Hiles > Essays > Listening >

Hiles, Jeffrey A. Listening to Bike Lanes. September 1996.

Chapter 8
Listening to Bike Lanes

So far in this paper I have diligently tried to cast doubts on assumptions that shape both sides of the bike lane debate. One of my goals has been to take a crow bar to the hard shells of dogma with which we bicycle advocates shield ourselves from bicycling’s complexities and contradictions. I have also tried to paint an accurate picture of bicycle transportation, warts and all.

Still, I have stayed on the surface, have examined the more common subjects of discussion and points of contention. In this chapter, I part company with most of what has been written about bicycling and explore some different approaches to understanding bike lane issues.

Variation: threat or asset?

In the previous chapter, we saw that studies measuring the consistency, or inconsistency, of how closely cars pass bicyclists make questionable arguments for bike lanes. Nevertheless, these studies can help us describe motorists’ movements if we view the data from a different perspective. First, we have to understand the concept of “shy distance.” When you pass an object or another vehicle on the road, whether you are in a car or on a bike, there is a certain minimum distance at which you feel comfortable passing; any closer just feels too close because it requires too much accuracy and vigilance to avoid a collision. This distance that you shy away from things is called the “shy distance.” As a bicyclist, you might be able to ride four inches from the curb, but it would be unnerving; one small mistake and you’re lying on the sidewalk. Most cyclists ride at least a couple of feet from the road’s edge, often farther, so they can give their attention to everything else that’s going on around them.

Shy distance is related to speed. The faster you go, the more distance you want between you and something you pass. At higher speeds you have less time to identify and respond to the obstacles in your path. The results of your responses are magnified by your speed, so you have to be more accurate. If you need to drive through a space that is only slightly wider than your vehicle, you will proceed slowly so you can thread that needle without scraping the sides. If you have lots of room, you can sail through. Conversely, in a given space you have more room to maneuver at slow speeds than at high speeds. That is, if you are driving a car at 25 mph past a bicyclist in a 15-foot lane, your shy distance on either side of you is smaller and takes up less of the lane than if you were doing 55. So you have more leeway at slower speeds; you can comfortably pass more closely by the bicyclist if you want, or move out nearer the on-coming traffic—again, it’s because at slower speeds you have more time to correct your errors or respond to changes in what’s happening around you, and you can maneuver more accurately when you do respond. At high speeds, your shy distance increases, taking up more of the available space and narrowing your choice of trajectories—again, it’s because at higher speeds you have less time to correct your errors or respond to changes in what’s happening around you, and you cannot maneuver as accurately when you do respond.

In short, the variation in lateral position the researchers observed can be interpreted as a measure of the motorists’ ability to respond. That response ability is not a bad thing. It partly explains why, in the real world, bicyclists feel more comfortable riding among slower motor vehicles than they do in faster traffic. Furthermore, if you drop the premise that bike lanes should reduce variations in lateral position and substitute the assumption that their channelization is most needed where motorists have the least ability to respond and correct for misjudgements, then you come to the more reasonable conclusion that it’s the higher-speed roads where the lanes “work best”—remember also that high speeds are a major component of those Type 13 fatalities, a kind of crash that bike lanes could help prevent. In addition to speed, hills and turns would also reduce response ability by reducing sight distances.

Response ability and vehicular cycling

To the vehicular cyclist, the more response ability motorists have, the easier it is to establish communication. Motorists’ response ability makes it possible for a cyclist to “negotiate” a lane change, for example. When experienced cyclists vary their road positions to signal their intended tracks through intersections or to keep from getting squeezed into hazards up the road, they depend on motorists’ ability to notice and respond to these nonverbal signals. Remember, Forester says that bicyclists must resort to being “road sneaks” when the motor traffic is moving more than 15 mph faster. Forty-mile-per-hour motorists have less ability to respond to cyclists’ body language than 25-mph motorists.

So, variation is a measure of response ability and bike lanes decrease variation. Of course, it doesn’t necessarily follow that by decreasing variation bike lanes decrease motorists’ ability to respond to cyclists. Even so, some bicyclists charge that bike lanes decrease motorists’ inclination to respond. They say they feel trapped, that motorists won’t let them out of the lane to get into position for a left turn. In a typical response, Oregon bike lane advocate Michael Ronkin acknowledges that left turns are often difficult when bicycling in traffic, but says:

This has nothing to do with bike lanes versus shared roadways either way, a bicyclist is going to have to be prudent and look over their shoulder repeatedly before merging to the left. A bike lane does not trap a bicyclist: they may cross over the white line to turn left, pass other cyclists or avoid road hazards (Ronkin, 1993).

Ronkin’s statement is probably true for many bicyclists who never try to negotiate with motorists. Cyclists who always think of themselves as road sneaks, who don’t make a point of establishing communication with motorists, would feel no more trapped with bike lanes than without them. But Ronkin’s reply doesn’t address the heart of the complaint against bike lanes. Effective Cyclists do not passively “look over their shoulders repeatedly” and simply wait until it is “prudent” to merge left, they actively use body language to create an opening into which they may move. Some cyclists feel that the thick, solid bike-lane stripe creates a psychological barrier that makes it harder to “ask permission” to move over.

