In this episode, we take a deep dive into a research paper about using fear or shock tactics to inspire behavior change when it comes to health and safety communication. Motor vehicle safety adverts are one of the more common references for this type of method of communication. We discuss what the research has found on the effectiveness of communicating in this way and why it sometimes has the opposite effect.
The reason we are talking about this today, is because this tactic is often used in workplace safety videos and we ask whether or not it works for everyone, how well it works for workplace safety and whether its even ethical in the first place, regardless of its efficacy.
Topics:
Quotes:
“Just because something is effective, still doesn’t necessarily make it OK.” - Dr. Drew Rae
“The amount of fear doesn’t seem to determine which path someone goes down, it just determines the likelihood that they are going to hit one of these paths very strongly.” - Dr. Drew Rae
“Communication which gives people an action that they can take right at the time they receive the communication is likely to be quite useful. Communication that just generally conveys a message about safety is not.” - Dr. Drew Rae
Resources:
Griffith University Safety Science Innovation Lab
The role of fear appeals in improving driver safety (Research Paper)
David: You're listening to the Safety of Work podcast episode 78. Today we're asking the question, do shock tactics work? Let's get started.
Hi, everybody. My name is David Provan. I'm here with Drew Ray, 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. Drew, what's today's question?
Drew: David, as you know, we've had a lot of requests to talk about shock tactics in safety advertising. It's been a topic that I've been very keen to talk about but we've been slow to do an episode, mostly just because the evidence is really mixed. We don't like to cherry-pick papers and just talk about one single study when there might be other studies that say the opposite. We've eventually decided to give it a go.
What we're talking about in this episode is generally called threat appeals. The idea is to present someone with the negative consequences that they might experience as a result of a particular unsafe or illegal behavior. The idea is that a threat appeal results in fear response and that fear will motivate people to adjust their behavior the way we want them to.
Probably the most well-known examples of these are where the threat is really explicit, realistic, and sudden. We're trying to exaggerate the fear response. I think the shock check tactics you see in road safety campaigns, there are also nonroad examples. We see them sometimes in electrical safety, in pool and water border safety. You could say the same thing about putting graphic pictures on cigarette packets. You might remember those really old workplace safety training videos that used to be really common to show people losing eyes and limbs.
Workplace safety training still often uses the threat component. As far as I know, David, it doesn't usually use the exaggerated response that road safety tends to use.
David: I think, Drew, less and less, although I still feel like shock tactics have a big place in safety management strategies of organizations. We tell lots of stories of incidents, disasters, and the consequences of getting safety wrong—personal consequences, organizational consequences. We have these threats to create fear of people losing their job if they don't follow the safety rules. There are still lots of posters and public speakers who are quite prominent in safety circles who have been burnt or lost their eyesight because they didn't do a particular safety practice in their job.
I think we can even maybe extend this, Drew, to other tactics in safety where companies do these cultural programs where they ask workers to bring in photos of family members and say, here's a photo of someone who will miss me if I'm fatally injured at work. It's all just creating these fear and anxiety about not doing the right thing for safety. Drew, I don’t know if that's too much of an extension from these acute personal injury shock tactics. As I was thinking about all the different things that we do in safety to try to create motivation through fear, it's still quite common in our practices.
Drew: Yeah, I think all of these things rest on the same general theory of operation, which is, we create fear that fear creates motivation. Motivation changes behavior. Most of the actual evidence we've got is around particularly graphic-type campaigns. Where people are deliberately trying to increase the amount of fear in the campaign, the idea being that the extra fear creates some extra effectiveness. This is controversial for a few reasons.
The first reason it's controversial is just that the evidence is really mixed. Lots of academics just think on balance, they're not generally effective. The other alternate conclusion you could take from that mixed evidence is that they are effective, but they're risky. The reason why we get mixed evidence is they sometimes work and they sometimes don't work. When they don't work, it can genuinely backfire. It can create the exact opposite effect to the one that you intend.
The third reason, some people don't like them, is just that regardless of whether they work or not, they're considered to be a bit unethical. David, do you have an opinion on the ethical side of things, the use of fear as a motivator?
David: Yeah. I think, Drew, like with all ethical dilemmas, it's not a black and white issue. You can make an argument for both. How I think through this myself personally, I don't think it's okay to intentionally set out to make someone scared or anxious. I don't think that's a psychologically healthy thing for organizations to be doing. Even if we think that it's okay to create that fear in people if my intention is to keep them safe, I don't think the ends justify the means.
