Chapter 26. Bicyclists
Ian Walker
University of Bath, Bath, UK
Bicycling is an important mode of transport because it offers benefits for reducing congestion and for improving public health. However, there has been considerably less research to date on bicycling than on driving. This chapter begins with a short historical overview of how roads are shared between different transport modes and then discusses three emerging areas of applied psychological research that are related to how bicyclists interact with other road users in these environments: (1) the role of drivers’ cognitive processes, particularly mental models that guide attention, in collisions with bicyclists; (2) drivers’ limited stereotypical ideas about bicyclists and how these might affect driver behavior; and (3) the asymmetrical social nature of encounters between bicyclists and drivers. For each area, a case is made for how a better understanding of the area might help improve bicyclist safety and encourage the more widespread use of bicycles for personal transport.

1. Introduction

2. Bicycles, Cars, and Public Acceptance

These attacks—which in Britain were one of the driving forces behind the formation of bicycle clubs, with riders seeking safety in numbers as much as companionship—were of course illegal. However, the fact that these attacks were against the law should not be taken to imply that the world’s legislatures were on the side of the bicyclist. In the nineteenth century, cities from New York to Berlin and Moscow swiftly introduced laws to restrict bicycling, often banning it altogether. Moscow’s 1881 ban in particular comes across as a trifle hysterical with hindsight, given that there were perhaps only five bicyclists in the city at the time (McGurn, 1987).
Then, as is the way of such things, the fuss subsided and after a while bicycling came first to be tolerated, then accepted, and then popular. This state persisted for some time until, a few decades later, motor vehicles appeared and the entire process repeated itself: The car was initially loathed until, over time, economic and social changes meant that it established itself in the niche once held by the bicycle.
The initial reaction against the bicycle was primarily motivated by conservatism coupled with class prejudice: Velocipeding was a well-to-do activity, unaffordable to the typical working family, and it was distrusted on those grounds as much as any other (McGurn, 1987). Similarly, the subsequent resistance to the car arose because by that time, mass production and increased popularity had worked to reduce the price of bicycles dramatically. The public had come to love the personal mobility that bicycles provided and did not wish to share the roads with these new motorized contraptions, particularly because high prices again made car ownership unattainable to most.
Once more, the public aversion to the new transport mode was shared by officials who by that time were mostly ardent converts to bicycling. The initial instinct of nonplussed authorities when faced with motor vehicles was to protect all the other road users from what they saw as a new and unprecedented threat. Of the various road safety procedures introduced in the very early days of motoring, certainly the most infamous was the system of red flag bearers, obliged to precede traction engines as they motored around Britain’s lanes, sometimes at speeds as fast as a brisk stroll. This law was rescinded soon after petrol-driven cars were introduced (ironically, often by bicycle makers wanting to sell vehicles with more cachet than the now commonplace bicycle; Wilson, 1973). However, as the red flag law was repealed, so it was replaced with a raft of new measures, many of which, such as speed limits and compulsory driver licensing, remain with us today (Hindle, 2001).
It was in this spirit of seeking to protect the public that, as motor vehicle ownership expanded, well-intentioned governments abandoned the millennia-old system in which roads were shared equally by everybody in favor of an approach whereby roads were divided, typically with a large central portion for people using vehicles and wholly separate areas for pedestrians (Hamilton-Baillie, 2008a and Hamilton-Baillie, 2008b).
This ultimately led to the situation found in most developed countries today, in which different travel modes automatically confer different sets of rights and limitations on their users. Pedestrians know that they are not permitted on roads as a matter of course; they are expected instead to take themselves to designated places if they wish to apply for permission to cross. Motorists know that they cannot drive along the zone reserved for pedestrians—hence the dramatic effect of this behavior when it is shown in action films.
This sort of system, in which each mode’s permissions are clearly demarcated, provides a great deal of certainty: A pedestrian can walk along a road without constantly needing to search for potentially hazardous motor vehicles; similarly, a driver can travel in the knowledge that a pedestrian should not appear in front of him or her without warning, and that if a pedestrian does, it will likely be the pedestrian who is held to account for any collision (Carsten, Sherborne, & Rothengatter, 1998).