This may seem like a small point, not something likely to create great chasms within the bicycling community. But if we examine these views from the perspective of contemporary stress theory, we can better understand just how deep the division runs.

Cognitive stress theory, system image, and perceptions

According to the cognitive theory of stress and coping developed by Lazarus and colleagues, the amount of stress a person feels in a situation depends on what that situation means to the person. That meaning arises through two kinds of appraisal. In “primary appraisal,” the person judges the degree of threat he faces. In “secondary appraisal,” the person takes inventory of his resources and options for dealing with the threat (Folkman, 1984, p. 840). In other words, the more a situation appears to threaten your well being, the more stress you feel. But you can mitigate this effect if you believe you have ways of dealing with the threat, if you believe you have some control. (Although the theory just stated will serve our purposes, secondary appraisal has actually proven to be more complicated. Sometimes, for example, it’s not clear how much control you have, in which case you might read into that ambiguity either that you do or you don’t have control; your choice reflects your personality. Also, sometimes taking control requires us to break our usual ways of thinking or doing things, which can be more stressful than not having control.)

Using this stress theory as an analytical framework, I will now describe two bicyclists. Call these bicyclists hypothetical, stereotypical, archetypal; each is a composite of many bicyclists I have interviewed. Stress theory can help us understand bicycle facility issues on a level that is closer to the heart of what bicyclists think and feel, closer than we could ever get by looking at crash statistics alone.

Gus the ordinary guy

Gus’s primary and secondary appraisals make him feel pretty uncomfortable in the presence of overtaking motorists. Getting hit by a speeding car is a threat to a bicyclist’s health (primary appraisal); no one would dispute that. While Gus is riding in heavy traffic, he notices that it’s the overtaking vehicles that pass him most closely, most often, and at the highest speeds. So he naturally worries most about the cars that come from behind. What’s more, once in a while a motorist inflames Gus’s fear of being rear ended by passing too closely, blasting the horn, shouting, or tossing a pop can out the window. Plus, it’s almost always overtaking motorists who do these things. It’s only while overtaking that motorists have time to think up such abusive deeds. In any other situation, motorists’ interactions with Gus are much too fleeting. Although the vast majority of motorists pass Gus carefully and politely, the rare transgressions make an impression:

If there were a thousand similar events, we would tend to remember them as one composite prototype. If there were just one discrepant event, we would remember it, too, for by being discrepant it didn’t get smudged up with the rest. But the resulting memory is almost as if there had been only two events: the common one and the discrepant one. The common one is a thousand times more likely, but not to the memory; in memory there are two things, and the discrepant event hardly seems less likely than the everyday one.

So it is with human memory. We mush together details of things that are similar, and give undue weight to the discrepant. We relish discrepant and unusual memories. We remember them, talk about them and bias behavior toward them in wholly inappropriate ways” (Norman, 1988, p. 118).

So what Gus sees around him when he’s riding—the system image of the cycling environment—makes him wary of overtaking motorists. To him, they pose an in-your-face threat. Crossing and turning collisions, on the other hand, are out of sight and out of mind. He spends most of his cycling time between intersections, so his crossing and turning interactions with motorists are few and far between compared with being overtaken. When he does turn, he tries to wait until he is nowhere near a speeding vehicle. If it’s the motorist who is turning, the vehicle is usually moving slowly when it crosses Gus’s path, and that car rarely comes anywhere near as close to Gus as do overtaking motorists. So Gus’s primary appraisal is that crossing and turning is not a big threat. (Of course, crossing and turning collisions are so common precisely because they take both bicyclists and motorists by surprise.)

So far, we have only looked at Gus’s primary appraisal. He also believes he has little control over overtaking collisions. This secondary appraisal arises from two things. First, Gus has no rear-view mirror and rarely looks behind him while he’s riding for fear he will swerve if he turns his head to glance over his shoulder. As a result, overtaking motorists approach unseen, giving Gus the impression that if he is struck from behind, it will be without warning and so without any possibility for evasive action. Second, Gus does not know of any action he could take to help prevent an overtaking collision; he believes he is totally at the mercy of motorists, including some who are careless or even hostile. In contrast, crossing and turning interactions happen in plain view in front of him or from the side. It appears to Gus to be much easier to see these threats coming and so easier to avoid them.

If, as Forester asserts, many bicyclists fear overtaking collisions more than they should, we can now understand why this might be. For Gus, the key is right there in the system image. The real-life experiences of riding in traffic, the messages carried in the sights and sounds, conspire to convince him that he has plenty to fear from the rear. What’s more, this same dynamic shapes motorists’ perceptions. Their most frequent and stressful interactions with bicyclists involve trying to squeeze by in shared lanes. So motorists, too, are apt to see overtaking as their primary threat to cyclists.