Do I really want to actually make someone either physically or emotionally upset through thinking about something that could happen to them and create some trauma and intentional scare tactic to try to get them to follow the safety behaviors that I want them to follow? Personally, I'm not sure that it sits well ethically with me, Drew, but your thoughts?
Drew: Yeah, I'm on similar lines. I think that it is genuine harm to someone to make them afraid. Anytime that we're setting out to cause harm to people, we should have very, very good reasons for doing it. There are times when harm is ethical. Surgery is a great example of ethical harm. I think that if we're going to do something that causes harm, we've got to be very, very, very sure that that is worthwhile.
More importantly, we need the consent of the person that we're doing it with. If the person agrees that causing that harm to them is worthwhile, then that's a very different case to when we involuntarily inflict that harm on them. Almost always, in the case of this shock advertising, there is no consent. Half the point is to actually surprise the person with what you're doing. I think we're very ethically dodging through territory when we're trying to surprise people by causing harm to them.
David: Yeah, and we don't know what the extent of the emotional response that will create in each individual based on their individual circumstances, life experiences, and things. We could genuinely be creating harm, even though our intention is to improve safety. Drew, I don't know that we've had that many ethics of safety conversations, but this one was a great opportunity for us to think about not just the evidence base behind what we're doing, but also the moral basis of the approaches that we're taking in safety.
Drew: Yeah, I thought that was worth touching on. As this story goes on, we're going to see that in fact, this is a tactic that has some effectiveness behind it. I wanted before we did that, to recognize that just because something's effective, still doesn't necessarily make it.
David: Yeah, absolutely, Drew. Let's talk about this paper because even though we're more in the occupational safety space, we're leaning in an adjacent field of research of road safety for this. Do you want to introduce the paper and maybe talk about why we've gone to this particular body of research to answer this question?
Drew: The paper is called The Role of Fear Appeals in Improving Driver Safety: A Review of the Effectiveness of Fear-arousing (threat) Appeals in Road Safety Advertising. It was published in 2007 in the International Journal of Behavior Consultation and Therapy. There are three authors: Associate Professor Ioni Lewis, Professor Barry Watson, and Professor Katherine White. They're all at the Queensland University of Technology Centre for Accident Research and Road Safety.
They were there back in 2007, they're still there now. That center has an incredibly contrived acronym. It's called CARRS. Apart from the acronym, it is a genuinely world-famous center for road safety. All three of these authors are experts in road safety campaigns. They do all sorts of interesting work, including different types of messaging such as looking at shock at the one end and then looking at the other end of things like using humor to promote safety.
The reason why we've picked this particular work is really that it's in road safety that people have most rigorously studied the idea of these campaigns. Even though they've done lots of places outside road safety, they're not often evaluated nearly as well as they are within the road safety space. They've got access to lots more theoretical work, lots more evidence behind the different interpretations of that evaluation.
I'll just give a quick reminder on why we even bother talking about the authors of literature reviews. Really, if you pick up a literature review, chances are it's done by two types of people. The first one is that they've done the review because they are new to the area and they want to understand the area themselves, so they're doing a literature review. For example, at the start of a Ph.D. Or they do it because they're an expert in the area and they're trying to explain that area to other people.
Obviously, if you want to get a good handle of infield yourself, you want the second type of review. You want to know that the person who's written it is an expert in the area, understands everything that's been written, and is giving you a fair interpretation of the current state of the field and the current state of the evidence. That's why we look to the authors. In this case, the authors gave us a lot of confidence that this is going to be a good representation. We do need to be cautious though because 15 years is too old for a literature review.
This was published in 2007, there can be a lot in between. In fact, there has been a lot in between on this topic. The reason I picked this review is because I really liked the way they break down and discuss the issues and discuss the different things that we need to think about. We can use that as our framework and then do a bit of currency checking along the way. Every single thing of this that I've checked is really more of the same.
The authors back in 2007 said, it tentatively looks like this. Research that's happened in the meantime has just given us a stronger picture of that same impression we had back in 2007.
