On today’s episode, we discuss the visibility of high-vis clothing. Are they actually “high-vis”?
We came across this topic, because of a conversation happening on LinkedIn. Thus, we thought it would be a good idea to dig into this subject and discuss it further.
Listen in as we discuss what actually makes humans most visible in unsafe situations and what some studies have concluded.
Topics:
Quotes:
“The general goal of this, is they just want to compare a whole heap of different factors.”
“The ability to just spot high-vis and the ability to spot a human wearing high-vis, seem to be actually two different mental tasks.”
“There’s been some suggestion in the research that we should actually standardize a human high-vis color.”
Resources:
David: You’re listening to the Safety of Work podcast episode 72. Today, we’re asking the question: How visible is hi vis clothing? Let’s get started.
Hey, everybody. My name is David Provan. I’m here with Drew Rae, and we’re from the Safety Science Innovation Lab at Griffith University. Welcome to the Safety of Work podcast. In each episode, we ask an important question in relation to the safety of work or the work of safety, and we examine the evidence surrounding it. Sometimes we ask really big questions about safety, and sometimes we ask more specific questions about practical situations. Today, it’s one of those more specific questions. Drew, what question are we answering today?
Drew: I really don’t think we can say this is necessarily an important question—at least I don’t think anyone who is crying out for the answer, David—but there was a discussion going on LinkedIn. I’d love to be able to give the right people credit. The way LinkedIn feeds work, I sort of saw the discussion and I can never find it again afterwards.
Someone was talking about seeing a worker in high-visibility clothing against a backdrop of a roadside site with lots of other high-visibility signs and markers. They said it took a while to see the person just because they weren’t standing out, because everything was hi vis. I thought, yeah that’s interesting. I wonder if anyone’s done any research about that. I decided we’d do an episode just about the high-visibility clothing and how well it works.
David: It sounds like a big question but come on, safety’s all about high-visibility, like you don’t have to go too far to find hi vis. It’s probably one of the most dramatic safety changes in the last 20 years.
Drew: Well, it’s visually dramatic. We don’t argue about it nearly as much as we argue about Safety I and Safety II, or Resilience, or Behavioral Safety. Everyone seems to sort of quite happy to get on with wearing this ridiculous-looking stuff.
David: I know. It was a big deal for fitters, mechanics, and electricians to go from navy blue to fluorescent orange. That was a very big deal.
Drew: Tell us about your own hi vis wardrobe, David. Mine consists of two vests which are very pristine that I’m always tempted to throw down in the mud and kick a little bit to make them look more authentic.
David: Now, the problem with hi vis is changing companies because everyone embroids their logos onto high vis. Unfortunately, being a safety person—this is probably not a good thing to say—my hi vis really doesn’t get much use. That’s a terrible thing to say, but my hi vis doesn’t usually get very dirty, maybe I’ll say it that way.
You’re not getting all these sets of hi vis clothing for a particular company and then you change jobs, and it’s kind of useless because it’s got the wrong logo in it. There’s an opportunity to have tear-off patches with company logos, so that as companies get bought and sold and people change jobs, that they don’t have to replace all their high vis. I do have a couple of pairs of pants, I do have some shirts, some jackets, some fire-retardant overalls. I’m pretty well [00:03:19] out.
Drew: I have to admit that one of my vests says Telstra across the back even though I’ve never worked for Telstra. My wife used to work for a law firm which in turn used to work for Telstra, and she once had to go somewhere that needed a hi vis outfit, so that was added to the wardrobe. I probably wear more hi vis stuff for cycling than I ever do actually working as a safety researcher.
David: I did try to chase down some articles that we haven’t introduced what we’re going to talk about today, but I did try to see if I could find some other research as well. There is a little bit on the visibility of cyclists as well, so it’s probably a good thing to be wearing a hi vis while you’re cycling, Drew.
Drew: Yeah, that one actually doesn’t seem to be that controversial. As usual, what we’ll do is I’ve picked out one sort of main paper for our discussion. David, you obviously looked at a few other things and I’ve obviously looked at a few other things. I don’t know if we both dare go down the same rabbit hole of, should you wash your hi vis clothing and what effect does that have on its luminescence. We’ll have a little bit of chat as we go along.
