Successive Approximations

Is Facebook More Like Cigarettes or Alcohol?

Is Facebook More Like Cigarettes or Alcohol?

It's becoming increasingly clear that infinite-scroll, algorithmic social media in its current incarnation is harmful to both the users that over-indulge, as well as countries filled with obsessive users. For an overview of some evidence, here's a snippet fron Jonathan Haidt's interview with Joe Rogan. If that piques your interest, his full interview with Rogan is worth a listen, as well as his 2019 article in the Atlantic about how social media is corroding our republic, or his 2015 article with Lukianoff about how this is bad for students.

Credit for calling this phenomenon coming all the way back in 2012 goes to Eli Pariser in his book The Filter Bubble that reads like it was written yesterday:

"The filter bubble tends to dramatically amplify confirmation bias—in a way, it’s designed to. Consuming information that conforms to our ideas of the world is easy and pleasurable; consuming information that challenges us to think in new ways or question our assumptions is frustrating and difficult. This is why partisans of one political stripe tend not to consume the media of another. As a result, an information environment built on click signals will favor content that supports our existing notions about the world over content that challenges them."

Still worth a read today.

The mechanism of danger here is two-fold. First, the algorithmic: by observing what posts you like, favorite, share, retweet, click on, or even just linger on for a few seconds, Facebook, Twitter, Instagram, et al figure out what type of content you want to see, and then shows you more of it. Their goal is not to make you happier or better informed, but to get you to spend more minutes looking at their page instead of something else, and therefore give them more eyeballs to sell to advertisers.

Second, the infinite scroll, non-linear feed. Because the feed is shown in a random order, it's not like reading a newspaper where you can read all the content and be done. Because the page is always a little different every time you look at it, your brain is never satisfied that you've "caught up" or you're "done" for the day. There's always that itch to check "one more time" for some new content. One by one, each app slowly got away from the chronological feed and into an algorithmic feed, not by user request, but by business request to increase time on the site.

The result is a slot machine, where each time you refresh the page you have a small chance of being shown something that will elicit a reward response—or outrage, which is unconfortable to hold, but pleasurable to vent. So every time you have fifteen seconds without other stimulation, you absent-mindedly pull out your phone and pull the handle on the slot machine.[1]

Which metaphor?

So if we accept the idea that algorithmic, infinite-scroll social media is an unhealthy indulgence when taken to excess, we need to figure how to think about its damage and how we mitigate it (assuming we don't quit cold-turkey). Another question to consider is what is an appropriate age at which to allow children to have their first taste.

One of our human struggles with technology is that we have no norms that have been passed down through the generations of how to handle novel technologies, so I've been thinking about a spectrum of vices from sugary drinks to cocaine as comparison cases to social media. Perhaps they can help us find a good comparison and point the direction we should go.

First up, sugary drinks like soda and fruit juice.

  • Short-term downsides: jitteryness, craving for more, increased appetite.
  • Long-term downsides: diabates, inflammation, metabolic syndrome, shortened life, decreased sensitivity to sugars in all forms.
  • Benefits: none.
  • Regulations: almost none; maximum sizes in certain municipalities.
  • Alternatives: diet soda (other risks), seltzer water.

Next, alcoholic beverages.

  • Short-term downsides: impaired judgement and dexterity, risk to self in accidents, risk to others via drunk driving.
  • Long-term downsides: In moderation, limited. In excess, liver damage, cirrhosis, early death, etc.
  • Benefits: salutory relaxation, enhanced sociability.
  • Regulation: unrestricted above a certain age; limits on activities (driving, carrying weapons) while inebriated.


  • Short-term downsides: bad breath, yellow teeth, cost.
  • Long-term downsides: large increase in susceptibility to respiratory infections, large increase in cancer risk, decreased nicotine sensitivity and dependence.
  • Benefits: nicotine rush, until dependence is established.
  • Alternatives: vaping.
  • Regulation: unrestricted above a certain age; banned indoors almost everywhere; taxed heavily.

Cocaine (mostly guesses):

  • Short-term downsides: impatience, lack of consideration for actions.
  • Long-term downsides: dependence, cost, harm to self and others of reckless actions.
  • Benefits: sense of well-being and productivity.
  • Regulation: prohibited.

You can use this framework to compare social media to other things as well, such as gambling and marijuana, but I think these four are the best reference cases.


The two comparisons that I think best approximate social media are cigarette smoking and alcohol consumption.

When you take a ten minute break in the middle of the workday to scroll through social media, are the effects more like taking a cigarette break where you unwind for a bit and come back somewhat refreshed? Or is it like pulling out a flask and taking a swig where it impairs your focus and productivity even once you've resumed work?

When you wake up in the morning and the first thing you do is grab your phone to check your notifications, is that more like a smoker having their first cigarette of the day to satisfy a craving, or an alcoholic keeping a fifth on the bedside to help them get out of bed?

Can social media be used for salutory effects only at the end of the day or on weekends, comparable to having an occasional drink but only getting truly drunk a few times a year? (Does any smoker limit their use to a cigarette or two at the end of the day?)

Is it tempting to overdo it and regret it the next day, such as getting blackout drunk? (Is binge-smoking a thing?)

Do people who are addicted try to hide it like an alcoholic drinking liquor out of a coffee cup, or just accept it's part of who they are like a smoker?

Do we want to keep kids away from it because it's harder to break an addiction built early in life like smoking, or because it can stunt brain development and lead to life-long consequences (pregnancy, injury, criminal record) like alcohol?

Is a smartphone with Instagram, Twitter, and Facebook apps installed in your pocket more like a pack of cigarettes, or more like a flask of vodka?


So what then do we do with this?

