[From Bruce Gregory (2004.01.20.1037)]
On the off chance that the following from today's NYT might be of some relevance:
"The new research stands out because it offers a detailed window into how the Bayesian thought process works, showing the point when uncertainty becomes great enough to give past experience an edge over current observation.
Each participant in the experiment sat down and placed a hand on a tabletop. A projection of a computer screen blocked their view of the hand. The goal was to guide a cursor, which followed the movement of the hand, from one side of the screen to a target on the other side.
Adding to the uncertainty, the cursor usually appeared slightly to the right of the hand, and the participants caught at most a quick glimpse of it when it was halfway across the screen. Sometimes, the cursor appeared as a discrete point; other times, it was an ill-defined cloud.
The researchers found that when no cursor flashed, people relied on what they had learned during 1,000 practice runs before the experiment: namely that the cursor was, on average, one centimeter to the right of the hand. When a cloud flashed, they considered it, but only somewhat, in a pattern that followed what Bayes's formula predicted. When a distinct cursor flashed, they relied on it and not past experience."
Bruce Gregory
Bruce,
"On the off chance" you say. It makes sense to me. What else would you expect? But, again "On the off chance" such as the possiblity that I might be mis-interpreting what the experiment means, and embarass myself by walking off a cliff or something else that was suggested by a BAyesian analysis, or foolish misinterpretation thereof, could you explain it a bit more fully? Is there a psychological/ evolutionary basis for this-- silly question I suppose.
If you would be so generous as to provide a non-sarcastic clarification/ expansion , I faithfully promise, that under no circumstance will I say that your explaination "needs work."
Bill Williams
[From Bruce Gregory (2004.01.21.0930)]
If you would be so generous as to provide a non-sarcastic
clarification/ expansion , I faithfully promise, that under no
circumstance will I say that your explaination "needs work."
I though there was some chance that this result would be of interest to
the PCT modelers. As to what its significance might be, I leave that to
the experts.
Bruce Gregory
Everyone is entitled to his or her own opinions; everyone is not
entitled to his or her own facts."
Daniel Patrick Moynihan
"Everything that needs to be said has already been said. But since no
one was listening, everything must be said again."
Andre Gide
"Everyone is entitled to his or her own opinions; everyone is not
entitled to his or her own facts."
Daniel Patrick Moynihan
"Everything that needs to be said has already been said. But since no
one was listening, everything must be said again."
Andre Gide
ยทยทยท
On Jan 21, 2004, at 7:30 AM, Williams, William D. wrote: