The cure for doomscrolling is not what you think
Dear Reader,
There is a specific behavior that most people I know have engaged in over the last decade, and that most of them are embarrassed about. It is called doomscrolling. Sitting late at night with a phone, scrolling through news, or social media feeds, or reports of ongoing disasters, unable to stop, feeling worse with each item, and still unable to stop.
The word was coined around 2020, when the pandemic and the political situation of that year made the behavior visible in a specific way. But the phenomenon predates the word. And in the years since, it has not diminished. If anything, it has become more integrated into the texture of daily life. The specific pattern of consuming bad news past the point of usefulness is now baseline behavior for a substantial portion of the population.
I want to write about what doomscrolling actually is, why the usual advice for stopping it does not work, and what actually helps. This is a piece for people who have tried the standard techniques and found that they did not touch the problem. If you have never doomscrolled, this piece will probably not be useful to you. If you have doomscrolled and been ashamed of it, I want to offer a different framing.
Start with why doomscrolling exists. It is not a modern phenomenon in any deep sense. It is an ancient behavior pattern being triggered by modern technology.
Human brains evolved with a specific bias toward negative information. In the environment we evolved in, negative information was much more actionable than positive information. A rustle in the grass that might be a predator required immediate attention. A ripe fruit that had been there yesterday could wait. The specific asymmetry of survival stakes meant that brains that paid more attention to threats had better survival outcomes than brains that did not. Over evolutionary time, this produced what psychologists now call negativity bias, a well-documented tendency for negative information to be attended to more, remembered better, and given more weight than positive information of equivalent objective importance.
Negativity bias was adaptive when information was scarce and locally relevant. If you noticed a threat, you could respond to it. The attention paid off in survival. In a world where you might encounter a few threats a day, all of them potentially relevant to your immediate situation, negativity bias was a genuine advantage.
The modern information environment inverts this. Threats are not scarce. They are streaming continuously from every direction. Most of them are not locally relevant. Most of them cannot be responded to. And yet the negativity bias is still active. Your brain is treating every alarming news story as a threat that requires attention, because that is what your brain was built to do. It cannot know that the story is not local, that you cannot respond, and that continuing to consume such stories will not help. It just knows that something alarming is happening, and that attention is required.
This is the specific vulnerability that doomscrolling exploits. It is not a moral failing to have negativity bias. It is a feature of the human brain that used to be adaptive and is now being weaponized against you.
The specific mechanism by which doomscrolling gets going is worth understanding.
The first step is exposure. Something alarming comes across your feed. This activates your threat-monitoring system, which orients your attention to the alarming thing. So far, this is what your brain evolved to do.
The second step is search. Because your brain has been alerted to a threat, it wants more information. It wants to know if the threat is worse than initially indicated. It wants to know if the threat is coming closer. It wants to know if there are related threats. So you seek more information, either by scrolling further or by clicking through to related content. This is also what your brain evolved to do.
The third step is where the pattern gets pathological. The specific information environment you are searching in has been optimized to hold your attention, using the mechanisms I wrote about in previous pieces. So the search does not terminate when you have enough information. It continues indefinitely, feeding your threat-monitoring system a stream of alarming content that keeps activating the search behavior. The behavior is self-perpetuating because the environment is designed to sustain it.
The fourth step is the specific quality of what you are consuming. Modern news, particularly online news, is optimized for engagement. Which means it is optimized to activate emotional responses, particularly negative ones. The specific stories that go viral are the ones that produce the strongest emotional reactions, and the strongest reactions are usually to fear, outrage, and despair. So the content you are consuming during a doomscrolling session is disproportionately weighted toward the most emotionally activating material available, which is the material most likely to sustain your search behavior.
The result is a specific pattern that has all the hallmarks of a compulsion. It is triggered by exposure to alarming content. It sustains itself through search behavior. It escalates as the search finds increasingly emotionally activating material. It resolves only when the person is exhausted, or the phone is put away, or the specific external interruption occurs. And it typically leaves the person feeling worse than when they started, with no meaningful information gained.
Let me tell you about the specific version of this I have struggled with.
I run a company, I run a law practice, and I write publicly under my own name. Which means I am unusually exposed to news, both because I need to be informed for professional reasons and because I have a specific interest in the direction of the industries I work in.
For years, I checked the news several times a day. I told myself this was informational. I needed to know what was happening. What I actually did was open news apps, scroll through headlines, click on the most alarming ones, read them without absorbing much, feel slightly worse, and then check something else. The specific texture of the behavior was not learning. It was consumption. The specific outcome was not information. It was low-grade anxiety.
I noticed the pattern most clearly during periods of large events. Major political developments. Major geopolitical crises. Major market movements. During these periods, my checking behavior would escalate. I would check every hour, or every half hour, or every fifteen minutes. Each check was small, but the cumulative effect was substantial. And the state I was in during these periods was measurably worse than the state I was in during less eventful periods.
