koganbot | Entries tagged with daniel kahneman

This is an edited-down excerpt from a reply I made to Dave on the Elephant Call thread. In my edits I've taken out some of my sharp opinions because they aren't relevant to the point of this post, but by all means click the link, for the sharpness. And I'm going to sneak the actual point of this post down in the comments, so look there as well:

Suppose Sam says to Chris, "Let's get together for lunch. Are you free Thursday?" Chris replies, "I'm pretty much swamped for the next couple of weeks. How about after that?" Sam says, "Actually, as I think about it, I'm swamped too and Thursday was overoptimistic. Let's say around the end of the month." Chris: "That sounds good."

Now, I would say that they're each implying that they'd like to see the other, though they're also implying that they have more immediate (though not necessarily more important) priorities. Neither of these implications may be true, but each is definitely implied. Even if Sam and Chris are lying — perhaps they're secret embezzlers who plan to see each other the next afternoon to plan their latest chicanery, and the whole conversation is a charade to mislead potential undercover agents — they've nonetheless implied that though they want to see each other they have more immediate priorities.

I'd also say that this interchange reveals a hunk about Sam and Chris and their world. It may not reveal what they really want or need, but it tells me what they want to convey and the social forms they use to convey it. Of course, I myself know something about their world (we'll say it's contemporary America, and Sam and Chris are socially more or less like me). E.g., "lunch" isn't the same commitment as "dinner," the latter implying (again, not necessarily correctly) a stronger friendship.

Okay, we can ask questions of this interchange. For instance, "What sort of friendship will Sam and Chis end up having?" "What sort of friendship would Sam and Chris like to have?" "What sort of friendship should Sam and Chis have?" Is this the sort of thing you have in mind when you [i.e. Dave] use the phrase "implicitly posed question"? If so, "implicitly" is the wrong word, since neither Sam nor Chris implied the question. In some ways, their current and subsequent behavior may "answer" such questions, but that doesn't mean that Sam and Chris are either asking them or implying them. I'm the one who's asking them. I'd say the questions are there to be asked, simply because the world has a future and we can try to predict it, and, barring a sudden calamity or unexpected events, Sam and Chris are likely to at least have the opportunity to interact in said future. (I take it that neither is expected to be sentenced to prison in the next day or so.)

( We can ask similar questions about people in anime/VG fandom and people in rockwrite )

Current Music: Jefferson Airplane "Bear Melt"

Say I want to understand human financial behavior. The behavior is actually complex, but if I make the following simplifying assumptions I can at least have a place to start, and can make some calculations, predictions, etc.

(1) In the financial decisions (e.g., job choice, hiring, shopping, selling, investing) all people are trying to maximize their monetary profit or "financial" value and to cut their losses.

(2) Given the available information, people go about this pretty well (e.g., this apple at this store costs $2.00 a pound, the identical apple at that store costs $3.00 a pound, both stores are equally easy to get to and I'm going to both stores anyway for other reasons, so I buy the apple at the first store not the second).

(3) When things don't turn out so well, people modify their understanding of the information and (subject to the caveat in the footnote) they seek new, better information.*

We can say that, given assumption number 1, in doing numbers 2 and 3 people are being rational and that when people don't do 2 and 3 they're being irrational. But this all rests on the simplifying assumption number 1, that in all their financial decisions they're trying to maximize income or financial value etc. But to want to maximize money and financial value in the first place is neither rational nor irrational. And we know, or ought to know, that number 1 in itself is not altogether true, that maximizing profit is not the only motive in play: we posited it as a simplifying assumption so we could get a grip on economic behavior. Motives — such as loss aversion and brand loyalty — that run counter to profit maximizing are no more irrational (or rational) than maximizing profit is, and are no more or less emotional either.**

This is on my mind for two related reasons.