David: I think the UV has a bigger impact in the detergent. Drew, do you want to introduce the paper that we’re going to review?
Drew: This paper is called, The roles of garment design and scene complexity in the daytime conspicuity of high-visibility safety apparel, from the Journal of Safety Research 2008. The authors are Dr. James Sayer and Mary Lynn Buonarosa. They’re still both working and were working at the time for the University of Michigan Transport Research Institute.
David: The title of that paper was a bit of a mouthful. It’s basically, can you see people during the day when they’re in a busy environment. As I was reading this article and we talked about titles and wrote a few papers with Drew, we tend to put a bit of an effort in the title. Here’s a perfect opportunity to call this paper something like ‘Where’s Wally?’ Where’s Wally wearing his hi vis vest, because exactly what they’re trying to talk about, seeing complexity and how much something stands out.
Drew: I think that would have been an improvement. I was going to something more like, how visible is hi vis.
Let’s talk a little bit about how people study visibility. This paper gives a pretty good summary of the range of types of the studies that exist. Generally, it comes down to a trade-off between researchers having lots of control over the variables and having very natural settings. This is important because it seems that the results actually change between lab environments and the real world. You have to be really careful. You can get quite good results in an artificial environment that doesn’t actually mean anything when you take it into a real setting.
Some of the things people do, the really most simple one, is they give people a picture of someone in hi vis and say, how conspicuous is this person? How conspicuous it is on a scale of 1 to 10? They use that for testing out things like different colors, different patterns, different stripes, or different situations. Just give people batteries of pictures and ask for their opinion.
I guess slightly more scientifically, they measure how much light is coming off different things. They’ll test different garments and different layouts just in terms of how conspicuous it is in technical or light terms.
The next thing they do is they do a cascading series of tasks that get more and more realistic. A really simple task might be you flash a picture on a computer screen and someone has to click on the person and they measure how long it takes. Or we might make it more realistic. We might get someone actually going around in a car sitting in the passenger seat. You have the equivalent of a blindfold on it, it flashes open, how quickly can you spot the person. Or we get someone to actually drive around and point out to the people what they see. The most realistic is when we get people actually out in the real world driving around looking for people, and we measure how quickly they can see the people and when they miss any of the people.
David: In this study it was more the latter there about this naturalistic environment. Participants in the study drove along a 31-kilometer route—I think I actually did that twice—had no idea where or how many workers they would need to spot. They’re simply told, you’re going to drive along this route. Researchers had situated themselves—researchers were pretending to be workers standing about one meter from the edge of the road—in places where there was at least a 500-meter or half a kilometer clear line of sight to the car, so they weren’t hiding behind trees or anything like that.
These fake workers—let’s call it the researchers—had one of four different hi vis outfits on, so there were four different hi vis outfits, and they were either stationary—just standing there—or swinging their arms around. They were either in low or medium complexity settings.
What that meant was, the low complexity setting was basically just an open road, occasional signs, houses but a fairly open street sort of scape. The medium complexity had shops, it had traffic lights, it had parked cars, so while they were still sort of that clear 500-meter line of sight to the fake workers, there was more going on in the field of vision of the drivers. Drew, I’ve sort of explained that research design and that set up.
Drew: I think that’s a fair description. They’ve got some photos in the paper if you’re interested in seeing the difference. The medium complexity environments really were visually complex. They were multi-lane roads, there was traffic on the roads, so there was a lot to look at and see to try to spot out the person standing there in the hi vis.
All the participant has to do is drive around and when they see someone in hi vis they just have to say worker. Someone in the back of the car clicks a button, records exactly what time that happened, and what position it was in so they can measure how quickly their researchers got spotted.
The general goal of this is they just want to compare a whole heap of the different factors, so compared to low complexity, high complexity, and compare the four different types of hi vis, which was basically either short vests or long coats, and that yellow-green fluorescent color versus the red-orange fluorescent color. Hopefully, listeners can visualize those two; they’re the two main hi vis colors that people use.
David: The Australian standard, particularly on high-visibility clothing, is (I think) there are a few more colors now, but the most traditional clothing is that really lemony colored, lemon lime yellow type of color, and that burnt orange, orange-red color. Those are the two colors.
