Generalizing From One Example

Related to: The Psychological Unity of Humankind, Instrumental vs. Epistemic: A Bardic Perspective

"Everyone generalizes from one example. At least, I do."

   -- Vlad Taltos (Issola, Steven Brust)

My old professor, David Berman, liked to talk about what he called the "typical mind fallacy", which he illustrated through the following example:

There was a debate, in the late 1800s, about whether "imagination" was simply a turn of phrase or a real phenomenon. That is, can people actually create images in their minds which they see vividly, or do they simply say "I saw it in my mind" as a metaphor for considering what it looked like?

Upon hearing this, my response was "How the stars was this actually a real debate? Of course we have mental imagery. Anyone who doesn't think we have mental imagery is either such a fanatical Behaviorist that she doubts the evidence of her own senses, or simply insane." Unfortunately, the professor was able to parade a long list of famous people who denied mental imagery, including some leading scientists of the era. And this was all before Behaviorism even existed.

The debate was resolved by Francis Galton, a fascinating man who among other achievements invented eugenics, the "wisdom of crowds", and standard deviation. Galton gave people some very detailed surveys, and found that some people did have mental imagery and others didn't. The ones who did had simply assumed everyone did, and the ones who didn't had simply assumed everyone didn't, to the point of coming up with absurd justifications for why they were lying or misunderstanding the question. There was a wide spectrum of imaging ability, from about five percent of people with perfect eidetic imagery1 to three percent of people completely unable to form mental images2.

Dr. Berman dubbed this the Typical Mind Fallacy: the human tendency to believe that one's own mental structure can be generalized to apply to everyone else's.

He kind of took this idea and ran with it. He interpreted certain passages in George Berkeley's biography to mean that Berkeley was an eidetic imager, and that this was why the idea of the universe as sense-perception held such interest to him. He also suggested that experience of consciousness and qualia were as variable as imaging, and that philosophers who deny their existence (Ryle? Dennett? Behaviorists?) were simply people whose mind lacked the ability to easily experience qualia. In general, he believed philosophy of mind was littered with examples of philosophers taking their own mental experiences and building theories on them, and other philosophers with different mental experiences critiquing them and wondering why they disagreed.

The formal typical mind fallacy is about serious matters of mental structure. But I've also run into something similar with something more like the psyche than the mind: a tendency to generalize from our personalities and behaviors.

For example, I'm about as introverted a person as you're ever likely to meet - anyone more introverted than I am doesn't communicate with anyone. All through elementary and middle school, I suspected that the other children were out to get me. They kept on grabbing me when I was busy with something and trying to drag me off to do some rough activity with them and their friends. When I protested, they counter-protested and told me I really needed to stop whatever I was doing and come join them. I figured they were bullies who were trying to annoy me, and found ways to hide from them and scare them off.

Eventually I realized that it was a double misunderstanding. They figured I must be like them, and the only thing keeping me from playing their fun games was that I was too shy. I figured they must be like me, and that the only reason they would interrupt a person who was obviously busy reading was that they wanted to annoy him.

Likewise: I can't deal with noise. If someone's being loud, I can't sleep, I can't study, I can't concentrate, I can't do anything except bang my head against the wall and hope they stop. I once had a noisy housemate. Whenever I asked her to keep it down, she told me I was being oversensitive and should just mellow out. I can't claim total victory here, because she was very neat and kept yelling at me for leaving things out of place, and I told her she needed to just mellow out and you couldn't even tell that there was dust on that dresser anyway. It didn't occur to me then that neatness to her might be as necessary and uncompromisable as quiet was to me, and that this was an actual feature of how our minds processed information rather than just some weird quirk on her part.

"Just some weird quirk on her part" and "just being oversensitive" are representative of the problem with the typical psyche fallacy, which is that it's invisible. We tend to neglect the role of differently-built minds in disagreements, and attribute the problems to the other side being deliberately perverse or confused. I happen to know that loud noise seriously pains and debilitates me, but when I say this to other people they think I'm just expressing some weird personal preference for quiet. Think about all those poor non-imagers who thought everyone else was just taking a metaphor about seeing mental images way too far and refusing to give it up.

