People wrongly believe their friends will protect them from COVID-19

During the two years of the COVID-19 pandemic, many people have become accustomed to spending time with their closest social circle, which may have unintended consequences, said study authors Hyunjung Crystal Lee, PhD, assistant professor of marketing, and Eline De Vries, PhD, associate professor of marketing, at the Universidad Carlos III de Madrid in Spain.

"Friends and family can provide a sense of comfort, but it's irrational and dangerous to believe they will protect you from being infected by COVID-19," Lee said. "This tendency that we call the 'friend-shield effect' could intensify a false sense of safety and contribute to future infections."

De Vries and Lee conducted five online experiments with U.S. residents in a study that was published online in the Journal of Experimental Psychology: Applied.

The experiments found that individuals engaged in fewer health protection behaviors when the COVID-19 infection risk was associated with close friendships, including situations when people thought of a friend while reading COVID-19-related news, believed a friend was the source of a prior COVID-19 infection or noted a friend's presence while dining at an indoor restaurant. Under such circumstances, study participants decided to purchase fewer health protection items, such as masks and hand sanitizers, and perceived less likelihood of infection, even when the infection risk could stem from strangers in crowds.

The friend-shield effect was more prominent among people who identified themselves as conservatives than those who said they were liberals, arguably because conservatives tend to have clearer boundaries between people whom they hold as close friends and those they regard as distant others.

In an experiment with 495 participants, one group was asked to write down memories of a close friend while the other group wrote about a distant acquaintance. All participants then read a news article stating that unhealthy snacks can increase risks of more severe COVID-19 infection symptoms, while the use of hand sanitizers, face masks and disinfecting wipes can reduce the likelihood of infection.

The participants then chose either a junk food item (candy bars or chips) or a health protection product (face mask, hand sanitizer or disinfecting wipes) from an online store. Participants who wrote about a close friend were more likely to pick junk food over a health protection product than those who wrote about a distant acquaintance.

Another experiment divided 262 participants with no history of COVID-19 infection into three groups. They were told to imagine they had been infected with COVID-19 either by a friend, an acquaintance or a stranger. Participants were then asked how much they planned to spend on health protection products over the next two months. Those who imagined infection by a friend planned to spend less than half as much on health protection items ($9.28 average) than those who imagined infection by an acquaintance ($18.84 average) or a stranger ($21.36 average).

The next experiment comprised 109 participants who had been previously infected with COVID-19 and knew the source of their infection. Participants infected by friends or family members were less likely to think they would get infected again than those previously infected by acquaintances or strangers.

The last experiment divided 301 participants into three groups whose members were told to imagine they were going to a coffee shop either alone, with a friend or with an acquaintance. Then they were asked about how crowded they expected the coffee shop to be and about their political orientation. Conservatives expected the coffee shop to be less crowded, and thus estimated a lower likelihood of COVID-19 infection if they were going with a friend rather than an acquaintance or on their own. There was no similar association for liberals.

Based on these findings, the authors stated that COVID-19 public health campaigns should caution against individuals' tendency to engage in less protective behaviors when the infection risk is associated with friends and family, even if only tangentially. Health campaigns also should consider tailoring messages for different audiences, such as conservatives or liberals.

"We think health safety campaigns should make greater efforts to inform the public regarding the friend-shield effect and aim for a more holistic response to future pandemics by taking both physical infection rates and psychological risk perceptions into account," De Vries said.