David: Drew, I think there are some nice models and frameworks for how we think about why the shock tactics would work and how they work. Do you want to just lay down if you’ve taken for granted questions or conclusions in the article to frame the next part of what we'll talk about?
Drew: Sure. I guess the most obvious one is that it's apparently very well known in the field of road safety campaigns and in persuasion generally, but few campaigns have very mixed results. Sometimes they work, sometimes they don't. Often, if you compare two campaigns—one that worked and one that didn't work—they look very similar to each other. Two campaigns with the same amount of resources behind them, the same amount of shock value—one works, one doesn't.
That's not really the question. The answer to the question, do fear campaigns work is sometimes and maybe. What we really want to know is why and how do some of them work? Do we understand well enough that we can just do the ones that work? Can we use our knowledge of them to create ones which reliably work?
With that background, we've got some general theory that's been established. David, I think I'll get you to walk us through the theory developing it as we go and I'll jump in where I understand and where appropriate.
David: Yeah. Drew, we want to understand how we get from this fear to behavior change. What we’re looking at is, how does fear connect with persuasion for someone to adopt change in behavior or adopt a behavior? A number of different models for how this relationship might be created in the cognitive processing of individuals.
We've got these linear models. These are drive theories, which you mentioned earlier. We create more fear, that creates more arousal, that creates more motivation, that is more likely to create behavior change. It's almost this linear model where the more fear we can create, the more motivation that we'll get. Some of the early theories said that's the way it works. I think in some health promotion areas like weight loss and addiction, more fear maybe is believed to create more motivation.
Then we've what they call a curvilinear model where there's almost like an optimum amount of fear or in sports, you might think of an optimum amount of performance arousal. The more shock value that we can create influences behavior up to a point, but then it becomes too great and almost too overwhelming, too anxiety-provoking for people. They then enter into avoidance reactions or maladaptive behaviors as ways of not actually reducing the threat, but just reducing their emotional response to the threat.
Drew, there are lots of different models. I don't know whether you want to add anything more to the linear and curvilinear models before we get into slightly more complex models.
Drew: I guess the main thing I should just say is that the reason why we're trying to fit these models is that none of the early models actually match the data so we postulate this linear thing. Then we have some studies that say, yes, there's a linear relationship, and others that say, yes, there's a linear relationship, but it's the exact backward. That as we decrease the amount of fear, it increases the amount of persuasion.
It's trying to reconcile, how come some studies say it goes up, some studies say it goes down, that's when you get to the curvilinear model, which says, this is why it goes up to a point and then down after that point. Then you do further studies trying to work out exactly what that curve looks like and none of them fit either. That's the point at which we need to say, okay, it would be great if we had a simple model, but none of the simple models fit the data. We need something more complicated.
David: I think Drew, you're exactly right. When we're talking about cognitive processing, decision making, and behavior of individuals, we need more complicated theories with more mediators and more relationships to understand how they might work. In 1975, Rogers—to better understand these fear appeals and how people cope with them—proposed this idea of protective motivation theory.
In the 1980s, he expanded it to more of a general theory of persuasive communication. This idea of protective motivation theory was created to help us understand individual human responses to these fear appeals in a slightly more nuanced way than these linear and curvilinear drive theories. It basically proposes that people will adopt safe behaviors or protect themselves from the threat promoted in the shock tactic based on two factors. Their appraisal of the threat itself and their appraisal of their ability to cope and adopt the behaviors.
Drew, if I just run through some of these, I think there are some dimensions to these threat appraisals and coping appraisals that people are going to resonate with when they think about safety. What does an individual do when they are confronted with this shock tactic? It might be a scene of a car crash, a serious workplace incident, or even someone who has been seriously injured at work. They're appraising for themselves both the severity of that threat and their individual vulnerability to that situation.
They're thinking through these factors that are the source of the threat, the degree of harm that it could cause, and the probability that they themselves might experience that harm. With that threat appraisal, they're also thinking about the positive aspects or the rewards of them actually then moving forward and changing their behavior. It's this cost-benefit trade-off. Is the threat real for me? Is there benefit in me actually doing something different? Drew, do you want to add anything else to the discussion on threat appraisals?
Drew: No, but it might be just worth throwing in an unrelated example, like climate change. There are two aspects to it. One of them is how much people believe that it's real and believe that it's serious. The second is how much they feel that they can personally do anything about it. You're only going to get people to change their behaviors, not just if they believe that there is a real threat, but they believe they've got the capacity to do something about it.
