ChatGPT changed how we interact with AI, but its linear chat interface fundamentally misaligns with how creative professionals actually think and work. Designers, artists, writers, and researchers don't move forward in straight lines—they explore multiple parallel paths, compare variations side-by-side, revisit abandoned ideas, and build on creative branches that lead nowhere initially but prove valuable later.
This case study documents Ooha, a spatial canvas interface that reimagines AI interaction as non-linear exploration. Rather than fighting against sequential chat conversations, Ooha enables creative professionals to generate multiple response branches from a single prompt, visually compare variations across different AI models (GPT-4, Claude, Gemini), and navigate their entire creative exploration spatially.
Through rigorous co-design workshops and evaluation studies with UCSC creative professionals, we proved that spatial interfaces dramatically enhance AI-assisted creativity—yielding 340% more creative variations, 62% faster solutions, and 43% reduced cognitive load compared to traditional chat interfaces.
The Problem
"My students are creating incredible work with AI, but they're fighting against the interface every step of the way."— Experimental Media Instructor, UCSC
Current AI interfaces force creative professionals into sequential conversations. Every interaction requires choosing one path, abandoning others, and losing the ability to compare scenarios simultaneously. A designer exploring color schemes for a logo can't see GPT-4's suggestions next to Claude's responses. They must choose one conversation, generate one variation, then start over to see anything else.
Visual thinkers also lose spatial context. Creative professionals rely on spatial memory, remembering ideas by their position on a page, in a sketchbook, across a wall of sticky notes. Linear chat interfaces remove this anchoring entirely, forcing users to rely on conversation history that scrolls out of view.
Our initial survey of 87 creative professionals found that 76% reported chat interfaces "significantly hindered" their creative process. This is not an AI capability problem. It's an interface problem.
My Approach
Phase 1 — Co-Design Workshops
Three iterative workshops with 10 creative professionals from UCSC to map spatial thinking patterns and design the branching interface paradigm.
Phase 2 — Prototype Development
Built a full-stack application integrating GPT-4, Claude, and Gemini APIs with a spatial canvas interface enabling multi-path exploration.
Phase 3 — Rigorous Evaluation
Week-long study with 20 participants comparing Ooha against traditional chat interfaces across quantitative metrics and qualitative insights.
Phase 4 — Open-Source Launch
Deployed at ooha-creator.github.io with active community engagement and ongoing feature development.
Understanding the Creative Process
The UCSC HCI program, with its emphasis on experimental and computational arts, provided an ideal context for exploring alternative AI interaction paradigms. The students and faculty there weren't just using AI—they were pushing boundaries of what creative AI interaction could become.
Our research began with a simple question: if creative professionals naturally work spatially—sketching thumbnails across pages, pinning inspiration to walls, arranging concepts in design software—why would we force them into linear chat conversations? The answer wasn't just about interface design; it was about fundamental cognitive misalignment.
Research Methodology
Co-Design Workshops
We recruited 10 participants from the UCSC HCI community for a series of three co-design workshops held over two weeks in June 2025. Each workshop lasted 3 hours and focused on progressively refining an interface concept.
.png)
The first workshop began with participants mapping their current AI workflows. What emerged was striking: every single participant drew branching tree structures to represent their ideal creative process. "This is how my brain works," one participant explained, pointing to the interconnected web of ideas. "I need to see all the paths I didn't take, because sometimes I want to go back and explore them later."
During the second workshop, we introduced low-fidelity prototypes based on the participants' diagrams. Participants could physically arrange and rearrange nodes representing AI interactions. The session revealed critical insights about spatial memory in creative work—participants consistently placed related concepts in proximity and developed personal organizational systems.
The third workshop introduced a functional prototype of the AI Branching Canvas. Participants were given creative briefs—design a poster for a fictional event, create a narrative for a short film, develop a concept for an interactive installation—and asked to use the prototype to explore solutions.
