By Emily Reynolds
An important facet of cognitive behavioral therapy is the questioning of “cognitive distortions”, imprecise thought patterns that often affect people with depression. Such biases can include jumping to conclusions, disaster, black-and-white thinking, or self-blame – and seriously distressing those affected.
But how do we track cognitive biases in people with depression outside of self-reporting? A new study published in Nature human behaviorresearches cognitive biases online and finds that people with depression are more biased in the language they use on social media.
Indiana University Bloomington’s Krishna Bathina and colleagues examined the language of over 6 million tweets from 7,349 Twitter accounts, some of which had previously tweeted that they were diagnosed with depression and some that were chosen at random. The researchers were specifically interested in how often these tweets contained 241 phrases that they considered to be the “building blocks” of various cognitive biases associated with depression. For example, the phrase “everyone believes” has been understood as part of the “mind-reading distortion” where people think they know what others are thinking. Other cognitive biases are disasters, over-generalization, and the devaluation of positive experiences.
The results showed that those who tweeted about a diagnosis of depression used significantly more cognitive bias phrases than those in the random state. This was true for almost all 12 types of cognitive biases, with the exception of divination (predicting a negative outcome), mind reading, and disasters, which were not significantly different between the two groups. The types of biases that were most common in the depression cohort compared to the control group were personalization (taking things personally) and emotional thinking (mistaking a feeling for a fact).
This could mean that depression can be tracked through language – especially online, where people are more open to what they think and feel. The team also suggests that the results could have an impact on the way cognitive behavioral therapy is actually performed: How certain types of language are used can reflect certain cognitive biases, and thus better insight for targeted, relevant treatment give.
Although all data has been anonymized, the team admits that wiping mental health data off social platform users poses difficult ethical problems. Some users have expressed serious discomfort about the fact that their personal information is being used without consent. The team suggests that in the future, “automated interventions” for depression could target people who use such language on social media. However, with so many people using social platforms to express themselves and find a space and community to support them, such interventions may be undesirable.
There are also questions about self-disclosure by those with depression. First, there was no way for researchers to verify the diagnoses. second, there may have been many in the random state with diagnoses they simply never thought of or wanted to share. Although funny tweets were filtered out (e.g. “This Game of Thrones episode gave me a diagnosis of depression”), the language of insanity is often co-opted or exaggerated – for example after political defeat. A look at the subtlety of online communication can also provide different and interesting answers.
– People with depression express more distorted thinking on social media
Emily Reynolds is an associate at BPS Biomedarticles