Someone dropped a screenshot in our class group. A headline, confidently worded, about a study that supposedly proved something about phones and concentration. Within four messages people were arguing about it. Within ten, two people had changed their study routine because of it.

Nobody in that chat had read the study. Including me. So on a Saturday, mostly out of stubbornness, I found the actual paper and decided I wasn't getting up until I'd read the whole thing.

It took about four hours. The first hour was genuinely humiliating. The last hour was the most useful thing I'd done that month.

---

Hour one: doing it completely wrong

I opened the PDF and started at the top. Title, authors, abstract. The abstract had maybe six words I recognised. I read it four times, understood nothing more on the fourth pass than the first, and felt that specific hot feeling of this is above me.

Then I kept going. Introduction, which was fine, then straight into a methods section full of measures and models, and I hit a wall so hard I nearly closed the tab.

Hour one. Reading top to bottom, understanding nothing, feeling stupid.

Here's what I know now: I was reading it like a textbook chapter, front to back, expecting to understand each sentence before the next. Nobody reads papers like that. Not students, not professors, not the people who wrote them. I wasn't too dumb for the paper. I was using completely the wrong method and blaming myself for the result.

---

The thing that changed everything

I went and looked up how researchers actually read papers, and found the answer is embarrassingly simple: you read it three times, and each time you're doing a different job.

I restarted with pass one. Ten minutes and I could already say what the paper was about, which was more than my first hour had achieved. The confusion hadn't been a comprehension problem. It was an order problem.

---

Hour two and three: where the actual story was

The methods section — the bit I'd found unreadable — turned out to be where everything interesting lived, once I knew what to look for. I stopped trying to understand every term and instead just answered six questions:

*Six questions. You can answer all of them without understanding a single equation.

That last one shocked me most. There's a whole section where the authors themselves write down where their work is weak — sample wasn't diverse enough, timeframe too short, this measure is imperfect. They say it plainly. And it's the section nobody quoting the study ever mentions.

Hour three. Six questions, one highlighter, and it finally opened up.

---

The part that made the whole thing worth it

By hour three I could state what the paper actually found. And it was noticeably narrower than the headline everyone in the group chat was repeating.

The study measured one specific thing, in one specific group, over a limited period. The authors said, carefully, that their results were consistent with a particular effect and that more work was needed. The headline had turned that into a universal claim about everyone, stated as settled fact.

The paper wasn't wrong. The headline wasn't exactly lying either. But somewhere between the limitations section and the screenshot in our group chat, every single qualifier had been quietly deleted. And two people had changed their routine over the version with the qualifiers removed.

That's the thing I actually took away. Not a fact about phones — a demonstration of the gap. Between what a study found and what people say it found, there is almost always a gap, and the only way to see it is to go look.

---

Hour four: writing it down

The last hour I spent writing a one-page summary in my own words, with the paper closed. This was harder than reading it, and it's where I found the three places I thought I'd understood something and hadn't.

That's the real test. If you can't explain the method without looking at it, you haven't finished. And yes — I could have asked an AI to summarise the paper in four seconds. I did, afterwards, to check myself. Doing it after is a check. Doing it instead is just having read a summary of a summary, which is roughly where the group chat already was.

---

Was it worth four hours?

For that one paper's conclusion? Honestly, no. I could have read a decent summary in five minutes and been roughly as informed about phones.

For everything after it, absolutely yes. The second paper took me about ninety minutes. The third one, under an hour. It's a skill with a horrible first attempt and a very fast improvement curve, and once you have it you can go check things yourself for the rest of your life — for a school project, a college application, an argument, a health claim someone in your family is about to act on.

Four hours to stop being someone who repeats studies and start being someone who reads them. That trade is not close.