RELEVANT RISK COMMUNICATION ISSUES

The discussion began with panelists sharing their experiences and expertise related to encouraging people to engage in protective behaviors. The conversation provided a brief summary of the field of risk communication¹ in general, and of risk communication issues related to the use of safety belts in particular. While each panelist used the terminology of his/her own field, the panelists had many overlapping comments pointing to commonalities of approaches across diverse disciplines. This section of the report represents an attempt to summarize the general “approaches” to risk communication discussed by the panelists. The next section of the report identifies risk communication issues that panelists considered especially relevant to promoting the use of safety belts.

A common theme throughout the discussion was that changing behaviors is difficult. Panelists emphasized the need for research to guide any communication effort. However, panelists also said that it is important to remember that the reasons for human behaviors are complex and often unknown even to the people engaging (or not engaging) in them. This is where the notion of unconscious defense mechanisms entered the discussion. While most panelists did not specifically think of themselves as experts in unconscious defense mechanisms, all were familiar with cognitive tendencies such as selective attention and selective perception (which often operate at a subconscious level) that allow people to depersonalize and minimize risks. Additionally, all panelists suggested techniques such as reframing as a way to get around these defense mechanisms, demonstrating that they had experience overcoming unconscious defenses. A more detailed summary of general risk communication issues discussed by the panelists is provided below.

Audience research is critical
Both the literature review and the panel meeting highlighted the importance of knowing as much as possible about the target audience. In fact, the very first thing panelists wanted to know from NHTSA was “how to characterize the subpopulation you are most concerned about.” Panelists identified several reasons why audience research is important. First, target audiences are likely to be more receptive to persuasive messages if the messages match their information preferences (e.g., some people are more persuaded by logical appeals and others by emotional appeals). Additionally, panelists stressed that successful campaigns use the emotions and perceptions people bring to the table to motivate change; they do not try to create new motivations. Thus, research is required to understand what motivators are pre-existing. As one panelist said: “The basis of the behavior is important, but it is not our basis, it is the audience’s basis [that is important]. So their basis [for wearing a safety belt] may not be fear. We are motivated by their likelihood to get in a crash. We are trying to protect them from killing themselves. We have got to put that aside when we think about how to make them do the right thing.” Audience research can also help to identify the “narrow windows where you can catch someone and deliver a risk message.”

This emphasis on tailored approaches and messages underlies the difficulty in creating strategies that will work for all people. Panelists talked about the fact that successful campaigns require multiple approaches, e.g., there are many types of people who do not wear safety belts “for a lot of different reasons.” Thus, it is unlikely that a “one-size-fits-all” solution exists for the remaining non-users².

Risk perceptions matter, but changing risk perceptions is not enough
Without any awareness of risk, people do not consciously act to reduce risks. For example, if people perceive no risk of getting in a motor vehicle crash, they will not wear a safety belt as a way to reduce that risk³. Thus, one strategy to increase the percentage of people who engage in any risk-reduction behavior is to heighten people’s risk perceptions. Raising risk perceptions can (and does) work; there is a link between risk perceptions and risk behaviors. For example, drivers report that they are less likely to use safety belts when driving at low speeds, and are more likely to use belts when they see others engaging in unsafe behaviors (Bradbard, Panlener, and Lisboa-Farrow, 1998). Thus, an approach that increases risk perceptions in driving situations that tend to be seen as “safe” may lead to more wearing of belts.

However, changing risk perceptions is complex. If the risk is perceived as too high, people may engage in fear avoidance behaviors (e.g., repression) instead of risk reduction behaviors. This is especially likely, according to panelists, among groups with less social power. In extreme cases, people who feel powerless may actually exhibit greater risk-seeking behaviors. Moreover, different social groups perceive and respond to risks differently. For example, white males are more accepting of risks than any other social group. This further complicates the task of altering risk perceptions. Moreover, risk information is hard to understand. As one expert said, “People have a really hard time with numbers.” Additionally, risks typically are not “communicated in ways that people can easily understand, visualize, or compare to other risks.” Also, the concept of cumulative risk is not well understood, even though it is helpful in explaining why it is worthwhile to engage in certain protective actions: “There are things we do that are very low risk [e.g., any one ride in an automobile]. Why do we worry about them? Because we do them over and over again … the idea of cumulative risk can help [make] … information be a little more meaningful.”

