Mind off driving: Effects of cognitive load on driver glance behaviour and response times

Introduction: Safe driving requires drivers to look at relevant information in the traffic environment and react in time in case a critical event arises. Concerns exist that cognitively loading tasks might interfere with drivers’ abilities to do this. Studies on the effects of cognitive tasks on driver behaviours are however ambiguous and incomplete. The recently formulated cognitive control hypothesis might be able to explain some of the inconsistencies. Objectives: The aim of this thesis is to better understand the effect of cognitive tasks on response times in unexpected lead vehicle braking scenarios and on glance behaviour in traffic environments with potential threats in off-path locations. Effects are studied both at aggregated levels and with higher temporal resolution. Method: A series of experiments were conducted in an advanced driving simulator. Results: Cognitive tasks increased response times in the non-automated, artificial Detection Response Task (DRT) but did not influence response times in an unexpected lead vehicle braking scenario. Also, drivers adapted their visual scanning behaviour to the traffic environment in the same way in terms of timing when doing cognitive tasks as when not, but to a lesser degree. Interestingly, the effect of cognitive load on the visual behaviour depended on gaze direction and the demand variations in the cognitive task. Conclusions: The results demonstrate the importance of context when trying to interpret effects of cognitive load on traffic safety and are in line with the cognitive control hypothesis. They also indicate that there is not a unidirectional and uniform effect of cognitive activities on driver behaviour. This calls for further exploration of the interaction between the cognitive task and the driving task