Johan Bollen
A multidisciplinary study conducted by researchers from the Luddy School of Informatics, Computing, and Engineering, the Department of Psychological and Brain Sciences, and Wageningen University in The Netherlands, suggests that people suffering from depression may benefit from avoiding social media in the evening because it may cause a disruption in their circadian rhythm.
The study, “Depression alters the circadian pattern of online activity,” which was published by the journal Scientific Reports, used a large sample of Twitter users who had received a clinical diagnosis of depression and tracked the times in which they did and didn’t tweet, and what they tweeted about. Using the data, the researchers obtained detailed information about their sleeping and waking patterns, including the topics they tweeted about at times that other people might be asleep, providing insight into what users were thinking about at that time of day and night.
“This article is the outcome of a larger effort to investigate the usefulness of social media data as a source of real-time, societally relevant information to model the complex dynamics of mental health,” said Johan Bollen, a professor of informatics and computing at Luddy. “During the COVID-19 pandemic, there is widespread concern about depression due to lockdowns and other social distancing measures. We already know that sleep and depression are closely linked, and that depressed individuals experience changes in their circadian rhythms.”
Researchers expected to observe a shift in activity between the individuals with depression and a random sampling of Twitter users that showed the first group to be going to bed later and waking up later in the day. Instead, the study found that sleeping and waking cycles weren’t shifted in time but that activity levels changed at specific times of the day.
“We found that the difference in activity at those times is directly linked to higher levels of emotionality and rumination,” said Marijn ten Thij, Bollen’s post-doc and first author of the study. “That was a telling observation because it provided insight into when depressed individuals were up at night and what may be keeping them up, potentially pointing towards approaches or highly targeted interventions to help people get more and better sleep.”
The results suggest that diagnosis and treatment of depression may focus on modifying the timing of activity, reducing rumination, and decreasing social media use at specific hours of the day.
“For instance, people who are depressed could try to minimize their time on social media in the evening and avoid worrying about things that can’t be fixed in the middle of the night,” Bollen said. “That’s more difficult than it sounds, but our research indicates it could make a big difference.”
The study also highlights the benefits of using social media data to study mental health. The researchers plan to further compare cognitive differences in individuals with depression using online data based on clinical experiences using cognitive behavioral therapy, one of the most commonly used methods of treating depression. These comparisons can be conducted for a wide range of mental health issues in real-time on social media and lead to highly targeted interventions to improve well-being.
“This type of research really showcases the power of interpreting data to improve lives,” said Kay Connelly, the associate dean of research at the Luddy School. “It’s the backbone of all the work that we do at our school, and Johan and his colleagues are making discoveries that can make a real impact in the day-to-day lives of people.”