3. Bicycling, Infrastructure, and Driver Attention

The involvement of motor vehicles in the more serious bicyclist injuries tends very often to follow a predictable pattern in which the bicycle is struck at an intersection by a turning car that fails to yield priority appropriately (Atkinson and Hurst, 1983 and Stone and Broughton, 2002). Because the mechanism underpinning these collisions appears in most cases to be a perceptual failure on the part of the driver (Räsänen and Summala, 1998 and Summala et al., 1996), similar attentional issues in drivers will even affect segregated infrastructure to some extent, given that each segregated cycleway must have at least entrance and egress intersections, and many are designed with far more conflict points. To illustrate the extent to which intersections disproportionately threaten bicyclists, I once asked the United Kingdom’s Department for Transport (personal communication) for some data. These showed that whereas 62% of all recorded road accidents happened at intersections, when the subset of collisions involving a bicycle and a motor vehicle are specifically considered, the figure rises to 75%. A lesson for infrastructure providers, then, might be that the fewer intersections one has, the better it is for bicyclists.
The perceptual and attentional lapses causing drivers to overlook bicyclists near intersections are not well understood. However, there is good reason to believe that the process involves a substantial top-down processing component in the drivers. “Looked but failed to see” errors, in which a driver fails consciously to perceive another road user despite being in a position to do so, are more common in experienced drivers than in novices (Herslund & Jørgensen, 2003). These attentional lapses particularly endanger bicyclists, and because they become more frequent when people are familiar with the driving task, the problem clearly does not simply lie in the bicyclists being difficult to see; if that were the case, it would be the novices who overlooked them more often. As such, it is far from obvious that the best solution to the problem of drivers hitting bicyclists is to make the bicyclists more conspicuous.
A plausible mechanism behind these perceptual failures, which so far appears to be untested, is that drivers often fail to notice bicyclists at intersections because of misplaced expectations about what they will encounter. Specifically, because bicycles are relatively rare in many settings, drivers, based on their experiences, might not expect to meet bicycles at any given junction. This expectation might cause drivers not to search for bicycles or not to attend to them when they are present. This argument—that bicyclists’ relative scarcity biases drivers’ expectations and so impacts on perception—has been suggested (e.g., Walker, 2009) as an explanation of the “safety in numbers effect,” in which the risk of being involved in a bicycle accident declines as the number of bicycles in a locale increases (Jacobsen, 2003; for a different view of safety in numbers, see Bhatia & Wier, 2010). A proper test of this idea, in which drivers’ attentional patterns in bicycle-rich and bicycle-poor environments are compared or in which drivers’ encounters with bicyclists are manipulated to determine how their expectations and attention change as the probability of encounters changes, would be very useful so that we can decide whether this account is likely to be correct.
Another reason to believe that bicycle collisions involve a top-down cognitive process in the driver comes from drawing an analogy with the motorcycling literature. From analysis of a substantial accident database, Magazzù, Comelli, and Marinoni (2006) found that people who could ride motorcycles were less likely to collide with other motorcyclists even when driving cars. The finding that drivers who have ridden motorcycles are more likely to perceive a motorcycle on the road and deal with it appropriately, even when in a car, suggests again that such collisions cannot primarily be attributed to the rider: Motorcyclists are there to be seen if only drivers are sufficiently prepared to see them. It is simply that many are not prepared this way.
Of course, were we being slightly flippant, we might be tempted to explain away Magazzù et al.’s (2006) data in Darwinian terms and suggest that given the extent to which motorcycles do not protect their users from mistakes (Lynham, Broughton, Minton, & Tunbridge, 2001), a motorcyclist who has survived for any length of time might simply have abnormally good perceptual ability and reflexes. However, such an argument certainly falls down when we consider the work of Brooks and Guppy (1990). These researchers found that merely having a motorcycling friend or relative was sufficient to make drivers better at perceiving motorcycles on the road. Again, these findings suggest that drivers’ failures to perceive motorcycles cannot simply be explained away by the suggestion that motorcycles are intrinsically difficult to see. Because the circumstances of motorcycle and bicycle collisions are very similar in many cases (Atkinson and Hurst, 1983, Harrison, 2004, Lynham et al., 2001 and Stone and Broughton, 2002), it would clearly be useful to determine whether, as in the studies of Magazzù et al. and Brooks and Guppy, direct or indirect experience of bicycling makes drivers more likely to perceive bicyclists on the road. As yet, however, this has not been studied. If such research is eventually carried out, we might gain insights into remedial actions to prevent collisions arising from drivers’ perceptual failures. We might also avoid the unfortunate state of affairs whereby vulnerable road users struck following drivers’ perceptual failures are blamed for failing to make themselves sufficiently conspicuous (Miller, Kendrick, Coupland, & Coffey, 2010).
In summary, mixing with motor vehicles around intersections is a particular source of danger for bicyclists and one that is not entirely removed even with segregated bicycling facilities. Moreover, it is likely that this is more a product of drivers’ perceptual failures than of anything to do with the bicyclists. More research is certainly needed to further explore these issues.

