ghettocruiser

Is this really a statistically significant sample? How often do police cars stop alongside live lanes of highways WITHOUT using emergency lights? I don't think I've ever seen one... the cops I see stopped without flashing lights pull way off the road and shoulder, and often behind some kind of obstruction.

I have, however, seen motorists plow into lots of easy-to see objects without flashing lights.

I'm not saying this phenomena doesn't exist, I'm saying we need better evidence.

vincentpaul

I agree that it's anecdotal, and indicated that in my first post. Any body here have access to any research databases? I'd be interested in seeing peer reviewed literature on the subject. My quick google search didn't pull up any studies, only mentions of changing opinions without cites to actual research.

SDRider

Also, how often do cyclists ride on freeways between the hours of 10PM and 5AM on Friday and Saturday nights? I suspect that's when the majority of these type of accidents occur.

serac

The flashing rear light has emerged as a standard. As such it signals to drivers that, "Hey, I'm a bicycle". I am curious about the statements surrounding flashing lights attracting impaired drivers and frankly it scares me.

I don't like flashing headlights on bicycles that much. I think it makes it very difficult for cars to judge distance, speed and trajectory. But I don't have a study to base this on.

BarracksSi

Studies have been done, and links have been posted before here in BF -- although not necessarily in the Commuting section.

Basically, a steady light is much, much better for distance recognition than a flashing light -- and two steady lights are even better than that. Honda did some studies themselves about how to better illuminate motorcycles and found that high- and low-mounted front lights helped greatly with distance recognition.

I run flash mode in the daytime, especially up front, and steady at night (although a Dinotte light has a couple flash-steady modes that might be a good blend of the two). My reasoning is that my light is the only real clue to my existence in the dark, and rather than confuse others with a flash pattern, I'll keep it on steady so they can judge my distance quicker & more accurately.

SSP

I don't understand how my flashing taillights could possibly "confuse" a driver. As a driver, when I see the typical flashing pattern of a cyclist's blinkie, I immediately think - "there's a bike in the road up ahead", and adjust my speed and lane position accordingly.

And with most cyclists flashing vs. steady is not an "either-or" proposition. Most of us have some flashers, but also steady state reflectors, and "up-down" pedal and crank arm reflectors. It all adds up to a clear message that says, "Hey, dummy! There's a bike in the road...be careful!".

IMO, a flashing blinkie is much more quickly identified as a bike in the road. A steady state light might be mistaken for a reflector on a mailbox, or some other unimportant light source. But a flashing LED is much more quickly identified as a bike (though my DiNotte 140L may sometimes be mistaken for a UFO landing site ).

bmclaughlin807

I generally run flash in low light or bad weather conditions... cylon mode in the dark (IF I remember to change it... otherwise it's on flash). I'm using a Cateye LD-600 in the back... and it's the dimmest I'd possibly be comfortable with. But I did get stopped by a car yesterday during the snow... she wanted to tell me I had a nice light!

As soon as I have some spare cash I'll be adding another tail light.

bmclaughlin807

Then how do you explain all the objects that drunk drivers hit that AREN'T attached to blinking lights? Or have no lights at all (And it's broad daylight?)

I've seen LOTS of anecdotal evidence that shows that a drink driver may hit ANY object... regardless of lighting, location, etc.

How about the one that drove through a yard and crashed through the guys living room? Doh. Damn, forgot about that blinking porch light.

BarracksSi

It's not that it's confusing in terms of identification that it's a bike, but confusing in terms of whether they think you're ten or twenty feet away. They may spot you, sure, but they may also guess that you're farther away than you really are, and might pull out right in front of you even though they're looking right at you.

iltb-2

Urban Legend may be a better description of this oft repeated anecdote, about an undocumented higher accident rate/risk due to an alleged attraction to crash into objects with blinking lights.

