For behavioral psychologist Ellen Peters, the COVID-19 outbreak presents a rich opportunity. Back in early February, with the potential extent of the novel coronavirus unclear, the author and researcher sprang into action. Before the virus had even arrived (officially) in the U.S., she had secured funding for a new project to study the public’s response to the outbreak and collected baseline data. These early survey responses now underlie her ongoing, nationwide study of how people perceive the threat of COVID-19 and how, in the face of it, they act.
Today, with the pandemic unfolding, she has found that our hyperconnected digital world allows scientists to study the human response to crises like never before. Rather than collecting data in hindsight, the global community of behavioral researchers can monitor reactions in real time. Cellphone data track our locations, detailed consumer information monitors what we buy, and internet-connected thermometers can pinpoint the spread of fevers in real time — all while social media posts record our opinions and emotions.
Peters is surveying 1,300 people over the course of the pandemic to track how media consumption affects emotional responses and, in turn, decisions about socialization and travel. She’s using a digital tool that was in its infancy during the H1N1 pandemic: Amazon Mechanical Turk, an online marketplace that crowdsources workers for discrete, on-demand tasks like writing code or transcribing audio recordings. In this case, they’re answering questions about how the virus has affected their state of mind. Peters hopes the outcome will inform future practices for how we communicate risk and other critical information during crises.
Currently the director of the University of Oregon’s Center for Science Communication Research, Peters has spent years studying how communication and emotion regarding issues like smoking and cancer influence personal decisions. Her book, Innumeracy in the Wild, which will be published by Oxford University Press this month, explores how people’s misunderstanding of information involving numbers impacts their choices and, ultimately, the outcomes of their lives.
She spoke with Discover about her latest research in the age of COVID-19.
Q: You’ve studied various information efforts like the anti-smoking campaign. What do those show us about using emotion to communicate in times of crisis?
A: We know that placing graphic warning labels on cigarette packets increases the emotional response to smoking and affects risk perception and behavior in exactly the same way we’re seeing with the coronavirus. The use of graphic warning labels suggests that increasing negative emotion can get people to behave in a particular way, but it also tells us that there’s a segment of people who are going to react negatively to it. It turns out that while these labels make a lot of people want to quit, there’s a subset of people — who tend to be more conservative — who react in angrier ways.
I suspect the same thing is happening with the coronavirus. When people see lots of negative news about the virus, some react angrily and say, “You’re just trying to manipulate me.”
Q: What do you see happening during COVID-19?
A: We thought people who are more fearful about the coronavirus would go with their gut when it came to assessing how dangerous it was and would perceive more risk; and, in fact, we see that. We’ve seen the proportion of people who were quite fearful doubling every two weeks.
Q: How is that fear linked to communication around the virus?
A: When you’re having a strong gut reaction, it affects your thinking. You want to know more about it and collect information. You’re going to think more about the negatives. That may relate to how the news media has tended to present COVID-19 in terms of the increasing numbers of infections and deaths — they haven’t talked as much about the proportion of people who had just mild to moderate symptoms or the proportion of people who do not die.
Q: How can we get through to the group that is angry and feels manipulated?
A: First, identify the communication goals: Don’t just give out all the information. Decide what really should be communicated. Reduce the cognitive effort needed to understand the information. Provide numbers to correct misconceptions and provide a more complete perspective on the situation. Also, don’t show the crowded beach in Fort Lauderdale; show people social distancing but still being perfectly happy and social doing it. And let people know that this beast can be stopped.
Q: How has technology affected your ability to research during a pandemic?
A: Fifteen years ago, you probably couldn’t have done this study. You would have been doing it through phone calls, and you know how much we like to answer phone calls and respond to survey questions. Today, using services like Amazon Mechanical Turk, we can do these surveys blindingly fast. We can program questions quickly, get out in the field quickly. People are responsive, and about 80 percent of them will respond when we follow up. That’s amazing. In terms of time and the ability to collect data at specific, spaced-out moments in time, it’s like the difference between traveling across the ocean in a rowboat and flying there in a supersonic jet.
Q: Does taking advantage of today’s lightning-fast, hyperconnected digital world present new logistical or ethical research challenges?
A: There are trade-offs. You’re dealing with people over the internet, and you don’t know how experienced they are as survey takers. Right now, I’ve seen that there are a number of COVID-19 surveys taking place on Mechanical Turk and elsewhere, and it’s possible that our respondents are getting experienced in the kind of questions that I’m asking. I don’t know how much of a difference it makes, though, because I suspect people are talking about this constantly in their personal lives. Every conversation I have is about COVID.
Anything with data raises the potential for ethical issues, but oftentimes data are de-identified. In our data, for example, we do not and cannot know who the individuals are. I suspect — hope — that’s the same for the studies using remote body temperature and cellphone data. If not, you could imagine marketers targeting ads that take advantage of people’s fears and misconceptions.
Editor’s Note: This Q&A has been edited for length and clarity.