Low complexity, medium complexity, four different types of hi vis clothing, vests, long sleeves, people moving their arms around, people standing still. What did the research find?
Drew: After all this work, really the only thing that they found was the very obvious, that in more complex environments, the workers are much harder to spot, which might seem a little bit disappointing, but it’s reassuring because if there was a huge difference between the different types of hi vis, then lots of us have been making lots of dangerous mistakes for quite a while.
It turns out to be a fairly consistent result, that the type of hi vis doesn’t hugely matter. In particular, the type of color doesn’t hugely matter. There are a few things they talk about in the literature review of this paper and that we found in some other papers that we’ve looked at. Why don’t we just sort of run through these as interesting findings and have a chat about each?
David: That would be great. Drew, you kick us off.
Drew: Okay. The first one is that the ability to just spot hi vis and the ability to support a human wearing hi vis seem to be actually two different mental tasks. That spotting, recognizing someone as a human is actually a little bit tricky, particularly if people aren’t expecting to see a human.
For that reason, what seems to matter is less about having hi vis on your body and more about people being able to see moving arms and legs. If someone is waving their arms and legs, that makes them much more recognizable, and if you put a hi vis on their arms and legs so they stand out. That seems to work as well as them waving it around.
In this study we summarized, they said that there’s not a lot of difference between just straight up coats and vests or moving and not moving, but when you cross-compare the two, the things that stand out is that either wearing a long coat or moving your arms and legs is better than just wearing a vest and just standing still.
David: I think that is an interesting finding, Drew, and it makes some sense. If someone’s wearing a full sleeve hi vis clothing, then you can see arms out to the side of a body which distinguishes it from a box, or a sign, or a bollard or something like. Whereas if you just see a vast, you are literally just seeing a box, just a square front if you’re looking at a person standing in front of.
That idea of people wearing vests in roadside environments, in [00:13:47] environments and things like, maybe, I don’t know, Drew. What do you think? Just fine? Does that mean that you’d be better off giving people pullover vests with sleeves in those environments rather than just the normal vests that people put over their work shirt?
Drew: It does at least provide a hint that the more important it is for someone to be visible. The higher the class of hi vis, then the more it is full body does seem to actually matter. That matches with cycling and motorcycling research, too, that says that having your reflective bands around your ankles and wrists might seem just like a little thing to hold your nice work clothes in but actually really does make a difference to people recognizing that you’re a fellow human.
Next interesting thing, which is not going to be that surprising, particularly to Australians, although this does seem to be coming around the world from the research. Even when drivers see roadworks, they don't tend to expect to see humans. People are very used to seeing roadworks and just assuming that there’s no one there, that it’s just an unoccupied site because we tend to leave roadwork setup even when people aren’t working there. The fact that there are obvious roadwork signs doesn’t actually get people looking out and getting better at seeing humans.
David: Say no surprise to Australians. I must admit I am one of those people that get a little bit frustrated of slowing down on a freeway to 40 at late in the afternoon or at nighttime for kilometer after kilometer after kilometer or not seeing a single piece of work being done but all of these traffic management set out. It really doesn’t surprise me that people don’t see a human, maybe they don’t slow down or something. We’ll talk about that again soon, I think.
Drew: I used to get really frustrated by the roadwork set-up with no one there, until I saw the statistics on how often people got hit by cars setting up or taking down the roadside markings. It really is quite a dangerous part of the job, so I really can understand people having set it up not wanting to take the risk of taking it down setting it up again.
David: That’s a good point. I hadn’t thought of it like that, but yeah, setting it up in the first place is quite a bit of risk.
Drew: The next one does seem to be fairly consistent, which is that there really doesn’t seem to be a preference for any particular hi vis color. There really doesn’t seem to be, hey we should all be wearing pink, or hey we should all be wearing lime green, or we should be wearing yellow instead of burnt orange. What people do find is that having colors that contrast is far more important than the particular color.
David, I don’t know if you ran into the study where—this is totally unrealistic—they’re basically proposing measuring the color background, and then if you’re working that day, you pick out hi vis that doesn’t match where you’re working. It showed that there was up to like a 50% improvement in spotability by picking out clothing that clashes with the background.
