Search

One of the tools that the United States Federal Highway Administration uses to evaluate highway infrastructure-signs, roadway markings, and geometric features-is a driving simulator. The nature of the research necessitates that important visual information is perceived similarly in the simulated world as in the real-world. Perceptual mismatches can lead to incorrect assumptions about driver behaviors and understanding of new or novel roadway designs. Because of limitations in simulator projector technology, and the lack of true depth in images projected onto a cylindrical scene, it is often necessary to adjust the size of projected objects. The objects may need to be scaled up or down relative to 1:1 equivalence of image projected on the retina in order to produce a detectable object or a desired apparent distance. For example, to achieve an appropriate legibility distance for highway signs, signs are scaled up to achieve a mean legibility distances of 1 inch of letter height per 60 feet of (simulated) viewing distance. On the other hand, to attain a 1:1 correspondence with real-world following distances between the subject vehicle and the vehicle ahead, simulated vehicles are scaled down to 75 percent of the correct retinal image size. The methods and theories for arriving at these scale factors are described. The need to scale up highway signs is believed to be related to resolution and contrast limitations of current technology projectors. The need to scale down vehicle sizes may be the result of a Ponzo illusion, in which the converging lines of highway lane markings distort the apparent size of vehicles ahead. Before using a driving stimulator to conduct studies of driver perception or behavior in response to out of vehicle stimuli, it is important to examine assumptions concerning the size of projected images and their detectability and perceived distance. Meeting abstract presented at VSS 2015. Language: en

Every year there are nearly 43,000 traffic fatalities and it is estimated that 25% of crashes involve some degree of driver inattention (NHSTA, 2005, 2000). A recent survey revealed 21% of crashes/near crashes reported by respondents involved at least one driver using a mobile phone (Seo & Torabi, 2004). The current study examined the effects of mobile phone use on drivers' attention and eye movements in a low-fidelity simulator. Sixteen Clemson University undergraduate students viewed 24 driving scenarios and responded to questions about vehicular events in the scenes. Eight participants simultaneously performed a language learning task (simulating a mobile phone conversation). The language learning group answered fewer questions about the driving scenes correctly (M = 9.3) than the non-language group (M = 16). Overall, participants' correctly responded to more scenarios with 4 cars (M = 7.3/12) than with 7 cars (M = 5.3/12). The total number of fixations on the vehicle(s) involved in the critical event in each scenario was greater for the non-language group (M = 471.7) than for the language group (M = 300.5). Additionally, participants in the language group who answered the event question correctly spent the same percentage of the total time looking at the vehicle of interest during the event (M = 13.5%) as those people who answered incorrectly (M = 12.4%). This finding provides support for the 'look but fail to see' phenomenon. The mean duration of total fixations was also greater for people in the non-language group (M = 9574.5 ms) than the language group (M = 6523.4 ms). This study supports previous findings that increasing mental workload (through mobile phone use, and/or increased traffic) decreases driving performance.