If people feel helpless, they feel there's nothing we can do, and they feel anything we do is not going to help anyway, then they're not going to be motivated to take action, regardless of how real they believe the threat is. Even to the point where they can then turn around and start denying the threat because they feel so helpless.
David: Exactly right, Drew, on what this model goes to explain. Our listeners might think, this sounds a little bit like a risk assessment and it kind of is an unconscious risk assessment that individuals are doing, okay, what's the severity of the threat when I see this particular graphic situation? What's the severity? What's the perceived probability of that occurring to me?
Then we move into the coping appraisal process through, like you mentioned, which is, what's my own belief that I can successfully enact what this recommended behavior is? What are the costs associated with me enacting that behavior? This theory would propose there's this subconscious risk assessment cost-benefit analysis going on in people's minds. If we think about speeding, drunk driving, quitting smoking, or following safety rules in the workplace, we understand that people are going through this constant risk assessment cost-benefit analysis in their minds.
Drew, I think the interesting one, the last part of this protected motivation theory model talks about the efficacy of the response. This is particularly whether people feel that the behavioral response will actually be effective in reducing the threat. This is like, well, do I actually think that driving 10 kilometers an hour slower is really going to reduce my risk of an accident? Do I really think that following this rule does actually stop me from being fatally injured?
When we're using shock tactics, I think what this model suggests is that one of the important mediators is that people can actually connect with the behavioral response as actually being directly and materially linked to preventing the threat.
Drew: Which goes a long way to explain why a lot of workplace safety campaigns are not as effective as you might think because we clearly point out the threat, but then the solution that we offer doesn't obviously counter that threat. You look at all of this danger, therefore make sure you fill out a risk assessment. It only works if people genuinely believe that filling out the risk assessment is going to protect them.
I think the worst one I've seen is, it showed a video of people having their hands cut off by a saw. The instruction was you make sure you wear your gloves. There's no way that wearing the gloves was going to protect against the type of accidents that were shown in the video.
David: Yeah, Drew. I think way back in episode 2 when we asked the question, why do people break rules or something, when we started looking at some of these things like live electrical work. I think if we run through this protective motivation theory that is showing people images of people being electrocuted doing live electrical work, and then telling them to always isolate and test.
A 30-year experienced electrician has a whole lot of experience around what they perceive those real risks to be and what they perceive those control measures to be, and when they're important to do and when they're not. To think that we can throw some shock tactics out there and some expected behavior, I think the evidence at least says that maybe that's an act of faith rather than a logical thing to do.
Drew: David, so far the theories we've talked about explain why fear campaigns don't work. Some of the things that we've talked about that could go wrong with them really just would reduce how effective they were. But it doesn't really explain why people can follow the rules about how to do a campaign and still have ones that actively backfire. We needed to continue to expand these models to explain how it is that people can have a campaign that follows that creating a threat, creating efficacy, but it still backfires.
David: Yeah, Drew. I think there's a couple of things that I'm not sure whether we will get into it later or not, but I think it makes sense now. We've got these different ways that the campaign can work or not work. Ideally, what we want to do is actually create this adaptive response to the message. People see the shock tactic, they see the expected behavior, they create that connection, and there's a level of change or a level of compliance with that behavior that we want. Then we've got this outcome, which would be what's referred to as maladaptive.
We actually connect with the shock tactics, but we don't want to necessarily do anything to prevent the threat. We actually want to just reduce our fear through denial, avoidance, or some other action that makes us, I suppose, less anxious about the threat, but not necessarily adopting the behavior that we want.
Sometimes those maladaptive things, this is where it really can backfire, where we can actually just drive different unintended behavioral consequences in our workplace—secrecy, a whole range of things. The last way it can probably backfire or just have no impact at all is that people just ignore the message. They just decide somewhere that either the threat or the message just isn't relevant to them.
Drew: The extension that we make to the models to account for the maladaptive behavior is really one that creates a couple of different paths that people can flow down. One of which has got the very rational adaptive response, and the other which has the more maladaptive, arguably less rational response. How do you explain why people go down that second path?
