The real story behind the research on ‘Twitter psychosis’

BY MIRI MOGILEVSKY

The perils and pitfalls of social media—getting fired from your job, having bad self-esteem because of all your friends’ selfies, fear of missing out—are probably familiar to most of us. According to a group of researchers, there may be another one to add to the list: Twitter-induced psychosis. According to a research article called ?Twitter Psychosis: A Rare Variation or a Distinct Syndrome?”, published this month in The Journal of Nervous and Mental Disease, a patient named “Mrs. C” developed psychotic symptoms in connection with her use of Twitter:

Prior to suffering the mental breakdown that caused her to be committed, Mrs. C never exhibited signs of a psychosis or any forms of personality disorder. However, according to her friends and family members, about a year prior, she had become obsessed with Twitter. ‘Sometimes, she would spend several hours a day reading and writing messages, neglecting her social relationships and, sometimes, even meals and regular sleeping hours,’ the study noted.

Mrs. C would spend hours a day on Twitter and was convinced that a famous actor was communicating with her via coded messages on the website. Eventually, she began to believe that other users were communicating with her in this way, too, and then that some sort of sect was trying to send her messages offline as well. Such delusions are typical of psychosis, for which Mrs. C was eventually admitted to a hospital. She made a full recovery and apparently lost interest in Twitter afterwards.

The researchers suggest that certain aspects of the Twitter experience may have triggered these symptoms:

The authors believe that the amount of symbolic language (caused by the limitation of 140 characters per Twitter message), the automated spam responses with seemingly related content, and the general interactive features of Twitter might combine several aspects that could induce or further aggravate psychosis.

The etiology of psychotic disorders such as schizophrenia is complicated. (While Mrs. C was not diagnosed with schizophrenia, as she probably did not meet the full criteria, her symptoms are typical of a person with that diagnosis.) There’s a substantial genetic component, but environmental factors matter, too. Living in urban environments has been shown to increase risk for schizophrenia, even when other relevant variables are controlled for.

For people who already have several risk factors, the type of communication that the authors attribute to Twitter may indeed be a trigger, but it’s extremely unlikely that such communication could—on its own—cause psychosis in someone who is not already strongly predisposed to it. By the way, the fact that Mrs. C had reached almost-middle age without having previously experienced psychotic symptoms is not unusual. In women, the average age of onset for schizophrenia (and, presumably, similar disorders) is later than it is for men—ages 25 – 34, compared to 15-24 for men. Mrs. C was 31 years old when her symptoms started.

The form that psychotic symptoms take can be influenced by environmental factors as well. For example, delusions involving religious themes, such as demonic possession or being spoken to by a deity, may be more common in cultures where religiosity is high. (They’re also probably unlikely to be experienced by a person who is not religious.) Paranoid delusions like Mrs. C’s were virtually unrecorded before the 20th century, and delusions involving technology that transmits signals—which are very common among people with psychosis—are clearly unique to societies and historical periods that have such technology. The content of delusions has to come from somewhere.

So, it’s not very surprising to me that, in a world where Twitter is well-known and ubiquitous, Twitter might play a role in someone’s delusions. However, it doesn’t mean there’s some sort of special type of mental illness that involves delusions related to Twitter. It just means that delusions change with the times.

In fact, the authors of the study seem to agree. On a blog post criticizing the research and the media response to it, one of them left a comment including the following caveat:

We did indeed use a provocative title in order to open a discussion on delusion phenomenology. Having criticized the explosion of diagnoses in DSM-5 ourselves, we certainly do not claim that syndromes constitute disease entities.

While “provocative titles” are useful for getting attention, they may not be so useful for spreading accurate, reasonable information about scientific research.

Unlike most other published psychological research, the study about Mrs. C and “Twitter psychosis” is a case study— a type of research in which researchers study one particular person, or case. Something you should know about case studies is that they’re the least scientifically rigorous experimental design possible. There’s obviously only one subject or participant, and a particular person’s psychology is so idiosyncratic and impacted by so many factors that we may or may not even notice that it’s difficult to draw any firm conclusions. Unlike other studies, that compare some group to some other group, case studies don’t allow us to see what happens if certain conditions are different.

This study was further an observational case study, not an experimental one. In experiments, researchers change something or do something to the participants and see what happens. In observational studies, they can only observe what’s already going on. This means that it’s impossible to tell what causes the observed phenomena to occur.

That said, case studies are useful sometimes. When researchers are first discovering a new phenomenon, or when people with a particular condition are very rare, there might be no choice but to study a single individual. Observational studies in particular are useful when it’s unethical or impossible to tweak some variables to see what happens. Twitter psychosis, if it’s a real thing, is probably quite rare. We would have to study thousands of participants to find cases of it. And if Twitter really can cause psychosis in certain people, it’s clearly unethical to purposefully expose them to it to see what happens. So, case studies, including observational ones, are often the first step of studying something new.

My main concern with this type of research—and with other recent warnings by mental health professionals that the Internet (and social media in particular) can cause or aggravate mental illnesses—is that people dealing with mental health problems may be pressured by friends, family, or doctors to stay offline. Of course, sometimes staying off the Internet (or off social media specifically) can be a wise choice for someone for any number of reasons. However, the general trend of anti-tech alarmism makes it likely that “stay off the internet” will be a piece of advice too often and too easily given.

People with mental illnesses can be vulnerable to persuasion and even coercion by those with authority over them, including therapists and psychiatrists. If a person with a Ph.D. says, “I think you need to stay off Twitter,” they may take their advice without any grains of salt.

You might ask why this matters. It matters because the Internet can also be an incredible source of support and information for people with mental illnesses. Tumblr, in particular, is known for its supportive community, but it’s not the only one. Reddit has subreddits dedicated to every major mental illness where users can post stories, ask for advice, and support each other. Twitter’s hashtags make it easy to find tweets about your illness, and mental health organizations and professionals are very active there, posting supportive messages, advice, and news about clinical research.

And Facebook is where many people “come out” about their mental illnesses for the first time, finding it easier to share with many people at once rather than with individuals—but without having to show it to the whole world. (Incidentally, Facebook is also where I run a support group for atheists dealing with mental health problems, which many of the participants have told me has been really helpful.)

It’s possible that Twitter can trigger psychosis in some people with other risk factors, and researchers should conduct more studies to find how whether, how, and why this happens, and how it can be prevented. But we should be careful not to cut suffering people off from a potentially vital source of support.

Miri Mogilevsky is a social work student who loves feminism, politics, New York City, and asking people about their feelings. She writes a blog called Brute Reason, tweets @sondosia, and rants on Tumblr.

Photo via jamesfischer/Flickr (CC BY 2.0)