Nature · 2 September 2026 · From issue 31
Explainable deep learning improves human mental models of self-driving cars
A real self-driving car explained its choices in plain terms, helping drivers predict its next move.
Eoin M. Kenny, Akshay Dharmavaram et al.
A new tool lets a self-driving car explain its choices using simple ideas, like "person walking nearby" or "stop sign ahead." Testing it in a real car showed drivers understood the car better after seeing these explanations. They got better at guessing what the car would do next, especially in confusing moments.
Shown to work on one real self-driving car, helping drivers predict its behavior better.
One real car tested, no numbers on how many drivers or by how much.
Integrity screen: passed (3 checks) Checked 16 September 2026. Retraction record: none. DOI resolves at doi.org. Metadata record found (Nature). Read the source
This screen checks process signals. It does not tell you the findings are true.