The Intuition
<rewatches Jewel's Intuition video>
I love it when second-tier (or first tier, I suppose) artists make fun of the stupid pop idols. But that's not the point of this post.
Daniel Kahneman came to Google today to give a talk on "The Marvels and Flaws of Intuition". It was really good, sounded right (hehe, intuitively, anyway).
Hehe, apparently they tried using some algorithm to do things human "experts" do (like evaluate how qualified people are, predict politics) and the algorithm did a lot better than the experts in the field (recruiters, pundits) when it came to anything beyond a short-term. The big things were that the experts weren't consistent - same data yielded different answers, and that the experts didn't learn from their mistakes.
Basic overview: there are two ways judgements and decisions are made, method 1 - intuitively or method 2 - deliberatively.
Basically, most decisions actions are made via method 1 with a little check by method 2. He had a lot of good examples:
A bat and a ball together cost $1.10. The bat costs one dollar more than the ball. How much does the ball cost?
The answer, of course is 05 cents. But most people say 10 cents. The idea is that your intutition substitutes a "similar" operation (substitution) for the more complex algebraic one, and that without the method 2 check overriding and correcting it, the wrong answer is given. Most people don't check this sort of answer. I imagine a lot of this check being used is related to what field someone is in - a CS or Math person is more likely to not say 10. Though, they will think it first, they will correct themselves before claiming it as the answer.
Other oddities:
At the airport you are offered an insurrance plan:
Chess masters have trained their intuition (apparently the time has been determined experimentally to be about 10,000 hours before this happens) to just intuit positions and gameplay. They don't have to think about it the same way as us lesser folks.
The substitution of a similar problem for the one at hand I find most interesting. That and the fact that some things just don't get checked, and some people in general don't check most of their intution (afterwards, and off-the-record, he did say that man and women differ a lot in how strong their checks are and what they check). I'm trying to be aware of when I switch in a different problem for the one at hand, but I suspect detecting this sort of subconsious behavior is an inherently difficult metacognitive task.
I love it when second-tier (or first tier, I suppose) artists make fun of the stupid pop idols. But that's not the point of this post.
Daniel Kahneman came to Google today to give a talk on "The Marvels and Flaws of Intuition". It was really good, sounded right (hehe, intuitively, anyway).
Hehe, apparently they tried using some algorithm to do things human "experts" do (like evaluate how qualified people are, predict politics) and the algorithm did a lot better than the experts in the field (recruiters, pundits) when it came to anything beyond a short-term. The big things were that the experts weren't consistent - same data yielded different answers, and that the experts didn't learn from their mistakes.
Basic overview: there are two ways judgements and decisions are made, method 1 - intuitively or method 2 - deliberatively.
| Intuition | Deliberation |
| Fast/Instant Accessibility | Slow Accessibility |
| Presented with one solution | Presented with many posible solutions |
| Emotion-based | Reason-based |
| Slow to train and learn | Fast to learn |
| Effortless | Effortful |
| Emotion-based | Reason-based |
| Slow to train and learn | Fast to learn |
| Present perceptions only | Past, present, future perceptions and concepts |
| ... | ... |
Basically, most decisions actions are made via method 1 with a little check by method 2. He had a lot of good examples:
A bat and a ball together cost $1.10. The bat costs one dollar more than the ball. How much does the ball cost?
The answer, of course is 05 cents. But most people say 10 cents. The idea is that your intutition substitutes a "similar" operation (substitution) for the more complex algebraic one, and that without the method 2 check overriding and correcting it, the wrong answer is given. Most people don't check this sort of answer. I imagine a lot of this check being used is related to what field someone is in - a CS or Math person is more likely to not say 10. Though, they will think it first, they will correct themselves before claiming it as the answer.
Other oddities:
At the airport you are offered an insurrance plan:
Plan A pays $50,000 in case of death by terrorism.Apparently, people are afraid of terrorism so they automatically pick the first one without doing the inclusion operation (I'm not at all afraid, hehe). Interestingly enough, when he asked the audience, every Googler I heard said "B", then he continued, "of course, the obvious answer is A... [here's why they are wrong...]".
Plan B pays $50,000 in case of death for any reason.
Chess masters have trained their intuition (apparently the time has been determined experimentally to be about 10,000 hours before this happens) to just intuit positions and gameplay. They don't have to think about it the same way as us lesser folks.
The substitution of a similar problem for the one at hand I find most interesting. That and the fact that some things just don't get checked, and some people in general don't check most of their intution (afterwards, and off-the-record, he did say that man and women differ a lot in how strong their checks are and what they check). I'm trying to be aware of when I switch in a different problem for the one at hand, but I suspect detecting this sort of subconsious behavior is an inherently difficult metacognitive task.
posted by Alden | 03:43
0 Comments:
Post a Comment
<< Home