Scoopfeeds — Intelligent news, curated.
agentic-ai

Critical Thinking as a Gym Schedule

LessWrong · May 15, 2026, 8:49 PM

I have an apparently novel n=1 experiment. I would like to start by making it n=2. Hello, how are you? I tried to make myself smarter and it worked. The basic result is that I used to make accurate predictions at roughly the 30% rate, and now it's over 99%. It was too low, and now it's high enough. It also does things like make video games easier. Unfortunately I'm too neurodivergent to get even remotely accurate IQ results, so I can't test that properly. For the record I test at 129. More on this later. I want to see if I can make you smarter too. I do recommend you read the health warnings, which are in the next section. Nevertheless I like [without further ado] so here we go. Set 0: definitions. Set 1: verbal predictions Set 2: intuitive, nonverbal, subjective predictions. Set 3: mind stretches. Set 4: contradiction thrifting. To do sets of set 1 with proper form requires precise definitions. Thus the first step is to practice defining things explicitly. Get into the habit of ensuring you know the exact meaning of every word in your prediction. Look for equivocations, and stop. The second step is to intentionally make predictions, and record them in some fashion. My memory is good enough that I simply remembered them, but if you are apt to forget or confabulate, write them down. Next, test the prediction. If you got it wrong, re-write the prediction such that you would have predicted correctly. Compare the two predictions to find the error. Fix your understanding of the underlying principles, and predict again. Repeat as necessary. The third step is doing the same thing, but using emotions as inputs instead of words and ideas. When you're in a situation, notice how the situation makes you feel, in as much detail as possible. Write it down if you can find the words. Look at how the situation turns out. Then, conclude: [feeling X] predicts [outcome Y]. Look for similar-feeling situations, and see if you can predict how they turn out. If you're wrong, update your fe

Article preview — originally published by LessWrong. Full story at the source.
Read full story on LessWrong → More top stories
Aggregated and edited by the Scoop newsroom. We surface news from LessWrong alongside other reporting so you can compare coverage in one place. Editorial policy · Corrections · About Scoop