First, I think the aspect of social media that makes it distracting and hard to concentrate even after you put away your phone requires treating it like alcohol. Don't use it during the day, and only in moderation at home. If you do binge, know it will probably have an effect on you for the next day or two.

Second, I also tend to consider underage use of social media more like alcohol than cigarettes.

Parents can responsibly let late-teenagers have some alcohol under their supervision, and even turn a blind eye to the occasional high school party that doesn't get out of control. This is because it is a much more realistic idea to teach and encourage responsible use of alcohol than require complete abstinence.

On the other hand, no sane parent would let a teenager try a few puffs of a cigarette so they can learn to handle it. Cigarettes are an unalloyed negative and do nothing but shorten your life at the expense of momentary pleasure.

The question then arises: at what age does it become appropriate to learn to use social media responsibly? Is it like driving a car (15-16 years old), cigarettes (18), or alcohol (21)? Jonathan Haidt has stated that his research points to no earlier than high school. My assumption is even that recommendation may be compromise because suggesting waiting until college is outside the Overton window currently.

In that vein, the third point is that social media, like alcohol, when used in moderation, at appropriate, non-work times it can be a net positive. Perhaps similar to limiting yourself to one drink a night, a limit like 20 minutes a day is safe. Hop on, see the latest pictures and updates from friends and family [2], maybe make a comment or two, and then shut it off.

Mobile notifications on your phone, however, are more like cigarettes with no redeeming value. I would go further and remove all the social media apps from your phone as well, to preclude the "nervous-smoker" habit of checking it any time you have five minutes to kill.

My practices

I've only recently started thinking about social media this way, but it seems to confirm the habits that I've triangulated on in the last year or so. (#humblebrag)

First and foremost, I removed any app with infinite scrolling from my phone. This meant Facebook, Instagram, Twitter, reddit, and news apps. All I have now for reading are Kindle, Feedly, and Instapaper. (More on those in a later post.)

Second, if I want to check Facebook, Instagram, or Twitter, I have to sit down at my desk, boot up my desktop, and open them in a web browser. It took a good couple of months to break the habit of pulling out my phone to check for notifications and new posts any time I had fifteen seconds to spare. If I do have a little time to fill, I now open Feedly or Instapaper.

This helps in two ways. First, it makes checking social media a decision and not a habit. It raises the barrier just enough to not be something I do reflexively.

Second, in some ways I can't really describe, a computer screen does not feel as "close to home" as a smartphone. The smartphone is the nerve center of my life: my calendar, my email, my messaging with friends and family, my GPS navigation, my notes. My phone feels personal. The computer screen on the other hand feels more external, like a TV. It's a lower-bandwidth pipeline to your limbic system. It's easier to detach and exercise skepticism on a computer, and not feel personally or existentially threatened by something that just happens to be on a screen but isn't "part of you."

My third practice is to contain checking social media to at most a few times a day, mostly towards the evening. Most days, I never see anything from social media before lunch and on work days it's usually not until after dinner.

Even when I am working at my computer, like when writing this blog post, I do not keep a Facebook tab open so my eyes can reflexively flick over to it to check for notifications. Each time I want to check it, it's a decision to stop what I'm doing and go do that instead.

Afterthought: Public Health Campaigns?

It's hard to say exactly why smoking in the US has steadily declined since 1954. The overt reasons might be public health campaigns about the dangers of smoking, increased education in schools, and warning labels on cigarettes. The subtle reasons might just be the growing understanding that smoking is bad for you, duh, in part by seeing older friends and family who smoke suffer and die. Was the public convinced of the dangers by studies and data? By government posters and TV ads? Or by observing the damage caused by cigarettes in their own life?

Also, what part was played by making smoking inconvenient by banning it virtually everywhere indoors? Or banning advertising of tobacco products in many contexts?

I'm not aware of any research that has tried to answer that question, but it would likely be relevant to this discussion.

If we agree that social media used to excess is harmful, but may be used in moderation for some positive outcomes, and that use should be socially (and perhaps legally[3]) constrained, how do we create that?

If we look to the tobacco example, perhaps more studies showing the harm of social media will help. Certainly most parents who know a teenager who has been enmeshed in social media since middle school will have doubts as to whether it's been a net positive on their development and mental well-being. Perhaps that one-two punch of research plus personal experience will be enough to get the word out.

Of course, where would the word get out? The social media companies themselves will fight such research the way cigarette companies did in the 50s. They are public companies. If their CEOs do not "protect shareholder value" (maximize revenue) they will be fired and replaced.

Meanwhile, news organizations have entirely reconfigured their business models to thrive not on subscribers, but on headlines and articles that get a lot of "engagement" on social media and therefore ad impressions on their site. Asking them to set fire to their own business model, as corrosive as it is for our republic, is not realistic.

Our modern information processing apparatus will discard this message immediately. Thus we're left with the handful of outsiders like Joe Rogan who will interview heterodox voices like Jonathan Haidt. Some people are publishing books and doing the podcast and YouTube press circuit. We have blog posts and face-to-face conversations. But none of these will reach a widespread audience.

Some kind of pervasive messaging of the harms of unchecked social media use is needed. But who will create it, and more importantly, how will it be disseminated?

  1. I also think other quasi-social media apps with infinite-scroll capabilities where there is no "end" to reach and each load of the page shows a different set of results like reddit fall into this category. ↩︎

  2. Helpfully, the algorithm showing you the content you will most likely engage with can work in your favor here ↩︎

  3. Haidt has suggested legal requirements that the site identify users with real names and addresses, even if their presence on the site is pseudonymous. ↩︎

Ben Berry

I don't have comments on my posts, but if you want to reach me, email me at my first name at