And the reason I'm posting this here is because it's rationality that helps us deal with these problems.

There's some evidence that the usual method of interacting with people involves something sorta like emulating them within our own brain. We think about how we would react, adjust for the other person's differences, and then assume the other person would react that way. This method of interaction is very tempting, and it always feels like it ought to work.

But when statistics tell you that the method that would work on you doesn't work on anyone else, then continuing to follow that gut feeling is a Typical Psyche Fallacy. You've got to be a good rationalist, reject your gut feeling, and follow the data.

I only really discovered this in my last job as a school teacher. There's a lot of data on teaching methods that students enjoy and learn from. I had some of these methods...inflicted...on me during my school days, and I had no intention of abusing my own students in the same way. And when I tried the sorts of really creative stuff I would have loved as a student...it fell completely flat. What ended up working? Something pretty close to the teaching methods I'd hated as a kid. Oh. Well. Now I know why people use them so much. And here I'd gone through life thinking my teachers were just inexplicably bad at what they did, never figuring out that I was just the odd outlier who couldn't be reached by this sort of stuff.

The other reason I'm posting this here is because I think it relates to some of the discussions of seduction that are going on in MBlume's Bardic thread. There are a lot of not-particularly-complimentary things about women that many men tend to believe. Some guys say that women will never have romantic relationships with their actually-decent-people male friends because they prefer alpha-male jerks who treat them poorly. Other guys say women want to be lied to and tricked. I could go on, but I think most of them are covered in that thread anyway.

The response I hear from most of the women I know is that this is complete balderdash and women aren't like that at all. So what's going on?

Well, I'm afraid I kind of trust the seduction people. They've put a lot of work into their "art" and at least according to their self-report are pretty successful. And unhappy romantically frustrated nice guys everywhere can't be completely wrong.

My theory is that the women in this case are committing a Typical Psyche Fallacy. The women I ask about this are not even remotely close to being a representative sample of all women. They're the kind of women whom a shy and somewhat geeky guy knows and talks about psychology with. Likewise, the type of women who publish strong opinions about this on the Internet aren't close to a representative sample. They're well-educated women who have strong opinions about gender issues and post about them on blogs.

And lest I sound chauvinistic, the same is certainly true of men. I hear a lot of bad things said about men (especially with reference to what they want romantically) that I wouldn't dream of applying to myself, my close friends, or to any man I know. But they're so common and so well-supported that I have excellent reason to believe they're true.

This post has gradually been getting less rigorous and less connected to the formal Typical Mind Fallacy. First I changed it to a Typical Psyche Fallacy so I could talk about things that were more psychological and social than mental. And now it's expanding to cover the related fallacy of believing your own social circle is at least a little representative of society at large, which it very rarely is3.

It was originally titled "The Typical Mind Fallacy", but I'm taking a hint fromt the quote and changing it to "Generalizing From One Example", because that seems to be the link between all of these errors. We only have direct first-person knowledge one one mind, one psyche, and one social circle, and we find it tempting to treat it as typical even in the face of contrary evidence.

This, I think, is especially important for the sort of people who enjoy Less Wrong, who as far as I can tell are with few exceptions the sort of people who are extreme outliers on every psychometric test ever invented.


Footnotes

1. Eidetic imagery, vaguely related to the idea of a "photographic memory", is the ability to visualize something and have it be exactly as clear, vivid and obvious as actually seeing it. My professor's example (which Michael Howard somehow remembers even though I only mentioned it once a few years ago) is that although many people can imagine a picture of a tiger, only an eidetic imager would be able to count the number of stripes.

2. According to Galton, people incapable of forming images were overrepresented in math and science. I've since heard that this idea has been challenged, but I can't access the study.

3. The example that really drove this home to me: what percent of high school students do you think cheat on tests? What percent have shoplifted? Someone did a survey on this recently and found that the answer was nobhg gjb guveqf unir purngrq naq nobhg bar guveq unir fubcyvsgrq (rot13ed so you have to actually take a guess first). This shocked me and everyone I knew, because we didn't cheat or steal during high school and we didn't know anyone who did. I spent an afternoon trying to find some proof that the study was wrong or unrepresentative and coming up with nothing.