The implications are profound. Whether or not we as bicycle advocates believe that overtaking motorists are a major threat to cycling, whether or not we think that protecting cyclists from these passing cars should be our highest priority, whether or not we believe that facilities such as bike lanes offer any real protection, we have to understand that the fears people feel are real and that those fears are fostered and perpetuated by the cycling environment, by the system image.

True, there are “bicycle safety” efforts that inappropriately reinforce the fear. But those efforts would have little power or credibility if life on the street did not seem to confirm the message. Gus does not suffer from, as Forester would put it, a “phobia” perpetrated by anti-bicycling interests. Gus’s concept of the dangers he faces while cycling may not match actual crash statistics, but he has nevertheless reached his conclusions through a rational interpretation of his observations.

Betsy the Effective Cyclist

Having taken an Effective Cycling course, Betsy feels comfortable riding in traffic. She’s much more comfortable than Gus with overtaking traffic. The program shaped both Betsy’s primary appraisal of the overtaking threat and secondary appraisal of her skills and options for dealing with that threat.

First, Effective Cycling taught her that motorist-overtaking collisions make up a tiny portion of all the ways in which bicyclists crash. So rear-end collisions are virtually insignificant, the story goes. Like a hypnotist, Forester repeats this message over and over in the course text. In addition to reducing Betsy’s primary appraisal of the threat, Effective Cycling taught her to communicate with overtaking motorists, to affect motorists’ behavior through her body language and road position. Betsy gets a sense of security not from staying out of the way of motor vehicles, but from actively establishing communication with drivers. Unlike Gus, she does not feel helpless. Her secondary appraisal is that she can join the flow and influence the traffic around her. Betsy has “knowledge in the head” that informs her experience of riding in traffic. Her experience is different from Gus’s because she interprets the system image differently. Also, she has a sense of competence while riding in traffic. To her, the cycling environment aids her in getting where she wants to go. For the most part, she sees traffic not as something she has to struggle through, but as something she rides cooperatively with.

Bike lanes: Knowledge in the environment

Gus loves bike lanes. That six-inch white stripe assures him that motorists will stay to the left when passing. It relieves him of uncertainty; he knows where the overtaking car will be. With the stripe he no longer feels helpless, it seems to afford him a simple technique for staying out of harm’s way: Stay on the right side of that line. Bike lanes also give him road space to call his own. Motorists can’t challenge him for being on the road. The pavement markings leave no question about his right to ride there.

Betsy, however, is frustrated and annoyed by bike lanes. The very premise behind bike lanes conflicts with her belief that overtaking traffic is not a significant threat. To her, the stripe is not only unnecessary, it creates a psychological wall that separates her from motorists and undermines the sense of security she would otherwise get from feeling that she is part of the traffic flow. It “says” that motorists and bicyclists both shall stay in their places. The area to the left of the bike lane is often referred to as the “car lane,” a term which implies that it’s for cars only, just as the bike lane is for bikes only. Being accustomed to using the entire roadway, Betsy does not share Gus’s feeling of having gained legitimacy on the road with bike lanes. Instead, she feels like she has lost ground, has been confined to one small part of the street. When moving out of the bike lane to set up for a left turn, she feels as if she is encroaching on motorists’ territory. Also, she finds that it’s harder to get motorists’ attention when she wants to negotiate her way across the roadway, and drivers seem less inclined to let her in when she does catch their eyes. It’s as though motorists pay less attention to cyclists, perhaps because they assume that cyclists will stay in their place. The motorists do not understand why bicyclists would want to ride anywhere but within what appears to be a safe zone.

Bicycle facility design guidelines, both at the state and national level, commonly recommend that bike lane stripes be dropped 50 feet before intersections to make it easier for bicyclists to get into the best position for a turn. But Betsy wants to begin getting into position 300 feet before the turn. Moreover, she sometimes needs to turn mid-block into driveways, not just at intersections.

Also weighing on Betsy’s mind as she pedals down a bike-laned street is Forester’s (1993) analysis that “all practical bikeway designs increase the number and difficulty of collision situations that produce some 30 percent of car-bike collisions while reducing the difficulty of only 2 percent” (p. 547). Forester, the man to whom Betsy feels an allegiance because his writings have helped her to ride comfortably and safely anywhere she wants to go, makes bike lanes sound downright diabolical:

So far as cyclists are concerned, the assumption is that bike lanes don’t impede competent cyclists while making incompetent cyclists believe they are kept safe. So far as motorists are concerned, they believe that bike lanes keep cyclists out of their way. These assumptions contain three errors. One error is physical: we don’t have, and probably will never be able to develop, bike lane designs that properly designate where cyclists should ride at the more difficult locations. The second error is political: such a system tells the world that cyclists should ride, for their own safety and because they are incompetent, in bike lanes and not on streets without bike lanes. Such messages justify the political opposition to competent cyclists and lower their social acceptability. The third error is moral: because bike lanes do nothing to reduce accidents to cyclists (and probably increase them), using bike lanes to attract people to cycling on the premise that they will be protected even if they are incompetent is immoral because it is deadly. It is deadly initially when incompetent cyclists use the roads; it continues to be deadly because it persuades people that becoming competent isn’t necessary for safety (Forester, 1994, p. 124).