Klaw

I use 2 lights at varying heights... one on my messenger bag, one on the bike... with differing blinking styles. If it's confusing I don't care, it's enough for a motorist not to ignore it and I'd prefer if they were annoyed and noticed it. I ride in the city so I don't need a one mile visibility, I need a 20-50 yard immediate recognition factor to stop someone from making a right turn into me or thinking twice about trying to bullrush past another car in traffic.

vincentpaul

My biggest unanswered questions are “what are rear lights on a bicycle protecting against?” And “what is the most effective light to use for that purpose?” It appears we research on several issues. 1) Can incidence of collision be reduced through the use of lights? 2) Does the use of constant flashing or constant light source increase/decrease the incidence of collision? 3) Does driver impairment alter the results above?

My experience from working with impaired drivers on a daily basis for the last 5 years is that the majority of accidents involving overtaking vehicles involve impaired motorists. If that is supported by empirical data, then what needs to be studied is what has the greatest impact in avoiding collisions between impaired motorists and bicyclists. Human behavior is normative and accident avoidance can be optimized with the right data. Here’s some studies I have found, but they don’t appear to accurately address my questions.


www.ctre.iastate.edu/PUBS/tsinfo/faq25l.doc:
“There has been some research into the effectiveness of flashing lights. A study by Cynecki and Sparks in the Institute of Transportation Engineers Journal was performed to determine whether flashers were successful in reducing illegal vehicle movements, primarily running red lights. This study found that speeds and violation rates of red lights actually increased after flashing beacons were added to advance warning signs at two school crossings. The study concluded that it is an incorrect assumption that flashing lights control what drivers view as safe behavior. In general, the actual presence of children was found to have the most substantial impact on vehicle speeds. It was also found that a flashing light blends into the existing scenery the longer it operates. This produces a situation in which the flashing light loses its value. Overall, the authors of the study believe that flashing lights are most applicable in a high-speed rural environment with unusual geometry characteristics, at locations with frequent pedestrian crossings, and along roadways with a large number of unfamiliar drivers.

“Another study that discussed the effectiveness of flashing beacons at intersections appeared in Transportation Research Record in January 1977. This study found that intersection control beacons had the greatest effect on the fastest vehicles and less of an impact on vehicles traveling at or near the average vehicle speed along the roadway. In addition, an analysis of the speed data revealed no significant change in the average speed at an intersection when a flashing beacon was added to a “Stop Ahead” sign. There was, however, a significant reduction in the amount that the speeds varied among vehicles.”


A 1996 Consumer Products Safety Commission report analyzed reflectors and a flashing LED light and determined that a flashing LED increased detection distances but not recognition distance. https://www.cpsc.gov/volstd/bike/BikeReport.pdf. The study did not compare a flashing red light source with a steady red light source. The results are problematical, because in the case of forward detection the most important factor for the driver to understand is that the object detected is essentially stationary from the standpoint of an overtaking driver. The driver needs to recognize that he is rapidly overtaking a substantially slower object. The report does recognize that the greatest risk to cyclist is impaired drivers, yet does no controlled study of impaired drivers to determine if their recognition and detection rates are the same as a non-impaired driver.

Here’s the report summary:
“CPSC staff determined that the risk of death and injury from nighttime bicycle riding is
considerably higher than riding during the day. Risk analysis shows that a cyclist riding during non-daylight hours is about four times more likely to be killed than a cyclist riding during the day. Alcohol use by either the motorist or the bicyclist was common in fatal incidents. Over one-half of the fatal incidents involved alcohol.

“Fatal incidents occurred at high posted speeds, over one fourth occurred at posted speeds
of 55 mile per hour or greater. Higher speeds require increased preview time for a driver to avoid a crash. At 40 mph, a driver needs to detect a bicyclist at 325 feet to avoid a collision according to stopping sight distance tables which are applicable to a crossing path collision. At 40 mph, a driver needs to see a bicyclist at 600-825 feet to avoid a collision, according to the decision sight distance table which may be more applicable then the stopping sight distance table in a rear run over crash.