Comments

sorted by
magical algorithm
Highlighting new comments since Today at 12:33 PM
Select new highlight date
All comments loaded

I'm glad to see someone bringing up the topic of seduction, and how it relates to rationality, and how attitudes inside and towards the seduction community relate to rationality and biases.

I'm going to give a big warning to everyone on this topic. The seduction community is an expansive and heterogenous phenomenon. Unless someone has some experience of the community (say 30+ hours of reading of multiple gurus with different philosophies, and they have gone out and tried the approaches the community advocates or seen real pickup artists in action), then it is virtually impossible to understand what it involves and describe it in a way that isn't skewed.

Elana Clift's honors thesis is a good place to start.

Yvain, you are right to take the mass perceptions of people of each sex as evidence (though evidence of what is unclear, so far). Let me unpack a few things:

There are a lot of not-particularly-complimentary things about women that many men tend to believe. Some guys say that women will never have romantic relationships with their actually-decent-people male friends because they prefer alpha-male jerks who treat them poorly. Other guys say women want to be lied to and tricked.

There are guys who think like this, but not all pickup artists do, and probably most of the men who think like this aren't pickup artists. Here's my quick availability-heuristicky impression of what pickup artists think on these subjects, and whether or not these beliefs are complimentary, based on more than half a decade of involvement with the community:

  • Female attraction to male friends: Pickup artists typically believe that if a woman sees a man as "just a friend," then it is unlikely that this perception will change, and that his efforts are best allocated elsewhere.

  • Alpha males: Pickup artists typically believe that women are attracted to "alpha males." What "alpha male" means is subject to intense debate.

  • Lying and trickery: Pickup artists typically don't believe that women want to be lied to or tricked. Pickup artists do present themselves selectively and strategically. Yet the modal point of view in my experience is that lying and trickery are looked down on, and seen as antithetical to seduction. If a pickup artist isn't looking for a relationship, then he will try to make that obvious, or even state it explicitly.

Well, I'm afraid I kind of trust the seduction people.

It's good to see someone caring what pickup artists think, but I would take their views with a bit more caution for several reasons:

  1. The availability heuristic. The seduction community has a pretty good model of young female extraverts with average IQ, because these are the women they encounter most often. As you look at women who differ more and more from the average extravert, the prototype of the seduction community becomes less and less correct. This is a point where I agree with Alicorn. This doesn't mean that the community's advice completely ceases to work, but it requires modification. Women who are nerdy, systemizing, bisexual, feminist, or in alternative subcultures are wired differently. (And to tie in to your post, women with those traits are going to be bad judges of the preferences of typical women due the Typical Psyche Fallacy, which I think is a special case of the availability heuristic.)

  2. Naive realism. Pickup artists often assume that because a theory produces results, then it is true. This isn't necessarily the case. Pjeby has correctly described how correct-enough theories will often be useful without being true. Having a model of women that lets you predict the behavior of say, 30% of women better than chance is actually really good for a guy who is completely in the dark about women and their preferences and behaviors.

(I wonder whether more complex models would necessarily be more useful; I think this varies. When you are a beginner, it may be best to understand typical women, and then later try to figure out how all the outlier types of women work by seeing their similarities and differences from typical women. Ultimately, the model that is most important to have is the model of the type of women you are compatible with.)

When you put these two together, you get pickup artists running around with oversimplified-but-nevertheless-useful models of women, who start to get some better results, confirming their over oversimplified-but-nevertheless-useful models of women in their minds.

I figured this out because I view the empirical approach as the core of the seduction community's teachings, so I often try out stuff that my gut tells me and break the rules of what is "supposed" to work or not work.

As for how much the view of women in the seduction community is complimentary or true, those are topics I'll have to save for another time.

