Link Roundup: July 2018

The Lifespan of a Lie – Zimbardo’s Stanford Prison Experiment is one of the most famous experiments in psychology.  I knew it was questionable in methodology and ethics, but this exposé gathers all the details in a more shocking picture than I had imagined.  Zimbardo is a crook.  He repeatedly lied about key aspects of his experiment, persuaded subjects to also lie after the fact.  And when other researchers failed to replicate, he accused them of being frauds.  It is a disgrace that people in psychology still respect him.

I know that people still like to cite the Stanford Prison Experiment, because it “proves” something they already know to be true, but have you tried instead just… making assertions without proof, because at least then we’re honest?

The World’s Highest Paid Musicians vs Average Salaries – This is an infographic showing how long it takes for popular music artists to earn as much as people with other jobs make in a year.  It’s based on a Forbes list.  That just makes me think of when I asked my brother how much money he made being a rock star (although “star” is rather overstating it), and the answer was basically nothing.  So, this is your reminder that capitalism doesn’t produce anything remotely resembling “fair”.

What’s interesting to me, is that this is easily explainable.  I honestly don’t understand why CEOs get paid so much–it kinda seems like companies just overvalue CEOs, or there are principal-agent problems–but it’s easy to see why top artists get paid so much.  The popularity of music follows a power law distribution, so artist’s salaries follow that too.  I think it would be an interesting exercise to think up a more utopian system of art monetization.

Do algorithms reveal sexual orientation or just expose our stereotypes? – I’ve always found “gaydar” research to be disappointing, because it’s fixated on what you can determine from facial features alone.  In reality, when I can guess someone is queer, that’s based entirely on cultural markers.  So I would say “gaydar” research isn’t really about gaydars at all, and is about some loosely related concept that really ought to be given a different name.

This article I’m linking is a detailed critique of a “gaydar” study that built an AI that can purportedly guess people’s orientation from their faces.  The critique is that the AI is clearly basing its guesses on cultural markers and grooming habits.  And to me, the cultural differences are the more interesting ones.  I would not have guessed that same-sex attracted men in the US are more likely to wear glasses, or that they take selfies from higher angles.

[Read more…]

Homeownership sure seems like a bad investment

It’s said that homeownership is one of the markets that millennials are killing. Well color me a typical millennial, because I don’t own a house, and I can’t see why I would want to. Renting housing just seems like a better deal. I’m going to explain why I have this intuition, then I’ll actually do some research to see common rebuttals.

Similarities and differences

Let me start with some caveats and basics.  While homeownership and renting fulfill the same need, they are not directly comparable. At least around here, if you rent housing then you tend to get a smaller apartment, and if you buy housing, you tend to get a larger house. If you want to spend more money for more space, or less money for less space, that may make your decision for you.

The other main difference is that homeownership is kind of a two-for-one deal. If you rent housing, you’re pretty much paying money for living space, and that’s it. If you buy housing, you’re not just paying money for space, but also investing some of that money. If you don’t want to invest money (say you can’t afford to), then homeownership is a bad deal for you. If you do want to invest money, then you could buy a house, but it’s not like that’s your only option. You could just rent housing, and then invest the money you saved by not buying a house. Thus, the comparison I’m making is not between a house that you own, and a house that you rent. The comparison is between investing in a house, and investing in, say, an index fund.

[Read more…]

Origami: Cube Plus Alpha

Cube plus alpha

Tomoko Fuse’s Simple Sonobe 12-unit Assembly Plus Alpha, from Unit Origami: Multidimensional Transformations

Fuse has a series of models that consist of basic polyhedra, with extra pieces of paper attached as embellishments.  This model is a cube (made of 12 pieces of paper) with a pyramid added to each face, and 3 spikes to each vertex.  All in all, that’s 42 pieces of paper.  This is a pretty neat idea.  Since it is made of three distinct types of units, it defies the usual convention of making modular origami from many identical units.

This one’s a fairly old model, apparently made in 2013.  I gave it away as a gift so I don’t know if it’s still living, or deceased.

The intelligence hierarchy

Never Let Me Go is a book about a bunch of unusual kids, who grow up without really understanding how unusual they are. They go to a special school, and the school is particularly intent on having the kids show creativity, and produce artwork. The reason the school wants artwork is… oh, I shouldn’t spoil it. The kids don’t really understand why. All they understand is that to have an artwork chosen by the school officials is an honor.

And therefore it is a basis for social hierarchy.

