Page 1 of 1

Thinking, Fast and Slow

Posted: Mon Oct 15, 2012 4:13 pm
by nafod
Book by Nobel Prize winner Daniel Kahneman. If you are at all curious about what goes on in your noggin, this book is a massively fascinating read.

I've read Part 1 which covers the dichotomy in your head, where he divides it into System 1 and System 2. System 1 is the always busy, intuitive part of the brain. System 2 is the part that does the real cognition, and does things like calculate hard math problems. System 2 is lazy. System 1 hates inconsistency but has no problem with incomplete data, and is with fine stopping asking questions once it has the answers it wants. System 2 has finite resources. System 1 is constantly computing what it can, and in fact more than is asked for. You can exactly ascertain just how hard System 2 is working by tracking the dilation of a person's pupils, and can tell when they are done or have quit a computation before they tell you. Etc.

Lots of great tidbits, all of it grounded in experiments and empirical observation either he or other leading psychologists have observed. A lot of it is surprisingly old research (70s and 80s).

Well-written too. Easy read.

Lots of others have written about it. Gets its own Wiki page.
http://en.wikipedia.org/wiki/Thinking,_Fast_and_Slow

Re: Thinking, Fast and Slow

Posted: Tue Oct 16, 2012 7:51 am
by JonnyCat
I'll have to give this a go. I love books like this.

The book, You are Not So Smart, takes on some similar topics and is a good read, as well.

Re: Thinking, Fast and Slow

Posted: Tue Oct 16, 2012 1:34 pm
by nafod
One of the interesting little tidbits is the lack of statistical intuition of humans. An example he gave is of the Bill & Linda Gates foundation dumping billion+ dollars into small schools, as the statistics showed that the better schools tended to be small schools.

But the statistics also showed that the worst schools also tended to be small schools, i.e., smaller schools had a greater statistical variance.

Another example was with regards to flight training, and instructors who yell. The author tried to explain to them that positive teaching vice screaming would get better results. The flight instructors disagreed, saying that when they praised, the students often did worse on the next flight, while when they yelled at a poor performance, the students almost always did better. The author showed them that that could be explained by the fact that people's performance varies, and so a bad performance will almost always be followed by something closer to the mean average for them, i.e., improving, and similarly for a good performance.

The section on anchoring is also cool. In one of the experiments, they took Realtors (experts at appraising value, theoretically) and showed them a house, then a price (not the listed price, just a number) and asked them two questions...
1. Is the price you would list it at higher or lower than this number?
2. What would you list it at?

In the experiment, they used two different numbers, one high and one low, for two different groups of realtors. What they found was that the number used in the first question had a HUGE impact on the answer to the second. If the difference between the low and high number was some value X, the estimates in the two groups maintained about 55% of that difference. This, by experts who swore up and down they were being totally objective. It is called anchoring effect in estimates. It is ubiquitous.

Re: Thinking, Fast and Slow

Posted: Tue Oct 16, 2012 2:11 pm
by Pinky
nafod wrote:An example he gave is of the Bill & Linda Gates foundation dumping billion+ dollars into small schools, as the statistics showed that the better schools tended to be small schools.

But the statistics also showed that the worst schools also tended to be small schools, i.e., smaller schools had a greater statistical variance.
This is similar to small businesses and job creation/destruction. Small businesses are not good at creating jobs if you count the jobs lost when the business closes.

Re: Thinking, Fast and Slow

Posted: Sun Mar 24, 2013 9:40 pm
by DrDonkeyLove
I decided that the book was longer than I wanted to read right now so I picked up an abridged version for Kindle. I've gone through it a couple of times and will go through it more. A lot's been discussed here but the section on choices particularly resonated with me. I've known for a long time that the aphorism "Fear of loss is greater than the desire for gain" is true, now I have a better understanding of why.

There's a lot of discussion about probability algorithms vs. the gut based predictions of experts. Lo and behold David Brooks of the New York Times had an article on some Kahneman's forecasting ideas put to the test and the results are impressive. I couldn't find it online but it's worth hunting down when it's posted if you're interested in this subject.