Interesting illustration of mental imagery (from Dennett):

Picture a 3 by 3 grid. Then picture the words "gas", "oil", and "dry" spelled downwards in the columns left to right in that order. Looking at the picture in your mind, read the words across on the grid.

I can figure out what the words are of course, but it is very hard for me to read them off the grid. I should be able to if I could actually picture it. It was fascinating for me to think that this isn't true for everyone.

Picture a 3 by 3 grid. Then picture the words "gas", "oil", and "dry" spelled downwards in the columns left to right in that order. Looking at the picture in your mind, read the words across on the grid.

Interestingly, I find the task much easier if I do it the other way: visualizing the words spelled across, and then reading off the words going down the grid.

If mental images consist of replayed saccades, this makes perfect sense. To generate the downward images of words and then read across would reasonably be harder than simply replaying the stored "across" patterns, and then reading them down. IOW, visualization is more like vectors and sprites than it is like pixels -- which reflects how sight itself works.

Maybe I'm just cynical but I think people vastly overestimate their own goodness. Often "goodness" is just a way to dress up powerlessness. Like an overweight man might say he's "stocky" or an overweight woman might say she's "curvy," so an undesirable or shy man or woman might emphasize the upside: "I would never cheat." There's a version of the typical mind fallacy in there: a person might genuinely think they would never cheat but be extrapolating from a position where the opportunity rarely presents itself. We can all talk about how, if we were in a position of political power, we'd never succumb to bribes or cronyism because we don't have any political power. It both makes us look good and, as far as we know, it's true. I think testimony, especially when it comes to ones moral worth, is the least valuable form of data available.

When I've taught ethics in the past, we always discuss the Nazi era. Not because the Nazis acted unethically, but because of how everyone else acted.

For example, we read about the vans that carried Jewish prisoners that had the exhaust system designed to empty into the van. The point is not how awful that is, but that there must have been an engineer somewhere who figured out the best way to design and build such a thing. And that engineer wasn't a Nazi soldier, he or she was probably no different from anyone else at that time, with kids and a family and friends and so on. Not an evil scientist in a lab, but just a design engineer in a corporation.

One point of the discussion is that "normal" people have acted quite unethically in the past, and how can we prevent that happening to us.

Very interesting post. Perhaps I should mention that there's a possibility to go to the other extreme; assuming you're different to everyone else. A lot of very bad pretentious teenage poetry stands as testament to this.

Very true. A typical reaction when reading advice or something about the typical flaws of people (biases, planning), is "Yeah but that doesn't apply to me". It often takes a deliberate effort to override the inside view and stop finding excuses.

Note that in both cases the mistake makes us look better:

  • "I know how others work from the experience of my own mind" sounds better than "I don't understand other people"
  • "I don't make that common mistake because I'm different from others" sounds better than "whoops I'm also likely to make that mistake"

Indeed, it's one of the interesting paradoxes about people. We think that everyone is the same as us (shown in examples like this), while simultaneously thinking that we're unique and special (for things like narcissism, the narrative fallacy, and even religion.)

It's actually a wonder we manage to accomplish anything at all, given the messy state of our brains...

Regarding differences in mental imagery: only this winter did I really understand that good musicians have vivid aural imagination, while I couldn't hear any sounds in my head, period. Immediately after this realization I started exercising. By now I can hear complete monophonic melodies, and (on good days) imagine two notes sounding at the same time. Classically trained conductors can imagine a complete orchestral sound while reading sheet music. I don't see any reason why visual imagination can't be similarly trained.

My experience in my non-academic work life, is that many programmers can't visualize verbal descriptions of subsystems, but they learn how to make convincing "I got it" noises to mollify their coworkers. It's not just programmers, it's all sorts of coworkers. I have no idea how an adult can avoid this pitfall.

At any given time, I always have some song or another playing in my head, and I can recall songs I've memorized and "play them back" at will. Usually it's just the melody, though; the harmony usually doesn't seem to get captured as easily. (I've taken piano lessons for most of my life and I'm told I'm rather talented, although I'm nowhere near as good as professional musicians.)

