I’m working with ChatGPT to build a conceptual framework for a chatbot that would support teachers in designing inquiry lessons using the Inquiry Design Model. My work began with an exchange about design parameters, features, and the structure of a chatbot to scaffold teachers through the challenges of designing an inquiry.
We’ve learned over the years that there are major sticking points in the design process for anyone trying to design an inquiry, especially for novices. Our thinking is that we could use an AI chatbot to help teachers navigate the tough parts of this work.
To begin, I felt it was important to find some conceptual grounding that would root the chatbot’s design in an accepted, or reasonably acceptable, set of interconnected theories. These theories concern how we design instructional materials, consider student needs, apply disciplinary habits of mind, and purposefully weave in assessment and other related contingencies required for high-quality teaching and learning.
My first chat in this process aimed to establish the “design parameters, features, and structure of a chatbot that will support teachers to create inquiry-based instructional materials using the **Inquiry Design Model**.” At that time, I was trying to gain clarity on what I called “deep inquiry design” and “side-by-side recursive thinking.” Both of these concepts emerged from earlier conversations with ChatGPT.
In those chats, I was pursuing a line of thinking where we recognize that high-quality inquiry design necessitates deep, sustained engagement over time. This is something I wanted to call “deep inquiry design.” Over the years, I’ve experienced this firsthand while working with teachers at multiple points in the design process. We see challenges right from the start when trying to find a compelling question. It comes up again with **supporting questions**, particularly regarding how the designer expresses their understanding of content through these questions and their structure as they unpack the compelling question.
Of course, we also see challenges in the **selection of sources**. You need strong content knowledge to select high-quality sources, and then you have to be able to engage with those sources. If you’re unfamiliar with them, it’s not only time-intensive but also a significant intellectual endeavor that can really slow down the design process.
So, the goal of this work wasn’t necessarily to speed things up, but rather to make the design work more efficient and productive. It’s about enabling scaffolding that will help teachers make progress in the design process, especially when they might be novices in a particular area or simply overwhelmed with the sheer volume of content.
I also wanted to introduce the idea of a side-by-side, or even recursive, process for teachers engaged in deep inquiry design. While I’ve mainly discussed the content knowledge aspect so far, teachers must also simultaneously engage in the pedagogical thinking that, in my view, should drive inquiry design.
The question then becomes: how can we recursively and simultaneously balance content thinking with pedagogical thinking? Of course, the concept of pedagogical content knowledge (PCK), stemming from Shulman’s work and its subsequent developments, immediately comes to mind and is certainly relevant here.
However, I had something else in mind. Beyond the emergence of PCK as a field of knowledge, there’s a dynamic, recursive process that unfolds. As designers, teachers are constantly bouncing back and forth between content engagement and pedagogical thinking, striving to balance their involvement with these two ways of thinking within the domains of knowledge they’re working with.
So, I asked ChatGPT to provide some possible theoretical foundational anchors, and the list it generated was quite useful. As you go down the list, you’ll see some familiar names, such as Shulman and PCK (Pedagogical Content Knowledge), along with “reflection in action.”
đź“š Foundational Theoretical Anchors
Here are some relevant scholarly traditions to draw upon:
1. Shulman’s Pedagogical Content Knowledge (PCK)
- Shulman (1986, 1987) emphasizes the transformation of subject matter into instruction — a recursive act of representation.
- This is foundational for framing how content expertise and instructional design evolve in tandem.
2. Schön’s Reflective Practitioner
- Schön’s idea of “reflection-in-action” aligns with your vision: the designer is not simply executing a plan but constantly thinking while doing, adjusting based on new insights.
3. Ball & Cohen’s Work on Teacher Learning
- Ball & Cohen (1999): Teachers learn best through designing and revising curriculum materials—exactly the kind of embedded learning you describe.
4. Bereiter & Scardamalia’s Knowledge Building
- Their model of expertise centers on productive inquiry and recursive refinement of ideas, relevant to how teachers design and deepen content understandings simultaneously.
5. Design-Based Research & Learning Sciences
- The iterative cycle of design (e.g., Collins, Brown & Holum, 1991; Sandoval, 2004) mirrors what you’re doing. Design becomes a way of knowing, not just a way of doing.
It’s interesting that towards the end of the list, ChatGPT seemed to shift away from design thinking and the blending/intersection of pedagogy and content knowledge, turning instead towards design-based research. While design-based research is useful, this isn’t necessarily a research project I’m engaged in. It’s more of an effort to build a foundation or create scaffolding for design thinking. Nevertheless, all these suggestions were helpful and certainly moved me forward in the process. And importantly the iterative cycle design is suggested in these approaches to research is precisely what attempting to do here is to create some sort of iteration or recursiveness in teachers thinking about the designing of an inquiry.
With these initial ideas about design, its recursive nature, and a theoretical grounding in place, I continued my chat session with ChatGPT. My goal was to push the thinking forward and find terms that would effectively capture how I envisioned this chatbot functioning. So, I asked ChatGPT for some possible terminology to refer to my thinking, and the list it provided is what I received.
- Recursive Co-Design Thinking
- Dual-Track Inquiry Design
- Reflective Instructional Co-Development
- Inquiry-by-Design (already used in other domains; possible confusion)
- Design-Dialectic
Iterative Instructional Thinking
The idea of a dialectic really resonated with me, as I wanted teachers to seamlessly move between engaging with content and pedagogy. This led me to settle on the term “dialogic design.”
I was particularly drawn to “dialogic” as a modifier for the design process because it implies more than just a simple dialogue. My hope is that the user – the teacher interacting with our chatbot – would be engaged in a truly dialogic exchange. In this scenario, both the teacher-designer and the AI chatbot would be on equal footing, bringing ideas to the table and challenging each other, rather than passively accepting what the other says. It’s not just about sharing and receiving ideas; it’s about actively engaging with those ideas to move something forward. While it’s not really a “dialectical,” (we’re not looking to reconcile contradictions or synthesize ideas) it’s definitely aimed at achieving a productive goal that the teacher-designer has in mind and that the AI chatbot is committed to helping achieve those same goals.
So that’s where I am right now. I’m going to be continuing on this journey working more with chat GPT to move the ideas forward and will continue to post here with updates stay tuned!
