Dynamics of Mental Models: Objective vs. Subjective User Understanding of a Robot in the Wild
Ferran GebellÃ, Anais Garrell, Séverin Lemaignan, Raquel Ros
AI summary
Problem
HRI research typically evaluates either objective or subjective user understanding in isolation, rarely comparing how these distinct metrics evolve during long-term, real-world robot deployments.
Approach
We deployed an autonomous healthcare robot in a geriatric unit for five weeks and tracked nursing staff's weekly objective accuracy and subjective confidence to analyze their long-term dynamics.
Key results
- Notable divergence between objective and subjective understanding over time
- Subjective confidence increased with exposure despite stagnant objective accuracy
- Inaccurate mental models persisted due to limited system feedback
- Long-term wild deployment reveals limitations of short-term lab evaluations
Why it matters
Robot designers and HRI researchers must track both understanding metrics separately to avoid overestimating user comprehension and to effectively calibrate explainability measures.
Abstract
In Human-Robot Interaction research, assessing how humansunderstandtherobotstheyinteractwithiscrucial,particu- larly when studying the impact of explainability and transparency. Some studies evaluate objective understanding by analysing the accuracy of users’ mental models, while others rely on perceived, self-reported levels of subjective understanding. We hypothesise that both dimensions of understanding may diverge, thus being complementary methods to assess the effects of explainability on users. In our study, we track the weekly progression of the users’ understanding of an autonomous robot operating in a healthcare centre over five weeks. Our results reveal a notable mismatch between objective and subjective understanding. In areas where participants lacked sufficient information, the perception of un- derstanding, i.e. subjective understanding, raised with increased contact with the system while their actual understanding, objective understanding, did not. We attribute these results to inaccurate mental models that persist due to limited feedback from the system. Future research should clarify how both objective and subjective dimensions of understanding can be influenced by explainability measures, and how these two dimensions of understanding affect other desiderata such as trust or usability.