Stop Signs Are Dangerous

In the latest Atlantic, John Staddon, a professor of psychology at Duke, has an article on the dangers of road signs and speed limits:

The American system of traffic control, with its many signs and stops, and with its specific rules tailored to every bend in the road, has had the unintended consequence of causing more accidents than it prevents. Paradoxically, almost every new sign put up in the U.S. probably makes drivers a little safer on the stretch of road it guards. But collectively, the forests of signs along American roadways, and the multitude of rules to look out for, are quite deadly.

His argument rests on inattentional blindness, or the inability of the brain to focus to multiple things at the same time. Because all those traffic signs compete for the scarce resource of driver attention, he argues that drivers end up neglecting more necessary variables, like the car that stopped directly in front of them. Basically, when it comes to driving it's not the Magical Number Seven (Plus or Minus Two): it's the Magical Number One (Plus or Minus One). The conscious brain, while controlling a car, can only handle one (or perhaps two) things at the same time.

The article has a certain contrarian allure, but the statistics are not completely compelling. Staddon's evidence consists mainly of the fact that fewer people die from car accidents in Britain, which seems to have a slightly simpler method of directing traffic. (As an American in Britain, I found all those dashed lines and tiny roundabouts utterly confusing. I often felt like Chevy Chase in European Vacation.) He also cites a few examples of high-density European streets (Kensington High Street, etc.) where traffic signs have been completely eliminated. The end result is a reduction in pedestrian accidents, although he doesn't say whether there was also a reduction in overall traffic accidents.

Personally, I'd find the argument more compelling if Staddon had taken driver training into account. After a few months of driving, many of the basic skills of driving become utterly automatic. You don't have to consciously think about stopping at a red light because your brain automatically responds to the cue. (I always find yellow lights dangerous precisely because they require drivers to make a deliberate decision.) A stop sign doesn't consume a lot of computational space, since you know that the sign means "put foot on brake". You don't think about it, you just do it. While a world without traffic signs and speed limits might make driving more, um, interesting, I'm not sure that it's wiser to count on the constant vigilance of everyone on the road. I feel safer knowing that the guy in the big truck yapping away on his cell phone, or the driver eating lunch while shifting, doesn't need to be paying conscious attention to the road: their unconscious brain can obey the rules without them.

More like this

Perhaps you are proposing the question in a slightly backwards manner. Rather than asking what is it about engineers that might make them more likely to accept intelligent design, perhaps it would be more correct to ask what is about scientists that make then less likely to accept it? Scientists, if I am not mistaken, are rigorously trained to gather data in methods that remove as much error and bias as possible and are expected to be incredibly scrupulous in reaching conclusion that are isolated from any pre-existing expectations. This would, presumably, mean that a well trained scientist is less likely to accept intelligent design as it does not represent good science.

And that was posted to the wrong blog entirely.
Please delete at your convenience.

Spotweld - A proper bashing of ID or its adherents is never wrong.

re: Driving... Isn't it true that is GB talking on cell-phones and driving has been illegal? IMO, this would have a large effect on any comparison of US - GB driving and accident stats.

I would also be interested to see some stats on SUV drivers vs "normal-size car" drivers. Again, IMO, SUV drivers are the worst. And of course you have to consider age a factor.
Who would trust John McCain to drive a car, let alone a country?

My $.02. Thanks for giving me the forum to rant.

I'll have to buy this issue of Atlantic, but based on your item, it strikes me that Staddon may well have made the sort of broad-brush conclusions that give psychologists a bad name.

Simple example - two roads that cross at right angles.

Case 1 - no buildings or anything else to obstruct the view, 50 vehicles/day through the intersection. No signs? Probably not an issue.

Let's skip a few intermediate steps on our way to ...

Case 2 - buildings on all 4 corners, 5000 vehicles/day through the intersection. Still feel safe?

I'd be on his side if he were arguing that signs/lights should be tailored to the conditions at a particular intersection ... but in my experience, that's what's done today. Not perfectly, to be sure, but a broad conclusion such as ...

Paradoxically, almost every new sign put up in the U.S. probably makes drivers a little safer on the stretch of road it guards. But collectively, the forests of signs along American roadways, and the multitude of rules to look out for, are quite deadly.

... strikes me as the sort of pseudo-paradox beloved of psychologists (and yes, that's a broad-brush observation of my own!).

Have signs and other controls gone too far in some cases? Undoubtedly, based on the roads I drive on. Retrench a bit? Yes.

As a side comment, his focus on "inattentional blindness" would seem on the surface to provide compelling reason to outlaw talking on cell phones, listening to music, or talking to passengers while driving.

