Intelligence Traps
How being smart can make you stupid
The Marshmallow Test, Revisited: What Self-Control Research Really Shows
A larger 2018 replication found the marshmallow test predicts far less than the famous story claimed, and the "willpower as fuel" theory failed its own replications. What holds up: people with good self-control mostly avoid temptation rather than resist it.
Why Experts Are Often Wrong, and How to Get Better Advice From Them
Experts can give bad advice without being frauds. The curse of knowledge, confidence that outruns accuracy and incentives that bend advice are all well documented. When to trust expertise, and the questions that check it.
12 Cognitive Biases Costing You Money, and a Check for Each
Loss aversion, anchoring, present bias, the framing effect and eight more. The research behind each bias, how it shows up in your spending and investing, and a specific check that counters it.
Why Your Brain Is Bad at Predicting the Future
The planning fallacy, the impact bias and projection bias: three predictable ways we misjudge time, feelings and our future selves. The research behind each, and three checks that correct for them.
Does AI Make You Dumber? What the Research Shows So Far
Early studies point one way: when AI does your thinking, you learn less and often feel you learned more. What the evidence does and doesn't show, and the habits that keep your judgment in the loop.
Why Smart People Make Dumb Decisions
For many classic biases, intelligence offers little protection. What it changes is the story afterwards: smart people build better arguments for conclusions they reached for ordinary reasons. The research, and what actually helps.
High IQ Problems: The Real Downsides of Intelligence
High intelligence is mostly an advantage. Its real downsides are narrower: it helps you argue your own side more than see the other, it can feed overthinking, and it pulls towards analysis when a situation needs something else.
Why Being Wrong Makes You Smarter
The brain learns from prediction errors, and confident mistakes, once corrected, stick best. The research on learning from being wrong, its limits, and how to shorten the loop from mistake to update.