If these images make you squirm or want to click away, you’re not alone.

How, then, can this type of message change the choices you make? Can we really be motivated by something that turns us off, rather than on?

You’d think, perhaps intuitively, that the scarier the ad, the more powerfully it affects our behavior. And the research supports that argument. Indeed, since the classic 1964 Surgeon General report on “Smoking and Health” came out 50 years ago this month, that’s been the basic strategy for health communication around the issue. But there’s a catch. A BIG one.

While we’ve seen a significant drop in global smoking rates (down 25% for men and 42% for women) since those landmark reports in the 1960s demonstrated the link between smoking and lung cancer, many people continue to smoke: 31% of men and 6% of women. In the U.S., 18% of adults (down by half since 1964) continue to do something they know might kill them.

Public health agencies have spent years communicating the dangers of smoking. Their anti-smoking ads have grown increasingly disturbing, threatening us with graphic images of bulging tumors and holes in our throats — possibly to try to reach that last stubborn segment of the population that hasn’t kicked the habit.

Why aren’t these ads working?

Turns out, the most recent and comprehensive research on so-called “fear appeals” and attitude change says that this kind of messaging does work, but only if the person watching the ad is confident that they are capable of making a change, such as quitting smoking. Public health gurus call this confidence in one’s ability to make a change “self-efficacy” — and threats only seem to work when efficacy is high. (The reverse is also true.)

If someone lacks efficacy, ads with fear appeals don’t help. In fact, they make the behaviorworse. How? Many people engage in unhealthy behavior because it makes them feel better and relieves their anxiety.

If you threaten someone who has little to no confidence they can change their behavior, their anxiety goes through the roof. What do they do? Perhaps turn off the threatening ad, walk away, and light up a cigarette — the very behavior you were trying to prevent. This same principle applies to other coping behaviors, such as eating unhealthy types of food or just too much of it.

Unfortunately, anxiety is quite common in this country. According to arecent Atlantic article, 1 in 4 Americans is likely to suffer from anxiety at some point in life. Making big life changes is tough, and it seems as though fear and anxiety don’t energize people, they just paralyze them.

So what’s the solution?

A step in the right direction would be for ad campaigns to couple compelling threats with equally clear and specific paths to behavior change. Or why not apply the rewards built into reaching a new level in addictive video games to apps that people can use for real-life challenges? One great example of this is Superbetter, a social online game to help people build resilience and stay motivated while working to overcome injuries, anxiety, and depression.

Stand-alone threats implicitly assume that people don’t already know how bad their choices are, and can drive them to the very behaviors they wish they could change. Truly effective ad campaigns might still appeal to our fears, but they should also let us wash it all down with a confidence chaser that empowers the more anxious among us to act on our fears.