Friday, August 05, 2005

Other Brain Damaged States of Interest

PBryon writes: "One of the other major themes of "Blink" is that when these snap decisions are made, that those making them can't really put their finger on why they made them. It just gut instinct, without much to back it up.
Doesn't that fit our President too?


Yes, I would say that Bush does rely on gut instinct, especially about the people he chooses. And there is a large discussion in Blink of two psychologists who have characterized some 200 different facial expressions and what each reveals. One of these psychologists, who was a fan of Clinton in 1992, saw him on TV at the convention and pegged him for a liar right away based upon certain facial tics. He actually contacted the campaign to offer his services to help Clinton look less like a liar; the campaign, althought very interested (they knew their man), turned him down because they thought it would be more damaging to the campaign if people found out they were consulting a psychologist to make Clinton look truthful. Now, that psychologist was clearly an expert, but the stories related to his work reveal two things of interest: people vary in their ability to read facial language (autistics, for example, cannot discern any body language, and rely upon literal understanding of words alone), and people convey emotions via their facial expressions that they are not consciously trying to convey (like Bush in the first debate).

Anonymous writes: "The supposition in “Blink” as I read your Blog, is that decisions made without basis can be as often correct as those made with careful thought and a justifiable basis. I did not read the book but did they actually test the decisions made without consideration of facts versus those made with consideration of facts? Does this analysis only apply to decisions that can only be evaluated after the fact on a subjective basis?"

No, the blink supposition applies to opinions that can be factually refuted. In fact, the book begins with a discussion of a statue purchased by the Getty Musuem in California for $10 million. They had a team of scientists trying to verify its authenticity for a year prior to buying, but tragically they did not consult any art experts. There were a number of such experts who, upon laying eyes on the statue for the first time, declared it a fake immediately. But, again, one of the other themes, illustrated well with the psychologists who grew more confident but no more accurate in their prognisises as they gained more information is that at some point you can get information overload, and become supremely confident in a decision that is worse than it would be if you had less information.

Professor Vic writes: "One would never suggest that medical research be conducted on the basis of "I just feel this drug should cure cancer." Similarly, the Bush Administration's gut feeling that Sadaam was a threat due to WMDs turned out to be flat out wrong and they would have been better served to continue to collect objective information about the weapons. Maybe in subjective decision making where there is no strict ordering of what is best, for example who should replace Greenspan, this Blink principle works, but I would suggest it is completely ineffective in many other applications."

The "blink" title is slightly misleading, in that the book deals more generally with the mistaken notion that more information is always better. A useful analogy might be an economist using a regression analysis to predict some variable, and including every conceivable independent variable as a predictor. With so many irrelevant variables thrown in, the predictive power of the regression goes down. In fact, one of the stories of the book pertains to the efforts of Cook County hospital to manage the flow in its ER.

Most ER patients are there because they suspect they are having a heart attack. The resources required to do a bunch of tests and to understand the patients history and habits of the patient are enormous, beyond the constraints of the hospital. They ended up testing whether four objectively measured factors, scored in a certain way, would provide superior results to the usual work-up. They ended up reducing type I (false positive) errors and type II (false negative) errors by making their doctors ignore their typical training.

As for the WMDs, unless you call decades of intelligence gathering by most Western nations equivalent to Bush's gut feelings, the analogy hardly applies. Intelligence information, by its very nature, will never be subject to an academic level of scrutiny for its accuracy - that can only happen after we've acted on our understanding of that intelligence, as we did. There is a chance, as seems to have been true with Iraq, that the intelligence ends up being biased or incorrect, but we would have never known that in all likelihood if we hadn't gone in.

Now, for some other fun brain damaged problems I learned from a book called The Midnight Disease, which talks about the neuroscience of writing, and how I might improve my own writing via depression, epilepsy, and hallucinogens. I leave it to you readers to apply these problems to your understanding of contemporary politics.

First, Wernicke's aphasia: Wernicke's aphasia is caused by damage to an area (Wernicke's area) of the temporal lobe of the brain. The speech of someone with Wernicke's aphasia is at least partially nonsensical, but those with the condition are generally unaware that they don't make sense. An example of the speech is as follows:

...oh hear but that was a long time ago that was when that when before I even knew that much about this place although I am a little suspicious about what the hell is the part there is one part scares, uh estate spares, OK that has a bunch of drives in it and a bunch of good googin, nothing real big but that was in the same time I coached them I said hey stay out of the spear struggle stay out of trouble so don't get and my kids, uh, except for the body the boys are pretty good although lately they have become winded or something... what the hell... kind of platz goasted klack...

It reminds of that jailhouse educated character that Daymon Wayons played in Living Color, where he talks with these very complex juxtaposed words that make absolutely no sense, but he delivers his soliloquies as if he makes perfect sense. Kind of like Jesse Jackson.

Broca's aphasia is caused by damage to an area of the frontal lobe, and leads to difficulty with normal use of langauge, either spoken or written. Broca's aphasia patients can comprehend language much better than they can produce it, struggling with each word. They have more trouble with syntax and grammar than with semantics. Here is a patient with Broca's aphasia telling the story of Cinderella:

Patient: Cinderella ... poor ... um 'dopted her ... scrubbed floor, um, tidy ... poor, um ... 'dopted ... Si-sisters and mother ... ball. Ball, prince um, shoe ...

Examiner: Keep going.

Patient: Scrubbed and uh washed and un ... tidy, uh, sisters and mother, prince, no, price, yes. Cinderella hooked prince. Um, um, shoes, um, twelve o'clock ball, finished.

Examiner: So what happened in the end?

Patient: Married.

3 Comments:

Blogger Victor Matheson said...

There's a difference the two ideas you describe in Blink. One idea suggests that snap decisions are flat out better than ones where more information is collected. This is a really interesting claim and appears to be backed up by anecdotes like the Getty Museum story.

The other idea is that collecting more information sometimes doesn't give you enough additional new knowledge to make it cost effective. This is the idea in the Cook County example. This is not a novel idea at all. The economics term is "bounded rationality," i.e. sometimes it's rational to make decisions based on incomplete information. It's why you don't shop at 20 different stores for apples looking for a better price.

Certainly the second concept applies in nearly every circumstance, but again, I'm not so sure about the first concept in many cases.

7:28 AM  
Blogger Hatcher said...

You're absolutely right. As I said, "Blink" is not really an accurate title for the book, and I agree, there is nothing novel about the notion that more information is not always better. But I wouldn't necessarily say that this is an example of bounded rationality - which I think of as an issue of cutting your information gathering due to how costlt that can be.

And, there are many examples where truly "blink" decisions are bad. Also, some are better at it then others, and circumstances matter. For example, your ability to make good blink decisions goes to pot if your heart rate is high due to the stress of the situation. He talks about the Diallo shooting in the Bronx from this perspective.

7:35 AM  
Blogger pbryon said...

Back from a few days of watching baseball in Kansas City.

You didn't mention the other very interesting part of Blink--the whole Middle East war planning tabletop exercise. In short, grizzled retired general (playing rogue dictator) destroys current military planners (playing themselves), using unconventional (i.e. non-predicted) plans.

They re-did the whole thing later, with the rogue dictator under severe constraints about what he could do. Then the planners won, and they called the tabletop exercise a success.

I can see where you might think of Bush in the first, gut instinct camp, but I wonder how many of those war planners have his ears or are making decisions about pretty important stuff.

6:52 AM  

Post a Comment

<< Home

Sign up for my Notify List and get email when I update!

email:
powered by
NotifyList.com