More importantly, however, risk decisions are not based solely on risk perceptions. How people respond emotionally (affect) is critical. As one panelist said, “You can’t tell someone that their emotional reaction is wrong.” Another said, “When things are strongly grounded in affect … simple information won’t do the trick.” In other words, people have more complex reasons for engaging or not engaging in protective behaviors than whether they “correctly” or “incorrectly” perceive the risk. For example, the literature review cited a study where high-risk individuals refused to use condoms before engaging in sexual activity because of the affect associated with condom use (e.g., what it said about them and their trust in their partners), despite an understanding of how diseases such as AIDS are spread.

People do not always know why they do what they do
Further complicating matters, human behaviors are motivated by a variety of factors, many of which we are simply unaware of. As one panelist pointed out, “If you could find out why people do things simply by asking them, we wouldn’t need psychotherapy.” Moreover, even though “people will come up with answers” when asked why they do things, panelists cautioned: “One should be skeptical of reasons that people give for their behavior.” Thus, people are quite capable of giving convincing yet false reasons for their behaviors. Moreover, they think these reasons are true.

While all panelists were familiar with this facet of human behavior, most were not comfortable using the terminology of unconscious motivators to explain these oddities. Thus, while all the panelists encountered this notion that “people do not always know why they do what they do,” most panelists (as noted) did not specifically think of themselves as working in the field of unconscious motivators.

Depersonalization of risk (society versus individual)
Interestingly, risks tend to be overestimated at the societal level. That is, for any given risk, most people assume that the societal risk is larger than it actually is. For example, most people think the likelihood of any given person being in a fatal car crash is higher than it is. However, risks are underestimated at the level of the individual. In other words, “If you ask people what their personal risk is compared to [society at large] it is always lower. So I may say, ‘Oh yes … these terrible things can happen to you … but they won’t happen to me.’”

This depersonalization of risks was a key theme throughout the meeting, and was often cited as a critical issue that had to be overcome. For example, panelists offered the following thoughts:

• It is hard to overcome the notion that while risks exist in general “it won’t happen to me;”
  
  
• The “big issue” is the discrepancy between “people’s view of the general and the specific;” and,
  
  
• “Trying to get across to people that their risk is high is a different thing from trying to get across to people that the risk is high.”

Cognitive skills allow people to downplay risks
The depersonalization of risks is one example of humans’ remarkable adeptness at resisting potentially threatening arguments via biased cognitive processing. Panelists summarized the mindset of someone employing this type of resistance: “There is something about me that makes me unique and different, invulnerable and safe. And it is a variety of things. I … exercise; I am very alert; I won’t get in an accident; I am a defensive driver.” In other words, while people are accepting of risks in general, and even overestimate societal risks, they systematically underestimate their personal risk by focusing on their uniqueness. This requires extensive cognitive effort: “People will sort of choose whatever it is that makes them uniquely less risky and focus on that as the reason they are not at risk.”

Strongly tied in with this denial of risk is the idea of personal control. Panelists talked about how there is a very strong perception at the level of the individual that we have control over what happens to us (i.e., “people think they have control, people want to have control”). For example, panelists talked about how people think they are good at selecting sexual partners without diseases, or how “people who don’t vaccinate their kids think they can control whether their kid is going to come into contact with a disease.” This notion of control is rarely challenged in routine situations, which is one reason panelists say the myth is so persistent: “In most of our lives, while driving or smoking, or having sex, you don’t get the perception that you are out of control.” Moreover, the sense of control is reinforced by repeated good outcomes. As one panelist (citing a poem) said, “The stair does end almost always when you think it is going to end.”

