How's My Driving? A Q&A With the Author of Traffic

Traffic

Traffic and congestion have come up a lot on this blog lately.

We even blegged for parking solutions and analyzed the effectiveness of traffic signs — according to Tom Vanderbilt, author of the book Traffic (due out July 29), they’re virtually ineffective and may even “allow us to basically stop thinking.”

But dozing cows, he says, can work better than speed bumps to slow traffic.

In his book, Vanderbilt explains how traffic works (including why the other lane always moves faster), why we drive the way we do, and what we can do about it.

Road rage isn’t all bad, he claims:

… it can be quite therapeutic to act like a yelling maniac once in a while, and the plush, leather-seated interior of a car provides a nice, semi-private environment in which to do that.

VanderbiltTom Vanderbilt, Photo by Kate Burton.

He also gives great driving tips, like: don’t drive with divorcees.

Vanderbilt is contributing editor to I.D. and Print, and writes for, among other publications, Wired, the Wall Street Journal, The New York Times and his own blog.

He has agreed to answer our questions about his forthcoming book.

(And if someone in a 2001 Volvo V40 ever cuts you off and looks like this ——————–>
just smile and wave.)

Q: A soft-spoken friend of mine turns into a yelling maniac when someone cuts him off in traffic. You say in the book that drivers “struggle to stay human” behind the wheel. What happens to my friend?

A: What happens to most of us, in most driving conditions, is that we’re losing some of the key attributes that facilitate human cooperation and, in a larger sense, society.

Eye contact, for example, has been shown in any number of experiments to increase the chance of gaining cooperation — that’s why when drivers give you what was called on Seinfeld the “stare-ahead,” your chances that they’ll let you merge in ahead of them are greatly reduced.

Then there’s the anonymity in traffic — there’s no one to spread rumors or gossip about you about how bad your behavior was — not to mention the lack of consequences for acting like an idiot. It’s all strikingly similar to the way we act on the internet, in what’s called the “online disinhibition effect.”

We can lurk behind screen names (our car and license plate), say nasty things about people (honk, give the finger), and then sign off and never be heard from again (drive away).

In a nice convergence of this, the website PlateWire allows people to anonymously insult others’ rude and unsafe behavior.

I personally think it can be quite therapeutic to act like a yelling maniac once in a while, and the plush, leather-seated interior of a car provides a nice, semi-private environment in which to do that. Remember, in traffic no one can hear you scream.

Q: How effective are road signs really?

A: Last week I just spoke to someone at the Chicago D.O.T. At a particularly curvy part of Lake Shore Drive, they’ve had trouble with crashes. They tried putting up signs, then tried larger signs, then tried larger flashing signs. Still people behave foolishly. Finally they tried putting markings on the pavement that trick drivers into thinking they’re going faster than they are (an example of the “choice architecture” discussed in Thaler and Sunstein‘s Nudge). When we actually see signs to begin with, as we often don’t seem to when doing something like talking on a cell phone and driving, it is a further mystery how and if we decide to act on that information.

“Children at play” signs and the like are absolutely ineffective in changing a driver’s behavior, and studies of drivers through school zones show they were driving much faster than they remember. It’s been argued that signs allow us to basically stop thinking, and in certain places experiments have been done in which they’ve been removed, with no negative safety effects.

Q: How about these?

Road SignPhoto: Ben Hayes
Road SignPhoto: Ben Hayes


A: The signs remind me of the “relax” messages emblazoned on traffic lights in Delhi – amusing, but with little hope of changing actual behavior.

Q: When I was in Delhi, I felt weirdly safer in an auto-rickshaw weaving past cars, motorcycles, and the occasional cow. Is there some logic to that or am I just insane?

A:
Statistically, you’re much better off in an auto-rickshaw than on a motorbike or on foot in Delhi, though this is a complex question with many variables.

Also, in any city you’re safer in a car than you are in rural areas, where the bulk of fatalities happen. On cows in the street — Delhi’s former top traffic cop told me he was of two minds; yes, they obstruct traffic flow, but they also act as a calming influence on Delhi’s “rash” drivers. A cow is only a hazard if you drive in a way that doesn’t take into account the cow.

