Researching for a debate
You have been given a side to argue and you already agree with it. You have twenty minutes to prepare.
How do you spend it?
Step 1: What Would You Do?
There is often more than one good answer. Pick one and see what happens.
Online, late
You have read fifteen posts saying the same thing about a news story. You feel very sure now.
What is worth doing?
Step 2: What That Was
The name for what you just worked out.
Once you believe something, examples that fit it stand out and examples that do not slide past. This is not stupidity. It happens to everyone, including people who know about it.
What you type matters
Searching "why X is bad" gets you reasons X is bad. You will find them whether or not X is bad.
Feeds make it worse
A feed shows you more of what you stopped on. After a week it looks like the whole world agrees with you, because you are being shown the part that does.
The fix
Go and find the best version of the other side. Not the silliest one — the best one.
Step 3: Quick Check
Two questions, then you are done.
Question 1 of 2
You want to know whether a claim is true. Which search is fairest?
Question 2 of 2
Everyone in your feed agrees with you about something. What does that tell you?
Step 4: Your Life Mission
Something to do away from the screen. Nobody is checking.
Steel-man one thing
- Pick something you feel strongly about.
- Find the best argument against it — the one a thoughtful person would make.
- Write it in one paragraph, fairly enough that they would agree it is theirs.
Why: If you cannot do this, you know your conclusion but not the question.
For a grown-up: Do the same exercise on something you disagree with your child about, and show them yours. It changes the conversation more than any rule about screen time.
Step 5: What You Learned
- You notice what agrees with you. Everyone does.
- How you search decides what you find.
- Go looking for the strongest version of the other side.
For the grown-up
The steel-man test in the tip is the most valuable habit in this chapter and the hardest to teach, because it requires giving an opposing view your best effort rather than your worst. It is worth practicing on low-stakes disagreements first.
Naming the bias after the experience is deliberate. Introduced as a term first, it becomes a thing students detect in opponents, which is the bias operating rather than being corrected.
Talking about your own feed openly — what it shows you, what it has stopped showing you — makes this concrete in a way that explaining algorithms does not.