Empathy gaps also allow for the denial of risks: “Because you are not experiencing something right now, it is hard for you to understand what that means, what the implications are, or what the feeling is like.” Empathy gaps thus allow for risk information to be safely processed as non-relevant. Thus, even if people know that bad outcomes exist (e.g., because other people experience bad outcomes), such outcomes can be minimized because they are happening to someone else. As one panelist observed, there is even something life affirming about witnessing a fatal car crash; the other’s death reinforces the notion that you are alive.

Framing matters
Panelists discussed the impact framing has on how risk information is understood. For example, people react differently when information is framed as a benefit (e.g., cash discount when buying gas) versus a loss (e.g., surcharge for paying with credit). Panelists spent a great deal of time discussing how belt wearing can be reframed as a means to change behavior; these ideas are discussed in more detail in the suggestions section. However, it is worth noting here that reframing is one way in which information can influence decisions. One panelist talked about the power of new information to change the focus of decision-making (e.g., consider carbohydrate content when making food choices): “Once you make people believe that that is important, for whatever reason, you are going to change the behavior. [Someone might say,] ‘Oh god, I just hate to give up that pizza. I mean, it is my favorite food, but look at the carbohydrates in it.’ So, if you can deflect them to some other consideration that is different from the basis of what they are doing, then you have a better chance than if you try to hit them and say, ‘Well, it doesn’t taste so good.’”

Risks should not be exaggerated
Part of the discussion on risk communication at the meeting included cautionary advice. Chief among this advice was that risk communicators need to be careful not to exaggerate their messages. First, if risks are seen as widespread and unavoidable, exaggeration might lead to a “what the heck” attitude that encourages risk taking. Second, “choosing very extreme outcomes” backfires in some cases because it discredits the message and/or the messenger. People know that extreme outcomes are rare, and also see such outcomes as easier to avoid via personal competence. Thus, showing less extreme outcomes (e.g., mild symptoms of STDs) can be more effective. Finally, because everyone is “told a lot of exaggerated risk messages,” when people find out actual risk levels “they feel very lied to.” The result can be something of a boomerang effect, so that the more the original risk was exaggerated, the more risk-seeking behavior may occur when people find out the actual risk level.

Our society engenders “risk fatigue”
Panelists talked about the fact that focusing on risks may not be the best way to get people to change their behaviors. For one thing, “everybody is competing in risk,” thus “we tend to discount [risk messages].” In addition, nonrisk messages are more effective at changing some behaviors. For example, panelists talked about how cosmetic arguments are often more persuasive than health arguments in getting people to quit smoking, and how social norms (i.e., the desire to “do what everyone else is doing”) can be another powerful “nonrisk” motivator.

It is hard to reach the remaining non-wearers
Finally, panelists talked about how difficult it is to reach the last portion of the population that is not engaging in any particular safety behavior. The panelists believed NHTSA has already done an excellent job promoting belt wearing (i.e., it is a common behavior); however, this means that the people who still do not wear belts are extremely difficult to reach (panelists referred to them as the “high-hanging fruit”). Panelists also emphasized that even if these people can be reached and persuaded to wear their belts, continual reinforcement is required: “Maintenance [of behavior change] is very, very difficult, in part because you really need … [to reinforce] the thing that caused change initially … the people who are successful at maintaining these behaviors somehow construct their lives so that they are constantly reinforced.”

---

¹While panelists represented many different disciplines (e.g., psychology, communication, and sociology), we characterize this as a discussion about risk communication. Risk communication has been described as “any purposeful transfer or exchange of information” about risks among interested parties (Covello, 1988), and risk communication is a broad enough term to encompass all these fields.

²This point recurs later in the report. For example, panelists identified dramatically different approaches to reach people who do not wear safety belts for reasons of repression, denial, rationalization, and fatalism.

³They might wear a safety belt for some other reason, of course. Examples of such “other reasons” are provided in the suggestions portion of this report. Panelists also discussed when a “nonrisk” approach may be advisable.