To the kids, art isn’t an outlet for creativity, it isn’t a matter of amusement, but a matter of achievement. There’s one kid who drew a silly picture one of an elephant one time, and as a result the kids perceive him as lagging behind. For years, he gets teased, pranked, and throws huge tantrums. He finds peace when a teacher takes him aside and tells him it’s okay to be less creative. The other kids are scandalized by the very idea.

I’ve been thinking about this story, and how it resembles our own situation. The truth is, we grew up into a system that we didn’t really understand, but where we understood that being smart was an achievement. We’ve had personal experience building social hierarchies around perceived intelligence, before even understanding what a social hierarchy is, or why intelligence is important. We’ve pinned the labels “smart” or “stupid” on other people, and had them pinned to ourselves, often on flimsy evidence, and these labels have governed the early years of our lives.

[Read more…]

Emotions on the internet

Cultural differences

Recently, I’ve talked about the book How Emotions are Made, by Lisa Feldman Barrett, and I explained the basics of the theory of constructed emotion. I would like to go a step further, and discuss some of the implications.

If emotional categories are socially constructed, I’d expect different cultures to have different categories. Perhaps we have a lot of categories in common, since our cultures are all in contact with one another, and different cultures might be fulfilling similar needs. But the construction of emotions predicts that there must be some exceptions–emotional categories that only exist in some cultures and not others.

Dr. Barrett gave many examples of emotions that exist in other cultures, but not in US culture. For example, in Czech culture, “litost” is described as “torment over one’s own misery combined with the desire for revenge”. In the Ilongot tribe in the Philippines, “liget” is described as a feeling of exuberant aggression, usually felt by a group of people competing against another group. While these concepts are intelligible to us, we rarely think or talk about having exactly those combinations of feelings, and we have few expectations for how we would respond to those feelings.

I found these examples to be quite compelling, and not just because of the sheer number of examples that Dr. Barret described. Once I understood what a new emotional concept looks like, I realized that we’re creating new emotional concepts all the time! Even without looking outside the US, you can find plenty of relatively recent emotional concepts created right here on the internet.

[Read more…]

Muddling the Dunning-Kruger Effect

The Dunning-Kruger Effect states that people with the lowest competence tend to overrate their competence, but people with the highest competence tend to underrate themselves. This was shown in 1999 paper by Dunning and Kruger.1 Here’s one of the figures from the paper:

A graph showing people's self-assessed ability, and actual test score. The bottom quartile gives themselves a rating in the 60 percentile, and the top quartile gives themselves a rating in the 75 percentile.

This figure shows results from a test on humor. People are scored based on how well their answers agree with those of professional comedians, and then they are asked to assess their own performance. There were similar results for tests on grammar and logic.

The Dunning-Kruger effect has entered popular wisdom, and is frequently brought up whenever people feel like they’re dealing with someone too stupid to know how stupid they are.  But does the research actually mean what people think it means?

Before reading into this subject, I must admit that I had a major misconception.  I thought that people’s self-assessment was actually anti-correlated with their competence.  I thought someone who knew nothing would actually be more confident than someone who knew a lot.  (This leads to an amusing dilemma: Should I choose to give myself a lower rating, because it would that increase posterior probability that I’m more competent?2)

But it is not true.  People who know nothing are less confident than people who know a lot.  People who know nothing are overconfident relative to their actual ability, but they are still not as confident as people who have high ability.

[Read more…]

FTA part 4: Anthropic reasoning

This is the fourth and final part of a series discussing the Fine Tuning argument (FTA). The outline is here.

Mundane multiverses

A “multiverse” is a set of multiple sub-universes, which together comprise a single super-universe. The idea is that the universe that we know is a single sub-universe, and there are multiple other universes like ours. So we can imagine another sub-universe where everything is the same, except that ever coin flip comes up on the opposite side.  Or a sub-universe where everything is the same, but we’re all evil and have goatees. Or another sub-universe where

Typically, when physicists talk about parallel sub-universes, what they mean are non-interacting sub-universes. So, you can’t ever talk to the evil goatee’d version of yourself. Although, people sure like to imagine that sort of thing in sci-fi.  So let’s talk about the kind of parallel universe that we could, in principle, interact with. What if I told you that this kind of parallel universe is one we already know exists?

To travel to a parallel universe, just hop in a space ship, and travel 4 lightyears over to Proxima Centauri. You will find a universe exactly like ours, except that the sun is different, and the planets are different, and the constellations are different.

[Read more…]