I found the following anecdote on Quora

Many years ago, I worked for a gas & electric power company, and had been a member of a crew installing a road-side gas main. Despite the signs, flashing lights, and traffic cones we'd set up well in advance of the work zone, the overwhelming majority of people would speed past us--sometimes within an arm's length, seemingly having no regard for our safety. Our supervisor showed up to check on our progress, and, witnessing the danger, he walked to our truck, took out a yellow compressor hose, and rolled it out across the road. Immediately & without exception, from that point on, every single car slowed down before driving over the hose.

This is the worker's interpretation of the behavior:

In my youthful mind, I concluded that people were simply more concerned about damaging their cars than they were about damaging other people, but in my now middle-aged mind, I believe it's more complex than that. Maybe it's one of those countless subconscious, split-second decisions our minds have to make throughout each day, assessing probabilities & adjusting accordingly: 'I see the road crew, I've never struck a person with my car before, it's much more likely than not that I won't strike a person today, and those people are aware of my car & have the ability to avoid it, therefore, carry on.' [versus] 'I see an object on the road, I see that I can't avoid the object, I've experienced damage to my car before when driving over objects in the road, and that object isn't going to move, therefore, caution is merited.'

Is this interpretation correct?

  • $\begingroup$ @Jeff, I believe you've said pretty much the same as I read in the interpretation. 'It's much more likely'.. hence 'I expect' I won't run him over, 'I see I can't avoid' hence 'I expect' to run over it. To green: Yes, I believe the interpretation is right. To learn more about this effect and this kind of decision making, you could browse through experiments on that, though I can not think of any particular right now. This decision seems natural to me: one may expect, that workers are aware of the risk and won't cross the road - on the other hand an object already on the road is sure hit. $\endgroup$ – Oriesok Vlassky Jan 14 '13 at 14:46
  • 1
    $\begingroup$ @OriesokVlassky ah yes, i think i misread. but i still don't think this is an answerable question. $\endgroup$ – Jeff Jan 14 '13 at 19:26

Presumably the decision of drivers to slow down in response to work zone signage is influenced by many factors.

  • Signage and road factors: Presumably there are a wide range of factors related to the nature of the signs and the structure of the road setting that influence whether people slow down. For example, I've seen road work signage on freeways that were often in place even when no actual road work is occurring. In these cases I've seen many drivers slow down little if at all. In contrast, I typically see much greater reduction in speeds when actual road work is occurring. Or for example, I imagine that a worker standing with a sign would be more effective than a static sign.
  • Cultural and norm factors: Presumably countries and regions differ in cultural attitudes to driving, the acceptability of speeding, and the general tolerance for risk taking. Road safety advertising employ a wide range of strategies for reducing speeding including targeting personal interest (e.g., you'll get fined if you speed) or highlighting consequences (e.g., speeding increases the likelihood of killing yourself and others).

A few empirical studies

Jarvis (1983) performed empirical research on Australian highways and found that

There is a clear indication that work area speed limits reduce vehicle speed on entry into the work zone; however, the large majority of drivers still exceed the speed limit posted.

Garber and Patel (1995) found that signs triggered by radars indicating excessive speed were more effective in slowing cars than static signs.


  • Garber, N. J., & Patel, S. T. (1995). Control of vehicle speeds in temporary traffic control zones (work zones) using changeable message signs with radar. Transportation Research Record, (1509), 73-81.
  • Jarvis, J. R. (1983). The Effectiveness of Road Work Speed Limit Signs. Australian Road Research, 13(3).

This short talk deals with issues of "cheating slightly" :Dan Ariely: Our buggy moral code. While it does not explain why a hose slowed down traffic, the speaker discusses cases of "I see others cheating, so it's ok to cheat a little more".

In case of road work, people slightly bend traffic laws. They see other drivers from the same state or same social class(ex:a bunch of cheap cars) cheating on traffic laws. It's similar to people from the same college cheating, as described in the talk. Following the talks logic, if a driver sees others speed In a work zone, the driver is much more likely to speed. And with a constant stream of traffic, this effect would not stop.

It would be interesting to hear if a single driver on an empty road would slow down for men in work zone.

  • $\begingroup$ Ariely's talk is interesting, but I disagree that it applies here. I realize you mention "this does not explain why a hose slowed down traffic", but that is the question, isn't it? $\endgroup$ – Jeff Jan 19 '13 at 5:02

It may also be due to errors in perception. Our brains did not evolve to perceive/judge objects and distances while traveling at such high speeds and we may misjudge how fast we are actually going to how fast we think is safe. There are some studies out there that ask people how long the white dividing lines on freeways are (people always under estimate the length). I think I came across these studies in Daniel Kahneman's, Thinking Fast and Slow.

  • 1
    $\begingroup$ You can add references to these studies. it should be fairly simple to look up some of these studies..Ideally, you should summarise the study and elaborate it a little more. $\endgroup$ – Software Mechanic Jan 24 '13 at 13:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.