Q: So why does the other lane always move faster?

A: Well, if you’re on something like the SR 91 “high occupancy tolling” lanes in Orange County, it’s because you’re paying more for the privilege. And sometimes, the adjoining lane may be losing vehicles because of an impending off-ramp.

But there’s a curious bias that plays out in oscillating shifts of traffic, observed by the researchers Donald Redelmeier and Robert Tibshirani, in which drivers, who are oriented towards observing things in the forward view much more than the rear, spend more time watching cars passing them than they spend watching themselves passing other cars.

Given the general findings that humans are more sensitive to losses than gains, it doesn’t seem a stretch to imagine that this sense of being passed — of the other lane being faster — would stick out in our brains. All you have to do is pick out a benchmark car in the adjoining lane to see how often we fall for this illusion. I’ve seen these cars pass well out of vision, only to find myself passing them again minutes later. Part of the reason this seesaw effect is happening in the first place is because of all the drivers ahead thought they could get a better deal, and basically ended up just shifting the equilibrium around temporarily.

There are rarely any better deals in congestion, but that doesn’t stop people from bargain hunting.

Q: I’ve thrown paper, yelled in my car, and revenge-cut people off (I learned to drive in New Jersey). What’s the best way to deal with festering road rage?

A: Because I’ve done everything you’ve mentioned myself, I’m probably not the best person to ask!

Honking is a cheap signal, like birds chirping to warn of an approaching predator, that arguably may do some good for the species by letting someone know that the way they’re acting is decidedly not cool.

The problem is the person doing the offending — and often that may be you — may likely not be the type of person to register your feedback, or accept it as any “teachable moment.”

It’s probably better to maintain the moral high ground and simply attribute other’s bad acts to their slovenly moral decrepitude, or my favorite tactic is to call those highway patrol numbers they have posted for this very purpose. Lior Strahilavetz at the University of Chicago has argued for universal “how’s my driving” stickers, so we could all exchange eBay-style feedback on the road, reporting cheaters — people who aren’t playing by the rules. I think this is a promising solution.

Q: In the book, you mention using game theory’s “asymmetry in communication” concept when negotiating your way through traffic. How is this done and it is wise/ safe?

A: This is the whole idea, articulated by Thomas Schelling, of making yourself “unavailable” for the receipt of messages, thus forcing your adversary to make a decision based on imperfect information.

It’s like sending a nuke-loaded bomber towards Moscow and shutting off ringer on the red phone. If, at one of Mexico City’s many unsignalized intersections, for example, you approach at roughly the same time as another car, you could quite possibly “win” the intersection by not looking at the other driver — as long as he knows you haven’t seen him, or is unsure you haven’t seen him.

Schelling argued you could most strongly signal your commitment to winning the intersection by throwing your steering wheel out the window, but I don’t recommend that form of deterrence!

Studies have actually shown things like that when pedestrians crossing in crosswalks (the mid-block kind where drivers are supposed to yield to those in the walk) do not look at drivers, drivers are more likely to grant them the right of way. I see this in New York City occasionally — a pedestrian just totally, obliviously crossing the street, not looking at me at all. Of course, I have no alternative but to slow.

As much I applaud any sort of impromptu “traffic calming,” I wouldn’t recommend not looking at drivers, if only because you can’t be sure they’ve seen you. Drivers are far from rational actors.

Q: Would you prefer being stuck in traffic with a bunch of narcissists or insecure people? Why?

A: Insecure, for sure. Narcissism has been implicated in aggressive driving behavior. People who drive expensive cars, for example, have been shown to do things like honk faster and drive closer to the car in front of them.

Actually narcissism as a cultural force may contribute to the sense out there that driving behavior has gotten progressively worse. This tracks with increases in self-reported narcissistic behavior in psychological tests. More people these days are likely to say “yes” to questions like “if I ruled the world, it would be a better place,” and more people seem to be driving in that spirit.