Sometimes, an earworm gets attached to the point where I can't tell the difference between what's in my head and what I'm hearing with my ears. This usually happens when I've been playing a video game with MIDI-like music for a long period of time. (On a side note, I must have no taste, because I find I prefer the MIDI-like sounds of the NES and SNES-era to the more elaborate music of today's video games. The FF6 soundtrack is my favorite music, ever.)

There's a lot of great music that's gotten into videogames. Anything that people can listen to for hours on end and not get sick of must have some merit.

(Anyhow, the only true measure of taste is what people like years hence. And even supposedly great musicians can be unreliable predictors.)

I think a lack of aural imagination explains a lot of mediocre musicians who are beginners, and who stay beginners, in traditional music. They are only trying to waggle their fingers in the right magical sequence to get the tune to somewhat come out. They're not hearing the tune in their head and letting it come out.

I think clever people are especially susceptible to the belief that their perceptions are typical. Let's say you can't visualize images in your mind, but your coworker insists that he can. Since you're not a brain scientist, you can't verify whether he's right or whether he's just misinterpreted the question. However, the last few times you had a disagreement with him on a verifiable subject, you were vindicated by the facts, so you can only assume that you are right this time as well. Add to that the fact that people's stated perceptions and preferences are frequently dishonest (because of signaling), and it's very easy to mistrust them.

One useful first step to overcoming this bias is to compare one's results on a test like UVA's Moral Foundations Questionnaire here to other segments of the population.

However, it's not enough to just learn the facts about how other people perceive the world; sometimes one has to experience them firsthand. I have always been an ambitious high achiever and used to get frustrated and confused by people who were not able to follow through with their goals. However, a few years back I had an adverse reaction to a medication, and experienced for a few hours what depression must be like. From then on, it all made perfect sense.

One day I wonder if it will be possible to alter my brain chemstry safely and temporarily so that I can experience what it is like to perceive the world as a conservative, a liberal, a luddite, a woman, a blue collar worker, a depression sufferer, a jock, an artist, etc. The impact on my emotional maturity and ability to empathize would be tremendous.

This reminds me of some of the literature on fallibility of introspection. (If you have time only for one essay, read "The Unreliability of Naïve Introspection" and try the experiment with the playing card.)

As far as generalizing about an entire gender: It's extremely likely that I know a wildly unrepresentative sample of women, but why would you assume that the pickup artists don't? I imagine they meet vast numbers of women, but if they find them all at parties and clubs and bars, they're going to meet the kinds of women who go to parties and clubs and bars, not the ones who spend their time gardening at home or who go to all-women gyms to avoid being hit on or the ones who play D&D with their brothers in the basement. Even if their statements are accurate about that sort of woman (which I am not yet prepared to believe), that doesn't make them applicable to the entire gender, and the stereotype remains wildly inappropriate and offensive. If you're hearing things about men as a group that don't apply to you or any men you know, then chances are you're not hearing from someone who has a really ideal sample. If a female friend of mine complains about her sixth boyfriend in a row being a jerk, I don't conclude that men are jerks, I conclude that she has terrible taste.

...which I am not yet prepared to believe...
...wildly inappropriate and offensive...

Alicorn, are you applying the virtue of evenness, and searching equally for evidence for and against your conclusions? I mean, is your aim solely to get an accurate answer?

For myself, I find that phrases like “not yet prepared to believe” are a tip-off, when I notice them in my own thinking, that... I’m looking for permission to keep believing a pleasant, socially useful, or otherwise convenient belief, rather than really trying to figure out what’s true. I’m thinking “but the evidence doesn’t yet force me to change my mind, or at least I can see it that way!” instead of asking “what’s most likely to be true? what clues can I draw from the evidence? what models are most likely to help me make accurate predictions in the future?”.