This is one of the areas where there is a lot more recent research. Looking into things like individual factors, particular things about people's psychology, their pre-existing beliefs, the exact tone of the message, or how directly the message appears to be critical of the person receiving it. There's a whole raft of research around that. All of that just goes to show how strong some of the defensive maladaptive responses can be.
The idea that someone simply dismisses the message and says, oh, this doesn't apply to me, that's at the weakest level. At the strongest level, people can get quite aggressive, can start fighting back against the person giving the message, and can counter messages. You're telling other people that they should ignore the message, that it's misinformation, that it's controlling behavior.
We really need to be genuinely careful about how we design the message to avoid people going down a maladaptive response path. The amount of fear doesn't seem to determine which path someone goes down, it just determines the likelihood that they're going to hit one of these paths very strongly. Beyond an initial amount of fear, we're not doing any good by adding in that extra fear.
Once we get people's attention, the key thing is shaping the message around things that can influence them into appropriate adaptive behavior in response to the message. David, shall I go down these four things and you just jump in?
David: Yeah, great. Absolutely, let's do that.
Drew: Four things that matter in how people respond to the message. Number one we've already covered, which is just the severity of the fear. That has some impact, but it doesn't decide which path they go down. The second one is the thing they call susceptibility. This is how much the particular threat matters to the particular audience.
We don't often consciously think about this, but there are all sorts of different versions of the threat we can offer. The most obvious one is a physical threat of injury or death. Lots of campaigns are based around more moral value threats that people will be judged harshly by other people, threats that are a little bit indirect like a threat to your family or loved ones instead of to yourself, a threat of punishment, a threat of being caught, a threat of a fine, a threat of a jail sentence, or a threat of losing your license. These are all different threats, and different people are scared of different things and respond differently to the messages.
One of the things that I find most interesting is that the target audience of a lot of road safety campaigns aimed at young males are the people that we really want to influence out of some of the most dangerous driving behaviors. That particular audience is the ones who are least susceptible to physical threats of injury or death. It's almost like humans are hardcoded that young males don't particularly care about dying. They might care about a lot of other things, but the threat to death doesn't particularly faze them. They're more likely to just ignore the message.
David: Yeah, Drew. I think that was one where the connection into workplace safety—particularly in many of our high hazard hardhat industries where I know a number of our listeners work—is that the target demographic for a lot of our occupational safety communication messaging is a similar demographic maybe, unfortunately, largely male-dominated between the ages of 20 and 40. This evidence around the shock tactics go, that's not a population that's motivated by a fear of hurting themselves.
Drew: That's susceptibility. The second big factor is relevance. The idea of relevance is that if people think that they're not going to experience the consequences or that the message isn't aimed at them, then they're not going to be activated to go down either path in responding to the message. If some people have just got a really strong it-won't-happen-to-me belief. That's particularly true in things that are relatively rare like road accidents.
Road accidents are a serious problem, but most people will never even experience a related accident. The most you'll see is something by the side of the road, something on TV, or flashing blue lights. Your chance of being in a serious accident is high compared to other industries, but low just compared to your average lifetime.
The other thing is that with messages like speeding, lots of people genuinely don't believe that they speed. Telling people don't speed, they're just going to say, well, you should have told those other people who speed all the time. The evidence that most people believe that they're better drivers than average. Most people believe that on average, they drive slower. They're slower than an average driver. Most people, if they're getting a speeding message, will think that the message is really targeted at other people.
There's some evidence that that also applies to workplace safety. Very often people will look at a message and say, yeah, that's a good message. I'm glad that the people who need to be told that message will be hearing it, or I think this message will be really persuasive for other people, but I'm not going to change my behavior because it doesn't really apply to me.
Then the fourth one is the one that we've already talked about, which is the efficacy. We need to give people a clear action to take. They have to feel that they're able to take that action. They have to believe that if they do take that action, then taking that action will prevent harm.
David: I think Drew, what was interesting in one of the results of the findings is when we think about speeding and drunk driving as being two prominent risks that are present in road safety campaigns, there's a difference between speeding and drunk driving. Some of the adaptive responses were where there were alternatives for people to have. If I need to get myself home at the end of the night, I've got alternatives to get that need met. I can have a designated driver, I can carpool, I can take public transport, I can take an Uber. There are ways of me actually changing to adaptive behavior.