Q: What about the way we drive leads to the most accidents?

A: That’s an easy one: distraction, speed, and human psychology.

We don’t pay enough attention when we drive, and our attention itself is a fragile entity prone to any number of gaps and distortions. To make matters worse, we drive fast. Speed is the fundamental cause of all traffic danger — not only do injuries and deaths rise with speed (places where people never drive above 35 mph have very low car fatality rates, full-stop), our ability to avoid injuries and deaths declines with increased speed.

And often we have precious little notion of how fast we would be able to react to something; in night driving, for example, we “overdrive” our headlights with shocking regularity. But we’re blind to just how blind we are.

Q: How does talking on a cell-phone really affect your driving ability?

A: A prominent human factors researcher put it simply to me: We can’t time-share. We can get away with doing very simple things in tandem, but with any activity with a certain range of difficulty — like driving or talking on a cellphone — there is always a cognitive detriment.

Speed reading comes at the cost of comprehension. The crawl on CNN disrupts our ability to focus on the actual images of the news as well as the crawl itself. Even people’s walking is affected by their talking on a cell-phone. Unlike alcohol, we have a positive social image associated with cell-phone use, even when in the car — it’s the very image of viral multi-tasking. We need to call home to find out if there’s milk in the fridge (somehow we got by without the ability when I first learned to drive).

But that positive image is an illusion, and it may be just one factor that helps explain the stubbornly high crash and fatality levels in this country.

I can hear you say, but I drive all the time on my phone. Most drunks make it home at night too, through sheer dumb luck, but think of all the death and damage done by those who don’t. Laws will be hard to enforce on this; what we really need is a strong social norm that says it’s just not a good idea to drive and talk — and hands-free offers no cognitive benefit over a hand-held cell-phone.

Q: What’s the most effective or promising traffic model you’ve come across?

A: One technology that is quite clever and productive is the concept of “variable speed limits,” as seen on highways in England and elsewhere.

Basically, if there’s a patch of congestion, it’s detected by sensors, and a new, slower speed limit is announced back “upstream.” That way, rather than having everyone drive at full speed into a traffic jam, which is neither good for safety nor traffic flow itself, the congestion shock wave is “damped.” Of course, people being people, you tend to have to have things like speed limit cameras there to make sure people go the suggested speed.


Q:
You have a whole chapter devoted to the fact that women cause more congestion. Why, and does this influence insurance prices for men and women?

A: There has been a historical shift over the past few decades in women’s participation in the workforce, so their driving, as a whole, is now almost equal to men’s. Men still drive more miles, but women make more trips.

The reason is they’re statistically doing more of the non-work trips — picking up the kids, etc. (and the more kids you have, the more miles you drive). Men still tend to do less what of is called “trip-chaining.” Women make more of these trips during the peak hours on the roads that are less well designed for heavy traffic flow; hence the “more congestion” argument.

Men’s fatality risks in traffic are still higher, but apart from young males, the gender gap in insurance has closed rapidly, in part because the exposure of women drivers has increased so much.

Q: It takes me an abnormal amount of time to parallel park. Am I somehow affecting traffic?

A: Urban street parking is one of those curious trade-offs. Some engineers hate street parking; they say it clogs roads (remember, a single double-parker on a street cuts the throughput in half) and causes crashes.

But others, and I view myself here, see it as an effective traffic calming device. People drive measurably slower on streets which are enclosed by rows of parked cars.

It’s searching for parking that’s more problematic, as Donald Shoup at UCLA shows us. Underpriced or free street parking causes significant amounts of excess traffic, as people “cruise,” or bargain hunt, for spaces. Meters should be set at rates, he argues, that ensure roughly 15 percent vacancy at any time. To paraphrase a cliché: there’s no such thing as free parking.

Q: What’s a surprising thing that can make roads more dangerous?

A: Sometimes, it’s the appearance of safety that makes roads dangerous.

Take four-way signalized intersections versus roundabouts, for example. Most people in the U.S. prefer the former, which give clear, simple, precise instructions on how to proceed; whereas we still tend to view roundabouts as these weird, confusing European imports.

