As the COVID-19 pandemic tears through almost every country, public health officials and governments continue to implore citizens to take preventative measures in the hopes of “flattening the curve.” Unfortunately, preliminary reports show vast differences in people’s willingness to embrace public health measures: In March, only 50% of individuals reported wearing face masks in public. And recent anti-quarantine protests represent an even more extreme resistance to complying with preventative measures. This pandemic presents a unique, generation-defining challenge — how do we encourage people to adapt and change their usual habits to adhere to preventative measures that can mitigate the spread of COVID-19?

Amidst the growing pandemic, some appeals have capitalized on a classic tried-and-true fear mongering approach. For example, a popular Medium article, which received over 40 million views in just a week after publication, leveraged fear language (“the coronavirus is coming for you”) in an attempt to get people to socially distance. There is good reason for adopting a fear framework, as research shows negative emotions influence attitudes and behaviors by increasing attention and perceptions of risk. In short, fear can be a powerful avenue for stimulating behavior change.

However, messages emphasizing positive emotions are also effective in public health campaigns, serving as a distinct contrast to fear-based appeals. One way to evoke positive emotions is to emphasize behaviors that can help the collective community, and recent research on COVID-19 interventions show that such appeals — for example, help protect others rather than the self — can be successful in garnering behavioral change.

With each new day bringing more infections and deaths from COVID-19, it is vital to understand how different emotional frames drive behavioral compliance. We tackled this question in a new study by developing two messages to motivate participants to self-isolate: One that leveraged threatening language, and another meant to evoke positive emotions through prosocial actions. Our results showed that both messages increased willingness to self-isolate, but not for the same reasons. While fear messages evoked stronger emotional reactions by making them more afraid, the success of the fear message did not depend on how afraid people felt. In other words, the fear mongering message was equally successful regardless of how much negative emotion it evoked. In contrast, the success of the prosocial message was more effective at boosting willingness to self-isolate if it produced a strong, positive emotional response.

If fearful emotions don’t encourage people to self-isolate, why did the fear message work? One possibility is that people’s sensitivity to negative information (e.g., “some may die”) is already quite high. This is an important factor to consider when designing public health messages. It suggests that creating a message with more graphic and emotionally evocative language is unlikely to generate greater behavior change. Conversely, our results suggest that designing an effective prosocial message has the potential to create even greater compliance if it evokes highly positive emotional responses.

The COVID-19 pandemic has taken a toll on everyone. In just the past few months, preliminary reports indicate a sharp rise in mental distress, including anxiety and depression. Even our public communications, such as Twitter messages, express more anxiety and sadness. While appealing to fear to mobilize society during global threats might be tempting, it can also exacerbate anxiety. Our results suggest a promising upside for changing behavior without resorting to fear mongering tactics. Prosocial calls to action not only create more positive emotions, but they also elicit just as much willingness to self-isolate. And who couldn’t use a little more positivity in their lives right now?

Oriel FeldmanHall is an assistant professor of cognitive, linguistic and psychological sciences at Brown University and a member of the Carney Institute Executive Committee.

Joseph Heffner is a Ph.D. candidate in the Department of Cognitive, Linguistic & Psychological Sciences at Brown University.

Marc-Lluis Vives is a postdoctoral researcher in the FeldmanHall Lab at Brown University.