Thematic analysis as interaction model and research method
What if we modeled human-AI interaction on thematic analysis rather than conversation? And what if we used that same method to discover themes in a knowledge garden?
Thematic analysis as interaction model and research method
Note from Maaike: “I actually made a typo here, I didn’t mean thematic analysis as interaction model, but research method. But now Claude suggested an interaction model that actually opens up new ideas for my ‘how to replace conversation as a metaphor’. Discover might be an interesting verb to explore.”
As a research method
In thematic analysis, you work through data iteratively: reading, coding, grouping, refining themes, going back to check. With a large enough context window, an AI can ingest an entire content base in one go. What would be a good method to discover existing and emerging themes across all that content? Thematic analysis is a conventional, well-researched method for exactly that.
As an interaction model
What if the interaction between a person and an AI isn’t a conversation, but a research process?
You’re not having a conversation, you’re doing collaborative analysis. The AI reads, you direct. It proposes patterns, you accept or redirect. The unit of work isn’t a turn, it’s a theme. The verb isn’t “talk” or “chat”, it’s “discover”.
Connects to is-conversation-still-a-useful-metaphor and the note-taking/metacognition metaphors explored there.