Developing a Culturally Grounded, AI-Augmented UX Research Point of View (POV): An Exemplar Case Study from Telemedicine Dementia Care
This paper presents a case study from Nigeria demonstrating how a culturally grounded, AI-augmented User Experience Research Point of View was developed to inform a telemedicine dementia care framework, offering reusable Play Cards and a Play to extend the UXR POV Playbook for future AI-powered research practices.
Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Imagine you are trying to build a bridge across a river to help people get to the other side. In this story, the "river" is the gap between dementia care and telemedicine (medical care done over video calls) in Nigeria. The "people" are family caregivers who are already stressed, dealing with stigma, and often have shaky internet connections.
This paper is a guidebook on how a team of researchers built a compass (called a "Point of View" or PoV) to make sure they build the right bridge, using a special kind of AI assistant to help them plan, but not to make the final decisions.
Here is the story of how they did it, broken down into simple parts:
1. The Problem: Building a Bridge in the Rain
The researchers knew that simply making a "user-friendly" app wasn't enough. In Nigeria, caring for someone with dementia is deeply tied to culture, family honor, and sometimes spiritual beliefs. Plus, the internet might cut out, and people might not trust a computer to give medical advice.
If they built a high-tech app without understanding these realities, no one would use it. They needed a Point of View (PoV). Think of a PoV not as a guess, but as a defensible stance—a clear, evidence-based "North Star" that tells the whole team: "We are building this specific way because of these specific reasons."
2. The Tool: The AI as a "Super-Organizer"
The team didn't let the AI write the story for them. Instead, they treated the AI like a super-organized librarian or a high-speed sorting machine.
They had a massive pile of messy information:
- Interviews with caregivers.
- Surveys about digital skills.
- Reviews of medical studies.
- Notes from community workshops.
How they used the AI:
- The "Ontology" Fence: Before letting the AI touch the data, the researchers built a "fence" (called an ontology). This was a strict set of rules defining what words meant (e.g., "caregiver," "stigma," "bandwidth"). This prevented the AI from making up wild guesses or getting confused.
- The Sorting Hat: They asked the AI to sort the messy pile of notes into neat categories, like "Emotional Burden," "Trust Issues," and "Internet Problems."
- The Human Check: The AI suggested patterns, but the human researchers had to say, "Yes, that's right," or "No, that misses the cultural nuance." The AI never got to make the final call.
3. The Process: Four Steps to the Compass
The team followed a four-step recipe (based on a "Playbook") to turn their messy data into a clear plan:
- Step 1: Gathering the Clues (Foundation): They fed all their research into the AI to find repeating themes. The AI helped them see that "trust" and "low internet speed" were bigger problems than "cool features."
- Step 2: Mapping the Players (Stakeholder Roadmap): They asked the AI to map out who needed to be happy for this to work. It wasn't just the patient; it was the family caregiver, the doctor, the local community leader, and the government. The AI helped show where their needs clashed (e.g., doctors want data, caregivers want simplicity).
- Step 3: Making the "Play Cards" (Insight Generation): This is the coolest part. The team turned their findings into four "Play Cards" (like cards in a strategy game) that any designer could pick up and use:
- Card 1 (Cultural Grounding): You can't have a usable app if it doesn't respect local culture and beliefs.
- Card 2 (Infrastructure): The app must work even when the internet is slow or the phone is old.
- Card 3 (Emotional Load): Caregivers are tired and stressed; the app must be calm and simple, not complex.
- Card 4 (Explainability): If the AI gives advice, it must explain why, or people won't trust it.
- Step 4: Telling the Story (PoV Articulation): Finally, they used the AI to help write different versions of the plan for different people. They wrote a version for the caregivers (focusing on emotional support), a version for the developers (focusing on technical limits), and a version for policymakers (focusing on big-picture trust).
4. The Result: A Shared Language
The paper claims that by using the AI as a "bounded collaborator" (a helper with strict rules), they created a shared language for everyone involved.
Instead of arguing about what to build, the team could point to the Play Cards and say, "We need to follow Card 2 because the internet is unstable," or "We must follow Card 1 because of cultural stigma."
The Big Takeaway
This paper isn't about a finished app that saves lives today. It's about how to think and plan when building technology for complex, sensitive situations.
It shows that AI can be a powerful tool to organize chaos, but it must be held on a short leash by humans who understand the culture. The AI helped them sort the puzzle pieces, but the humans had to figure out what the picture was supposed to look like. The result is a "Point of View" that is grounded in real-world reality, not just in technology hype.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.