4 ms·

Risk Compensation theory predicts this technology will increase risky behavior by drivers. It has been well established that people adjust their behavior in res

Risk Compensation theory predicts this technology will increase risky behavior by drivers. It has been well established that people adjust their behavior in response to perceived risk -- and if the car is going to handle the emergency braking it will cause more dangerous speeds and tailgating. See: https://en.wikipedia.org/wiki/Risk_compensation https://en.wikipedia.org/wiki/Risk_compensation

teej 3y ago
Theory is great. What has data shown?
- Drunk_Engineer 3y ago
  Among other things, data showed mandatory helmet laws may have had a perverse impact on motorcycle fatalities: http://john-adams.co.uk/wp-content/uploads/2006/motorcycle%20helmets.pdf http://john-adams.co.uk/wp-content/uploads/2006/motorcycle%2...
  - doodlesdev 3y ago
    Haven't read it completely or paid attention to the data yet, but the conclusion got me hooked: > Conclusion > The foregoing review of the statistical case for helmet legislation suggests that the faith of the BMJ in the reliability of the statistical evidence on the effect of helmet legislation is misplaced. But debates about sophisticated statistics should not be allowed to drown out the voice of common sense. The risk compensation hypothesis is essentially common sense. There is a wealth of evidence from everyday experience that suggests that people's behaviour is influenced by their perception of risk. People tend to be more cautious when up high ladders than when up low ladders. They tend to take more care when standing on the edge of a high precipice than when standing on a low kerb. They tend to slow down when they encounter bends in the road or patches of fog and to speed up when the road becomes straight or the visibility good-and so on. The possible illustrations of the phenomenon are countless. > There is an impressive amount of propaganda designed to encourage people to believe that they are very much less vulnerable when using seat belts and motorcycle helmets, and common sense suggests that driving behaviour will be influenced by safety devices that diminish the user's sense of vulnerability. What unaided common sense cannot predict is whether the behavioural changes induced by a safety device will partly, completely, or more than completely nullify the intended effect of the device. > The statistics which the NHTSA considers the most compelling evidence in support of helmet legislation are displayed in figure 1. The state which Watson et al (1980) consider has had the greatest increase in deaths as result of helmet law repeal is Utah. > When the distorting effects of the analytical methods employed by the NHTSA and Watson et al are removed, their data (with the caveats discussed above) suggest that the effect, if any, of helmet legislation on motorcycling fatalities is perverse. I'll have to read the paper with more attention later to see if I agree with this, but it's a fascinating hypothesis regardless. Also, this is tangential, but I'm always impressed how different conclusions can be based on the same sets of data. Statistics is a truly fascinating field of mathematics because it's still an objective topic (regarding numbers, in isolation) but becomes very human and subjective when it comes to the different conclusions, transformations, and visualizations for the same data sets.
    - lesuorac 3y ago
      I would challenge the common sense aspect of wearing a helmet makes you feel safer. Like if kevlar pants were mandatory on motorcycles would the same author claim it would increase accidents? It narrows your visibility and the main feeling of danger (imo) was all the cars that act as if you aren't there (ideally because they didn't/can't see you).
SkyPuncher 3y ago
I have it on my car. I think your concern isn’t an issue in reality. I have never seen it kick in while being close to someone’s bumper. It has only kicked in in three situations, all of which have been warranted. * I was turning onto a main road. Foolishly, I assumed the car turning in front of me went. I started making my turn. It alerted me and avoided a minor, but costly rear end. * in dense traffic where people cut in and out. In the time it takes to verify a lane change, someone has cut you off or traffic has slowed. It doesn’t not apply heavy braking, but issues an audible alarm and pretensions the brakes. * once in a situation where I misjudged the braking speed of the car in front of me. The vehicle was a bit smaller than I though, so it appeared further away at speed. In the latter two cases, the actual emergency braking did not engage. Only the pre warning and tensioning. In both situations, I would have likely avoided an accident without the system, but had to apply by brakes even harder. My understanding is emergency braking doesn’t necessarily avoid a collision, but it can bleed of speed in the final few feet before an impact. A collision ending 2 feet into your bumper is far safer than one ending 10ft into your car.
im3w1l 3y ago
Well first, people respond to perceived risk rather than actual risk, so those paradoxical instances where accidents increase comes when something seems safe - but isn't. Secondly, consider you have to consider what compensation buys you. People bringing up risk compensation often forget that. Doing a task faster and with less concentration, these are real benefits that makes peoples lives better. Third, people don't fully compensate for risk. Often a safety technology will both reduce accidents and let you get the job done faster.

teej 3y ago

Theory is great. What has data shown?

