Improving hazard perception and tracking through part-task training

Abstract

Safe driving requires wisely allocating focal attention among multiple changing events and comprehending events that are attended to. Research suggests that attentional skills can be improved by training. In this experiment, we are using a low-fidelity driving simulator to train participants using part-task training on two attentional subskills: identifying (comprehending) and tracking potential hazards; and detecting and avoiding imminent hazards. Following initial familiarization with the driving simulator, each participant will receive training in one of these two attentional subskills. Scene comprehension probes train (and measure) identifying and tracking potential hazards by having participants watch a moving driving scenario and then select the vehicle that behaved hazardously during the scene. In hazard avoidance probes, participants must make driving responses to avoid imminent hazards without hitting nearby vehicles. After the training phase, there is a test phase measuring near transfer, to hazards similar to training, and far transfer, to untrained hazards. We hypothesize that the participants who receive part-task training on identifying and tracking hazards should perform better at scene comprehension probes than the hazard-avoidance training group in both near and far transfer conditions. We also hypothesize that the group trained on avoiding imminent hazards will perform better on hazard avoidance probes in both near and far transfer conditions.