Not strictly evol-psych related, but you should know for future
reference that, stuck at a boring dinner party, you can create instant
chaos, and sometimes rupture long and close friendships, by
introducing the Monty Hall problem.
Charles Murray
> (snip) A few weeks ago I did one of my occasional "Math Guy" segments
> on NPR's Weekend
> Edition. The topic that I discussed with host Scott Simon was
> probability.
> Among the examples we discussed was the famous - or should I say
> infamous -
> Monty Hall Problem. Predictably, our discussion generated a mountain
> of email,
> both to me and to the producer, as listeners wrote to say that the
> answer I
> gave was wrong. (It wasn't.) The following week, I went back on the
> show to
> provide a further explanation. But as I knew from having written about
> this
> puzzler in newspapers and books on a number of occasions, and having
> used it as
> an example for many years in university probability classes, no amount
> of
> explanation can convince someone who has just met the problem for the
> first
> time and is sure that they are right - and hence that you are wrong -
> that it
> is in fact the other way round.
Not strictly evol-psych related, but you should know for future reference that, stuck at a boring dinner party, you can create instant chaos, and sometimes...
... A fairly extensive account of the Monty Hall problem is in Chapter 16 of my book, Game Theory Evolving. It turns out that the answer depends on whether...
... I am confused now - I reasoned that there is no gain in choosing again because it is still a random choice, with the one already chosen having the same...
Irwin Silverman
isilv@...
Jul 8, 2003 3:57 pm
A good discussion of this problem can be found here: http://tinyurl.com/gbx3 The answer (switch doors because this increases your chance of winning from 1/3 to...
... AHA!!! But Charles advised against asking the question at dinner parties (I have to be right about something)...
Irwin Silverman
isilv@...
Jul 8, 2003 5:26 pm
... -- It gets much more intuitive if you imagine a change in the numbers involved. Suppose I fan out an entire deck of 52 cards and ask you to pick out the...
... .. Or if describe the same sequence of evens this way: 1) You have selected two doors. 2) The host told you that one of them does not have the prize. ...
... This is not correct. You already know that one of them does not have the prize, so if MH is merely informing you of this, there is no reason to change. But...
Herb wrote: << if MH says, "I have randomly chosen one of your two doors, and it does not have the prize", then you should switch, because the prize is twice a...
... It's very real. Here's one correct way to see it. Let P mean Prize, and N mean No Prize. Let PNN mean "The prize is behind door 1", NPN mean "The prize is...
... My way of explaining it, having had to overcome the illusion myself, is - Monty couldn't pick the door with the prize and he couldn't pick the one you...
Irwin Silverman
isilv@...
Jul 16, 2003 6:53 pm
... Another way of thinking about this problem is provided by Gerd Gigerenzer in Reckoning With Risk. Imagine there are 100 boxes. In one box there is a prize;...
... It's still random, but the probabilities may (or may not) have changed. Before MH's move, your chances were 1/3 of winning. What are they after MH shows...
On Tuesday, July 8, 2003, at 5:02:26 PM, Herbert Gintis wrote: HG> It's still random, but the probabilities may (or may not) have HG> changed. Before...
... If the third guy asks the jailer the same question, and the jailer points again to the second prisoner, he is saying nothing new, so it can't change the...
... (IS) ... (HG) It's still random, but the probabilities may (or may not) have ... (IS) 1/2 - whether or not Monty knew what was behind the door he opened 2...
Irwin Silverman
isilv@...
Jul 8, 2003 5:25 pm
... My typing finger will fall off if I say this one more time, so this is the last. The answer depends on whether Monty randomly chose a door, or purposely...
I'd like to look at this question a different way. Of 999 consecutive contestants making random guesses, about one third should guess correctly on their first...