Forester (1994) claims to have produced a study with statistical confidence levels of 95 to 99 percent, that “compared the behavior of cyclists in cities with bike-lane systems against the behavior of cyclists in cities without bike lanes” (pp. 119, 136-141). Because he saw bicyclists in the bike-laned cities of Davis and Palo Alto deviate from his list of acceptable cycling techniques more often than the cyclists he observed in the non-bike-laned city of Berkeley, he concluded that he had shown the detrimental effects of bike lanes. Unfortunately, all we can conclude from the study is that the cyclists he observed riding in Berkeley scored better on his proficiency test than those he observed in Davis and Palo Alto. To draw a general conclusion about bike-laned cities as opposed to non-bike-laned cities he would need a study sample of more than just three cities. It seems reasonable to expect some bicyclist behavior changes when you change the system image, when you change the apparent affordances and constraints. Forester’s results are consistent with that conclusion, but they don’t lead to it. It may be true that bike lanes encourage unsafe cycling and discourage competent cycling, but as one Transportation Research Board referee put it, the conclusions Forester drew in this case “are far beyond what is warranted by the meager and biased data” (Forester, 1994, p. 137).

Forester’s charts and tables make his arguments appear authoritative and logical. His words carry extra weight in Betsy’s mind because the League of American Bicyclists appears to endorse his expertise by promoting the education program Forester initiated.

Gus rides his bike unencumbered by such knowledge. His gut responds intuitively to what he sees going on around him. His neighbor Betsy has talked to him about Effective Cycling techniques. But to Gus, Betsy’s habit of fearlessly charging through heavy traffic seems a little nuts. He feels satisfied to let his common sense be his guide in what appears to be a very simple and easy-to-understand activity.

The ABC’s of the “design cyclist”

Wilkinson, Clarke, Epperson, and Knoblauch would call Betsy an “A” bicyclist and Gus a “B” bicyclist (1994a, pp. 1-3; 1994b, pp. 4-6). They define these classifications as follows:

Group A–Advanced Bicyclists

These are experienced riders who can operate under most traffic conditions. They comprise the majority of the current users of collector and arterial streets and are best served by the following:

Group B–Basic Bicyclists

These are casual or new adult and teenage riders who are less confident of their ability to operate in traffic without special provisions for bicycles. Some will develop greater skills and progress to the advanced level, but there will always be many millions of basic bicyclists. They prefer:

This classification system is akin to Forester’s dichotomy. Instead of “A” cyclists, Forester talks about “club cyclists” or “vehicular cyclists.” Then there are his victims of “cyclist inferiority phobia.” Calling this group “B cyclists” is less insulting, but no less simplistic. (Wilkinson, et al., also recognize “C”cyclists, or children, but consider their needs to be essentially the same as “B” cyclists.) Both classification systems are based on cyclists’ experience and assume that with experience comes a change in the kind of environment cyclists prefer.

Forester maintains that we should only accommodate vehicular cyclists’ preferences because everyone else’s preferences are misguided and unsafe. Wilkinson, et al., (1994b) consider “A” cyclists’ preferences—wide curb lanes, mostly—to be the least, not the best, accommodation a city should make for bicyclists:

The Bicycle Federation of America estimates that less than five percent (of U.S. bicyclists) would qualify as experienced or highly skilled bicyclists. Therefore, as the goal is to increase bicycle use and as new users will be predominantly novice riders, any plans must meet the needs of both experienced and less experienced riders (p. 4).

They call Forester’s emphasis on education a “sales approach,” where advocates have a product, a training program, that they have tried to sell, and that the public hasn’t shown much interest in buying. These authors recommend what they call a “marketing approach,” where “the needs of the intended users are determined and a product or service offered that they will find attractive and use” (1994b, p. 4). For “B/C” cyclists on urban streets with posted speed limits of 30 mph or greater, or with average daily traffic over 10,000 vehicles, the product they propose is bike lanes (1994a, pp. 19-20; 1994b, pp. 103-104). (Incidentally, these recommendations by Wilkinson, et al., that bike lanes be used on higher-speed roads contradict the conclusion they drew from their own “variation” studies, that bike lanes work best at slower speeds. The recommendations are, however, in accord with a “response ability” analysis.)

Limitations of dichotomous paradigms

Are all cyclists either “A” or “B”? More importantly, do cyclists progress, as the description implies, from a preference for “well-defined separation” to a preference for shared lanes? This is one of those firm convictions that “mark our limitations and our bounds.” Perhaps if we question those convictions we can expand the bounds of our understanding of bicycle transportation.

One reason to question the experience/preference assumption is that there are plenty of experienced cyclists among the proponents of separate facilities. Bicycling magazine, for example, gave a nod to bike lanes after it commissioned a poll to find out what would encourage more people to bike to work: “The results make it clear: Much of mainstream America is ready for bicycle commuting. The only thing holding it back is a lack of bike lanes, showers at work, and other basic amenities” (Pena, 1991, p. 44). Bicycle advocacy organizations, such as the Santa Cruz-based People Power, New York City’s Transportation Alternatives, and the national Bicycle Federation of America, frequently publish articles in favor of bike lanes.