Evaluation Study
Following the co-design phase, we recruited 10 new participants from UCSC for a formal evaluation study conducted over one week in September 2025. This group included 3 self-identified Artists, 4 graduate students, and 3 advanced undergraduates.
Each participant attended two 90-minute sessions. The first session introduced the AI Branching Canvas through a brief tutorial followed by free exploration time. We used a minimal instruction approach to observe how intuitive users found the interface. The second session involved structured creative tasks designed to test specific aspects of the system.
Key Findings
Spatial Memory Enhances Recall
8 out of 10 participants could accurately describe content from nodes they had created 48 hours earlier by referencing their spatial location. "The spatial layout becomes a map of my creative journey."
Parallel Exploration Changes Strategy
Participants generated 67% more prompt variations but each prompt was 34% shorter. Instead of crafting "perfect" prompts, they adopted a more experimental, divergent approach.
Visual Comparison Accelerates Decisions
"When I see all the options together, I immediately know which direction feels right. It's intuitive, like choosing between sketches on a wall."
Design Evolution
The feedback from UCSC participants directly influenced several critical design decisions. The initial prototype used a rigid grid layout for nodes, but artists consistently broke this structure, manually repositioning nodes to create meaningful spatial relationships. This led us to implement a fully free-form canvas.
The branching mechanism underwent significant refinement. Our original design required users to explicitly create a branch through a menu option. However, 7 out of 10 participants instinctively tried to click or drag from existing nodes to create connections—leading to the hover-based branch button.
Color and visual hierarchy proved more contentious than expected. Artists had strong, conflicting preferences. Our solution: a subtle, monochromatic design with optional color customization hidden in advanced settings.
Design Iteration
The journey from initial concept to the final Ooha interface involved multiple design iterations, each informed by user feedback and technical constraints. Our early sketches explored fundamentally different interaction paradigms.
Our first major iteration focused on structured use case organization through a dropdown interface. While this approach provided clear organization, co-design participants immediately identified a critical limitation: artists don't work in discrete categories—they blend text, images, and video in fluid exploration.
The second iteration shifted toward horizontal growth patterns with distinct processing paths. Participants appreciated the visual branching concept but noted that the separation between video and image paths felt artificial. Creative professionals wanted to explore all content types within a unified spatial context.
The final design synthesized insights from earlier iterations while addressing their limitations. We created a unified spatial canvas where text, video, and images coexist naturally. The branching mechanism supports multi-directional exploration while maintaining the "neat progression" principle.
Quantitative Results
We collected comprehensive metrics during the evaluation phase. Participants completed identical creative tasks using both the AI Branching Canvas and a traditional chat interface.
Qualitative Insights
"The branching canvas turned my creative process from a linear march toward a single goal into a genuine exploration."— MFA Student, Digital Arts
Beyond the quantitative metrics, qualitative feedback revealed profound shifts in how participants conceptualized AI-assisted creation. The spatial interface didn't just change how they interacted with AI; it changed how they thought about the creative process itself.
A graduate student in interactive media drew parallels to her existing practice: "This feels like how I work in After Effects or Touch Designer—I can see my entire node graph, understand the relationships, and jump between different parts of my project instantly. It's the first AI interface that feels native to how visual artists think."
Challenges & Limitations
Three participants hit a wall early on with the freedom of the spatial canvas. The problem wasn't the interface itself; it was the absence of a starting point. Without any prompt or scaffolding, they would open a blank canvas and stall. One undergraduate sat for nearly four minutes before typing anything. Another kept creating nodes and immediately deleting them. I noticed this wasn't decision paralysis in the conventional sense; they weren't overwhelmed by too many choices so much as unsure what shape their thinking should take.
I responded with two changes. First, I added a set of "seed prompts" that appear on an empty canvas, suggesting broad starting moves ("Try branching from a single idea," "Compare how two models respond to the same question"). These disappear as soon as the user adds their first node. Second, I introduced a lightweight onboarding flow, three short interactive steps that walk new users through creating a branch and comparing two outputs side by side. It takes about ninety seconds and is skippable. In follow-up sessions with the same three participants, none of them stalled.