4. The Minority Status of Bicycling, Stereotypes, and Driver Behavior

I can understand the pedestrians’ point of view because of the amount of times I’ve been in Oxford city centre and I’ve just walked out and thought “Hell, this is actually a road!” but have just walked straight out. I can understand that—I’m always aware of that—because of the amount of times I’ve done it. So I can forgive pedestrians, but cyclists I cannot.
This statement, from a professional bus driver (Walker, 2005a), illustrates the problem of what we might as well call modal empathy. The driver expresses a willingness to forgive the lapses of a group with which she identifies but refuses to tolerate the same behaviors from a group to which she does not feel belonging. Bicycling is a minority behavior in many countries, and there are surely none left on Earth where bicycling is a more aspirational activity than driving. In a context whereby most road users do not identify with or aspire toward bicycling, might the outsider status of bicycling be working against bicyclists, generating resentment or even threats to their safety? To date, as noted previously, there has been little research specifically examining whether drivers’ lack of bicycling experience directly affects their treatment of bicyclists. However, there is a small body of work on nonbicyclists’ internal representations of bicyclists that might be relevant.
Basford, Reid, Lester, Thomson, and Tolmie (2002) appear to be the first group to have seriously considered the idea that bicyclists might be viewed somehow as “other” by the majority of road users, discussing driver–bicyclist interactions in terms of the in-group and out-group effects well-known in social psychology. Gatersleben and Haddad (2010) probed into these grouping effects using a factor-analytic approach to search for consistent patterns in people’s concepts of bicyclists. Specifically, a group of bicyclists and a group of nonbicyclists were presented with short descriptions of bicyclists, each of which mentioned a behavioral, a motivational, or a visual characteristic (e.g., “They wear tight clothing” or “They bicycle to work”). By assessing how often descriptions were selected together, the authors could identify groups of characteristics that seemed to fit together in people’s minds. They found, among other things, that the typical driver perceives only a very limited range of bicyclist stereotypes, including the “die-hard” bicyclist (who rides as fast as possible, helmeted, on an expensive bicycle) and the “necessity” bicyclist (who rides for functional transport and does not enjoy it). These perceptions appeared in many cases to act as a barrier to the uptake of bicycling, with nonbicyclists apparently finding it difficult to view themselves bicycling because they did not share the identities and motives they perceived among existing bicyclists.
Interestingly, a slightly earlier study of mine suggested that such stereotypes might cause measurable behavioral changes in drivers that could affect riders’ safety (Walker, 2007). Using an instrumented bicycle, which kept accurate records of how close vehicles passed to it, I was able to log more than 2200 instances of vehicles overtaking me on city streets, all the time keeping my riding behavior as constant as possible while manipulating two key variables: my position on the road and whether or not I wore a helmet. All other things being equal, donning a helmet was, on average, associated with a significant reduction in the space left by overtaking drivers. Why should simply putting a helmet on my head have led to drivers overtaking more closely, and why should hiding the helmet have led to them leaving more space? Both Basford et al. (2002) and Gatersleben and Haddad (2010) found that bicycle helmets were seen, by many nonbicyclists, as an index of experience and skill. As described previously, Gatersleben and Haddad found the helmet to be associated with the die-hard type of bicyclist for many nonriders. Similarly, Basford et al. found that “pictures of cyclists wearing helmets were generally considered to be more serious and sensible on the road than those without” (p. 9) and often “it was felt that people who had arranged appropriate and/or specialist cycling equipment and clothing were more likely to have also the experience and/or training to employ correct cycling behavior” (p. 9). I am saying nothing here about the efficacy of bicycle helmets, merely that many nonbicyclists seem to have a certain schema invoked by seeing one, which apparently leads to measurable changes in their behavior that might endanger bicyclists. It is clear that more research on these issues, as well as the more general issue of how out-group status might influence the way bicyclists are treated, would be highly valuable.