First, a couple of Paul Krugman posts about the parts of macroeconomics that have no "microfoundations" but nonetheless seem to describe macro results better than the alternatives: actually, I'm as ignorant of microeconomics as macro, but I do think that micro has simplifying assumptions that include something like my numbers 1 through 3 above, and that's exactly (or maybe not exactly) why the macro that's micro-based can't explain sticky wages and so on. (I don't pretend to understand the posts to any depth, by the way. But note the word "hyperrational" in the first post and "rational expectations" in the second, which I'm guessing mean that it's taken for granted by some people that assumption number 1 is in itself rational.)

Second, I recently got Daniel Kahneman's highly worthwhile Thinking, Fast And Slow from the library for the second time (I didn't finish it 11 months ago when it first became due), and I think if asked he would, or at least ought to, subscribe to my paragraph above beginning "We can say that..." But actually his language slips a lot, and "rational" and "emotions" sometimes float by in his text without explaining themselves. (I do believe he talks explicitly about at least one of them in the part I haven't read yet; but that doesn't mean he knows what he's doing when the words creep in earlier.)

[Error: unknown template video]

( Footnotes )

Current Music: The Kinks "Shangri-La"

Proposal for a social psychology experiment:

We'll use four separate, sizable groups of people, say 75 people in each group. (Not that I know if that amount is any good or not, or if we want our overall pool to be similar socioeconomically. I'm not a statistician.)

Ask each member of Group One:

What arguments would you use to try and persuade an atheist to consider that there might be a God after all?

[It's likely that at least a few people in each group will be atheists, but that's no reason they shouldn't try to answer the question.]

Ask each member of Group Two:

Cheryl tells you she is an atheist. What arguments would you use to try and persuade her there might be a God after all?

We're trying to see if by giving our atheist a name, so a potential personal, individual history, we elicit responses here and there that are different in type from what we generally got in Group One.

Ask two questions of each member of Group Three:

Group Three Question 1: Cheryl says she is an atheist. What arguments would you use to try and persuade her there might be a God after all?

It's important that the subjects complete the first question before seeing the second.

( The crucial question is under the cut )

( How to have fun in groups )

( Four hypotheses )

( Fishing expedition )

Current Music: Trina ft. Nisha "Hell Yeah"

Another one! Nate Silver cites Kuhn in a footnote, Silver probably** being unaware that his own passage (Nate Silver, The Signal And The Noise, p. 260) not only runs opposite to a couple of Kuhn's major ideas, and not only isn't in the same ballpark as Kuhn, it's barely in the same sport. Again, I'm not giving you the answer, this being a quiz:

The notion of scientific consensus is tricky, but the idea is that the opinion of the scientific community converges toward the truth as ideas are debated and new evidence is uncovered. Just as in the stock market, the steps are not always forward or smooth. The scientific community is often too conservative about adapting its paradigms to new evidence,⁶⁴ although there have certainly also been times when it was too quick to jump on the bandwagon. Still, provided that everyone is on the Bayesian train,* even incorrect beliefs and quite wrong priors are revised toward the truth in the end.

*And that they don't hold priors that they believe to be exactly 100 percent true or exactly 0 percent true; these will not and cannot change under Bayes's theorem.

64. Thomas S. Kuhn, The Structure of Scientific Revolutions (Chicago: University of Chicago Press, Kindle edition).

A couple of hints:

(1) Incommensurability
(2) Darwin

But this passage is a botch in whole hunks of other ways as well, e.g., the word "the" in the phrase "the scientific community."

Look, I've read enough philosophy to know that Kuhn is not hard, though he vagues out too much and he leaves some difficult problems in his wake. That near everybody gets him wrong isn't due to a fundamental ideological barrier or to any drastic unfamiliarity/novelty in his concepts.

( More griping )

**"Probably," since I don't know how much of Structure he read, and I myself had only read about half my nephew's copy of the Silver book, skipping around, before it was time to fly back to Denver.

