Facebook Deactivation and the Nocebo Effect

Source: Alexas_Fotos/Pixabay

In recent years, scientists have been paying a lot of attention to how social media platforms (e.g., Facebook, Twitter) may be associated with negative outcomes, including political polarization and psychological distress. There exists a popular viewpoint that social media apps are very bad for mental health and for societal harmony. As I’ve written before, I think this viewpoint is unfounded and misguided.

An influential paper in this area of research ostensibly shows evidence for those claims based on data sampled just prior to the 2018 American midterm elections. It reports results from an experiment in which the researchers paid some people to deactivate Facebook for about a month. The researchers compared this group on several key outcome variables (such as happiness) to a control group that used Facebook normally. This paper has been cited 392 times (as of June 2022) and is often mentioned in conversations among nonacademic journalists, pundits, and commentators about the supposed damages that social media apps do to our culture. But most of the commentary is based on misunderstandings about what the data actually reveal.

Myth #1: The treatment group (those who deactivated Facebook) was less politically polarized than the control group (those who used Facebook normally).

The study authors used several different measures of polarization. This is important, as there is a big difference between cognitive polarization and emotional polarization. Cognitive polarization is the extent to which people disagree on specific issues (e.g., abortion policy) whereas emotional polarization is how much we dislike each other (i.e., how much Democratic voters say that they hate Republicans, and vice versa). Social media apps have long been blamed for increasing the second type of polarization (emotional) by pushing people’s buttons and making them more “outraged” at their political rivals.

But in the experiment, the treatment group was not significantly different from the control group in terms of emotional polarization. Strike one against the claim that if we only turned off social media, we’d hate each other less.

However, the researchers did find that there was a decrease in the other type of polarization (cognitive)—meaning that there was less disagreement on specific issues when people stopped using Facebook. We’ll return to this point later.

Myth #2: There was a difference in the objective amount of time spent on offline social activities for the treatment group compared to the control group.

A popular notion is that social media apps have a "technoference" effect on our social lives. That is, apps like Facebook make people unhappy because they cause us to spend less time with others (particularly our loved ones) in person. The study authors reported that participants in the treatment group who weren’t using Facebook spent more time doing social activities (spending time with family/friends) compared to the control group.

The problem with this interpretation is that most people are notoriously bad at estimating how much time they spend doing various activities, and this is especially true for online activities like social media use. Some studies show that less than one-third of participants accurately reported their screen time. We know this based on research that objectively measures social media use with smartphone apps that keep track of how much time people spend on those apps. This discrepancy between actual and perceived usage is a big problem if we want to know whether social media use actually displaces the time we would normally spend with other people offline.

Given this, I’m just a wee bit skeptical that the study authors’ data actually show an objective increase in time spent doing offline activities for the treatment group. It would be more accurate to conclude that participants felt as if they were spending more time doing offline social activities. The study authors used an additional measure of social interaction, in which they asked people to list the number of people they saw in person and the number of offline activities they engaged in. Those variables were not significantly different for the treatment and the control groups. Only perception of time spent was different.

Myth #3: The study shows evidence that social media apps are addictive.

The study authors report a finding that, on average, the treatment group (who deactivated Facebook) was happier compared to the control group. I have no qualms with this. Subjective well-being was different across the two conditions. The authors also reported that it was challenging for participants in the treatment group to maintain their deactivation after the study ended, even though they thought doing so would be good for their mental health. After just one week, about 90 percent of the people in the treatment group resumed their Facebook use. The authors noted that because people from the treatment group reverted back to using a product that they were aware made them feel less happy, this suggests an addictive quality to these apps.

This part irks me the most because this is not even close to how addiction actually works. Addiction is a highly complex phenomenon that stems from genetic predispositions, the physiological and chemical nature of substances (often drugs), interpersonal factors such as neglect or abuse, and environmental factors such as economic deprivation or adversity.

Placebo Essential Reads

When the Story Is the Cure

Placebo Effects and Direct-to-Consumer Advertising

The study authors also found that those in the treatment group were significantly less knowledgeable about current events and the news. This could easily explain the boost in happiness for those who stopped using Facebook, and it could also explain the decrease in issue polarization. When participants weren’t paying attention to what was happening around them, they felt better, and they had fewer topics to disagree about with other people. It’s very strange to blame social media apps for having a detrimental effect on people’s mental health when what’s really causing people to feel distressed is all of the terrible things happening in the world (e.g., climate change, prejudice, violence).

Let’s imagine that I took a large group of high-school students who are in the habit of doing lots of stressful or boring homework assignments for hours every single day, and I paid half of them (the treatment group) to stop doing homework entirely for one month. I could document how these students felt significantly happier compared to the other half of students (control group) who kept doing homework normally. Then, after the study period ended, the students in the “homework deactivation group” reverted back to doing about as much homework as before the study began, even though they admit that doing less (or no) homework would make them feel happier. As a researcher, should I then conclude that homework is “addictive”? Of course not. That would be a ridiculous claim to make. Human behavior is much more complicated.

Sometimes we do things that make us feel happy even though those things are bad for us in other ways. And sometimes we do things that we don’t enjoy because they may be good for us in other ways. And in many cases (as with homework), students have no choice but to obey, lest they face punishment.

And then we have social media, which is likely in a unique category of nonaddictive activities that we engage in to connect with others virtually, and we feel lots of different kinds of emotions in the process. But our beliefs about social media may end up influencing how we feel about them. If we believe that we’re consuming an unhealthy product every day, and then this unhealthy product is removed from our daily lives, we may feel better due to the power of suggestion. Scientists call this a “nocebo effect,” which is like the inverse of a placebo effect.

The nocebo effect is based on the power of suggestion, and it's particularly strong for electronics and technology. This was popularized on the show "Better Call Saul," in which the character Chuck claims to have “electromagnetic sensitivity,” which causes allergic reactions to things like light bulbs or batteries. But scientists have shown that these allergic symptoms aren’t actually caused by the electricity. They’re caused by belief that electricity is unnatural and unhealthy. The same thing may be happening when people quit using Facebook (or other social media apps) and report feeling healthier. It’s basically a nocebo effect.

In conclusion, the study authors of the 2020 paper reported an ambitious and in some ways very valuable experiment. It’s unfortunate that their work has been misinterpreted and misunderstood. Their study does not conclusively show that social media is addictive, nor that quitting it makes people objectively more social or reduces emotional polarization. It may be that people feel happier when they stop using apps like Facebook, but this can be explained by the power of suggestion. Moreover, the mainstream discourse around these topics has gone off the rails and is riddled with factual inaccuracies, misperceptions about the research, and, in some cases, a fundamental lack of basic common sense and logic about psychological processes. Let’s try to, as the kids say, do better.