Speeding was like, there weren't that many options. If you want to get somewhere fast, you don't speed, it’s just don't do something. That there's no other way of getting that need met. I don't know if I've read into that efficacy piece right or what, but I found that interesting if you’re thinking about safety. Just telling people not to do something is maybe different than telling people, look, we want to avoid this situation through one of these ranges of options for us.
Drew: They talk about that a lot in the paper that that's one reason maybe why we get such mixed results is that traditionally, anti-drink driving campaigns have been quite successful. Traditionally, anti-speeding campaigns have been quite unsuccessful. Perhaps one of the reasons for it is that drink driving campaigns give you all of these positive actions. They're quite good at creating that efficacy. You can imagine what drink driving campaigns would be like if the only message they gave you was don't drink [...] without the Australian audience.
I didn't know that that was going to work. Whereas offering people all these alternatives that let them continue to drink, let them continue to drive, and just create strategies that don't involve the mixing of the two.
David: Yeah, I think that's really insightful. Again, it is a generalization. If we think about health and safety, communication, messaging—the difference between messaging and don't do something versus messaging a whole range of positive behaviors that you're looking to create in your organization might actually change the effectiveness. Whether it's shock tactics or not, I think even if we go beyond that might change the effectiveness of your communication.
Drew: Something that we've seen from other research in workplace safety is that communication which gives people an action that they can take right at the time they receive the communication is likely to be quite useful. Communication that doesn't generally convey a message about safety is not.
David: Drew, are there things that matter in this research?
Drew: We've got these big four in the theory and then they've got this category of even taking into account the big four things, there are probably lots of other factors that can have an influence. I have to admit, I haven't checked the recency of these ones. I haven't looked up to see how those have developed, but you could imagine that there's lots of stuff just hidden in this space because the way people think and behave is complex.
Two particular things they call out is the possibility of a wear-out effect. There's some evidence that if people get too much exposure to graphic images and messages, they just start to generally lose their effectiveness. The very first time you see a shocking ad, it might really grab your attention. By now, people are just so used to road safety ads that are shocking that lots of us just look away the moment we suspect what the ad is about. We don't even want to watch to the end where there's a car crash because we're sick of seeing it.
The second one is just that it's important that these things be credible. Sometimes if we make it too graphic or too extreme, then people aren't actually going to believe that this is a credible story that could happen to them. I remember watching one old workplace safety video and the message I came away from was, oh my gosh, do not work in this place because if you work, you're going to die. No matter what you do, some accident is just going to jump out of nowhere and kill you. You're either laughing or you're ignoring it, you're not actually thinking that it's real.
David: Yeah, a bit like wearing your gloves to stop your hand getting cut off by a saw. Drew, we know that people don't get hurt every day. If people are engaging in certain behaviors in the workplace—let's just call it work practices—that maybe the organization doesn't believe are safe enough, people aren’t probably getting hurt engaging in these practices every day. Suddenly, telling them one day that if they do this particular practice, then they're going to lose their arm or be killed.
When they're saying, but I do this every single day for the last 30 years nor I nor any of my colleagues have ever actually experienced that, then you're right. Particularly if the behavior we're trying to push is one of a behavioral or procedural type of control, we can quickly see how some of this communication would fall apart when we think about efficacy and threat.
Drew: The fourth thing that they go through in this paper is just some general problems with research in this field. Most of these aren’t all the research is shoddy. They're just showing how hard it is to do good research around such a complex topic. One of the things they point out is that lots of the studies of these things show that stimulus, but they forget to actually test whether the stimulus even generates fear and what other sort of emotions it might create. I think that's when we get into ethically dubious territory. If we're exposing people to something that's designed to create strong emotions and we don't consider the full range of emotions that might be creating and the harm it might be doing to people.
The second thing is the one that we've touched on, which is the behavior that you're trying to change itself matters. We're going to get different results if we're trying to change drink driving, if we're trying to change speeding, if we're trying to change glove-wearing, if we're trying to change people doing checklists. Any different behavior might have a different type of message that actually increases or decreases the efficacy. We can't just make blanket statements about what works and doesn't work.
Maybe you could explain this one, David, because I can't accept it without fully understanding it. They say that message acceptance and message rejection are like two different processes. You can't just assume because people follow the acceptance process they don't also follow the rejection process.
