Thinking Clearly Online

Most of what you now learn about the world arrives through systems designed to hold your attention, not to inform you. That single fact reshapes reasoning in ways that are easy to miss, because they feel like your own free choices. The stories that reach you, the emotions they provoke, and the certainty you end up with are all partly products of an environment tuned for engagement. None of this means the internet is making everyone stupid; that claim is itself an overreaction. But thinking clearly online takes more deliberate effort than doing so offline, because the medium quietly pushes in unhelpful directions.

Engagement, not truth

Recommendation algorithms mostly optimise for engagement — clicks, time, shares — because that is what sustains the business. Content that triggers strong emotion, especially outrage and moral anger, tends to travel further than calm or nuanced material. The result is a feed subtly weighted toward whatever provokes a reaction, which is not the same as whatever is true or important.

Filter bubbles and echo chambers

The popular story says algorithms trap each of us in a sealed bubble of agreeable views. The evidence is more mixed. Several studies find that many people actually encounter a wider range of opinion online than in their offline lives, and that the strongest echo chambers form among a smaller, highly engaged minority. Self-selection — whom we choose to follow — often matters as much as the algorithm.

Virality is not accuracy

What spreads is what is novel, surprising, and emotionally charged; truth has no special advantage in that race. A message can also travel far beyond the audience it was written for, a phenomenon called context collapse, so a remark meant for one group is read by strangers who lose its original meaning.

Better information habits

You cannot rebuild the platforms, but you can change how you move through them. Draw from a few independent, higher-quality sources rather than whatever the feed serves. Slow down before sharing, since the urge to share is strongest exactly when a story is engineered to provoke. And watch the gap between feeling informed and actually being able to explain the issue.

A striking claim crosses your feed: a screenshot, an outrageous quote, thousands of shares. You feel the pull to repost. Slowing down, you ask three questions — who posted this, what is the original source, and does the strength of my reaction match the strength of the evidence? Often the screenshot has no source, the quote is stripped of context, and the outrage is the point.

The environment is not all downside, and pretending it is would be its own error. The same networks let a distant expert correct a myth in real time and expose people to challenges they would never meet in a like-minded neighbourhood. The problem is not connection itself but the attention economy layered on top of it.

A 2015 study in the journal Science by Eytan Bakshy, Solomon Messing, and Lada Adamic, then working at Facebook, examined how millions of users met political news in their feeds. It reported that people were exposed to a meaningful amount of cross-cutting content — views from the other side — and that an individual's own choices about whom to follow and what to click limited that exposure more than the ranking algorithm did. The study drew fair criticism: it looked only at users who had declared a political affiliation, and the researchers worked for the company whose product they studied, so its precise conclusions are debated. Even so, it remains a widely cited reason to be wary of the simplest filter-bubble story. The algorithm shapes your feed, but it is not the whole explanation — your own selections matter too.

• "The algorithm decides everything I believe." It shapes what you see, but your own choices of whom to follow and what to click matter just as much.
• "Everyone is trapped in a filter bubble." The evidence is mixed; strong echo chambers are real but concentrated among a committed minority.
• "If a post went viral, there must be something to it." Virality tracks novelty and emotion, not accuracy.
• "Reading a lot online means I am well informed." Feeling informed and being informed are different things.

Before reposting a viral claim, run it through a quick source-and-emotion check and see how many "obvious" stories quietly fail it.

Think Like a Maester: The stories that most demand to be shared immediately are often the ones most worth pausing on — urgency is frequently engineered, not earned.

1. What do engagement-driven algorithms mainly optimise for, and why?
2. Why is the popular filter-bubble story an oversimplification?
3. Why does emotionally charged content tend to spread faster than measured content?
4. What is context collapse?
5. What is the difference between feeling informed and being informed?

The online world is not a neutral window onto events but an environment tuned for attention, where outrage travels well, novelty beats accuracy, and messages escape the context that gave them meaning. The tidy story of everyone sealed in a filter bubble is only partly supported by the evidence, and the doom-laden version of these claims is its own distortion. What holds up is simpler: choose your sources deliberately, slow down before you share, and keep asking whether you could actually explain what you think you know.

• Spotting Misinformation and Disinformation
• Steelmanning and Changing Your Mind