Current Music: Bob Dylan "Desolation Row"

Here are the worst five sentences from what's otherwise a pretty good book. The sentences are in no way essential to the book, and didn't need to be there. So I'm just giving you the sentences without the book title. My point in printing them is that most everybody gets Kuhn wrong. There's a mass mental block.

Historians of science have often noted that at any given time scholars in a particular field tend to share basic assumptions about their subject. Social scientists are no exception; they rely on a view of human nature that provides the background of most discussions of specific behaviors but is rarely questioned. Social scientists in the 1970s broadly accepted two ideas about human nature. First, people are generally rational, and their thinking is normally sound. Second, emotions such as fear, affection, and hatred explain most of the occasions on which people depart from rationality.

That passage doesn't mention Kuhn or Feyerabend as his "historians of science," but if the author wasn't thinking of either of those two — but he likely was! — he was thinking of someone else who was thinking of them. In any event, if you think you know something about Kuhn, and that passage doesn't strike you as way wrong, you gotta go back and read Kuhn again (or at least click the Thomas Kuhn tag and read our discussion).

I will say a little about the two "broadly accepted" ideas, since they're not particularly relevant to my Kuhn quiz: there were still Marxists and Freudians* running about in the 1970s, and whatever they did or didn't believe regarding the soundness of human thinking, they most definitely would not have considered the phrase "emotions such as fear, affection, and hatred" to be at all adequate to what's going on in ideology and oedipal dramas. (But that's a side issue.)

(I imagine that someone reading this might say to herself, "Frank falls into the category 'somebody'; so if everybody misunderstands Kuhn, Frank too must misunderstand Kuhn." Well, I think there's a way that I veered wrong in the past. But I think I've now substantially got the guy right. May be a subject for a future post, what I got wrong.)

*Yeah, I know the passage uses the word "most," and Marxists and Freudians were never the majority of social scientists. But the word "most" is one of the very features that cause the passage to careen off into wrongness.

(Also don't know if Feyerabend is considered a philosopher or a historian, but he definitely knew plenty about the history of science, whatever field he was officially in.)

Current Music: Mott The Hoople "Violence"

The greatest challenge in understanding the role of randomness in life is that although the basic principles of randomness arise from everyday logic, many of the consequences that follow from those principles prove counterintuitive.... In the mid 1960s, [Daniel] Kahneman, then a junior psychology professor at Hebrew University, agreed to perform a rather unexciting chore: lecturing to a group of Israeli air force flight instructors on the conventional wisdom of behavior modification and its application to the psychology of flight training. Kahneman drove home the point that rewarding positive behavior works but punishing mistakes does not. One of his students interrupted, voicing an opinion that would lead Kahneman to an epiphany and guide his research for decades.

( regression toward the mean )

The issue of regression to the mean is interesting in itself, and it's the motive for Mlodinow's anecdote, but I'd like to focus on the claim of behavioral psychology, that rewarding good behavior works but punishing bad behavior doesn't. Is this true? If so, what do I do with this principle? How do I apply it? On my mind today is that, as I've often said ~~in a punitive tone of voice~~, music critics don't know how to sustain an intellectual conversation. And my assumption is that I'm not really going to have many sustained intellectual conversations unless I and people like me teach others how to do it. More immediately, I'm wondering if there's a way to have an impact on the gross dysfunctional behavior that sinks a lot of music discourse - a current example is the stupid commentary at Jezebel and Autostraddle about Taylor Swift, which Alex O. and Erika do a good job of taking apart. Basically, Autostraddle and Jezebel project a virgin-whore dichotomy onto Taylor that Taylor's actual words and behavior don't support at all, then excoriate Taylor for perpetuating the virgin-whore dichotomy. But the real dysfunction in criticism isn't the making of a false inference on the basis of too-little evidence and being too thoughtless to look for further evidence or to notice what contradicts the inference - who doesn't do that at some point (and to be honest I only skimmed the Autostraddle piece myself)? - but rather what comes after, the inability of the overall conversation to take care of this, the many voices being unable to make up for the limitations of the single voice.

( Further reflections )