5. The Human Nature of Bicyclists

I once, in an exploratory study with no particular hypothesis, took a series of photographs showing street scenes and asked participants to describe, in their own words, what they saw in the pictures (Walker, 2005b and Walker, 2005c). When the descriptions were analyzed, a surprisingly clear story emerged. Whenever a picture showed a motor vehicle such as a car, the language people used to describe it was always inhuman: The words chosen were “car,” “vehicle,” and so on. So viewers would say “A car is turning left” or “A car is waiting for pedestrians to cross,” not “A driver is waiting for pedestrians to cross,” which would of course be more logical. In contrast, when a picture showed a bicycle and its rider, the words used were human: “A cyclist is …,” “A man is …,” and never “‘A bicycle is ….” This effect remained even when the driver of a car was clearly visible: Even when the driver could easily be seen, the words people chose referred to the car and not the driver, perhaps suggesting that the most salient component of the scene, for the viewer, was the machine and not the person controlling it. However, in the case of the bicycle, the words suggested that the salient component was the person, not the machine the person was piloting. Might this matter? Might bicyclists be treated in a qualitatively different way on the road, as they are in people’s descriptions, simply because they are so clearly human to other road users?
Some follow-up experiments attempted to explore this further, and they found intriguing signs that because bicyclists remain clearly human—undisguised by a vehicle, as motorists usually are—this might indeed affect how other road users interact with them. For example, it has long been known from laboratory studies that making eye contact with another person causes distinct and powerful neural responses (Gale, Spratt, Chapman, & Smallbone, 1975), indicative of eye contact’s important status as a social signal. Accordingly, an experiment suggested that eye contact between a driver and a bicyclist led to the driver making decisions substantially more slowly than in equivalent situations in which there was no eye contact (Walker, 2005d). Somehow, the eye contact seemed to interfere with, or delay, drivers’ decisions, perhaps because it invokes an extra, involuntary stage of cognitive processing. Bicycle users will clearly invoke eye contact in other road users far more than motorists.
In addition, when we later used eye tracking equipment to examine drivers’ attentional patterns as they made decisions about a bicyclist, we saw a powerful tendency for drivers to fixate immediately on the bicyclist’s face the moment the bicyclist appeared and to linger on the face longer than any other part of the scene (Walker & Brosnan, 2007).
These data on the human appearance of bicyclists, and how this might influence driver attention and decision making, are still at the point of being intriguing but inconclusive. They suggest that there might be a fundamental asymmetry among people sharing the road: When a driver and a bicyclist meet, it might be that the bicyclist largely has the experience of interacting with a machine, whereas the driver largely has the experience of interacting with a person. The consequences of this asymmetry, combined with the additional asymmetry arising from the car’s greater ability to cause damage in a collision, certainly seem worthy of additional study.

6. Summary

In principle, bicycling is just as diverse an activity as driving, but its many facets have received less behavioral research than other areas of traffic psychology. (The engineering aspects of bicycling, in contrast, have enjoyed a considerable amount of attention; Wilson, 2004.) In this chapter, instead of attempting to outline a clear narrative or present definite findings about the psychology or behavior of bicycling, I highlighted three areas of bicycling research that I believe are sufficiently intriguing and important to merit further attention. I hope that researchers take up the challenge posed by such topics and work toward providing the scientific literature needed to encourage higher levels of bicycling in the future.