Why Doomscrolling Is So Addictive

It is past midnight. You opened your phone to check one thing, and now you are reading about a disaster on the other side of the world, then a frightening health story, then an argument in the comments. You feel worse with every swipe, yet your thumb keeps moving. This habit has a name, doomscrolling, and many people recognize it in themselves. Why is something that makes us feel bad so hard to stop?

The research in this article does not include brain scans of people doomscrolling. But studies of how the brain handles reward, uncertainty and threat offer some plausible explanations.

What researchers mean by doomscrolling

In research, doomscrolling is defined as persistent or repetitive engagement with mostly negative, threatening or distressing online information, often with difficulty disengaging. It overlaps with, but is not the same as, problematic social media use, general exposure to negative news, or broader problematic internet use. The key is the combination: bad news, plus a pattern of not being able to put it down.

A brain built to notice surprises

To understand why scrolling can be hard to stop, it helps to understand dopamine. Dopamine neurons in the midbrain do not simply fire when something good happens. Most of them signal a reward prediction error, the gap between what we expected and what we got. A review of this research describes how these neurons are activated by more reward than predicted, stay at baseline for fully predicted rewards, and dip when there is less reward than predicted. Prediction errors like these are crucial for basic forms of learning about rewards.

One interpretation is that a feed is full of small surprises. You never know whether the next post will be boring, funny, shocking or important, so each swipe is a tiny gamble on what comes next. This idea fits what we know about dopamine, but it has not been tested directly with doomscrolling.

There is evidence that reward learning shapes social media behavior more generally. A study that analyzed over one million posts from more than 4,000 people found that their behavior followed the principles of reward learning: people spaced their posts in a way that maximized the average rate of likes they received. An online experiment with 176 participants confirmed that social rewards causally influenced behavior. That study looked at posting and likes, not reading bad news, and the authors note that their data do not speak directly to whether intense social media use is maladaptive or addictive.

Why bad news is so hard to look away from

If scrolling were only about reward, why would people keep reading things that upset them? Part of the answer may be that information itself can feel rewarding. In a classic experiment, monkeys offered water rewards of uncertain size chose to get advance information about how big the reward would be. Strikingly, the same dopamine neurons that signaled the expected amount of water also signaled the expectation of information. The study involved rewards, not threats, so it cannot show that the same system drives checking bad news. Still, this suggests a possibility: when the world feels uncertain, the urge to find out what happens next may pull us forward even when the answers are unpleasant.

Negativity also grabs attention. In a large analysis of headline tests on a news website, covering about 105,000 versions of stories and roughly 370 million impressions, negative words made people more likely to click. For a headline of average length, each additional negative word increased the click-through rate by 2.3%. Researchers studying doomscrolling have proposed that threat monitoring and intolerance of uncertainty may connect repetitive checking of negative information with anxiety. Put simply, a brain that evolved to watch for danger may struggle to stop scanning a feed that keeps offering new things to worry about.

What it may cost

Long exposure to distressing news can affect how people feel. After the 2013 Boston Marathon bombings, researchers surveyed 4,675 adults from representative samples in Boston, New York City and the rest of the United States. After adjusting for people’s mental health before the attack, six or more hours a day of bombing-related media in the following week was associated with higher acute stress than being directly present at the bombings. The authors suggested that repeatedly engaging with trauma-related media for several hours a day may prolong acute stress. That study measured bombing-related coverage across television, radio, print, online and social media, not phone scrolling specifically.

A 2026 meta-analysis of doomscrolling studies found it was associated with anxiety, depression, psychological distress, lower well-being and lower life satisfaction. But the authors rated the certainty of the evidence as very low for all five outcomes, and noted that the studies cannot determine whether doomscrolling comes before poorer mental health, follows existing distress, or reflects shared vulnerabilities. In other words, anxious people may doomscroll more, doomscrolling may make people more anxious, or both.

Is it really an addiction?

The word “addictive” is common in everyday speech, but many researchers are cautious about it. A review of research on smartphone addiction found that, although much of the field treats it as real, there was not enough support from the addiction perspective to confirm that smartphone addiction exists. The authors argued the behaviors were better described as problematic or maladaptive use, with consequences that do not reach the severity of addiction. Another group of researchers warned that some studies risk identifying an unlimited list of “new” behavioral addictions and called for research that focuses on the psychological processes involved rather than simply checking off criteria.

The dopamine comparison has limits too. The dopamine review describes how drugs of addiction generate, hijack and amplify the dopamine reward signal. None of the studies discussed here shows that scrolling does the same. Using a reward system is not the same thing as being addicted. At the same time, heavy social media engagement has been likened to an addiction in parts of the research literature, so the debate is partly about evidence and partly about which words best fit it.

What we still do not know

Most doomscrolling studies are surveys that capture a single moment, and the meta-analysis found that results varied widely between studies and were concentrated in a few countries. The brain explanations in this article, involving prediction errors, information seeking and threat monitoring, are reasonable hypotheses drawn from wider neuroscience, not findings from scans of people doomscrolling.

The takeaway

Doomscrolling may be hard to stop because it taps into several powerful brain tendencies at once: learning from unpredictable rewards, a drive to resolve uncertainty, and an alarm system tuned to notice threats. That does not make it an addiction in the clinical sense, and the evidence that it harms mental health is suggestive rather than settled. Noticing the pattern, and how you feel afterward, may be a useful first step toward deciding when to put the phone down.