But four-way signalized intersections physically have many more potential points of “conflict” than roundabouts; also, roughly 1,000 people per year die in intersections because of red-light running. People tend to reduce their caution going through signalized intersections because they think they “have the light.” At a roundabout, you’re less sure of your right of way, and in any case, everyone has to slow to enter a roundabout, so by its very nature it prevents the dangerous t-bone collisions of a four-way signalized intersection.

Q:What’s the best way to decrease traffic fatalities?

A: One answer would be to get men off the roads. World War II, when the bulk of men were overseas, was the golden age of traffic safety in the U.S. (granted, there were a lot fewer miles driven as well).

The next largest drop came in 1974, when the 55 mph speed limit was enacted (the recession helped somewhat too, as it may this year). Fatalities dropped by more than 9,000 — we’ve not managed anything remotely similar in a single year since (and in some years they’ve gone up).

Then there’s the painfully obvious things — don’t drink and drive, and wear a seat-belt (it’s amazing how many people who do the former don’t do the latter — not that wearing a seat-belt excuses drunk driving). But speed really lies at the heart of fatalities. When a car hits a pedestrian at 20 mph, the chances of survival for the pedestrian are roughly 9 in 10, when the speed is 30 mph, that drops to half.

In a world of a maximum 20 mph speed limit, even impaired driving would produce a fraction of the fatalities it does today. Crash damage to cars (and thus their occupants) is also non-linear — the faster you go, the worse it gets, and all the airbags in the world can’t overcome certain physiological limits of what forces humans can tolerate. Certain vehicle technologies like electronic stability control will help, though arguably not by the optimistic rate the regulators say they will, as people will invariably drive more recklessly.

Q: In the book, you give many examples of how human nature works against us when driving. Does it ever work in our favor?

A: Most of us do move about every day without incident, and I think there’s something pretty incredible in the fact that these complex systems work to begin with. Whether it’s cultural, or a product of evolution, I do think the human impulse for cooperation — even when we’ve lost most of the human interaction — does aid in getting traffic to work as well as it does.

Most of us, I believe, want to do the right thing on the road; unfortunately, those that don’t are the ones that tend to stick out in our minds.

Leave A Comment

Comments are moderated and generally will be posted if they are on-topic and not abusive.

 

COMMENTS: 60

View All Comments »
  1. Adam says:

    If I can drive a manual transmission, and talk to someone in the passenger seat at the same time, how is that different than talking on a cell phone without a hands-free device?

    Thumb up 0 Thumb down 0
  2. John says:

    Adam,

    A passenger in the car with you will see potential danger and adjust his/her speech pattern. They are aware of your muti-tasking, and will most likely pause while your attention is diverted, say during a merge or lane change. The person on the other end of your cell phone does not, and will speak to you as if they have your full attention.

    Thumb up 0 Thumb down 0
  3. RubyTues says:

    Adam, I know your being a smart@ss here but I’ve actually heard this explained before.

    When someone is in the car with you they are more likely to see that moron dangerously changing lanes in front of you and stop talking while you take evasive action. If you are on the phone they don’t actually see what is going on and chatter on happily.

    Driving a manual transmission isn’t really a distraction from driving since it is technically part of driving. If anything it keeps you focused on the fact that you are in fact driving.

    In fact now that you mention it I wonder if there are any studies out concerning the correlation between the level of a vehicle’s automation and the propensity of driver distraction. By that I mean that the more complex a car is to drive the more likely you are to pay attention to what you are doing. I know that nothing focuses your attention on driving more than a brief adrenaline boosting moment of hydroplaning.

    Thumb up 0 Thumb down 0
  4. prklypr says:

    @adam – big difference. The person in the seat next to you is aware of your surroundings – if they see you headed into a tough driving situation, they can stop talking and allow you to concentrate fully. The person on the other end of the phone call has no idea.

