Active prostheses have many adjustable parameters—more than a typical user can or should tweak blindly. Most previous studies only captured what users chose, not why they chose it.
We want to:
Uncover the decision-making process behind tuning preferences
Identify usability needs and emotional responses to self-adjustment
Make recommendations for designing intuitive tuning interfaces
We conducted studies with two groups:
Able-bodied participants simulating prosthesis use (Study 1)
Amputee participants using powered robotic prostheses (Study 2)
Each participant:
Watched an instructional video and completed a practice tuning task (picture editing with 4 parameters)
Used a remote-controlled interface to adjust prosthetic knee settings in real-time while walking
Spoke aloud their reasoning as they changed settings
Rated their preferences after each tuning round
Repeated until they reached a “preferred profile” for daily use
All sessions were recorded, transcribed, and coded using thematic analysis.
To dig deeper into preference formation, I designed think-aloud prompts such as:
“What makes this setting feel better or worse?”
“What are you using to compare this against?”
“Would you be happy using this every day?”
These helped surface not only usability concerns, but also emotion, memory, and embodiment.
1. “Natural” Means Different Things to Different People
Non-disabled users often compared to their intact leg
Amputees more often compared to their passive prosthesis
Some prioritized balance, others speed, others ease of mind
2. Emotional Cues Drive Behavior
Words like confident, comfortable, worried, off-balance came up repeatedly
These emotional responses were closely tied to physical perception
Trust and stability outweighed perfect gait mechanics
3. Exploration Needs Structure
Users appreciated having a visual tuning interface, but needed context for what each control did
They reached “saturation” when no more improvements felt worth the effort
Some made trade-offs: “I could walk more casually, but I’d lose stability”
Our findings directly influenced:
Future iterations of tuning interfaces, adding more accessible guidance on control points
Clinical insight into how autonomy affects mindset—shifting from “just get through it” to “this could be mine”
A better understanding of how long-term use experience (like with passive prostheses) shapes tuning expectations
Think-aloud studies are powerful in hardware UX—they reveal real-time reasoning and emotion
Let users compare against their own baselines, not yours
Comfort, trust, and workload are just as important as performance
Design tuning systems that support confidence, not just control
Yuan, J., Bai, X., et al. (2023). Finding a Natural Fit: A Thematic Analysis of Amputees’ Prosthesis Setting Preferences During User-Guided Auto-Tuning. [Read the publication]
Yuan, J., Bai, X., et al. (2022). Understanding Preferences for Lower-Limb Prosthesis: A Think-Aloud Study During User-Guided Auto-Tuning. [Read the publication]