By Scott Belyea (not verified) on 11 Jun 2008 #permalink

In many cases, the purpose of signs isn't the prevention of accidents per se -- it's the improvement of traffic flow. This is particularly true with stoplights. An intersection without stoplights often won't have all that many accidents, but if there's any significant traffic it will become a huge bottleneck, because everyone has to stop, look around, wait for the person with right-of-way to move (and sometimes, try to figure out who has right-of-way to start with), etc.

their unconscious brain can obey the rules without them

Though it would be harder to train (because it's harder to simulate), the brain can learn and unconsciously obey the rules of "look out for other cars and what they're doing" just as easily as it can the rules of signals.

It's something experienced drivers know anyways - people don't signal when changing lanes, people brake for nothing in front of them, people ignore YOUR signal when you want to change lanes, often accelerating right into your blind spot, plus your "decision" at that yellow light is also made based not just on whether or not you think your car can make it, but also whether or not you think the car in front of you decides to go for it (and whether or not you think the new traffic camera has caused local authorities to shorten the length of the yellow, as they got caught doing in some regions like San Diego).

So GOOD drivers have already learned that its more important to "know" other cars than the specifics of the road...though that's combined with the benefit (that you see) in knowing the signs subconsciously as well.

Indeed, the Brits with their sign-less streets don't really have it the "better", as the drivers of those towns still need to know the signs when they're somewhere else.

Traffic circles will never work in America, not because they're confusing, but because they assume that traffic is coming from most sides equally. The reality (that anybody who's tried the circles on Wisconsin and Connecticut Ave in Chevy Chase outside DC has discovered) is that one particular direction so floods the circle that the other entry points simply can't get a car in at all.

yeah, out in rural areas like rarely used interstate stops it might make sense (and be cheaper than cloverleafs), but suburbia has unusual driving practices that don't translate well (and are constantly changing as more houses are built up).

By Joe Shelby (not verified) on 11 Jun 2008 #permalink

It is good that there are some things that the brain does automatically while driving. I once attended a meditation weekend two hours the other side of Washington DC. To commute back and forth involved being on I-95 and the DC beltway. The driving went fine until Sunday night. It took twice as long to get home because my brain had been so constrained by meditation and taking verbatim notes that it would not stick to driving on the way home (new meditators are basically suppressing thoughts.)I kept missing the I-95 exit and then would overshoot in the other direction. It was a wonder that I managed to stay in a lane and not hit other cars all on autopilot.

"And that was posted to the wrong blog entirely.
Please delete at your convenience.

There's an inattentional blindness joke in there somewhere.

J-Dog -- yes, talking on a mobile (cell) phone while driving is illegal in the UK unless you are using a hands-free; but given that the best research I've seen indicates that using a hand-held mobile increases the risk of accident to about 4.5x that of a normal, unimpaired driver, whereas using a handsfree set reduces that to only 3.9x the risk, I wouldn't say that this accounts for the difference entirely.

I would also be interested in a state-by-state breakdown of accidents. I know that driver testing in the UK is standardised at what would count (in the US) as quite a high level. Driver testing in the states, however, varies wildly according to where you are. Driver testing in Boston or New York seems to involve some complicated maneuvering and signalling; driver testing when I was in Colorado was, could I drive around the block and park. What effect does that have later?

I would also say, the SUVs for sale here are a fraction of the weight of the SUVs for sale in the US. Last I checked, SUVs were disproportionately involved in rollover accidents both in the US and the UK, but the people killed in these accidents were usually in any smaller cars involved, and were disproportionately high numbers in the US -- which my immediate impulse is to say has to do with the relative weight and construction of the SUVs.

I'm sure that the issue doesn't have simple answers, anyway.

By Luna_the_cat (not verified) on 12 Jun 2008 #permalink

I was wandering around a small southern town that is in a
1930's-1950's time warp still, seeking thought provoking photographs to take. I came upon a stop sign that was faded to a pale, pale pink with rusted, dented edges. It
was knocked slightly askew - and as it was on a back road
I sure it had been there (Not interrupting traffic for at least fifty years). I sat down in the grass and laughed before I took several close up shots. I matted and framed the 11x14 photograph and entitled it - "What For?"

There's my point of view on that!

By Lee Pirozzi (not verified) on 12 Jun 2008 #permalink

You approach an intersection with a STOP sign and see traffic rolling your way from an intersecting side, at a distance of 300 yards. It's a very visible intersection in all directions and you could comfortably roll through it, however, there is a STOP sign, so you have to stop in order to avoid a potential ticket. Now you stopped and the approaching cross traffic is at 150 yards. Your most recent memory was that you had enough space and time to make it across in a save manner, though, and in this moment of contradicting stimulus you mght go still thinking there is enough time and space. Stop signs should be illegal, especially since they only seem to serve the revenue of local municipalities and contribute nationally to the revenue of all car makers. The constant stopping and going we are being forced to do increases wear and tear on all components of a vehicle.....the whole thing actuaqly seems like a true conspiracy, when you think about!!!! Make STOP signs illegal, I say, and follow to your God-Given sensory organs instead, your eyes!!!!

By TJ Rosenblatt (not verified) on 21 Jul 2010 #permalink