The interesting thing is that my actual actions during these periods were not different. I was not making better decisions because of the frequent checking. I was not more prepared for events. I was not more able to respond. The information was accumulating in me, but it was accumulating as anxiety, not as usable knowledge. And it was displacing other things I would rather have been doing.
At some point I realized that I was not consuming the news to make better decisions. I was consuming it because my nervous system had been trained to want to. The behavior was serving the platforms, not me. And once I could see this clearly, the specific question changed from how to make my news consumption more effective to how to reduce the behavior itself.
The standard advice for stopping doomscrolling does not work well, and it is worth understanding why.
The most common advice is willpower. Just decide not to do it. Notice when you are doing it. Put the phone down. Choose better activities. This advice fails because it treats the behavior as if it were the result of a choice, when in fact it is largely automatic. By the time you notice you have been doomscrolling for twenty minutes, the pattern is already well underway, and the specific state you are in makes it hard to interrupt.
The second most common advice is media diet. Consume less news. Follow fewer alarming accounts. Curate your feed to include more positive content. This advice fails because it does not address the underlying pattern. If you remove some negative content, your negativity bias will attach to whatever negative content remains. If you add positive content, your negativity bias will filter past it to whatever negative content is nearby. The bias is doing the work, not the specific content.
The third most common advice is time-boxing. Only check news at specific times. Set limits on how long you spend. Use app timers to enforce boundaries. This advice works better than the others but is limited by the fact that timers only apply if you are aware of them and choose to comply. When you are in the middle of a doomscroll, the specific mental state you are in is not one that respects your earlier decisions about what you would do.
None of these approaches address what actually maintains the behavior, which is the interaction between your evolutionary hardware and an environment that is exploiting it. You cannot willpower your way out of an evolutionary hardware issue. You can only change your relationship with the environment that is doing the exploiting.
The specific intervention that has worked for me, and that seems to work for others when they try it, is stranger than the usual advice.
The intervention is to notice, and honor, the underlying need that doomscrolling is a corrupted expression of. Which is the need to know that you are safe.
Doomscrolling is not really about information. It is about threat-monitoring. And threat-monitoring is not really about learning what threats exist. It is about resolving the question of whether you are safe right now. When you doomscroll, your nervous system is asking, over and over, "am I safe? am I safe? am I safe?" The answer it is getting from the content stream is "not sure, keep checking." The behavior continues because the question never gets resolved.
What resolves the question is not more information. It is a somatic confirmation of safety. Your body knowing, at a level below cognition, that you are okay right now. Which requires attention on the body, not on the phone.
The specific practice that produces this is nervous-system regulation. Slow breathing. Grounding attention in the physical present. Noticing the immediate environment, the specific texture of the chair, the specific quality of the light, the specific sounds nearby. This is not meditation in any esoteric sense. It is the specific act of returning attention from the abstract space of the news feed to the concrete space of the actual body in the actual room.
When you do this reliably for a few minutes, your nervous system gets the answer it was seeking. Not "no threats exist in the world." That would be false. But "no threats exist here, right now." Which is the specific answer your threat-monitoring system was actually built to check. And once your system has that answer, the compulsion to keep checking dissolves. Not always. Not perfectly. But reliably enough that this intervention actually changes behavior.
There is a specific implication of this framing that I want to draw out.
Doomscrolling is not evidence that you are shallow or weak. It is evidence that you have an intact nervous system responding to a nervous-system-hijacking environment. The behavior is what your body is doing to try to make sense of a threat environment that has been artificially amplified by systems designed to exploit exactly this response.
Which means the shame that most people feel about doomscrolling is misplaced. You have not failed. You have been targeted. The specific ways your evolutionary hardware works are being used against you by systems more sophisticated than any individual defense. Recognizing this is not an excuse for the behavior. It is a more accurate diagnosis of why the behavior is so hard to change.
The right response is not to double down on willpower. It is to change your relationship with the systems that are triggering the response. Fewer news apps. Fewer alerts. More time offline. More time in physical presence with physical people and physical objects. These are not small interventions. They require actual changes in daily life. But they are the interventions that actually address the underlying pattern.
Let me tell you what happened when I actually made these changes.
I removed all news apps from my phone. I unsubscribed from every news newsletter I had. I stopped following news accounts on the social platforms I still use. I kept one weekly news source, delivered by email, that I read on Monday mornings. That was it.
The first two weeks were uncomfortable. I felt like I was missing something. I checked my phone reflexively and found nothing to consume. I felt slight anxiety about not knowing what was happening. This was withdrawal. It resolved.
By the end of the first month, my baseline anxiety was measurably lower. I was sleeping better. I was working more effectively during the day. I was less reactive in conversations. And critically, when major events occurred, I found out about them through other channels, usually through people I knew, and I found out with enough delay that the acute panic phase had passed. What I got was information without the emotional hijacking that had accompanied it before.
I have not returned to consuming news in the way I used to. It has been about two years. I am, in some technical sense, less informed than I used to be. What I actually am, in practice, is less reactive, more focused, more available to the people around me, and more able to do the work I care about. The tradeoff has been enormously favorable.