“The CPSC's field study compared driver perception of various reflector and light
treatments. The study simulated a potential rear run over crash and a potential intersection crash. Treatments were selected to improve detection and recognition of the bicycle. The results of the study showed that there was no statistically significant difference in performance among the various reflector treatments and the current CPSC regulation reflectors. Therefore, improvement to the existing reflector requirement does not appear warranted.

“The parallel path crash simulation showed that the flashing red LED taillight used in the
evaluation could significantly improve a driver's detection distance over the CPSC regulation reflectors. Recognition distance was not improved with this or any of the treatments tested. CPSC staff believes that this type of lighting has potential to improve nighttime bicycle safety. Currently, rear LED lights are available in a variety of styles. Additional research is required to determine minimum requirements for rear lighting.”

There’s an excellent critique of the 1996 CPSC study at this website: https://www.johnforester.com/Articles...s/cpsc9701.htm

A discussion of the moth effect:

"Whether or not the moth effect explains roadside crashes continues to be
debated. Dr. Stephen S. Solomon, an ophthalmologist who has studied extensively
this issue, pointed out in his 1999 book, Emergency Vehicle Accidents:
Prevention and Reconstruction:
Because bright flashes act as a visual stimulus, some drivers have
a tendency to continue to look at them after the lights attract their
attention. As a consequence, they steer where they are looking.
The attraction has been referred to as a hypnotic or like a moth being
drawn to a flame. This appears to be a good explanation as to
a number of the accidents involving parked emergency vehicles. "

From: https://www.theiacp.org/documents/pdfs/LESS/ACF29B.pdf

Ed Holland

This has already been covered, but, I use flashing and continuously illuminated rear lights. I hope to get the best of both worlds.

Besides, I think the letter of the law requires at least one rear light with continuous output, physically attached to the bicycle.

Ed

iltb-2

More telling are the few paragraphs before the opthamologist's "explanation." On page 4 of the referenced document:

The California Highway Patrol detected in the 1970s an increase in the frequency of roadside crashes involving its cruisers and consequently studied such issues as “the moth effect,” “a little understood but widely accepted behavior . . . [describing] the tendency of drivers to focus on an object outside their intended path of travel and steer involuntarily into it,” and the value of different types of emergency lighting.

The report concluded that explanations of the moth effect had been based upon human perceptual difficulties; that no comprehensive study relating the theory to actuality existed; that even though injury crash rates were higher for vehicles displaying emergency lights, the overall collision rate for emergency and non-emergency vehicles was the same; and that, “[N]o evidence was found to suggest that the [four different] lighting systems utilized in the test produced hazardous driving, or reactions leading to a conclusion that they produced, or influenced accident prone behavior or reactions.”

BarracksSi

Whether this is applicable or not is up to the reader --

I just thought of how difficult it can be to catch a lightning bug, especially when it's doing a quick-flash "I just found a mate!" mode.

You can detect its presence with no problem, of course. It gets difficult, however, when it's dark enough that you can't see the bugger when it's not lit -- it effectively disappears in between blinks.

Imagine how much easier it would be to catch a lightning bug if it were glowing all the time.

When an onlooker sees a bright light blink "on", they can see it right at that moment. But during its "off" phase, they can't see where it really is anymore, and can only guess that it might be near the same spot as the "on" blink.

Catch my drift? That's why distance recognition -- "How far away is that light?" -- isn't improved at all by a blinking light.

vincentpaul

Yes, my point exactly. There is no evidence that flashing lights do squat to avoid roadside collisions. Thanks for pulling the quote for me. I didn't have time to find it.

Mr. Underbridge

I'd rather they were wrong as to how far away I am, as that implies that they actually see me at all. I think that's the battle that more of us face on a daily basis during dark commutes - being seen. There have been a few threads here to that effect lately - cyclist almost gets hit because car didn't register his dim, steady light.

I'm good if they see me at all.

