One-shot intervention reduces online engagement with distorted content

Abstract

Depression is one of the leading causes of disability worldwide. Individuals with depression often experience unrealistic and overly negative thoughts, i.e. cognitive distortions, that cause maladaptive behaviors and feelings. Now that a majority of the US population uses social media platforms, concerns have been raised that they may serve as a vector for the spread of distorted ideas and thinking amid a global mental health epidemic. Here, we study how individuals ( $n=838$ ) interact with distorted content on social media platforms using a simulated environment similar to Twitter (now X). We find that individuals with higher depression symptoms tend to prefer distorted content more than those with fewer symptoms. However, a simple one-shot intervention can teach individuals to recognize and drastically reduce interactions with distorted content across the entire depression scale. This suggests that distorted thinking on social media may disproportionally affect individuals with depression, but simple awareness training can mitigate this effect. Our findings have important implicasstions for understanding the role of social media in propagating distorted thinking and potential paths to reduce the societal cost of mental health disorders.

Keywords: cognitive behavioral therapy; depression; large language models; mental health; social media.

Fig. 1.

Study design. The top panel depicts the four-step stimuli generation procedure: (i) large language model stimulus generation, (ii) clinical validation, (iii) sentiment analysis, and (iv) final validation. The middle panel depicts the identification and interaction trials in our simulated social media environment. The bottom panel depicts the randomization and experimental procedure. Participants who were assigned to the “Interaction After Training Condition” did the Training-Identification block before the Interaction Block and the individuals in the “Interaction Before the Training Condition” did it in the opposite order. Image in the top panel used from the clipart library: https://clipart-library.com/clipart/1682104.htm.

Fig. 2.

Correlations between the different measures of interest ($n=838$). Variables involving liking and retweeting only use the participants assigned to the interaction before the training condition ($n=418$). Three key observations are (i) Depressive Symptoms are weakly correlated with accuracy, (ii) Twitter score is moderately negatively correlated with accuracy, and (iii) Depressive Symptoms and TUS are moderately positively correlated with a higher ratio of liking and retweeting of distorted content to total content. The number of stars indicate the significance *$P<0.05$, **$P<0.01$, ***$P<0.001$.

Fig. 3.

The impact of cognitive distortion psychoeducation on liking and retweeting distorted and nondistorted content. The thick dark lines depict the mean tendency. The light lines depict separate statements in the experiment. We observe a consistent reduction in the liking and retweeting of tweets with distorted content after the intervention compared to before the intervention.

Fig. 4.

The impact of cognitive distortion psychoeducation across depression severity, and TUS ($n=838$).