![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Thinking, Fast and Slow
Jun. 18th, 2013 04:01 pmI am reading Thinking, Fast and Slow, by Daniel Kahneman, and I like it so far, but I have reached one particular part in the discussion of risk aversion, and particular, the difference in how risk-averse people are with prospects of winning something, vs. with prospects of losing something.
Kahneman presents how his prospect theory views this choice in terms of the reference point, an improvement over Bernoulli's utility theory. He then goes on to say most people never notice the gift of $1000 or $2000 that they've been given - it's included in most people's reference points. People see this in terms of winning money in Problem 3, and losing money in problem 4, and, as he had demonstrated earlier in the chapter, and verified through studies, when all of the choices are bad - for example, when all the choices involve losing SOMETHING - then people are more inclined to go ahead and take the gamble.
But there's where apparently I am way different from most people. That I had been given money was the first thing I noticed, and how I immediately thought of both problems is that hey, there are no losses here - in no case am I not winning at least $1000! Problem 4 is as much about winning as problem 3 is, despite the misleading wording of losses. For me, I interpreted problem 3 as "I definitely win at least $1000 no matter what; I have a 50% chance of winning $2000 instead, or a certainty of winning $1500. So I picked the gamble, because hey, even if I lose the so-called gamble, I have still won $1000! And in problem 4, I interpreted it as "I definitely win $1500 if I choose the second option, or I have a chance that I only win $1000 - of course I'm going to take the definite win of $1500!" In other words, because I saw that I win $1000 no matter what I do, my perceptions of which gamble was worthwhile was exactly the reverse of most people's.
This isn't to say that Kahneman's prospect theory is as wrong as Bernoulli's utility theory - but it is to say that it's less complete than Kahneman thinks it is, because there are people who do see that their "reference point" is what they had before they were ever offered the opportunity to gamble, and that a gift given to you as soon as you agree to participate, before you even make any choices, does not change the starting point - the real gamble here is deciding whether to accept the offer to participate in the study in the first place! Once you make that decision, you have already won, if you decide to participate, or not-won (lost an opportunity, though not lost any real money) if you decide not to participate.
One of the things this shows is that studies based on academic settings with limited and fixed choices are never going to elicit the kinds of behavior that people will exhibit when they actually have money in hand and a real-world setting - and furthermore, which Kahneman should certainly have perceived, based on lots of earlier stuff in this book, how they have been primed by prior life experience that is not under the control of the experimenter, usually not even known by the experimenter nor does the experimenter even realize that he should somehow find about about those things. A person who has spent time previously in their life thinking about risk aversion, about gambling, and about decision theory, even in an amateur way, is primed to read these problems differently than someone who has never spent any particular amount of time thinking about risk and gambling. Even if asked to make a quick decision, these people will be more inclined to think that thinking it through is FUN rather than work, in that it's pursuing a personal interest, and so will expend the effort to think about it for a few seconds even if not encouraged to do so.
Am I wrong here?
The observation of contrasting attitudes to risk with favorable and unfavorable prospects soon yielded a significant advance: we found a way to demonstrate the central error in Bernoulli's model of choice. Have a look:
Problem 3: In addition to whatever you own, you have been given $1000. You are now asked to choose one of these options: 50% chance to win $1000 OR get $500 for sure
Problem 4: In addition to whatever you own, you hav e been given $2000. You are now asked to choose one of these options: 50% chance to lose $1000 OR lose $500 for sure.
You can easily confirm that in terms of final states of wealth - all that matters for Bernoulli's theory - problems 3 and 4 are identical. In both cases you have a choice between the same two options: you can have the certainty of being richer than you currently are by $1500, or accept a gamble in which you have equal chances to be richer by $1000 or by $2000. In Bernoulli's theory, therefore, the two problems should elicit similar preferences. Check your intuitions, and you will probably guess what other people did.
- In the first choice, a large majority of responders prefered the sure thing.
- In the second choice, a large majority preferred the gamble.
Kahneman presents how his prospect theory views this choice in terms of the reference point, an improvement over Bernoulli's utility theory. He then goes on to say most people never notice the gift of $1000 or $2000 that they've been given - it's included in most people's reference points. People see this in terms of winning money in Problem 3, and losing money in problem 4, and, as he had demonstrated earlier in the chapter, and verified through studies, when all of the choices are bad - for example, when all the choices involve losing SOMETHING - then people are more inclined to go ahead and take the gamble.
But there's where apparently I am way different from most people. That I had been given money was the first thing I noticed, and how I immediately thought of both problems is that hey, there are no losses here - in no case am I not winning at least $1000! Problem 4 is as much about winning as problem 3 is, despite the misleading wording of losses. For me, I interpreted problem 3 as "I definitely win at least $1000 no matter what; I have a 50% chance of winning $2000 instead, or a certainty of winning $1500. So I picked the gamble, because hey, even if I lose the so-called gamble, I have still won $1000! And in problem 4, I interpreted it as "I definitely win $1500 if I choose the second option, or I have a chance that I only win $1000 - of course I'm going to take the definite win of $1500!" In other words, because I saw that I win $1000 no matter what I do, my perceptions of which gamble was worthwhile was exactly the reverse of most people's.
This isn't to say that Kahneman's prospect theory is as wrong as Bernoulli's utility theory - but it is to say that it's less complete than Kahneman thinks it is, because there are people who do see that their "reference point" is what they had before they were ever offered the opportunity to gamble, and that a gift given to you as soon as you agree to participate, before you even make any choices, does not change the starting point - the real gamble here is deciding whether to accept the offer to participate in the study in the first place! Once you make that decision, you have already won, if you decide to participate, or not-won (lost an opportunity, though not lost any real money) if you decide not to participate.
One of the things this shows is that studies based on academic settings with limited and fixed choices are never going to elicit the kinds of behavior that people will exhibit when they actually have money in hand and a real-world setting - and furthermore, which Kahneman should certainly have perceived, based on lots of earlier stuff in this book, how they have been primed by prior life experience that is not under the control of the experimenter, usually not even known by the experimenter nor does the experimenter even realize that he should somehow find about about those things. A person who has spent time previously in their life thinking about risk aversion, about gambling, and about decision theory, even in an amateur way, is primed to read these problems differently than someone who has never spent any particular amount of time thinking about risk and gambling. Even if asked to make a quick decision, these people will be more inclined to think that thinking it through is FUN rather than work, in that it's pursuing a personal interest, and so will expend the effort to think about it for a few seconds even if not encouraged to do so.
Am I wrong here?
