What makes people change the way they behave? It’s a question that cuts to the core of human nature, and one that has been approached by researchers from many different fields.

Answering that question through the lens of one of the biggest public health crises of last several decades—smoking—has led to a collaboration between researchers at the Perelman School of Medicine, the Annenberg Public Policy Center (APPC), and Annenberg School for Communication (ASC), who are investigating the effectiveness of anti-smoking ads.

The study was led by Daniel D. Langleben, a psychiatrist in the Center for Studies of Addiction at Penn Medicine, and An-Li Wang, a post-doctoral researcher at the APPC, in collaboration with colleagues from the Department of Psychiatry and the ASC. By applying neuroscience methods to communications concepts, the researchers aimed to increase the scientific rigor of the study of health communications.

Earlier behavioral research on public service ads suggested two key parameters of the ads’ effectiveness: content—the persuasiveness of its core message—and format—the way that message is presented. For their study, the researchers selected 32 anti-smoking TV ads from a collection of hundreds of similar ads that were previously scored by the strength of their content and format.

Content scores were generated by independent raters, who determined an ad’s core arguments and then assessed the ad's strength by surveying smokers. Format scores were determined by looking at visual and audio effects, such as the number of cuts, special effects, sound saturation, and surprise endings in a given ad.

The researchers then recruited 71 smokers who were not trying to quit and showed them a series of anti-smoking ads while scanning their brains in a fMRI machine. These participants were split into two groups. One group watched ads that had exclusively weak content scores, while the other group watched only ads with strong content scores. Each group’s set of 16 ads contained both strong and weak format scores.  

To test whether these ads had an effect on the smoking behavior of participants, the researchers measured cotinine—a metabolite of nicotine—in participants’ urine before the experiment and then again a month later.   

The researchers found that the participants who had reduced their cotinine levels the most were the ones who watched strong content ads. Among brain regions associated with content and format interaction, activation in the dorsomedial prefrontal cortex, a part of the brain associated with planning future behaviors, predicted reduced urine cotinine levels a month later.

“The reason I’m encouraged about this finding is that we’re not the first people to find a connection between medial prefrontal cortex activation and future behavior, but we are the first to find this in the context of real life anti-smoking ads and a reliable biological measure,” Langleben says. “Other groups either did not look at complete ads, or used less reliable measures than cotinine to determine smoking severity."

The researchers say that getting a more rigorous look at what facets of an ad make for better outcomes will spur future research, and, hopefully, make for more persuasive appeals to quit smoking.   

“The bottom line is that content is what makes the difference,” Langleben says, “If you want to invest your ad production money wisely, invest in a good screenwriter.”