Assessment of the SEEV Model to Predict Attention Allocation at Intersections During Simulated Driving

The authors attempted to model attention allocation of experienced drivers using the SEEV model. Unlike previous attempts, the present work looked at attention to entities (vehicles, signs, traffic control devices) in the outside world rather than considering the outside world as a unitary construct. Model parameters were generated from rankings of entities by experienced drivers. Experienced drivers drove a scenario that included a number of intersections interspersed with stretches of straight road. The intersections included non-hazard events. Eye movements were monitored during the driving session. The results of fitting the observed eye movement data to the authors' SEEV model were poor, and were no better than fitting the data to a randomized SEEV model. A number of explanations for this are discussed.