Ditto for terms like “offensive”, if applied to peoples’ anticipations about the world (matters of truth and falsity) rather than to peoples’ non-belief actions. If what you mean by “offensive” is that you suspect folks’ beliefs here are stemming from emotional biases, it is okay to say that, and to explain the causes of your beliefs about their biases. If what you mean by “offensive” is that having statements like this around may make women uncomfortable, it is okay to say that, to explore why, and to start a dialog on how (without ceasing to seek accurate beliefs, but while perhaps taking special pains to include other facets of the story) we can make LW a comfortable place for women. But a belief’s “offensiveness” isn’t directly relevant to its truth, and so it’s confusing to include it in an argument about what’s true, or in an argument about what we should say and believe.

I agree that women and men sometimes vary (though I'd love a better model of the details). It isn’t really your conclusions I’m trying to talk about here; it’s how to talk about potentially mind-killing topics, as a community, in a way that helps true conclusions come to the fore.

They don't find them all at parties and clubs and bars. There's a whole raft of material on 'day game' - approaches in non-obvious places like bookshops, grocery stores, museums, the high street, etc. which are designed in part to reach women who are unlikely to be encountered in clubs and bars.

Not to mention that they're only talking about a specific subsection of a specific subsection, namely the women they are actually successful with. I'm assuming their batting average is well below .500, though I could be wrong. Thus, a small subsection of a small subsection of women conform to those particular stereotypes, or at least that's all you can say from that evidence.

Other examples suffer somewhat similar problems; all men may seem like chauvinistic jerks because chauvinistic jerks are quite noticeable and quite memorable. Thus, women may encounter more jerks because they get around more, rather than because most men are jerks.

Post is overall excellent, but some of those vaguely anecdotal counterexamples may well suffer from skewed reporting due to other biases.

The response I hear from most of the women I know is that this is complete balderdash and women aren't like that at all. So what's going on?

I think asking people directly is the wrong approach. Both men and women are good at rationalizing and you never hear someone admitting: "Yes, I'm an asshole." You really have to observe how people actually behave and the more I open my eyes I see that there is a lot of wisdom in the seduction community.

John T. Molloy once paid actors to go into bars and try to get women's phone numbers. One group of actors he asked to act confident. A second group of actors he asked to act arrogant. The actors asked to act arrogant were more successful. (Described in Molloy's 1975 book Dress for Success.)

Of course, as Alicorn says, the population of women who go to bars and talk to strange men might not be representative of all single women.

/me wonders what percentage of phone numbers received were fake

It struck me that I think you can still see the imagination debate playing out today. Consider the following conversation, which most people will have encountered a variant of at least once+:

-- Mr. Highbrow: It is better to read books than watch movies based on them. The movies limit you to someone else's perspective on the material, but the book gives maximum reign to your imagination.

-- Mr. Lowbrow: What are you smoking? The movie is an immersive experience that makes me feel like I'm really in the story. The book is just somebody else's description of the story.

Having thought about it, my highest-probability hypothesis is now that Mr. HB has more vivid mental imagery than does Mr. LB. Further introspection led me to realize that when I read fiction, I often have very specific images of places and scenery, but usually only vague impressions of faces. When I watch film adaptions, I'm often struck that the setting is "wrong," but rarely have that feeling about the appearance of people (unless the actors are grossly divergent from the description of them in the book).++

+The correct response to this

++ I considered putting this in the "How is your mind different" thread, but I don't know how typical or atypical I am. Which is, I suppose, the point.

This post completely takes the wind out of the sails of a post I was planning to make on 'Self-Induced Biases' where one mistakes the environment one has chosen for themselves as being, in some sense, 'typical' and then derives lots of bad mental statistics from this. Thus, chess fanatics will tend to think that chess is much more popular than it is, since all their friends like chess, disregarding the fact that they chose those friends (at least partly) based on a commonality of interests.

A worse case is when the police start to think that everyone is a criminal because that's all they ever seem to meet.

The women I ask about this are not even remotely close to being a representative sample of all women. They're the kind of women whom a shy and somewhat geeky guy knows and talks about psychology with.

This is why I find pickup theory so incomprehensible. It all seems to be aimed at people looking for sex in bars. I don't know anyone who does this (at least to my knowledge), so I have no mental model for how it works. I'm pretty sure the methods advocated would not work on me or most people I know, but I trust pickup artists to be right about how it works on people who hang out in bars.