Since its inception nearly 30 years ago, the Miami Valley Regional Bicycle Council has been a successful advocate for bike paths in the Dayton, Ohio, area. Yet Jerry Hopfengardner, who was MVRBC chair for many years, was not only an Effective Cyclist but a trainer of Effective Cycling instructors. He also served as chairman of the League of American Bicyclists’ Education Committee. “I’ve bought into John Forester’s philosophy of riding—that the safest riding style is to assume your fair share of the road and ride predictably,” Hopfengardner has been quoted as saying. “It has helped to make me more confident and with lane positioning and city riding.” Apparently, though, he has not bought into the anti-bikeway venom that comes through Forester’s writings, for Hopfengardner believes that “there is a place for both bikeways and surface streets that accommodate bicyclists” (MVRBC, 1995, p. 10).

Then there is Palo Alto, California, city council member Ellen Fletcher, who is well known for her bike facilities advocacy in that city, which now has bike lanes and bicycle boulevards. Writer Janet Else Basu (1990) describes cycling around Palo Alto with Fletcher as her guide:

“Stay in the middle of the lane at a stop light,” Fletcher coaches as we pause at an intersection waiting to turn onto busy El Camino Real. “Don’t hug the curb, and at a signal assert yourself so a driver doesn’t occupy the lane and then turn into you without noticing.” Be brave, be bold. “Whatever you do, don’t be timid,” Fletcher adds. “Drivers need to see you and understand your intentions” (p. 72).

Fletcher does not sound like a victim of “cyclist inferiority syndrome.” Experience and training may give cyclists a greater sense of security riding in traffic, but that doesn’t mean that experience erases all desire for some sort of separation from motor vehicles.

Replace “experience” with “traffic tolerance” or “preference”

Clearly, Gus and Betsy do not represent the full spectrum of bicyclists’ personalities. I have met young and novice cyclists who ride boldly on major thoroughfares. I have talked to experienced cyclists who have told me they can’t remember ever having much fear of traffic. I also know some long-time bicyclists who routinely take longer routes to avoid major roads.

One way to get beyond the misleading dichotomy of “skilled” or “experienced” cyclists versus “novice” riders is to replace those terms with levels of “traffic tolerance.” That is, some cyclists have high traffic tolerance, feel comfortable riding in heavy traffic; some have low traffic tolerance, stick to less-traveled routes; and there are many levels of traffic tolerance between. Traffic tolerance, then, would refer to the relative level of stress a bicyclist feels while riding amidst motor traffic. The lower the stress, the higher the traffic tolerance. Betsy has high traffic tolerance. Gus may have low to medium traffic tolerance.

Stress is a complex psychobiological phenomenon. We have already discussed cognitive theory relating stress to primary and secondary appraisal. Studies have also shown relationships between stress and many other factors. For example, the amount of stress experienced by motorists commuting to work is influenced by the degree to which they feel they had a choice in selecting where they live and by whether they tend to perceive events as consequences of their own behavior and therefore under their control or whether they tend to perceive the things that happen around them as independent of them and out of their control (Novaco, et al., 1979, p. 365). A correlation has also been found between commuter stress and “Type A” (no relation to “A” cyclists) “coronary-prone behavior” (Stokols, et al., 1978). The degree to which people react to stressors has been linked to such personality characteristics as introversion and extroversion (Stelmack, 1990). DeLeeuwe, et al., (1992) identified “stress-resistant” and “nonstress-resistant” personalities. Moreland (1993) describes different styles of dealing with stress that correspond to the personality types identified by the Myers-Briggs Type Indicator. Even air pollution has been shown to contribute to stress (Bullinger, 1990)—the motor vehicles with which bicyclists share the road are the world’s single largest source of air pollution (Lowe, 1990, p. 9)—and some people are clearly more sensitive than others to airborne chemicals (Lawson, 1993).

Also, a cyclist’s ability to create mental images could affect his traffic tolerance. Dean and Morris (1991) show relationships between skill on spatial tasks and some aspects of imagery ability. Riding in traffic is a complex spatial task. It’s more complex for cyclists than for motorists because, as discussed before, cyclists have more choices, have fewer environmental clues to guide those choices, and must try to fit in among faster-moving vehicles. A cyclist must create a mental image that keeps track of the surrounding motion and that helps him anticipate where motorists will be in the next moment, even while those motorists are beyond the cyclist’s angle of view. Cyclists with less acute imagery may have more difficulty making the minute-by-minute decisions that traffic cycling requires and would therefore find the experience more stressful and intimidating.

Finally, Carpi (1996) reports that a trauma or series of stressful experiences can “sensitize” a person to stress. That is, the brain gets “rewired” to react more strongly to stress. He quotes Jean King of the University of Massachusetts Medical School: “We must remember that there are physical reactions in our bodies when we are under stress and the extent to which we endure these reactions may be dictated by our past. Telling someone to ‘just take it easy’ is of no help” (p. 70). Likewise, telling someone to “just learn to ride in traffic” may be of little help.

