It is well established that a large proportion of child pedestrian accidents involve victims who, for one reason or another, failed to detect the approaching vehicle (see Thomson, Tolmie, Foot & McLaren, 1996). Such errors may arise from a failure to look for traffic at all, as in 'dart out' accidents (e.g., van der Linden & Goos, 1975); from an inadequate looking strategy, for example where the child looks for traffic approaching from some directions but not others (van der Molen, 1981); or from an inability to recognise that certain road sites are intrinsically dangerous because they offer an obscured view of the road (Ampofo-Boateng & Thomson, 1991). In addition, a significant proportion of accidents involve children who did, in fact, look for traffic but nevertheless failed to 'see' it (Grayson, 1975). Such accidents persist even though educational measures place considerable stress on the importance of 'looking' for traffic before crossing. Clearly, there is a need for deeper understanding of the factor that underlie such accidents. Whilst failure to detect approaching traffic is often assumed to reflect motivational deficiencies on the part of the child, there is a considerable amount of evidence suggesting that these difficulties may be at least partly attributed to immature visual search strategies and defective attentional control (Thomson et al., 1996). For example, the pedestrian must isolate relevant information about the movement of vehicles that are approaching the intended crossing point (often from several directions) whilst temporarily filtering out information about vehicles that are receding, stationary or otherwise not relevant to the crossing task. Pedestrians must also ignore a wide range of interesting but irrelevant information about, for example, the activity of other pedestrians, dogs, cats, ice cream vans and so on (save where this directly affects their own behaviour or that of approaching vehicles). Very little direct research has been conducted on children's ability to do this in a roadside context, but in general terms it is known that children are often poor at separating relevant from irrelevant features and are highly susceptible to the interfering effects of salient but irrelevant stimuli (e.g., DeMarie-Dreblow & Woody-Ramsey, 1988). Given that children's visual search is poorly structured to start with (e.g.,Vurpillot, 1968), the distracting effects of irrelevant but attractive stimuli are likely to be even more marked. Children also show difficulty in sorting out their priorities when allocating attentional resources to competing tasks (Wright & Vliestra,1975). We argue that attentional problems of this kind are a major contributor to the phenomena of roadside distractibility and impulsivity. Children's attention appears to wander, not simply because they have an inherent inability to concentrate attention, but because they lack a clear grasp of what they should focus on, especially under unfamiliar and/or complex circumstances. As a result, they fall back on what were functionally appropriate strategies at an earlier stage of development, namely directing attention to the novel and the socially important. Similarly, children may exhibit 'impulsive' behaviour, not because they are necessarily more prone to act without planning but because, faced with the conflict between making sense of a complex array of information that is too much for them and the imperative to get from A to B, they resort in the end simply to launching themselves into action. There are many identifiable circumstances under which adults behave in a similar fashion. Naturally, children can be expected to behave in this way more frequently. The present project explored these issues by means of three experiments. In Experiments 1 and 2, cross-sectional studies were undertaken to investigate how attuned children of different ages are to traffic-relevant information when presented with a range of traffic situations. We also looked at the effects of increasing informational complexity, introducing distractors and limiting the time available for extracting information from the scenes. Both experiments used simulated traffic environments; computer-generated in the case of Experiment 1 and video simulations in Experiment 2. In both cases, performance on the simulation was compared to performance at the roadside. In Experiment 3, an intervention procedure was introduced to see if children's sensitivity to traffic-relevant information might be improved and, if so, to determine which aspects of the intervention were most effective in doing so. The intervention used a version of the computer task developed in Experiment 1, suitably modified in the light of the findings of Experiments 1 and 2. Children were asked to solve a series of traffic problems, working either individually with an adult or co-operatively with a group of other children. It was hoped that the dialogue that took place during these sessions would provide insight into the dynamics of the learning process, thereby providing a foundation for the production of optimally-structured training programs at a later date. We also examined the extent to which children's judgements at the roadside improved as a result of these interventions. (Author/publisher)
Samenvatting