David: It makes sense. Literally, it makes sense. The contrast is important, and just thinking through, if you think about some of the environments that we’re working, some of the mining type environments where there’s so so much yellow—yellow vehicles, plant, signs, sand, and so much yellow, and then all of the workers are in yellow shirts as well.
Recently, we’re seeing many more dynamic colors. Like you said, pink shirts and that sort of thing. Just thinking back to that road environment, if you’ve got lots of red and yellow, all of the bollards are red and all of the streets signs in yellow, if every worker in that roadside environment was in pink or a completely different color from all of the other stationery signage and equipment, maybe that’s a sensible thing to do, or even a safer thing to do based on this finding.
Drew: There does seem to be a pretty strong case for picking one color for humans and making it separate from the color that we use for other things because of this challenge of identifying humans rather than just identifying something that we should be paying attention to.
David: And particularly this safety yellow, we’ve got the safety yellow everywhere—handrails are yellow, stair treads are yellow, our plant and equipment is yellow—everything is yellow, so when the people are yellow too, they just don’t stand out.
Drew: It’s funny because the paper I read actually said that we should standardize on yellow because that’s the color that people associate as human [00:18:56].
David: They also associate everything else in an industrial setting with yellow as well. One of the sites that I used to be involved with had a separate hard hat for visitors. Everyone else had a white hard hat on site. If you’re a visitor, which meant you didn’t usually work on that site, you had this most horrifically unfashionable watermelon-colored helmet. It was this cross between red and pink. The color was called watermelon.
It stood out, unbelievably. You could see these hard hats walking around the site and you knew that was a person who wasn’t usually on the site. It was designed for the emergency response system to know if someone was on a site who may not understand the emergency response [00:19:50]; it was a major hazard facility. I’ve always remembered that because I always thought in that environment there was something that is so different to everything else and other people who are wearing it in the background setting, it really did stand out.
Drew: So do we make a deal now, David, just to standardize on something and try to convince everyone that we need to [00:20:13] color and stick to it?
David: Yeah, Drew. Well, what’s your favorite color for people?
Drew: We don’t tend to paint that many trucks pink. I’d be inclined to go for the pink myself.
David: Pink’s a good color, so all of our listeners should only buy their workers bright pink high-visibility from now on. That’s our color. The Safety of Work podcast recommends pink high-visibility clothing. It’s going to make your worker stand out and be more visible.
Drew: Well, not necessarily be more visible but more recognizable as humans because humans wear pink.
David: So more recognizable as humans in medium complexity settings.
Drew: We’ll make that the slogan: real men and real women wear pink.
David: Yes. Be seen and be in pink.
Drew: Sorry, I just realized I really don’t want to be gender-exclusive there, so real workers wear pink.
A couple of other fun facts, David. The first one is I finally discovered why we wear colored hi vis at all, which always sort of confused me is why these ugly colors. Because you can make anything fluoresce. With modern technology, you don’t actually have to have yellows and oranges, 70s disco colors to fluoresce.
It turns out that there’s a real scientific effect that colors seem brighter to humans than just plain reflective white. It’s called the Helmholtz‐Kohlrausch effect if anyone wants to look it up. Although, again, there does seem to be some ambiguity about whether it works as well in the real world as it works in lab tests.
The other fun fact is that now that everyone is wearing hi vis, it’s not just humans that can recognize other humans. Most of the actual recent papers on hi vis are not about humans at all. They’re about using computers to spot workers. It turns out that we’ve neatly labeled all of the humans on at work sites using these motion capture clothing, so we can use drones and cameras to spot the humans because they’re wearing these particular shapes of hi vis.
David: Oh, really? You can program these cameras in that to say that’s a person?
Drew: Yeah, it’s much easier if the people are wearing standardized hi vis that if they’re just wearing work clothing.
David: It would be even easier, I suspect, for the developers if they’re all in pink as well.
Drew: Absolutely.
David: This final thing I think we’ve already covered. Actually, we’ve dealt with this, but we’ve dealt with this in a different way. As a final point here you make that there’s been some suggestion in the research that we should actually standardize a human hi vis color and not have it the same as other roadside market type or signage colors, just so that people do distinguish those humans are as different from signs or other sort of markers or pieces of equipment. I think we’ve covered that with our recommendation.
