The blind date: the effects of change blindness, passenger conversation and gender on looked-but-failed-to-see (LBFTS) errors

Abstract

This study examined situations where drivers looked-but-failed-to-see (LBFTS) hazards, and whether passenger conversation and gender affected hazard detection rates. To reliably produce LBFTS errors, 40 young drivers (M=20.3) encountered motorcycles and pedestrians while making left turns in the University of Calgary Driving Simulator (UCDS). Prior to turn initiation the UCDS screens flickered using an extension of change blindness methods. In addition, drivers either drove alone or conversed with an attractive confederate passenger. Measures of LBFTS errors, hazard detection and social factors were analyzed. Higher rates of LBFTS errors and hazard detection occurred while conversing than while driving alone. A discriminant function analysis (DFA) using conversation and gender as predictors accurately classified LBFTS errors. Higher passenger attraction and higher extroversion were related to critical events being missed. The basis of LBFTS errors in divided and selective attention and classification implications are discussed.