Considering all the possible ways that cyclists might differ in their responses to stressors, it is not sufficient to describe traffic tolerance as simply a function of skill and experience. To do so slights bicyclist who have lower traffic tolerance. Their problem, it implies, is that they are unskilled and inexperienced, a derogatory characterization that does not really get to the heart of the matter. It also puts those with high traffic tolerance on a pedestal. It makes it appear as though, through experience, they have achieved something to which others should aspire. In at least some cases their boldness may be due less to wisdom than to a reckless disregard for their own safety.

Worst of all, defining traffic tolerance as a function of experience encourages bike lane opponents to dismiss bike lane advocates as simply inexperienced and therefore unenlightened. When “vehicular” or “club” cyclists become smug and condescending, they lose their ability to understand and appreciate those cyclists who have no use for the latest clipless pedals. Who is the more experienced cyclist, anyway, the upstart gearhead who does interval training on a thousand-dollar composite frame or the senior who has rambled to the coffee shop and drug store each week for half a century on a rusty-but-trusty mount? Whose devotion to bicycling is more sincere?

Yes, we can expect cyclists to increase their cycling confidence through training and experience. This is true of almost any activity. Ultimately, though, the level of bicyclists’ traffic tolerance will be determined by many other things as well. Moreover, we might expect that cyclists with inherently high traffic tolerance would be more likely to become avid cyclists than would cyclists with low traffic tolerance. So, is high traffic tolerance an effect or a cause of more cycling experience? It could be both. We can avoid the whole chicken-or-egg problem by accepting that cyclists have differing levels of traffic tolerance, and that this is because they are human.

We could take this concept even further and use the term “traffic preference.” That is, even some bicyclists who have high traffic tolerance prefer a distinct separation between bicyclists and motorists. There is no need to psychoanalyze their reasons. In practice, it doesn’t matter if bicyclists can’t tolerate traffic or just prefer separation; either way their decision to ride a bike or take a car can hinge on the system image of the cycling environment.

Hoque (1990) compared actual bicyclist fatality figures from Australia with the results of a study by the Bicycle Institute of Victoria, which asked bicyclists to report their perceptions of the relative danger of a car-bike collision from different directions (see Figure 8). The bicyclists’ perceptions mirrored the fatality figures rather closely. So these perceptions differ from the actual likelihood of being hit as measured by non-fatal collisions, but they do seem sensitive to the danger of serious injury as measured by fatalities. At least in this case, what might be considered a distorted perception of risk may not be as off-base as it appears, providing another example of why it’s best to respect bicyclists’ varying perceptions.

Figure 8
Daytime crashes (outer circle of numbers) compared to bicyclists' perceptions (inner circle of numbers) of risk of colliding with vehicles from various directions in percent of total collisions. Source: Hoque, 1990.

Percentage of crashes by direction.


Linking stress and design

There is nothing new about looking at user stress to evaluate environmental design (Perin, 1970, p. 76). Bicycle-related research, though, has tended to focus on crash statistics and bicyclist behavior. Debates over bicycle facility issues frequently hone in on what’s safe, or what’s not. This is beginning to change, however. One reason is that bicyclists, and bicycle transportation planners and engineers, are becoming increasingly aware that separate facilities have limited potential for improving bicyclist safety. Side paths can be awkward or even downright dangerous, rail trails require select locations, and bike lanes haven’t been proven effective for reducing crashes either. So facilities advocates have begun to shift their arguments away from safety and to talk more about encouraging people to use bicycles. This shift has been bolstered by a movement throughout our society to put less emphasis on catering to cars and more on fostering a diversity of transportation options, a philosophy codified in the Intermodal Surface Transportation Efficiency Act of 1991.

One consequence of this shift is that the concept of stress has finally made its way into bicycle-related research, most notably in the work of Sorton and Walsh (1994), who have devised a rating system by which to assess the “bicycling stress levels” of different kinds of streets. Their system is still in the developmental stages and, unfortunately, uses experience-based categories for bicyclists. I won’t describe it in detail here. However, while explaining the stress level concept, Sorton and Walsh make a significant point:

It is well known that bicyclists choose routes that will cost them the least amount of effort. They save energy by following the flattest route, one that will enable them to avoid stopping and slowing as much as possible. However, conserving physical effort is only part of the story.

Bicyclists also seek to avoid conflict with motor vehicles, harassment from heavy traffic and the strain of having to concentrate for long periods on riding along narrow, high-speed, high-volume roads. In other words, they want to reduce not only physical effort but mental stress as well (p. 3).

This concisely describes two key factors that bicyclists weigh when choosing routes, and when choosing whether to travel by bicycle or by motor vehicle. It is also a respectful description because it makes no judgment about whether the “mental stress” is justified or overblown. Finally, it is a departure from much of the research of the past. That is, it frames the bicycling problem in terms of the bicyclists’ subjective experience. Past research has had an almost obsessive preoccupation with crash statistics.

