B-Alert Metrics Shown to Predict Driving Accidents

Download Full Publication

Results appeared in both normal and neurologically compromised drivers.

Early results from a collaboration with the UCSD HIV Neurobehavioral Research Program, exploring the impact of HIV-related cognitive decline on performance of regular activities such as driving, suggest that the B-Alert Cognitive State Metrics may be effective predictors of human performance errors.  Specifically, EEG-based Distraction Classifications were higher on both HIV+ participants and in poor drivers throughout a 30-minutes drive in a simulator.  EEG Distraction was elevated 7.14 sec prior to a crash, while EEG based workload dipped 10-16 seconds before the crash.  HIV+ participants were also shown to be more vulnerable to distraction-associated crashes than the HIV- controls.

These results further support prior findings that establish the B-Alert Cognitive State Metrics’ ability to predict human errors, as reported in this Frontiers in Neuroscience publication (2011) and these findings at The University of Arizona (2012).

ABM’s Director of Psychophysiological Research, Dr. Robin Johnson, will present these findings (Poster PDF) with our partners from UCSD this summer at The 7th International Driving Symposium on Human Factors in Driver Assessment, Training and Vehicle Design.   Funding for the study was provided by the US National Institute of Mental Health STTR contract 1R41MH097303-01.