges through coordination between:
people,
digital tools,
media systems,
and collaborative environments.
Rather than treating creativity as an isolated internal process, the work emphasized distributed interaction across human and technological systems.
At the same conference,:
introduced early ideas surrounding:
and collaborative computational art systems.
These early investigations laid the groundwork for later theories involving:
interaction-centered cognition,
creative trajectories,
perceptual attunement,
and co-creative AI systems.
The research trajectory expanded into creativity support systems and human-computer interaction.
Work during this period explored how computational systems might:
augment human creativity,
support ideation,
and participate within creative workflows.
Key publications included:
and:
These works investigated how interactive systems shape:
exploration,
cognitive engagement,
and creative decision-making.
A major conceptual milestone occurred in 2013 with:
presented at AIIDE 2013.
This paper formally proposed that computational systems could function as:
collaborators,
improvisational partners,
and co-creative participants
rather than merely autonomous generators or passive tools.
At the time, this represented a substantial conceptual shift within computational creativity research.
The work introduced early prototypes such as:
which explored collaborative artistic interaction between humans and computational systems in real time. (ojs.aaai.org)
This period marked the beginning of:
as an interaction-centered paradigm.
A major theoretical breakthrough occurred in 2014 with:
presented at the International Conference on Computational Creativity (ICCC). (computationalcreativity.net)
This work introduced:
which synthesized:
enactive cognition,
ecological psychology,
distributed creativity,
improvisation theory,
and participatory interaction
into a new framework for computational creativity and co-creative systems. (turn0search6)
The paper proposed that creative AI systems should function as:
capable of:
interaction,
adaptation,
improvisation,
and collaborative participation.
This work became one of the earliest formal articulations of:
In 2015, these theoretical ideas became operationalized through:
introduced in:
at ACM Creativity & Cognition 2015.
The Drawing Apprentice became one of the earliest true co-creative AI systems.
Unlike conventional drawing software or autonomous generators, the system:
interpreted user sketches,
generated collaborative responses,
adapted dynamically during interaction,
and participated in reciprocal artistic improvisation.
The Drawing Apprentice became both:
a co-creative drawing partner,
and a research platform for studying collaborative interaction itself.
This same year also saw the publication of:
published by Springer.
This work formally extended enactive cognitive science into:
The next phase of the research focused on understanding:
This culminated in:
presented at Intelligent User Interfaces (IUI) 2016. (dl.acm.org)
The paper represented one of the earliest empirical investigations of:
within human-AI interaction.
The research demonstrated that:
interaction rhythms,
turn-taking,
improvisation,
coordination,
and mutual adaptation
play central roles in co-creative collaboration.
Rather than evaluating creativity solely through final outputs, the work shifted attention toward:
This transition became foundational to later frameworks involving:
creative trajectories,
quantified co-creation,
and Creative Sense-Making.
As the Drawing Apprentice evolved, it became increasingly clear that co-creative systems generated:
Every interaction produced measurable histories of:
collaboration,
adaptation,
timing,
coordination,
and creative evolution.
This realization led to:
presented at ACM Creativity & Cognition 2017. (dl.acm.org)
This work introduced:
as a cognitive framework for quantifying co-creative interaction dynamics.
Major contributions included:
creative trajectories,
interaction trace analysis,
quantified collaboration frameworks,
and:
which modeled co-creative interaction continuously through time.
This work helped establish a new subfield focused on:
Subsequent research expanded co-creative AI into:
conceptual blending,
novelty generation,
explainable co-creation,
and computational models of design creativity.
Important works included:
and:
This work explored how co-creative systems could:
intentionally generate conceptual shifts,
support divergent ideation,
and evaluate collaborative creativity dynamically.
The research helped expand co-creative AI into broader design and hybrid intelligence contexts.
More recent work increasingly generalized these ideas into broader theories of:
interaction-centered AI,
adaptive systems,
hybrid intelligence,
and human-AI co-regulation.
The framework evolved beyond drawing systems into larger investigations of:
participatory cognition,
creative dynamics,
interaction viability,
and adaptive coupling between humans and AI systems. (turn0search0)
This work culminated in:
which won Best Paper at ICCC 2024. (computationalcreativity.net)
The paper formalized:
through five pillars:
autonomy,
sense-making,
embodiment,
emergence,
and experience.
The work positioned enaction as a foundational paradigm for understanding:
This trajectory culminated in:
published in the Handbook of Human-Centered Artificial Intelligence. (turn0search6)
The chapter argued that co-creative AI represents more than a niche subfield.
It represents:
Rather than treating AI systems solely as:
tools,
generators,
or autonomous agents,
the co-creative paradigm emphasizes:
participation,
adaptation,
collaboration,
improvisation,
and shared meaning construction.
This interaction-centered perspective increasingly aligns with broader developments in:
human-centered AI,
hybrid intelligence,
participatory AI,
embodied cognition,
and adaptive systems research.
Across this research trajectory, a consistent shift can be observed:
From:
isolated computation,
static evaluation,
and autonomous generation
Toward:
interaction,
participation,
co-regulation,
and emergent collaboration.
This body of work helped establish many foundational ideas now shaping modern co-creative AI research:
artistic computer colleagues,
participatory sense-making,
creative trajectories,
quantified co-creation,
enactive AI,
and human-AI co-creation as an interaction paradigm.
The central insight connecting these works remains consistent:
intelligence and creativity do not emerge solely inside isolated humans or machines.
They emerge through interaction between dynamically coupled participants engaged in shared processes of sense-making. (nickmdavis.com)