Technical limitations also emerged during intensive use. When participants created more than 100 nodes in a single session, performance degradation became noticeable on older hardware, which points to real optimization work that would need to happen before Ooha could scale to professional workflows at that density.
The evaluation sample itself is a limitation worth naming directly. Twenty participants, all from UCSC's HCI and arts programs, are a narrow slice. They came in with above-average comfort with open-ended tools and a baseline familiarity with AI. What looked like intuitive spatial reasoning may partly reflect that context. To validate the core findings more broadly, I would want to run a parallel study with creatives who have no HCI background, people working in fields like fashion, architecture, or film production, and compare the onboarding curves and output quality. A longitudinal study over several weeks would also matter; a one-week evaluation captures first impressions, not settled habits.
Future Directions
Participants consistently requested collaborative features, envisioning shared canvases where multiple artists could explore together in real-time. This aligns with the collaborative nature of many arts programs and could transform AI brainstorming from a solitary to a communal activity.
Integration with existing creative tools emerged as another priority. Seven participants explicitly requested the ability to export branching explorations to Figma, Adobe Creative Suite, or code repositories—the canvas serving as a creative preprocessing layer.
Design Principles
For AI Tool Designers
Spatial interfaces align naturally with creative cognition. The 43% reduction in cognitive load was about matching the interface to how visual thinkers naturally process information. Enable parallel comparison—the 340% increase came from eliminating the "fear of losing context."
For Educators
The branching paradigm offers pedagogical value. Students could visualize decision paths and understand how different prompts lead to different outcomes. They learned more through visual comparison than through text tutorials.
For Researchers
Interface paradigms significantly impact AI utility. The dramatic improvements suggest that AI capability alone doesn't determine effectiveness—interface design matters enormously. Co-design reveals fundamental design requirements that assumptions would miss.
Conclusion
Ooha demonstrated that reimagining AI interface paradigms can fundamentally transform how humans interact with AI systems. By aligning the interface with natural creative processes, branching exploration, spatial organization, and parallel comparison, we enabled artists to work with AI rather than against it.
Use in the Wild
Since the open-source release, I've been watching how people actually use Ooha, and the pattern that stands out most is how far the use cases drift from what I was testing for. The study was built around UCSC creatives doing defined creative tasks. What I didn't anticipate was the range of things people would reach for once the tool was public.
A meaningful portion of users are generating marketing content, branching from a single product description to explore tone, audience, and format simultaneously. Others are building out story structures, treating the canvas like a narrative map where each branch is a scene or plot direction they might return to. I've seen users describe using it for trip planning, branching from a destination into logistics, activities, and itinerary drafts in parallel. None of these were tasks I designed for or even thought to ask about during the study.
What's interesting is that these use cases share an underlying shape: they all involve exploring a space of possibilities before committing to any single path. That's what the original UCSC participants needed too, just in a different domain. The interface turned out to be more general than the research context suggested. That's worth building on deliberately rather than just observing after the fact.
As AI capabilities continue advancing, interface design will become an even more important differentiator. Paradigm-level thinking about interaction can create improvements in effectiveness that incremental feature additions cannot. The gap between what AI can do and what people actually get out of it is, in many cases, an interface problem.
The next step is understanding which of those wild use cases are stable enough to design for explicitly, and whether the onboarding and spatial affordances that worked for UCSC arts students hold up when the person sitting down is a marketing manager or a travel blogger with no interest in AI at all.
Skills & Methods Demonstrated
UX Research, Co-Design Workshops, Usability Testing, Qualitative Analysis, Mixed-Methods
Interaction Design, Interface Design, User-Centered Design, Prototyping, Information Architecture
Full-Stack Development, React & Node.js, LLM Integration, API Development, Open Source
Quantitative Analysis, Think-Aloud Protocol, NASA-TLX, Statistical Analysis