David: If you recall, we did an episode on trust where we said that trust and mistrust then is not part of the same continuum. The processes to trust someone aren't the opposite of the processes to mistrust them. When I read this, I read this in the same way that there are processes that happen for us to accept a message and take an objective response. Then there are different sets of processes for us to happen that reject messages, be maladaptive, or ignore messages.
This is where I think it wasn't much clear beyond that for me, but that's what I read into it, that there are these two different processes. You can both be engaging with both processes at the same time. You might accept a part of a message but reject a part of it. You might accept the threat or the fear but might reject then maybe the behavioral component of the message.
Drew: Thanks, David. That's helpful. That particularly matters then if we're doing a study where we show someone a message, and then just ask them, do you think this message is likely to make you change your behavior? We need to ask some questions that probe around that and check in both directions as well. What makes them think about the fear? What makes them think about what they're likely to take their view when they can't take the behavior?
The final one that's worth just mentioning is that doing these sorts of studies, people tend to do them in labs, which is not where people actually receive the messages. We often test TV messages in a lab instead of with people just at home happening to see it on the TV. We don't necessarily get the full idea of how people would actually respond to this in real life. Very often in a study, people will consciously look at the message and think about it. Whereas if it's a popup window on their phone, they just click ignore, it disappears, and they've never actually absorbed it.
David: Drew, I don't actually think of this until just now. I haven't checked any studies. I don't know how well you know this area. A study that actually I had, for example, a stretch of road and a really graphic message about speeding and had a series of speed detection devices before and after that particular sign to see. In the five kilometers after that sign, were people traveling slower than the speed they were traveling in the five kilometers leading up to that sign or something like that? Is that the thing that would give us a better understanding of whether there was that translation between the shock tactic and the behavior?
Drew: David, I think the answer is yes, but you've probably picked on the one example of where it's possible to actually easily do that. I know they've definitely done these studies with things like speed cameras. You can think of actually a speed camera sign as a fear response trigger. They've shown that in the area around seeing a speed camera sign, people do in fact slow down. You could do the same thing with a graphic billboard around that sign.
That's not going to test things like, does it change people's behavior for one week or two weeks? It certainly is not going to test things like messages on Facebook, messages on LinkedIn, on their phone, or training videos at work. We've also got the problem that we do a lot of these studies with students. Students are a very particular population. Particularly psychology students tend to skew female and very often a lot of these workplace behavior things have a deliberate young male target audience.
There's clear evidence that females respond better. They're more susceptible to threats of physical harm when they're at college age, whereas men are very poor at responding to threats of physical harm at college age. A lot of these studies were self-reported outcomes rather than measuring the behavior. They're doing things like asking people after seeing this, do you think you're likely to change your behavior, or do you think you're more or less likely to change your behavior rather than actually measuring the change in behavior?
Those are all caveats, but I don't think those caveats take away from the fundamentally mixed results anyway. Even with these caveats, some interventions clearly do work. Even with these caveats, some interventions clearly don't work. The reasons why they do and don't work are really, really complicated.
David: That being said, Drew, practical takeaways. Let's see if we can give our listeners some direction about what to translate out of this research into maybe some of our safety practices. Do you want to kick us off with some practical takeaways?
Drew: Okay. The first one, let's give a direct answer to the question. Having some form of threat in your safety campaign does help. In particular, it helps to attract initial attention and get people thinking about the threat and what they're going to do about it. But ramping up the size of that threat doesn't increase the likelihood of your campaign working the way you want it to do. It's not actually worthwhile deliberately creating shock value once you've given people some believable threat to start with.
David: Telling people that, hey, what's going to come next is important because it's about safety and can stop you from getting hurt, and getting hurt is bad. You don't then need to go and show people graphic images of people who have been involved in workplace fatalities.
Drew: There's no particular reason that's going to make your message more effective. What will make it more effective is choosing a threat that the target audience is susceptible to. Pick your threat carefully. It may in fact, not be that you're going to die is in fact the best message. Focus on increasing the efficacy. Give people something to do, make them believe that they can do it, and make them believe that doing it will keep them safe, and put your attention into that part of the message.
David: If I can just share a story, Drew, I was involved in redesigning a safety induction program with a colleague in a previous organization, and it was for a high hazard industry. We ran this pilot program to get some feedback from the workforce around the induction program. We had this heavy message in the end, which is you need to basically adopt what we're saying in this induction program to keep yourself safe because our family relies on you to think about all the things you'd miss if you were killed and how much pain you'd cause your family. It's creating that fear around it. That type of fear.