    Thumb up 0 Thumb down 0
  5. Mike B says:

    I was able to answer an interesting question using game theory during my daily commute. A 30 month bridge replacement causes a 3-lane highway to funnel into a single lane. The question is, which lane is faster, the mering lane or the mergee lane?

    The intuitive answer is the mergee lane is faster because the merging lane must “ask permission” to merge in and the mergees have no incentive to grant it. However it became extraordinarily clear that the merging lane has the advantage because the mergee lane has a distinct inability to collude to block out the merging traffic. In a stop and go situation it is infeasible to stay close enough to the driver ahead of you to prevent a merging vehicle from nosing in.

    At best this creates a situation where the lanes move equally, but in practice the advantage goes to the merging lane because there is not just a single merge point. Merging traffic can nose in at any point they choose. The more risk adverse the driver the farther in advance of the “no return” point they will merge and as soon as they merge in they join the mergee lane and can create new gaps for merging traffic to nose into.

    My observations in the terminal area before the “no return” point slowed about 3 merging cars would get into the cattle chute for ever one mergee car. One would predict that over time the demand for lanes would adjust and more drivers would choose the merging lane until wait times equalized, but remarkably this was not the case. A surprising number of drivers would sit in the mergee lane and sit there crawling along at 10-20mph in a queue stretching some 3000 feet from the merge point while drivers in the mergee lane (myself included) shot past them at 40-50 mph before slowing for the 250-500 foot long merging queue. Wait times were perhaps 1 minute for the merging lane and 5-10 minutes in the mergee lane. This phenomena persisted for the 18 months I was subjected to the construction.

    At this point I should probably consult a psychologist, but I would guess it has something to do with a general aversion to confrontation (which in this case is the act of having to nose one way into traffic). I will admit to this aversion myself, but after having sat in the mergee line for 10 minutes while merging traffic zipped by I decided to deal with it and relied on my faith in game theory that the mergee cars would never be able to collude and I would be able to nose in without much hassle.

    I also incorporated game theory into my driving after I merged. If drivers behind me in the merging lane tried to leapfrog me in a greedy way I would either try to “punish” this behavior by closing the gap to prevent a nose-in (often impractical) or better yet straddling the line to head off any leapfrog attempt in the first place.

    For one last personal note, I generally have the biggest problems with slow “safe” drivers than what I would consider aggressive drivers. Some people I know get very angry at fast drivers, but while they are acting in an unsafe manner they do know how to keep traffic moving (and also bait cops). Slow drivers are usually the cause of delays, congestion and most of my consternation incurred while driving. In fact I make most of my worst driving decisions when trying to find some way around a slow driver.

    My two goals when driving are to maintain a safe following distance and to avoid having to use the brake (which turns valuable gas into waste heat). Nothing makes me feel better than to be passed by an aggressive driver only to watch him have to slam on the brake. He might end up one car length ahead of me, but I’ve won :-)

    Thumb up 1 Thumb down 0
    • TacticalDriver says:

      Hey Mike,

      Wowww… .. what an interesting, point of view and way of expressing it. It would be interesting to know what is that you are doing for living and if you job/profession has something to do with vehicle dynamics and/or driving psychology/behavior.

      How do I get in touch with you?
      (send me an email: marinpa(at)msn(dot)com

      Thumb up 0 Thumb down 0
  6. Justin says:

    You would agree that human error is most often responsible for congestion and for accidents, so why not take the driver out of the equation through an automated system where vehicles are powered by a guideway, for example?

    Thumb up 0 Thumb down 0
  7. Imo says:

    Adam-

    There’s also the fact that when you are talking on the phone, part of your brain’s attention is diverted to mentally visualizing the person with whom you are speaking. This is distracting enough to make it very dangerous to phone and drive. When the person is in the car, much less “processing power” is required to converse with them, as you see them and don’t have to visualize them.

    Thumb up 0 Thumb down 0
  8. Stacy says:

    Re: Adam

    This recent article in the NYT has a good explanation.
    http://www.nytimes.com/2008/06/03/health/research/03brai.html?_r=1&oref=slogin

    Thumb up 0 Thumb down 0