Certainly, bicycle transportation planners need to understand the car-bike dynamics that cause collisions. But their efforts to reduce those collisions will have limited effect if planners do not also understand bicyclists’ desire to keep both the physical and the mental stress to a minimum. If, in the name of safety, you plan elaborate networks of circuitous bike routes that wind through quiet residential streets and you build bike paths in vacant and remote corridors, you are likely to find, as many communities have, that bicyclists largely ignore your efforts and take the more direct routes anyway. In surveys, people often say that fear of traffic is the number one reason they choose not to bicycle. But other factors play strong roles, too. In a survey of downtown workers in Missoula, Montana, for example, Bickell (1992) found that “time/distance constraints” was the top deterrent to bicycle commuting, followed by a closely-related problem: the need to run personal or work-related errands. “Safety concerns,” which included darkness in addition to traffic fears, was next; it garnered barely more than a third of the points that time/distance got (p. 4).

On the other hand, there are those who oppose virtually any kind of separate facility in the belief that it is far more important to accommodate bicyclists to the streets than to make the streets accommodate bicyclists. You may come to this conclusion by analyzing crash statistics. But it’s a cold analysis that lacks appreciation for the depth and breadth of human perception and behavior. In Sorton’s and Walsh’s preliminary studies, even “experienced” cyclists reported higher stress levels when streets were narrower, busier, and faster (pp. 8-9). Cyclists who tolerate traffic well may feel exhilarated by the stress. But for many others, city cycling means either taking a route that is too far and time-consuming, or one that is too nerve racking; in either case, the bike stays home. So far, we have seen two possible remedies for this. We can use a program like Effective Cycling to train cyclists and build their skill and their trust in traffic. Or we can change the system image of the streets themselves with bike lanes.

Bike lane design and mental models

Whatever bike-facility choices we make, we need to be aware of what the system image of our streets—and this includes the entire street system, not just a single avenue in isolation—says to bicyclists and motorists about how to behave. Streets are more than just smooth surfaces on which to travel. Their nonfixed, semifixed, and fixed-feature elements are nonverbal cues, mnemonic devices. As Rapoport points out, the various elements of the built environment have meanings that we can infer by observing how people behave (or don’t behave) in a setting. We can also learn a lot by listening to what people say about a setting. Do they feel welcome there? Does the setting support or hinder what they want to do?

Of course, nonverbal cues cannot tell bicyclists everything they need to know. Any cyclist needs some knowledge in the head in order to ride safely, just as any motorist does to drive safely. Ideally, though, the street system would communicate in a way that would help bicyclists and motorists form accurate mental models of the safest and most efficient ways in which to interact:

The power of mental modes is that they let you figure out what would happen in novel situations. Or, if you are actually doing the task and there is a problem, they let you figure out what is happening. If the model is wrong, you will be wrong too (Norman, 1988, p. 71).

Some elements of bike lane designs commonly used today aid bicyclists in understanding how best to get through traffic. Unfortunately, there are other elements that seem to contradict a good mental model for how bicyclists and motorists should mix. Among the good things is that bike lanes can help cyclists move away from the curb when that’s the safest thing to do. For example, a bike lane that runs to the left of a right-turn-only lane helps cyclists and motorists understand that a bicyclist riding straight through the intersection needs to move out into the through lane. When there is no bike lane, many cyclists hug the curb and try to cross the intersection from the right of the right-turn-only lane, which puts them in conflict with right-turning motorists. Likewise, a bike lane that runs to the left of a bus bay keeps cyclists from getting stuck between the curb and busses that are stopping. This is a blessing not only for cyclists and bus drivers, but also for bus passengers as they get on and off the buses. Without a bike lane, few cyclists would have the nerve to ride way out between the bus lane and the car lanes. Moreover, cyclists who did have the guts to do it might get a cool reception from motorists; without the bike lanes it just doesn’t look like bicycles belong there.

Direction arrows on bike lanes would appear to promote a good mental model by discouraging wrong-way riding. In the chapter on bicyclist behavior we saw that this appeared to be the case in one Oregon study, but in another part of the country bike lanes seemed to invite wrong-way cycling. For some cyclists—this may depend a lot on the local cycling culture—the seemingly-protected space in a bike lane creates an affordance that encourages wrong-way riding more than direction arrows discourage it.

Perhaps the worst way in which bike lanes allegedly contradict a good mental model has to do with the behavior of left-turning bicyclists and right-turning motorists:

Bicycle lanes tend to complicate both bicycle and motor vehicle turning movements at intersections. Because they encourage bicyclists to keep to the right and motorists to keep to the left, both operators are somewhat discouraged from merging in advance of turns. Thus, some bicyclists will begin left turns from the right side bicycle lane and some motorists will begin right turns from the left of the bicycle lane. Both maneuvers are contrary to established Rules of the Road and result in conflicts (American Association of State Highway and Transportation Officials, 1991, pp. 18-19).

