The current fashion among keen cyclists for all black cycle-wear has always struck me as a step backwards. If road workers, dock workers, rail worker and building workers all wear hi-viz to be safe and seen, might not the same apply to cyclists? A look at the academic research in the area of cyclist clothing and safety, as ever, does not provide a definitive or simple answer not least because the whole issue of cyclist safety is quite complex.
Firstly, consider the encounter of a moving cyclist or pedestrian by a motor vehicle. The vehicle driver needs to see the cyclist or pedestrian, interpret what has been seen and then take action according to their training or other value system. This process of seeing cyclists and pedestrians has been tested in a number of papers and it is clear that the greater the 'Conspicuity' of the cyclist or pedestrian, i.e. the extent to which they conspicuously stands out from the background, the sooner they are 'seen' by the motorist. This means that the motorist has more time to process the sighting and act upon his/her interpretation of the seen object.
It has been noted by a number of experimental studies12, 14 that older drivers were slower to react to pedestrian and cyclist conspicuity than younger drivers, with some comments being made about older drivers potential for poorer vision. It has also been suggested10 that the responsive action of the motorist is, in part, likely to be a function of the extent to which cycling or walking is common in the area and that the encounter is not unusual for that particular driver. Some of the research7 into road worker safety suggested that their safety wear should be to a national standard colour so that drivers would become habituated to associate that colour with road workers. It is possible that a 'standard' hi-viz colour for cyclists and pedestrians might help to modify drivers' recognition and response behaviour deficiencies.
Does that not mean that hi-viz is better than black? In daylight, the answer is yes and no. The key issue is the background against which the cyclist is to be recognised. The research has found black to be best in some parts of Australia where there is bright, 'big sky' with everything around in bleached colours. In other contexts its conspicuity can be zero as it merges with the shadows and dark colours of the environment12. In the UK, in daylight hours, hi-viz appears to have higher conspicuity than black but other questions are “when worn where?” and “in what format?”
The research suggests that motorists recognize uncertain distant objects as human/animal though bio-motion markers12. Apparently humans are naturally programmed to recognise certain movements as being a human or animal even if the whole object is not visible. Thus hi-viz knee and ankle bands are more effective than a hi-viz waistcoat since their movement signals 'animal' whereas the waistcoat is just a bright blob. Of course, the distant bright blob may catch the eye of the driver in a way that a black blob might not.
Visual 'clutter' can be an issue in that multiple visual stimulii, such as in a busy city street, could mask the cyclist. However, it was found that bio-motion markers are identified readily despite visual clutter11. Road workers are probably at the extreme end of the 'at risk' category in such an environment of clutter. Investigation of the use of more extensive bio-motion markers compared to the standard hi-viz waistcoat found that the bio-motion markers were significantly more conspicuous to the subjects despite the clutter13.
In the same context, blinking light systems have been shown to improve the distance at which motorists recognise motor cycles4. The same is true for recognition of cyclists8, but not all the research agrees on this12. The lights need to be bright enough to be seen from a considerable distance if the motorist is to have time to process the information and react to it.
The context in which the cyclist and motorist are operating also has been shown to have an impact10. In the countryside where motorists speeds are higher it has been suggested that it is possible that the time between recognising the cyclist, then processing the recognition into action can be longer than the time it take to close with the cyclist. In an urban environment, despite lower speeds, a key problem is that the cyclist is in the left margin of the driver's field of view when the motorist is overtaking. If the cyclist has a low level of conspicuity it becomes easy to see how the driver can look but not “see” the cyclist. Pedestrians are in an even worse position on pavements as they are further out of the mostotist's field of view.
When a vehicle is waiting to pull out into the path of a cyclist there is a classic context for a 'looked but failed to see accident'. It has been suggested10 that improving cyclists' conspicuity will reduce the risk of a 'looked but failed to see accident'. However, it has also been discovered that motorists tend to rate cyclists as 'lower risk' traffic6, i.e. not putting the driver's own life at risk, and thus are more willing to pull out in front of the cyclist at T junctions or on roundabouts.
