Individuals struggling with mental wellness often inadvertently curate an environment detrimental to their recovery by exposing themselves to unfavorable online content, according to a series of studies conducted by researchers at the Massachusetts Institute of Technology.
The team responsible for the analysis has created a tool designed to help individuals safeguard their mental wellness by making more informed choices about the content they consume.
The research findings were previously published in a publicly available paper by Dr. [Name], an adjunct professor of cognitive neurosciences at MIT and Professor at the College School London, along with co-author Christopher A. A former visiting PhD scholar from Sharot’s Affective Mind Lab, now a postdoctoral researcher at Stanford University’s Institute for Human-Centered Artificial Intelligence. The findings had been
“Our findings demonstrate a reciprocal, causally linked association between online behavior and overall wellbeing.” Individuals exhibiting initial psychological wellbeing symptoms are more likely to engage in online browsing, which often leads them to encounter unfavorable or fear-inducing content, notes Sharot. As symptoms worsen with continued consumption of this content, It’s a suggestions loop.”
The research examined the online searching habits of over 1,000 participants by employing natural language processing to assign an unfavorable rating and a favorable rating to each webpage visited, along with scores for anger, fear, anticipation, trust, surprise, disappointment, pleasure, and disgust. Participants also completed self-assessment questionnaires to gauge their mental wellbeing and recorded their emotional state immediately prior to and following online browsing sessions. Researchers found that people experienced a boost in mood following exposure to less negative online content, while those already in a poor state of mind were more likely to seek out similarly pessimistic digital environments.
Subsequently, participants were asked to absorb information from two websites selected at random from either six unfavorable websites or six neutral pages. Prior to reviewing the pages, they reported their emotional states, and subsequently, they revealed their temper ranges both before and after perusing the content. A study found that people exposed to unfavorable websites reported being in a poorer mood afterwards, and then proceeded to visit even more negative sites when asked to surf the internet for 10 minutes.
The findings fuel ongoing discussions about the intricate link between psychological wellbeing and online behavior, according to the researchers. Despite prevailing research focusing on the quantity of social media usage – namely, display time and frequency – a mixed bag of findings has resulted. By focusing on the type of content consumed, researchers find a causal and reciprocal link between its affective properties and both psychological wellbeing and emotional state.
To investigate whether interventions could influence users’ web-surfing choices and boost their mood, researchers presented participants with search engine results pages displaying three search outcome options for each query. Participants were assigned labels for each search result, ranging from “feel better” to “feel worse”, while others did not receive any labeling. Individuals provided with labels were significantly more likely to opt for unfavorable content and more inclined to select constructive content. A subsequent study found that individuals who viewed additional positive content experienced a significantly elevated mood.
Based largely on these findings, Sharot and Kelly devised the “Digital Weight Loss” framework, categorizing Google search results into three classes: emotion (whether users generally perceive content as positive or negative), information (how well webpage data helps users understand a topic, on average), and actionability (the extent to which webpage data is useful, on average). Jonatan Fontanez, an MIT electrical engineering and computer science graduate student ’24 and former undergraduate researcher in Professor Sharot’s lab at MIT, also collaborated on the development of the device. The device has been publicly unveiled this week, concurrent with the release of a supporting research paper.
According to Kelly, individuals with poorer mental wellness are more likely to seek out and engage with disturbing and anxiety-provoking content, thereby perpetuating a self-reinforcing cycle of declining mental health. “Our ultimate goal is to empower users by providing a tool that fosters greater control over their mental inputs, ultimately breaking the cycle of negative patterns.”