Drew: I think we need a second recommendation, which is no pink traffic cones, no matter the incentive. Once we’ve colored our humans pink, traffic cones just have to be boring orange colors.
David: Boring reds and orange. And then the road signs, black and yellow.
Drew: I do have to admit failure because I tried to find a study looking at actually the original question that prompted this episode, whether there’s any evidence that we’ve got so much signage on roadworks the humans blend in. I couldn’t actually find a paper looking at this.
The only paper I found actually says the opposite for kind of strange reason. This was by the same authors on a closed course, and they had humans sometimes standing in the roadworks and sometimes standing on the opposite side of the road. What they found was that the humans were much more identifiable when they were standing with the roadworks.
I was wondering why that was until I realized that they’ve got flood lights in the roadworks. The humans who were in the roadworks were all lit up and humans on the other side of the road were standing in the dark. Of course, humans are more visible when they’re a part of floodlit roadworks than when they’re in the dark. But we don’t really have an answer to the question of whether humans blend into the back of roadworks.
There is, though, one study and it is just one study from Japan, that was looking at whether we should put light emitting devices onto our people. They found that there were so many flashing lights on roadwork sites, that if you put flashing lights onto the traffic controllers it doesn’t actually help make them visible. What works much better it is to shine a floodlight onto where the worker is. That makes a worker really stand out from the rest of the site. If they’re standing in an area lit by floodlights, it’s much better than putting flashing lights onto the person than they just blend in with everything else that’s flashing.
David: And even if they blend in with everything else that is flashing, I think also it comes back to this initial thing you said, the difference between the task of spotting a particular color and recognizing that as a human. I think seeing a flashing light isn’t necessary something that at least I would associate with that’s an indication that’s a person. For me, that would be an indication that’s a part or a piece of equipment, a post or a sign or something like that as opposed to, flashing light, must be a person.
Drew: Yeah, and I think this is one of the reasons why on cyclists it doesn’t work that well. That people just see a flashing light, it’s very hard to spot what it is. Whereas if they can see the legs pumping round and round, that becomes fairly obvious quickly that it’s a cyclist.
Final out for you, David. This one is from QUT, our friends in Queensland. They just made a list of what causes drivers to slow down around roadworks. It was just a questionnaire and they gave people a list of 12 different factors, reasons why you might slow down. Would you slow down when you see a sign? Would you slow down when you see a warning sign that the roadworks are coming up? Would you slow down when you see traffic lights?
The single, biggest factor by a long way was if they could visibly see workers on the site, which is reassuring that that is genuinely what people care about, is they slow down if they think it’s an occupied side where there are people there.
David: Drew, when I saw this takeaway, it did make me feel good that the biggest factor by the long way was this presence of workers and then the second fact that would make people slow down was the presence of police. We know that there are times when you see the flashing police lights at the roadwork site.
Now that I know, now that you explain that it was a self-report questionnaire, the cynical part of me guys, I wonder whether that is actually what happens when drivers are driving. If they see police at roadworks or if they see workers at roadworks, whether they slow down. I’d love to see that study done in a naturalistic kind of way.
Drew: So would I, David, but for the moment I choose to believe that even the [00:27:46] the police is because people care about the safety of police as much as they do about other workers, so they’re slowing down to be careful of not hitting anyone.
David: Okay, maybe today’s my episode day to be the cynical one, Drew, or maybe I’m just revealing a little bit too much about my own behavior.
Drew: Is there anything else you wanted to throw in before we go on to takeaways?
David: No. Let’s go into practical takeaways. Do you want to kick us off?
Drew: Okay. I should say that all the papers we looked at are to do with stuff on the roads. Obviously, people wear hi vis in a lot of other places, and we haven’t got into the usefulness or otherwise of the spread of hi vis lots and lots of places, even where it seems impossible that anyone needs to be visible. But let’s assume that we’re either on the road or somewhere where it is important that we care about visibility.