I share this because it might work for you too. Not because I have a program to sell. Because the specific intervention is available to anyone who wants to try it, and because the benefits, in my experience, have been larger than the standard advice suggested was possible.
There is a specific connection to AI that I want to draw.
The systems that are producing the doomscrolling pattern are AI systems, in the sense that recommendation algorithms are AI. Every year, they get better at figuring out what will hold your attention. Which means, over time, they get better at hijacking your threat-monitoring system. The specific effectiveness of the exploitation is increasing.
Which means the intervention that worked for me two years ago may not be enough for someone starting now. The systems are more capable. They know more. Their optimization is more precise. And this pattern will continue as the underlying AI continues to improve.
The implication is that individual defenses will have to become more aggressive over time to keep pace with the systems they are defending against. What was sufficient in 2020 was not sufficient in 2024. What was sufficient in 2024 will not be sufficient in 2027. The systems are moving faster than most people's defensive practices are.
The alternative is systemic change. Regulate the platforms so that they cannot exploit users this way. Restructure the incentives so that engagement is not the primary metric. Build alternative systems that do not have this dynamic built in. These are big political projects, but they are the only interventions that would actually stop the pattern rather than just letting individuals cope with it.
I am not optimistic about the political timeline. Which means individual coping is the practical necessity for the foreseeable future. But I do not want anyone to think that individual coping is a solution to the underlying problem. It is a survival strategy. The problem remains.
I want to close with a specific reframing.
Doomscrolling is often described as a bad habit. This framing is not helpful. Bad habits are things you do for no good reason. Doomscrolling has a reason. It is your nervous system doing what it evolved to do, in an environment that has been optimized to exploit exactly that function. The behavior is not irrational. It is a rational response to a specific input, produced by a system that cannot know how to handle the specific input in the specific quantity being delivered.
Understanding this changes what you do about it. You do not fight against yourself. You change the input. You reduce the environment's exploitation of your evolutionary hardware. You give your nervous system fewer opportunities to be triggered by content it cannot properly respond to. And you give it more opportunities to confirm, through physical presence, that you are safe right now.
This is not a full solution. Nothing about the modern information environment is fully solvable through individual practice. But the specific texture of your life is under some of your control. And the specific practice of not doomscrolling, once you find the version of it that works for you, changes that texture more than anything else I have found.
Next month I want to write about the algorithmic god, which is the specific way that recommendation systems have taken over functions that used to belong to religion, philosophy, and community. The specific ways this substitution has proceeded, and what has been lost in the process, is worth taking seriously. Stay with me.
— Transmission Sent —
Niklas Hanitsch
Reference materials
- Roy F. Baumeister et al. — Bad Is Stronger Than Good (Review of General Psychology, 2001)
- Rick Hanson — Buddha's Brain: The Practical Neuroscience of Happiness, Love, and Wisdom (2009)
- Amanda Ripley — High Conflict: Why We Get Trapped and How We Get Out (2021)
- Cal Newport — Digital Minimalism (2019)
- Jenny Odell — How to Do Nothing: Resisting the Attention Economy (2019)
- Peter Whybrow — The Well-Tuned Brain: Neuroscience and the Life Well Lived (2015)
- https://www.pewresearch.org/journalism/2022/06/29/journalists-sense-turmoil-in-their-industry/
- https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8020389/
Continue reading
- Dopamine loops: How social media hacked human motivation
- The attention economy is eating consciousness
- The algorithmic god: How recommendation systems became our post-religion
- Ritual is not superstition. It is technology for the soul
- Waking Up Without Numbing Down
Frequently asked questions
What is doomscrolling? Doomscrolling is the compulsive consumption of negative news or social media content past the point of usefulness. The behavior is characterized by sustained scrolling through alarming information, difficulty stopping, and feeling worse after the session than before. The term was coined around 2020 but describes a behavior that has become widespread with the rise of continuous news feeds and social media.
Why do I doomscroll if it makes me feel bad? Because the behavior is driven by your threat-monitoring system, not your conscious preferences. Your brain treats alarming content as information about threats that require attention. It cannot easily distinguish between local threats you can respond to and distant threats you cannot. As long as new alarming content is available, your system will keep seeking it, regardless of how you feel.
Does willpower help stop doomscrolling? Not much on its own. The behavior is largely automatic and driven by systems that operate below conscious deliberation. Willpower interventions are quickly overwhelmed by the specific mental state that doomscrolling produces. Structural interventions, like removing news apps from your phone, tend to work better.
What actually helps with doomscrolling? Two things. First, changing your environment so that alarming content is less immediately available. Second, learning to satisfy the underlying need through nervous-system regulation rather than through more information. The need being expressed is safety-checking, and it can be resolved through physical presence more reliably than through further reading.
Is doomscrolling a real addiction? By behavioral criteria, yes. It shares the functional pattern of behavioral addictions: compulsive engagement, difficulty stopping, worsening outcomes over time. Whether it counts as addiction in the same clinical sense as substance addiction is disputed. The behavioral pattern is real regardless of the label.