That was really interesting because again, the young male population, and the feedback that we got at the time—the direct feedback was, can you stop saying those types of things to us? Don't tell us we could get hurt. Don't tell us our life's going to be over. Don't tell us that we should love our family more to work safely.
What they said is, why don't you just tell us that you have an amazing job with amazing pay? You probably didn't finish high school. You're covered head to toe in tattoos. If you don't learn how to work safely and follow the rules, then you won't keep your job for the rest of your career.
I don't know if that's how you think about that example, but in that population, it was very much about working safely is your opportunity to keep this amazing job you've got, as opposed to working safely is a demonstration that you care more about your family.
Drew: No, I think that's a really good example of picking a threat that people actually care about. If I can just trade that for a direct personal example, I don't have a lot of physical threat response. I don't particularly care about dying that much, but it would be incredibly embarrassing for me to die in a workplace accident as a workplace safety specialist, particularly if I was breaking a safety rule while I did it.
That fear of embarrassment, that fear of you're a safety person and you broke a safety rule, that's a serious threat that I care about. I think all of us have different things that we're likely to actually genuinely be salient and responsive to. We shouldn't just assume that you're going to die or your loved ones will miss you is necessarily what everyone will respond to.
David: For listeners who want Drew to follow their workplace safety rules, just follow him around with a film crew, with a post ready to go on LinkedIn, if he steps out of line.
Drew: Yup, perfect. That'll make me totally compliant. The next interesting one is that, even if fear matters. Point one was that fear does in fact work as the start of the cognitive process. It's not actually the only emotion that can do that. This is something that's been studied a lot since this review came out. There are lots of other emotions, which fill that same role of getting people's attention, getting them to start thinking about the threat.
Humor is one of the obvious ones that gets used a lot. It's got that same caveat, that it's not about the emotion and making that emotion as big as possible. There's no point in making a safety ad that is just completely funny and misses out on the efficacy side of it either. You use the emotion as the hook, and then you concentrate on the efficacy.
The final one is that nowhere in there is a promise that we can use these theories to predict in advance what's going to work. The biggest thing that you should get out of this is we should be really humble about any particular campaign that we've come up with. Don't invest big money in big production values without testing whether our message works. That testing needs to be more than just do people have a positive or negative response. It means, checking how the target audience is going to actually act after they have been shown this message in the format that we plan to give it to them in.
David: Drew, the question that we asked this week was do shock tactics work? The answer has probably been a little bit clear, but do you want to have a go at the answer to that?
Drew: I think you actually wrote the bit into the scripts, David. I'll just read your answer and blame you for it. A big, it depends. I think we can certainly say that deliberately going for a shock tactic might not be the best place to put your safety efforts. But also, don't just knee-jerk, criticize shock tactics. Look forward, do they have those other elements of the campaign and maybe criticize them for missing those other elements, rather than simply say shock tactics don't work?
David: Drew, there are a couple of sentences in the abstract of this article that I think are worth just reading out at the end of this episode, if you don't mind. The authors say in the abstract, at the front of the paper, “While fear-arousal appears important for attracting attention,” like you 8 said, Drew, we've got to get the clickbait at the start. “Its contribution to behavior change appears less critical than other factors.”
Those other factors are perceptions of vulnerability, effective coping strategies, and furthermore, threatening appeals targeting young males, which is the high-risk group in road safety, but also a high-risk group in occupational safety as well have traditionally relied on the portrayal of physical harm.
However, the available evidence really questions this relevance, hence the effectiveness of strong physical threats with this group. Like in the stories that we told earlier, Drew, getting the threat right is more important for shock tactics than getting the threat big.
Drew: David, that's the point at which I'm just going to hook back to the ethics that even effective threats are still not necessarily an ethical way to go.
That's it for this week. We do hope you found the episode thought-provoking and useful in shaping the safety of work in your own organization. There have been some really interesting chats on recent episodes on LinkedIn. Do look out for the post being posted on LinkedIn, read the comment section maybe, and please update yourself. As always, we really appreciate your direct feedback to us, as well as any ideas you have for future episodes. The more specific the question, the easier it is for us to answer at feedback@safetyofwork.com.