In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
will defend his dissertation proposal
Biofeedback for Controlling Distractions in Transitional Vehicle Technology
Distractions account for an increasing number of crashes and fatalities on the roads. Here we introduce a biofeedback method for controlling both physical and esoteric forms of distractions while driving. The method is based on detecting the sympathetic state of the driver through perinasal perspiration. The perinasal perspiration signal is extracted via thermophysiological imagery and is monitored for significant persistent increases through a statistical filter. An over-arousal alert from this filter is communicated as a pink light in the emblem of the steering wheel, an item strategically placed in the peripheral vision of the driver. This is a direct notification to the driver not only that s/he is distracted (physically or esoterically), but that s/he is clearly exceeding her/his capacity of driving safely. The suggested action is to disengage from the stressor, applying mindfulness or a short relaxation technique, eventually driving down the sympathetic signal and switching off the pink light. From the theoretical point of view, this is a superior method, not only because it captures all forms of distraction, but also because it enables the application of Cognitive Behavioral Therapy, a potent method for bringing about permanent behavioral change.
In collaboration with the Texas A&M Transportation Institute, we ran two parallel group experiments – one on a high fidelity driving simulator and one a test track facility with an instrumented vehicle. In both experiments, one group were control subjects that underwent cognitive and physical distractions while driving without the benefit of any feedback mechanism; the other group were interventional subjects that were given automated over-arousal notice by the biofeedback system and had to respond accordingly. We have analyzed the simulator dataset, and documented the positive effect of biofeedback both in bringing down arousal levels and improving driving performance. We are still to analyze the test track dataset. The results thus far demonstrate the feasibility and promise of the method, opening the road for longitudinal experiments where not only the short-term but also the long-term behavioral effect would be measured. This may prove to be the ultimate safety technology until full car automation.
Date: Monday, December 11, 2017
Time: 1:00 PM
Place: HBS 302
Advisors: Dr. Ioannis Pavlidis
Faculty, students, and the general public are invited.