When two cars get into an accident the cars around them need to slow down to avoid hitting the cars involved in the collision. This causes a backup in traffic. People have often told me that traffic around an accident is caused by people 'rubber necking' or slowing down to see what is going on.

In my experience most drivers will quickly speed up as soon as there is open road in front of them, regardless of if there is a car off the road nearby. Have there been any studies that prove that people slow down just because there is a car off the road? Or is the traffic just due to the initial slow down filtering itself out?

  • 1
    I think anecdata would be that traffic in the opposite direction (even with a central reservation) also tends to slow down. (Another aspect is possibly just people being briefly reminded of their own mortality, and the fact that they are driving a potentially lethal vehicle...)
    – Benjol
    Apr 8, 2016 at 13:53
  • I used to commute on a very congested road and when there was an accident or even a cop giving someone a ticket, traffic would slow to a crawl. I recall weaving around people in frustration as they continued to drive 20 miles an hour (limit: 55 - typical speed: 70) with 4 completely wide-open lanes in front of them. There was nothing in the road to block traffic.
    – JimmyJames
    Jun 15, 2016 at 21:22

1 Answer 1


According to a page on rubbernecking from the UK's Highways Agency:

Congestion caused by incidents is a significant cause of delay on the Strategic Road Network. In many cases, incidents can cause secondary congestion on the opposite carriageway due to road users slowing down to look at the incident scene (known as 'rubbernecking').

The Agency believes the problem to be so bad that it suggests putting up portable screens to block the accident scene from view:

Blocking the passing drivers' view of the scene with an incident screen helps address this issue; work carried out by the TIM team has demonstrated the benefit of incident screens in reducing secondary congestion caused by 'rubbernecking'. Incident screens remain a key project within TIM and have been featured in the Bulletin on several occasions, the last article in June 2008 focused on a recent trial of a free-standing incident screening system in Area 5 .

  • Apparently they put those screens up in Germany.
    – MSpeed
    Jun 6, 2011 at 10:17
  • I heard on the radio last week that the UK's Highway Agency has just got around to purchasing some of these screens. 4 1/2 years from idea to implementation. Civil Service efficiency at it's finest.
    – Ian
    Jan 4, 2013 at 17:13
  • 1
    Wait, so a question is answered by posting excerpts from a page citing it is their belief that this reduces congestion. I don't see how this adds anything to the discussion except reiterating the notable claim. Suggest that this post be moved to the 'Blind Trust' Stack Exchange
    – Mauser
    Mar 17, 2014 at 15:15
  • 1
    The article cited does not simply state a general belief, but provides references to specific incidents and studies.
    – barbecue
    Dec 6, 2014 at 0:07

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .