Agentic data science with the Data Agent Kit
cloud.google.com·7h ago
AI Research
arXiv CS.AI·Jul 29·5 min read
From Dyad to Triad: Eliciting XAI Requirements in Stroke Rehabilitation
TL;DR
A video-based protocol was developed to elicit explainable AI (XAI) requirements from stroke survivors and caregivers, revealing diverse needs and biases in responses. This approach aims to enhance the design of trustworthy brain-computer interfaces for rehabilitation.
✦ Why It Matters
Implement video-based protocols to better understand user needs in AI systems today.
Key Takeaways
- Utilize video-based methods for user requirement elicitation.
- Incorporate analogical bridging to enhance understanding.
- Be aware of biases that may affect user feedback.
- Focus on diverse user needs in AI system design.
How It Works
The protocol employs four scaffolding techniques: analogical bridging connects AI concepts to familiar systems, projective personas help discuss sensitive topics without personal bias, binary forcing simplifies decision-making, and extended response time allows participants to articulate their thoughts more clearly.