The first takeaway is that spotting people is a lot more than just them wearing hi vis clothing. The fact that it doesn’t vary when you vary the clothing, says very clearly that it’s actually the visibly uncluttered environment which is the most important thing, not the wearing of hi vis. If we care about people not being hit by equipment, do we care about people being spotted? Then we should be caring as much about the background we’re seeing at rather than the particular clothes we’re asking them to wear.
David: I think also recognizing them as a person. Something that’s a box with legs and arms. We’ve got the color and then we’ve got the fact that it is actually a person, and obviously the contrasting environment as well, which goes back maybe to that idea of having visitors long sleeve pullover tops as opposed to maybe just visitors vests at your site and things like that.
Drew: The second one—I realize now that it’s possibly a little bit of a contradiction to what I’ve just said—that exactly how hi vis is designed does seem to matter to some extent. In particular, the idea of having a color that contrasts with the environment matters. It doesn’t have to be the case that one particular color is always best. If you have a fleet of yellow trucks, then you probably want to go for a burnt orange hi vis. If you got a fleet of red stuff, you probably want to go for a yellow hi vis.
The choice of whether it is a vest or full-length seems to really matter. That’s gotten me interesting because most of the policies I’ve seen, at least for visitors, don’t tend to specify. They tend to just say you must be wearing either hi vis clothing or you must put on a vest over the top. I think the evidence shows that if a situation actively needs hi vis, then it is probably worth taking the time to work out what’s the best hi vis rather than just saying you must have something that’s hi vis. Your thoughts on that, David?
David: Yeah, I think it’s something that, again, what I start off thinking is a very simple question, is quite central to how much of an evidence-based approach we take to our practice. Like you said, in particular, I just reinforce that point. If a situation generally warrants high visibility, then it probably genuinely warrants going to links beyond just having visitors vests in the drawer.
Drew: The third one related to that, given the amount that we spend on hi vis and the size of the industry that sells hi vis stuff, there really is a good market opportunity for a little bit more research. What we’ve got doesn’t show any particular hi vis having a clear advantage, but mostly that’s because we haven’t done a lot of research. I reckon if we spent a little bit of money comparing different things, you could have a real advantage doing studies that work out what is actually better than something else, and then use that as a marketing opportunity if you’re in the business of selling safety services.
David: Drew, is that an invitation for anyone who wants to prove the efficacy of pink high visibility on humans in non-pink background environments until I get in touch with you at the university?
Drew: Absolutely. Once you get started there’s so many things, like is it better to have a splash of yellow and your pink is it better to have your reflectors be silver or a different shade of pink? I think it would be great to be able to say we have scientifically tested that this is the hi vis you should be wearing. We’ve patented this particular shade of pink. It’s the only one that’s been properly studied.
David: Pink, yeah. I’m looking forward to being able to go around and see which site. That being said, there is a lot of high-visibility pink in a number of different sectors now, particularly also in part to the mining sector, [00:32:54] organizations like that. We see a lot of pink increasingly in industry.
Drew: We should have picked a less common color to standardize on, David, just so that we could see the spread of, what you’d say, watermelon yellow?
David: Watermelon. It’s a color. It’s the real deal, Drew. Fluorescent watermelon. That’s going to be the color. Drew, invitations for our listeners now.
Drew: This might not have been the deepest podcast, but I think we enjoyed looking up the papers. What we’ve tended to find is that a lot of the listener questions we get, our challenge is so big that we can’t answer them in an episode. My inflection today is just find some listener questions that are small and might have just simple factual answers that we can find out for you. Something you’d like to know that maybe there’s a paper out there that David or Drew can dig up the paper and tell you what the factual answer to your question is.
David: So Drew, the question we ask this week is how visible is hi vis clothing? Do you have an answer?
Drew: I guess I didn’t phrase that question very well because the answer is yeah, hi vis clothing is literally visible. But it’s got a lot of interesting little complexities around whether we actually spot the people once we’ve spotted the clothing.
David: Yeah, and the importance of the contrasting background environment is definitely part of the answer to that question.
Drew, that’s it for this week. We hope you found this episode thought-provoking and ultimately useful in shaping the safety of work in your own organization. Send any comments, questions, ideas for future episodes to us at feedback@safetyofwork.com.