Presentation
Navigating AI Errors: The Role of Mental Models in Human-AI Team Performance
SessionWednesday Poster Session
DescriptionPerformance of human-AI teams depends on both AI’s overall accuracy and human’s ability to detect errors to form an accurate understanding of system behaviors. Less prior research has compared how different patterns of AI errors (i.e., when and how they occur) shapes users’ mental models that underpin human oversight. To address this gap, we conducted an experiment (N=380) exploring how the error patterns (between subjects) influence individuals' ability to detect AI errors and develop effective mental models for system accuracy in a phishing email classification task. Our findings demonstrate that while different error patterns did not significantly influence participants’ ability to detect AI errors, it did shape their perceptions of how the AI performed over the interaction. These findings contribute to the HFES literature by demonstrating that error pattern effects are more evident in users’ interpretations of system behavior than in their detection performance. This distinction advances theoretical models of monitoring and mental model development in AI-supported decision environments.
SNL is managed and operated by NTESS under DOE NNSA contract DE-NA0003525. SAND2026-17363A.
SNL is managed and operated by NTESS under DOE NNSA contract DE-NA0003525. SAND2026-17363A.
Event Type
Poster
TimeWednesday, October 21st5:30pm - 6:30pm PDT
LocationSummit Pavilion