Drunk_Engineer 3y ago
Among other things, data showed mandatory helmet laws may have had a perverse impact on motorcycle fatalities: http://john-adams.co.uk/wp-content/uploads/2006/motorcycle%20helmets.pdf http://john-adams.co.uk/wp-content/uploads/2006/motorcycle%2...
- doodlesdev 3y ago
  Haven't read it completely or paid attention to the data yet, but the conclusion got me hooked: > Conclusion > The foregoing review of the statistical case for helmet legislation suggests that the faith of the BMJ in the reliability of the statistical evidence on the effect of helmet legislation is misplaced. But debates about sophisticated statistics should not be allowed to drown out the voice of common sense. The risk compensation hypothesis is essentially common sense. There is a wealth of evidence from everyday experience that suggests that people's behaviour is influenced by their perception of risk. People tend to be more cautious when up high ladders than when up low ladders. They tend to take more care when standing on the edge of a high precipice than when standing on a low kerb. They tend to slow down when they encounter bends in the road or patches of fog and to speed up when the road becomes straight or the visibility good-and so on. The possible illustrations of the phenomenon are countless. > There is an impressive amount of propaganda designed to encourage people to believe that they are very much less vulnerable when using seat belts and motorcycle helmets, and common sense suggests that driving behaviour will be influenced by safety devices that diminish the user's sense of vulnerability. What unaided common sense cannot predict is whether the behavioural changes induced by a safety device will partly, completely, or more than completely nullify the intended effect of the device. > The statistics which the NHTSA considers the most compelling evidence in support of helmet legislation are displayed in figure 1. The state which Watson et al (1980) consider has had the greatest increase in deaths as result of helmet law repeal is Utah. > When the distorting effects of the analytical methods employed by the NHTSA and Watson et al are removed, their data (with the caveats discussed above) suggest that the effect, if any, of helmet legislation on motorcycling fatalities is perverse. I'll have to read the paper with more attention later to see if I agree with this, but it's a fascinating hypothesis regardless. Also, this is tangential, but I'm always impressed how different conclusions can be based on the same sets of data. Statistics is a truly fascinating field of mathematics because it's still an objective topic (regarding numbers, in isolation) but becomes very human and subjective when it comes to the different conclusions, transformations, and visualizations for the same data sets.
  - lesuorac 3y ago
    I would challenge the common sense aspect of wearing a helmet makes you feel safer. Like if kevlar pants were mandatory on motorcycles would the same author claim it would increase accidents? It narrows your visibility and the main feeling of danger (imo) was all the cars that act as if you aren't there (ideally because they didn't/can't see you).

SkyPuncher 3y ago

I have it on my car. I think your concern isn’t an issue in reality. I have never seen it kick in while being close to someone’s bumper. It has only kicked in in three situations, all of which have been warranted. * I was turning onto a main road. Foolishly, I assumed the car turning in front of me went. I started making my turn. It alerted me and avoided a minor, but costly rear end. * in dense traffic where people cut in and out. In the time it takes to verify a lane change, someone has cut you off or traffic has slowed. It doesn’t not apply heavy braking, but issues an audible alarm and pretensions the brakes. * once in a situation where I misjudged the braking speed of the car in front of me. The vehicle was a bit smaller than I though, so it appeared further away at speed. In the latter two cases, the actual emergency braking did not engage. Only the pre warning and tensioning. In both situations, I would have likely avoided an accident without the system, but had to apply by brakes even harder. My understanding is emergency braking doesn’t necessarily avoid a collision, but it can bleed of speed in the final few feet before an impact. A collision ending 2 feet into your bumper is far safer than one ending 10ft into your car.

im3w1l 3y ago

Well first, people respond to perceived risk rather than actual risk, so those paradoxical instances where accidents increase comes when something seems safe - but isn't. Secondly, consider you have to consider what compensation buys you. People bringing up risk compensation often forget that. Doing a task faster and with less concentration, these are real benefits that makes peoples lives better. Third, people don't fully compensate for risk. Often a safety technology will both reduce accidents and let you get the job done faster.