Drivers' attitude to 'lower risk' road users is one of the reasons why there is a big campaign in Scotland to introduce into Scottish Law the concept of Presumed Liability. Britain is one of the few EU countries (others are Romania and Cyprus) that does not use Presumed Liability. What is Presumed Liability? It is a “no-fault liability” where a person is held responsible because they are in control of a source of danger (normally a vehicle) to other people’s lives, health or property, not because of their failure to display the due diligence of a reasonable person.
How does it work in practice? A moving vehicle is, inherently, a source of danger to pedestrians and particularly children. Thus in any accident involving a pedestrian it is assumed that the vehicle is responsible for the incident. This places a much greater duty of care on the vehicle driver than applies currently. The system also creates a hierarchy of vulnerability, in that pedestrians are more vulnerable than cyclists, cyclists are more vulnerable than motor cars, and motor cars are more vulnerable than heavy goods vehicles.
Moving back to conspicuity, the situation is rather different in the dusk and at night. The European Cyclists' Federation (ECF) report15 on the options for the standardisation of cycle lights list some fairly damning statistics. Nearly 40% of night or dusk riders in most EU countries operate with defective lights or without any lights at all while in some localised areas the proportion is over 70%. Clearly, the use of lights of adequate brightness correctly mounted by 100% of cyclists ought to improve conspicuity. However, many cyclists rely on reflectors and upon the minimal reflective strips on their cycle clothing. The ECF report highlights that cycle reflectors are only effective at quite close range and within a fairly narrow angle of visibility.
Hi-viz wear with reflective strips is surprisingly ineffective at night. The fluorescent colours of hi-viz rely on infra-red light to make them glow and there are few sources of IR light at night. Conspicuity thus depends upon the reflective material incorporated into the garment. It has been shown1 that the greater the area of reflective (bright) material in the garment the more likely the motorist will see the wearer and yet typical cycle wear has very little. However, for bio-recognition in conditions of clutter, the research suggest that recognition is better where the reflective elements are widely distributed across the garment outlining the shape of the wearer rather than in a single big area. Thus for best conspicuity at night cycle wear should be made with significant additions of retro-reflective material all over.
The major problem with reflective materials are that they only 'light up' when they are in direct line of the vehicle's lights. Thus a cyclist with reflective clothing at the side of the road is going to achieve a lower level of conspicuity through being at an angle to the vehicles lights. This has been examined in relation to pedestrians3, who are even more out of the line of a motorist's headlights when on pavements. The researchers tested electrolumiscent materials which are typically small sheets or ribbon that emit light when a small current is applied. They found that use of such material for bio-motion markers significantly increased the conspicuity of the pedestrian. Use of such materials in cycle clothing should produce similar improvements in conspicuity at night.
Whilst conspicuity cycle clothing is clearly useful, research has shown that cyclists were rather less conspicuous than they believed when wearing these items13. The 2010 Transport Research Laboratory research on conspicuity and road workers showed that the same was true for those workers7. It also pointed out that at night when traffic is running with dipped headlamps, the effectiveness of reflective clothing was much reduced as the headlamp beam would not generate a good reflection until the vehicle was very close. In such circumstances the driver would have little time to recognise the wearer as being at risk and take appropriate actions. Clearly this research has implications for cyclists' conspicuity clothing.
The final question that should be raised is whether the use of conspicuity aids and clothing actually makes the user safer while cycling. Being conspicuious clearly increases the distance at which a vehicle driver will see the cyclist. Whether being seen makes much difference to safety then depends entirely upon the action of the motorist. However, the research into the extent to which users of conspicuity clothing believe they are conspicuous provides some interesting warnings. Essentially, users tend think that they are more conspicuous than they actually are and tend to act accordingly. It is possible that this may then put them at greater risk. Various researchers5,9,10 have suggested that conspicuity aids and clothing can, in some cases, be correlated with a higher level of accidents. It is possible that this may not be related to the conspicuity of the cyclists, but more down to the habituation of the motorist10. The research reports a higher level of accidents for cyclists using conspicuity aids in the city of Auckland, where cyclists were not at all common, and drivers do not expect to see them. By comparison, elsewhere in New Zealand where cycling is common place, there is no correlation between conspicuity aids and accidents.
