15 Ideas That Have Influenced Me

Inspired by a fellow poster, I listed the 20 ideas that have influenced me the most.

1. Homo hypocritus. People are able to hide their motives and beliefs from others by hiding them from themselves. People understand that this is possible on an intuitive level, but the scale at which this occurs is much larger than is realized: religion, the education system, politics, relationships…

2. Human biodiversity. Not race science per say, but the realization that humans are born with reasonably stable levels of intelligence, latent personality traits, and tendencies makes it easier to understand social dynamics. Particular traits (e.g. mathematical ability, cleanliness, chronotype, fertility) are generally less heritable, but the more latent a trait is, the more it approaches a heritability of 80%.

3. The Truth Towers are not what they are made out to be. Researchers (even physical scientists) are not as competent as most people think they are — peer reviewers concord closer to random chance than perfectly, statistical understanding in most fields is weak, and the true productivity of the average academic is quite disappointing. Most people who go into academia are either spergs, developmentally delayed normies, or people who are not cut out for the private sector. (Note: I don’t endorse all of the arguments in this link, but I think that most of the criticisms are relevant).

4. Results from one single study are not necessarily indicative what the literature communicates. But that’s true about meta-analyses as well…

5. p-value. This is the statistical concept where knowing the exact definition (the probability of having observed results as extreme or more extreme than the ones that were observed under the null hypothesis) is crucial. Many people, for whatever reason, think that it is the probability that a finding is true. Because of this misunderstanding, people can easily fall for p > .02 results, when really, the only result that should be believed in the absence of priors is p < .001 due to widespread p-hacking.

6. Narcissism and selfishness are the human default, in the sense that most people pursue their self-interest and view themselves as more important and valuable than they really are. Like motivated reasoning, it is something that exists, but is not something people can choose to overcome if they are not born with the right co-genes that can make you notice when you are engaging in motivated reasoning.

7. Linked to 6: most humans are status maximizing machines, but because people can’t really compete for status (or power) directly, it’s the factor that underlies most human interactions.

8. Most novel and politically relevant scientific findings are probably false. Politically relevant theories get tested more, and are more likely to be published and circulated if they hit p < .05. Researchers that optimize for novelty and political relevancy are also probably optimizing for the truth less than others.

9. Humans, for the purposes of convenience, communicate with an implicit understanding of absolute certainty. But this is obviously not possible, as evidenced by the existence of disagreement and changes in perspectives. This is easy to dismiss as obvious, but dawns on you if you dwel on too long.

10. Postmodernism is based. Progressivism, Leftism, Rightism, Libertarianism, Marxism, Neo-Conservatism, Reaction, Rationalism, Structuralism, Critical Theory, Liberalism, National Socialism, Objectivism, and Catholic Intergralism are all strongly believed in. Based on the prior that there are so many different incompatible meta-narratives, it should be assumed that almost everybody that holds a meta-narrative about the world that is at best incomplete or at worst untrue. I must add that some meta-narrativistic beliefs (e.g. HBD, homo hypocritus) are true or at least non-provably false.

1. Conformity is underrated and should be employed in the right situation. If you happen to see something totally differently from other people, you’re probably wrong.

2. Motivated reasoning. When it comes to political beliefs, people don’t reason the way they do when they are confronted with banal things. Notably, the tendency to engage in motivated reasoning does not correlate with intelligence; the finding that people with higher levels of motivated reasoning failed to replicate, but the effect itself does. This is rather horrifying, as people automatically believe that intelligent people can understand things better than others, when really, the political views of the median intellectual genius can be taken with as much salt as the ones of the village idiot.

3. If somebody is wrong with regards to several perspectives on a specific topic, they’re probably wrong about a lot more more than just those several that you notice. But they could still be right about something else. Consider a physicist who won the nobel prize, who nevertheless believes that the Holodomor never happened and that Communism fails because of a global CIA conspiracy. His political views can be safely ignored, but he can probably be trusted on whether a ball will reach the ground from 5m faster or slower than 0.7s.

4. The American standard of living did improve from 1970 to 2019. Decay that occurred during this time cannot be blamed on the state of the economy.

5. Revealed preferences. Because people can’t really understand their motivations, they must be inferred from actions and signals. It is worth mentioning that a disparity between actions and stated preferences is not necessarily indicative of a revealed preference, but of incompetence.

Admission of guilt: I gave an incomplete definition of the p-value in a prior blogpost”I defined it as the “the probability that the trend would be observed under the null hypothesis” and forgot the ‘more extreme’ part.