How to become smarter

• Only read at the highest (quality) level. Skim or give up as soon as possible.
• Avoid high frequency information sources. Read for patterns and the long view. Read a lot.
• Avoid the effects of social influence, that is, most media.
• Avoid most media because they are designed by attract attention by using the most outrageous, lurid, captivating, infuriating, intoxicating images or ideas, all of which could have happened, but very few of which matter very much to what you are doing here, now.
• Look for ideas and skills that accumulate and speed or improve later learning in the manner of compound interest. Use spaced repetition to build base of knowledge (Michael Nielsen).
• Focus on very powerful, very big ideas: Willpower and hyperbolic discounting, game theory/active adversaries/second-order effects, social signalling, Mercier/Sperber theory of argumentation.
• Rephrase what you just read. Write summaries. Avoid copying quotations.
• Recommendations for what fiction to read are almost always misguided because one's response to fiction is deeply personal and idiosyncratic. It's a lot like the idea of the best person to marry. There might be such a person for you, but not everyone shares your ideal. The same holds for "what book influenced you the most?"
• Try ignoring the emotionally felt effect of others' actions. Those emotional reactions may be important, but they may be strongly affecting your thinking, and moving it away from novel or better ideas. In particular, how much of what we think our self or our life is forced to lie within socially determined guardrails?
• Try ignoring intentional interpretations of actions and events. Contrast our interpretation of a person doing something bad or stupid with a machine doing the same thing. If you don't believe in free will, then these two interpretations are really the same thing.
• There is a natural inclination to believe either that a proposition is either true or false, but very often the best answer lies somewhere in-between and with a wide confidence interval. This makes most people uncomfortable.
• Heavily downweight any analysis of the habits of the highly successful to correct for the role of luck in their success.
• For fuzzy prediction tasks, start with the base rate, and the incorporate corrections for additional factors. (Tetlock)
• If it is not very important, don't do it.
• How did Newton discover the Law of Gravity: "By thinking about it all the time".
• Look at the highly evolved social and biological worlds for ways that combat that ills of technology, especially, of social and cognitive technologies.
• Two (contradictory) ideas about how to solve a new problem: (1) think hard on your own before looking at what others have done, (2) review books, papers, blogs (e.g., Kaggle winners) to see what has worked.
• Document your plan. Document where your plan went wrong. Iterate.
• Ask: is the problem zero sum or net enhancing (or detracting)?
• Ask: if this action is taken, then what new incentives are created and how will people respond?
• Look for paired high-minded and base motives that produce or justify the same action.
• Personal rules are useful and save time and mental energy. Be rule governed, pick your rules carefully, and feel confident about them (or get better rules).