Re: Thinking, Fast and Slow

Posted: Mon Mar 25, 2013 5:46 pm
by TerryB
nafod, you read too many books

Re: Thinking, Fast and Slow

Posted: Wed Mar 27, 2013 8:47 am
by JimZipCode
Wife liked this last year. It moved from her bedside table to mine, but I haven't dived deeply into it yet. (Still plowing thru Civil War books.)

She also liked You Are Not So Smart.

Re: Thinking, Fast and Slow

Posted: Thu Mar 28, 2013 1:07 pm
by Turdacious
DrDonkeyLove wrote:I decided that the book was longer than I wanted to read right now so I picked up an abridged version for Kindle. I've gone through it a couple of times and will go through it more. A lot's been discussed here but the section on choices particularly resonated with me. I've known for a long time that the aphorism "Fear of loss is greater than the desire for gain" is true, now I have a better understanding of why.

There's a lot of discussion about probability algorithms vs. the gut based predictions of experts. Lo and behold David Brooks of the New York Times had an article on some Kahneman's forecasting ideas put to the test and the results are impressive. I couldn't find it online but it's worth hunting down when it's posted if you're interested in this subject.
This one? http://www.nytimes.com/2011/09/16/opini ... llacy.html

Re: Thinking, Fast and Slow

Posted: Thu Mar 28, 2013 4:06 pm
by DrDonkeyLove
Turdacious wrote:
DrDonkeyLove wrote:I decided that the book was longer than I wanted to read right now so I picked up an abridged version for Kindle. I've gone through it a couple of times and will go through it more. A lot's been discussed here but the section on choices particularly resonated with me. I've known for a long time that the aphorism "Fear of loss is greater than the desire for gain" is true, now I have a better understanding of why.

There's a lot of discussion about probability algorithms vs. the gut based predictions of experts. Lo and behold David Brooks of the New York Times had an article on some Kahneman's forecasting ideas put to the test and the results are impressive. I couldn't find it online but it's worth hunting down when it's posted if you're interested in this subject.
This one? http://www.nytimes.com/2011/09/16/opini ... llacy.html
I think that's the follow up piece. I was referencing one called "Forecasting Fox". It's posted now but I can't grab it because I've exceeded my allotted number of free NYT articles.

Re: Thinking, Fast and Slow

Posted: Thu Mar 28, 2013 4:42 pm
by nafod
One of the later chapters talked about who you are making decisions for, your current self versus your future self. Some example questions

- If we told you that you could have a medical procedure that would leave you screaming in pain while it was going on, but that we could erase your memory afterwards so you'd have no memory of it whatsoever (versus a less painful one that you'd remember), would you choose this? Lots say yes.

- Similarly, if you could go on a fabulous vacation that would be 24/7 rip-roaring continuous fun but you'd not remember it afterward, would you do it? Most said no.

They found that memory of a pain event is interesting. They found that the memory of the pain in an event is the average of the worst instantaneous pain experienced and last pain experienced before the event stopped. What they expected was that the memory of the pain would be the pain level multiplied by duration. It wasn't.

So they did an experiment where they had guys put their hand in painfully cold water for some period of time (I forget...5 minutes?) and then take it out. A second group did the same thing, but then kept their hand in the water while they warmed it up so it was less painful, and kept it there for another 5 minutes. The second group reported less pain for the event than the first, even though they experienced the same pain level and duration as the first group plus some more. They made their decision on which future experiment to participate based on that, so more chose the second than the first when asked to repeat the experiment and given the option to choose which group to be in.

Re: Thinking, Fast and Slow

Posted: Thu Mar 28, 2013 4:45 pm
by nafod
protobuilder wrote:nafod, you read too many books
Thank you!

Re: Thinking, Fast and Slow

Posted: Fri Mar 29, 2013 12:51 am
by seeahill
I'm sorta confused about this book. (Only 1/3 through it.) Everything seems entirely obvious to me.

Re: Thinking, Fast and Slow

Posted: Fri Mar 29, 2013 12:27 pm
by nafod
seeahill wrote:I'm sorta confused about this book. (Only 1/3 through it.) Everything seems entirely obvious to me.
That's the section on thinking slow

Re: Thinking, Fast and Slow

Posted: Fri Mar 29, 2013 4:25 pm
by TerryB
nafod wrote:
seeahill wrote:I'm sorta confused about this book. (Only 1/3 through it.) Everything seems entirely obvious to me.
That's the section on thinking slow
=D>