ABSTRACT

To traffic psychologists and to cognitive psychologists alike, there are two categories of reply to the question: the internal, schema-driven control of attention determined by the road user’s driving goals and situation awareness, and the external, event-driven control of attention whereby some object in the environment captures attention by its unexpected activity. This dichotomy is sometimes referred to as the distinction between the endogenous and exogenous control of attention, and can be characterized by the distinction between a driver knowing that he should stop before entering a junction, in order to assess the oncoming traffic, and a driver noticing a fast-moving vehicle that requires evasive action. In one case the driver’s understanding of traffic dynamics requires inspection of the roadway, and in the second case an event in the roadway captures the driver’s attention. The dichotomy is too simplistic of course, and in many situations our attention is distributed according to our knowledge of where important events are likely to occur as well as being drawn to unusual or sudden-onset events such as a pedestrian walking out from behind a parked vehicle. On some occasions we do not inspect the roadway as we should, perhaps being distracted by another event, and other occasions we might rigidly scan the roadway locations where our situation awareness leads us to anticipate oncoming traffic, to the exclusion of a pertinent event in an unexpected part of the roadway. As an example of this interaction, imagine walking across an open pedestrian space – a city square, perhaps – while engaged in conversation with someone either on a mobile phone or who is walking with you. If something bizarre happened, would your attention be drawn to it? Suppose now a brightly dressed clown on a unicycle crossed your path (he was there to advertise a circus coming to town). Would your attention be attracted to the clown? If attention had been inevitably drawn by bizarre exogenous events, then the clown would be noticed. When navigating the environment, attention cannot be locked completely on to internal events or even a conversation, because the route must be maintained and our progress monitored, so there must be some notice taken of environmental events. This situation has actually been studied, in an experiment conducted by Hyman et al. (2010), who questioned walkers after they had crossed a pedestrian plaza on a university campus in the presence of a

unicycling clown. The answer is that individuals using a phone were half as likely to notice the clown relative to individuals not using a phone. This example of inattentional blindness, perhaps best known from the example of the undetected ‘gorilla’ amongst ball players (Simons and Chabris, 1999), illustrates that events that might be expected to be attentionally demanding in some circumstances can be invisible if attention is elsewhere. Endogenous and exogenous mechanisms are more likely to interact than to be in opposition in an ‘either/or’ fashion, but the distinction serves here to indicate our concern with the emphasis on external events in the control of a driver’s attention. Essentially the debate concerns the ways in which the prior allocation of attention can override bottom-up exogenous attentional capture and top-down endogenous attentional guidance. Evidence has suggested that both the goal-based searching for objects and the effects of unexpected conspicuous objects are influenced by the search goals from the recent past (for a review, see Awh et al., 2012). Both of these processes are relevant to driving of course, with the unexpected appearance of another road user being of potential importance and in need of processing, just as the volitional monitoring of the roadway is important at all times. We shall return to this distinction after considering the exogenous capture of attention in more detail.