So to conclude, it seems clear that hi-viz, reflective cycle clothing and flashing LED lights are much more conspicuous for motorists than dark clothing. It is also clear that users of conspicuity aids are mostly recognised by drivers sooner than users of dark clothing. There is however, a risk that users of conspicuity aids will overestimate the visual protective effect of those aids and hence put themselves at risk. Finally, the big problem remains as to what happens next in the driver's head after a cyclist is recognised. Perhaps, as more people cycle, drivers will become habituated to seeing them. If Presumed Liability could be adopted into English law drivers might also stop seeing cyclist as 'lower risk' road users.
© 2016 Cycling Otherwise
1. Cassidy P E, Brooks B E, Anderson N J (2005) “Size isn’t everything: the effects of size and brightness of Retroreflective materials on nighttime conspicuity”, Proceedings of the Human Hactors and Ergonomics Society 49th Annual Meeting, pp. 1931 – 1934.
2. Fekety, D K (2015) "A Comparative Analysis of Electroluminescent and Retrorefelective Materials as Nighttime Pedestrian Conspicuity Aids". All Theses. Paper 2134.
3. Fekety D K, Edewaard D E, Stafford Sewall A A, Tyrrell R A (2016) “Electroluminescent Materials Can Further Enhance the Nighttime Conspicuity of Pedestrians Wearing Retroreflective Materials”. Human Factors: The Journal of the Human Factors and Ergonomics Society, May 19, pp. (Not yet available)
4. Gershon P and Shinar D (2013) “Increasing motorcycles attention and search conspicuity by using Alternating-Blinking Lights System (ABLS).”, Accident Analysis and Prevention, Vol. 50, pp. 801 – 810.
5. Hagel BE, Romanow NT, Morgunov N, Embree T, Couperthwaite AB, Voaklander D, Rowe BH (2014) “The relationship between visibility aid use and motor vehicle related injuries among bicyclists presenting to emergency departments”, Accident Analysis and Prevention, Vol. 65, pp. 85-96.
6. Herslund MB, Jørgensen NO, (2003) “Looked-but-failed-to-see-errors in traffic.” Accident Analysis and Prevention, Vol. 35(6), pp. 885-91
7. Helman S and Palmer M (2010) “Road worker conspicuity daytime & night time”, Transport Research Laboratory Client Project Report (cpr1001)
8. Koo H S, Huang X, (2015) "Visibility aid cycling clothing: flashing light-emitting diode (FLED) configurations", International Journal of Clothing Science and Technology, Vol. 27 (3) pp. 460 – 471.
9. Miller P ( 2012) “The use of conspicuity aids by cyclists and the risk of crashes involving other road users: a population based case–control study” (PhD Thesis). University of Notthingham.
10. Tin Tin S, Woodward A, Ameratunga S, (2014) “The role of conspicuity in preventing bicycle crashes involving a motor vehicle”, European Journal of Public Health Advance Access, July 31, pp. 1 – 5.
11. Tyrrell, R A, Wood J M, Chaparro A, Carberry T P, Chu B, and Marszalek R P. (2009) “Seeing pedestrians at night: visual clutter does not mask biological motion.”,Accidents Analysis and Prevention, Vol. 41(3). pp. 506 – 512.
12. Wood J M, Tyrrell R A, Carberry TP, (2005) “Limitations in Drivers’ Ability to Recognize Pedestrians at Night”, Human Factors, Vol. 47, (3), pp. 644–653
13. Wood J M and King M J (2010) “Research initiatives to improve the visibility and hence safety of road workers at night-time”, Journal of the Australasian College of Road Safety, May, pp. 23 - 24
14. Wood J M., Tyrrell, R A., Marszalek R P., Lacherez, P F., Carberry T P., & Chu B S(2012) “Using reflective clothing to enhance the conspicuity of bicyclists at night”, Accident Analysis and Prevention, Vol 45(March), pp. 726 – 730.
15. Woolsgrove C (2012) “Requirements on Lighting (Light Intensity) and Reflectors of Bicycles”, Report Commissioned by ANEC, Brussels. [www.anec.eu/attachments/anec-r&t-2012-traf-002.pdf downloaded 26/6/16.]