(CNN) Your good friends may be able to help you get out of a jam. They may be good listeners and they may be good at keeping you company over a meal or drink. But they are decidedly not good at keeping you from getting sick with Covid-19, new research shows.
A study published Thursday found that while people in the friend zone are good for your mental health, when it comes to an infectious disease like Covid, your friends might make you even more vulnerable to it. It's what two scholars who happen to be BFFs found with the five studies they published in the Journal of Experimental Psychology. Hyunjung Crystal Lee and Eline De Vries are assistant and associate professors and marketing specialists who specialize in consumer behavior and business psychology in the Department of Business Administration at Universidad Carlos III de Madrid.
It's long been known that friendship, while psychologically beneficial, can warp a person's perception of risk. Risk perception comes from a person's ability to judge the severity and probability of a negative outcome. Past studies have shown people tend to feel safer when they have a close relationship with someone, and that can lead them to make emotional rather than rational decisions.
The researchers showed this through five different experiments with a wide variety of people throughout the course of the pandemic.
Lee said she and De Vries were interested in the work because as they were living through the pandemic, they started wondering what makes people take risks and what conditions would make people feel vulnerable or invulnerable.
"And then we went down the rabbit hole," De Vries added.
It's what they call the "friend shield effect."
"The idea was that we perceive our friends like a shield. We feel safe when Covid-19 is associated with friendship," De Vries said -- even if we shouldn't.
The first experiment involved junk food. The professors divided up participants into two groups. One was asked to think about a close friend. The other group was asked to think about a distant acquaintance. Both wrote down memories of those people. Then they were given an article that argued eating unhealthy snacks could increase a person's risk to develop severe Covid. The article also mentioned that hand sanitizers and masks were protective.
The groups were then allowed to shop online from a store that offered travel-size hand sanitizer and masks and Cheez-Its and king-size Twix bars and Mars bars. The group that thought about their close friends first were much more likely to buy junk food than protective items, despite the warnings.
A second experiment divided participants into three groups. None had ever had Covid. They then were told to imagine that they had been infected by a friend, an acquaintance or a stranger. Then they were asked how much they would spend on health protection in the next couple of months. Those who imagined they got sick from strangers or people that they weren't close to planned to buy about the same amount. Those who got sick from friends planned to spend half as much. The experiment confirmed that "positive emotions can make people relatively oblivious to risks and likely engage in risky behavior," the study said.
A third experiment involved people who had Covid-19 at one point in the pandemic and knew how they got sick after being exposed to Covid. Those who were exposed by a friend or family member were much less likely to think that they'd get it again when compared to those who got sick after exposure by an acquaintance or stranger.
The fourth study compared how people with a strong sense of boundaries felt about their risk of catching Covid when visiting a favorite burger joint. Those who clearly categorized others into a friend or acquaintance category were less hesitant to go out to eat with a friend rather than an acquaintance. Those with blurrier boundaries -- whether the person was a friend or an acquaintance -- didn't have their choice to dine indoors impacted in this kind of risky situation.
The fifth experiment looked at people's friendships and factored in political ideology. Earlier research has shown that politically conservative people draw sharper distinctions between who is a friend and who is an acquaintance.
In that experiment, people were asked to imagine going to a favorite coffee shop by themselves, with a close friend or an acquaintance. They were asked how crowded they thought the coffee shop would be and how likely they thought they'd get sick after being exposed to someone there. They were also asked how they would describe themselves politically. Conservatives thought the coffee shop would be less crowded and they'd be less likely to get sick if they were going with a friend rather than if they were going with an acquaintance.
"The people who had the clearest boundaries of who is a close friend and who they are distant from show the greatest friend shield effect and feel more invulnerable to Covid," De Vries said.
In total, these studies repeatedly seem to show that people just aren't good at perceiving risks when friends are involved, even if the risk was beyond this person in their social circle. This is what the study called an "irrational potentially dangerous bias," since limited interaction with others is the most protective behavior in a pandemic.
Kaileigh Angela Byrne, who did not work on these studies but has conducted research on risk taking in the pandemic, said these experiments made "really interesting reading" and build on work that shows "when trust is high, risk perception is low."
"Risk seems less threatening when it's associated with something positive, like a friend or friends, so it makes sense that going to a favorite coffee shop with friends, even in the height of a pandemic, would feel okay, even if it really isn't," said Byrne, an assistant professor of psychology at Clemson University.
Byrne's research has also found that people who identify as conservatives have a decreased perceived risk for engaging in social activities during the pandemic. In part, she said, this is because the pandemic was politicized, and their strong sense of boundaries about who is a friend further reduces their perceived risk.
The studies, she said, seemed to create realistic scenarios, and while they are experiments, "there is a fair connection between intention and actual behavior."
Byrne believes the designers of public health campaigns should keep this research in mind. It's good for people's mental health to stay connected with friends, but people should be encouraged to meet in safer spaces such as at a park or some other outdoor venue, she said.
"I think it is certainly possible to maintain social interaction in a pandemic, while still making efforts to reduce the risk of infection," Byrne said.
Some public health guidance encouraged people to limit interaction to close circles of friends, but De Vries and Lee hope their study will inform public health policy going forward. People should be reminded to be careful even with close friends.
"We would like a more holistic response," Lee said. "Risk perception was more neglected in the current pandemic strategy."
"Hopefully, we will never need this information in the future and we won't have another pandemic, but if we do, we should take this into account. Perception matters," Lee added.