GCRyder

The complete statement was:

"[E]ven though injury crash rates were higher for vehicles displaying emergency lights, the overall collision rate for emergency and non-emergency vehicles was the same."

In other words, they're just as likely to hit the light show, but they hit it harder. There are several ways to interpret that, and we can't really know how from what has appeared here. It seems highly unlikely that the non-emergency vehicles were struck under the same circumstances as the emergency vehicles. If emergency vehicles at accident and arrest scenes are getting hit at the same rate as non-emergency vehicles stopped at stop lights, that's not an indication that the emergency lighting is ineffective. The emergency vehicles tend to be stopped at more hazardous locations, where there's high-speed traffic moving by. That's why the injury crash rates are higher. If their lights had no effect, one might expect then to get hit more often than non-emergency vehicles, not at the same rate.

Obviously, though, I'm just guessing, since I have no way of knowing what controls went into the data collection or analysis.

BarracksSi

But, I'd rather not have them think that I'm 40 feet away and moving slowly when I'm actually 15 feet away and moving quickly.

In the few studies that have been mentioned on BF that focused on this, the most common errors when seeing blinking lights were thinking that the light was moving more slowly and being further away than it actually was.

Bright lights, of course, are better than dim lights -- but steady lights, IMO, are better than blinking, especially as the surrounding environment gets darker.

If regulations restricted lights to only one each on the front & back, I'd go with steady. Bikes equipped with such lights were remarkably visible (more than I expected) on the streets in the medium-sized German town where I spent Christmas.

Lamplight

If my taillight is annoying motorists, then that means they see me.

iltb-2

On my commute (posted 55mph roads) the difference between being first noticed at 15 or 40 feet is nil, since both are totally unacceptable. And my speed is totally irrelevant given the differential.

I am quite pleased that I can spot in my mirror most vehicles changing lanes at 1/2 mile and more behind me because of my blinkies. Unfortunately there are a handful of motorists, almost always driving a pickup truck, who still wait until the last second to change lanes at full speed but that is because they are JackAsses, not because they don't see me or can't judge my speed.

SSP

No...I don't catch your drift.

Your analogy is faulty...encountering bikes (even on a darkened roadway) is not like "catching lightning bugs". When a car encounters a bike on a dark road, the car is using a powerful system of illumination...the car's headlights, plus the steady state reflective material that most of us use, provides unambiguous feedback to the driver as to where we are, and what we are.

Thanks to the car's headlights, we cyclists are never "off"...at least, not at reasonable distances.

SSP

That would only be the case if you were riding a matte black bike, with dark clothing, and no other reflectors on your clothing or bike...and you were using a slow flashing on/off blinkie.

Thankfully, car headlights are very effective at illuminating us (especially when we augment with reflective vests, conspicuity tape, etc.).

And many of us use flashing lights that are never in "off" mode for very long - the PB Superflash and DiNotte 140L being obvious examples.

BarracksSi

I think we're talking about different directions for being seen --

Much of the risk I encounter is from people who would pull out in front of me, whether it's from a cross street, driveway, parking spot, or crosswalk.

That's when their headlights (and plain eyeballs) don't illuminate me at all, even if I were wearing enough reflective material to make a clown suit.

So the guy who's waiting at a red light (green for my direction) to turn from a cross street on my right side needs to have a chance to correctly guess my speed & distance, and it's harder for him to do so if my headlight's blinking. It'll be my luck that he guesses wrong, assumes that I'm not moving very fast and/or I'm twice as far away, then decides that he's in enough of a hurry to start turning into the intersection.

CommuterRun

On the occasion that this happens to me, most motorists seem to assume that I am traveling faster than I really am and will sit and wait. A quick headlight flash by cupping my hand over the light will, most often, communicate to them to go ahead and move on.

This usually happens in no ambient light situations, where I highly doubt they can tell I'm on a bicycle, except possibly by my lighting arrangement not matching any other vehicles on the road, and unless that particular driver has seen me at night before.