Cognitive tunneling mitigation device for driving

A driver monitor method and system for mitigating cognitive tunneling in a vehicle includes in-cabin video cameras, a heart rate monitor, an audio-visual device, and processing circuitry that detects eye gaze direction, eye lid position, and head position using images from the in-cabin video cameras, and measures heart rate variability using the heart rate monitor. A machine learning device uses the eye gaze direction, eye lid position, head position, and heart rate variability to predict whether a driver is transitioning into a cognitive tunneling state or a fatigue state. The audio-visual device outputs one audio-visual cue to mitigate the cognitive tunneling state and outputs a different cue to mitigate the fatigue state. The machine learning device learns by performing a reinforcement learning algorithm in which the audio-video device outputs a verification request and receives a response to the verification request that is fed back to the machine learning device.