Oregon Department of Transportation bikeway specialist Michael Ronkin’s bike-lane-advocate reply is that these problems “are dealt with by dashing the bike lane stripe before intersections and dropping the markings altogether across intersections” (Clarke & Tracy, 1995, p. 81). The Manual on Uniform Traffic Control Devices (MUTCD) (Federal Highway Administration, 1988) recommends dropping the line “not less than 50 feet” before an intersection (p. 9B-12). Once again, there are two problems with this remedy. First, a bicyclist or motorist who waits until he is just 50 feet from an intersection to merge is one who waits until the last moment, then cuts across at the turn; not a safe practice. Second, even if dropping the line before an intersection was helpful, it wouldn’t help bicyclists turning left into a mid-block driveway. Lott and Lott found that bicyclists making left turns from the right side of the road is a type of car-bike collision that increases with bike lanes (Wilkinson, et al., 1994b, p. 25).

The question, then, is whether it’s possible to create bike space on the road without creating a system image that contradicts the rules of the road. One possible way to minimize the problem is to change the way bike lanes are delineated. Typically, a lane is marked off with a solid white line, often an imposing six-inch-wide line. When not referring to bike lanes, the MUTCD recommends such a wide line only for “areas where it is advisable to discourage lane changing” (p. 3B-2). Usually, these hefty lines are reserved for chores such as marking the edge of a highway, or for intersection approaches where the goal is to discourage motorists from changing lanes at the last minute. Bike lane striping seems to have this concept backward. Contrary to the wisdom behind standard street lane markings, bike lane lines seem designed to discourage lateral movements mid block and to encourage mad-dash turns at intersections.

This is not to say that anyone means for bike lanes to have this effect. Surely, the solid line is there to give bicyclists a sense of protection from traffic. Also, the MUTCD does recommend solid lines for what it calls “special secondary lanes” (p. 3B-2), which include uphill truck lanes and transit bus lanes; so it could be argued that its a case of bike lanes being consistent with these uses. Overall, the practice of dropping the bike lane line before an intersection is simply meant to recognize that bicyclists need to match lane position to destination. Even so, bike lanes might promote better mental models for bicyclists and motorists if, instead of solid lines, the lanes were set off for their entire length with broken lines to indicate that bikes may merge out of the lanes to turn left and cars may merge into them to turn right:

… knowledge in the world is useful only if there is a natural, easily interpreted relationship between that knowledge and the information it is intended to convey about possible actions and outcomes.

Note, however, that when a user is able to internalize the required knowledge—that is, to get it into the head—performance can be faster and more efficient. Therefore, the design should not impede action, especially for those well-practiced, experienced users who have internalized the knowledge. It should be easy to go back and forth, to combine the knowledge in the head with that in the world. Let whichever is more readily available at the moment be used without interfering with the other and allow for mutual support (Norman, 1988, p. 189).

Some variations on the standard bike lane design have recently been proposed or tried that may offer more “natural, easily interpreted” relationships between design and desired behavior. The New York-based Transportation Alternatives has illustrated its Bicycle Blueprint with pictures of bike lanes that are delineated with double stripes, solid on the motor vehicle side and broken on the bicyclists’ side (Herman, et al., 1993, p. 32). That’s a good idea for left-turning cyclists, but does nothing for right-turning cars. Denver, Colorado, is trying out a promising concept dubbed “hybrid lanes” (Clarke & Tracy, 1995, p. 80; MacKay, 1994, sec. A-10; Zehnpfenning, Cromar & Maclennan, 1993, pp. 38-39). The city dropped the lane line and instead just paints the bike lane symbol (See Figure 9) in the middle of the zone where a traditional bike lane would be. This gives more freedom to bicyclists who felt hemmed in by traditional bike lanes, gives some reserved space to those who need a comfort zone, reinforces the proper riding direction, and reduces bike lane maintenance costs by using less paint.

Figure 9
Hybrid lane pavement marking.
Source: MacKay, 1994.

Figure 9: Hybrid lane marking.


Zehnpfenning, Cromar, and Maclennan (1993) describe a number of hybrid lane benefits:

Like so much that is written about the pros and cons of various facilities, the above list contains a fair amount of speculation. One further bit of speculation: it seems fair to say that bicyclists with low tolerance for traffic would not feel as protected while riding in a hybrid lane as they would with a solid stripe. However, the amount of stress bicyclists feel while riding in traffic increases on a continuum as the road gets narrower and the traffic gets faster and heavier. More than that, the relationship between bicyclists and motorists changes as the difference in their speeds increases. Remember, even Forester admits that in high-speed traffic bicyclists must play “road sneak” because negotiation becomes impossible. In other words, the degree to which bicyclists can communicate with motor traffic and participate in its flow is also a continuum. The bike lane system image could reflect this continuum if, for example, it employed hybrid lanes when the traffic was slower, dashed lines in medium-speed traffic, and solid lines where the motor traffic travels at full highway speed. I speculate that a bike lane system with this kind of flexibility would be less likely to frustrate vehicular-style cyclists and yet would change the system image enough to make it less stressful for cyclists with low traffic tolerance. The hybrid would also be less likely than current bike lane design to perpetuate mental models that inhibit bicyclists and discourage them from learning or practicing vehicular cycling techniques.

Previous | Contents | Next