Can you outsmart this logical fallacy? - Alex Gendler
video description
Date: 2020-08-22
Related videos
Comments and reviews: 10
CFSworks
You know, this one really doesn't feel like a logical fallacy to me, and I'm wondering if most people are simply misunderstanding the question. I'd love to see if this hypothesis has been tested:
The question is asked as a dilemma: an A-or-B-but-not-both-not-neither choice, and that could be leading people to believe that the speaker intends for the scenarios themselves (not just the numeric probabilities) to be mutually exclusive: she's in the category of poker-playing painters, or of plain painters, but not both. (That is, if she's in the latter category, she's just a painter and doesn't play poker)
Since painting is a common factor, the question simplifies down just to Is it more likely that she is or isn't a poker player? and at that point the listener's brain goes Well I really don't know. Of the roughly 8 billion people in the world, you singled out one individual, and your process for picking her is a total mystery to me. Since you control the criteria for who you picked to tell me about, that means you actually control those odds. Any a priori knowledge I might have is useless.
To have any hope of answering, the listener changes the question once again to, how likely is it that the person the speaker picked is a poker player? and looks for clues about where the speaker is going with this. Statistics was mentioned earlier, and the maxim of relevance means the speaker is likely about to make a point about statistics, so, the listener reasons, perhaps a poker player was chosen to make such a point!
. and that's how a misunderstanding turns it into a language/communication dilemma, when it was really intended as a probability question.
reply
You know, this one really doesn't feel like a logical fallacy to me, and I'm wondering if most people are simply misunderstanding the question. I'd love to see if this hypothesis has been tested:
The question is asked as a dilemma: an A-or-B-but-not-both-not-neither choice, and that could be leading people to believe that the speaker intends for the scenarios themselves (not just the numeric probabilities) to be mutually exclusive: she's in the category of poker-playing painters, or of plain painters, but not both. (That is, if she's in the latter category, she's just a painter and doesn't play poker)
Since painting is a common factor, the question simplifies down just to Is it more likely that she is or isn't a poker player? and at that point the listener's brain goes Well I really don't know. Of the roughly 8 billion people in the world, you singled out one individual, and your process for picking her is a total mystery to me. Since you control the criteria for who you picked to tell me about, that means you actually control those odds. Any a priori knowledge I might have is useless.
To have any hope of answering, the listener changes the question once again to, how likely is it that the person the speaker picked is a poker player? and looks for clues about where the speaker is going with this. Statistics was mentioned earlier, and the maxim of relevance means the speaker is likely about to make a point about statistics, so, the listener reasons, perhaps a poker player was chosen to make such a point!
. and that's how a misunderstanding turns it into a language/communication dilemma, when it was really intended as a probability question.
reply
Ruben
The misconception is caused in part by the use of the word 'also' which implies that the additional characteristic present in the second option 'plays poker' is not present in the first option logically it is unnecessary to add the word. So we imagine there is a reason for it. If the artists who play poker are already supposed to be included in the first option the question becomes a bit ambiguous. As one tends to infer that only artists who do not play poker are eligible to be in the first category because if they did play poker, they would be in the second. It should be clear in the first option that we are referring to artist who may or may not play poker. unless the intention is to mislead 80% of the readers. which is exactly what happened.
reply
The misconception is caused in part by the use of the word 'also' which implies that the additional characteristic present in the second option 'plays poker' is not present in the first option logically it is unnecessary to add the word. So we imagine there is a reason for it. If the artists who play poker are already supposed to be included in the first option the question becomes a bit ambiguous. As one tends to infer that only artists who do not play poker are eligible to be in the first category because if they did play poker, they would be in the second. It should be clear in the first option that we are referring to artist who may or may not play poker. unless the intention is to mislead 80% of the readers. which is exactly what happened.
reply
Tletna
Not only is the question misleading but the answer is as well. In fact all groups (people named Lucy, (people who are educated in statistics, (people who do art) and (people who play poker) could all be the same size. There could be 100 people named Lucy, educated in statistics, who like to do art and play poker all at the same time. So the likelihood of both choices presented in this video could be different but it also could be the same. One shouldn't assume that just because Lucy does one thing that she won't also do another. It is funny that the video talks about a logical fallacy, presents somewhat correct notation for an equivalency statement, but then incorrectly states that one set must be smaller than the other. Oh well.
reply
Not only is the question misleading but the answer is as well. In fact all groups (people named Lucy, (people who are educated in statistics, (people who do art) and (people who play poker) could all be the same size. There could be 100 people named Lucy, educated in statistics, who like to do art and play poker all at the same time. So the likelihood of both choices presented in this video could be different but it also could be the same. One shouldn't assume that just because Lucy does one thing that she won't also do another. It is funny that the video talks about a logical fallacy, presents somewhat correct notation for an equivalency statement, but then incorrectly states that one set must be smaller than the other. Oh well.
reply
Kraka
The reason is pretty simple and lies mainly in language:
A: being an artist
B: playing poker
When asked what is more likely?
Why dont think about it like
Is A or is (A and B) more likely?
But Is (A and B) or is (A and not(B) more likely.
With the given information we then conclude, that the probabilty for someone who is good with probabilities to play poker is higher than the probabilty that they dont play poker.
This probabilty is though probably still wrong and then comes the effect mentioned into play, but the way presented it is not entirely correct
reply
The reason is pretty simple and lies mainly in language:
A: being an artist
B: playing poker
When asked what is more likely?
Why dont think about it like
Is A or is (A and B) more likely?
But Is (A and B) or is (A and not(B) more likely.
With the given information we then conclude, that the probabilty for someone who is good with probabilities to play poker is higher than the probabilty that they dont play poker.
This probabilty is though probably still wrong and then comes the effect mentioned into play, but the way presented it is not entirely correct
reply
atimholt
I thought this was a trick question or something at first. I can't wrap my head around someone thinking that a subset of a set of scenarios could be more likely than the superset. I mean what? I'm as baffled as I would be if they'd instead claimed that 2+2=856 is a common misconception.
From the context of the other comments here (someone said something about the original study having 8 choices to be ranked, among a couple other aspects, I'm giving the study the benefit of the doubt that they actually used unambiguous language.
reply
I thought this was a trick question or something at first. I can't wrap my head around someone thinking that a subset of a set of scenarios could be more likely than the superset. I mean what? I'm as baffled as I would be if they'd instead claimed that 2+2=856 is a common misconception.
From the context of the other comments here (someone said something about the original study having 8 choices to be ranked, among a couple other aspects, I'm giving the study the benefit of the doubt that they actually used unambiguous language.
reply
Luke
It actually depends on how you interpret the problem. When you ask are they more likely to be a portrait artist or a portrait artist that also plays poker, it can be reasonably interpreted as are they more likely to be a portrait artist that doesn't play poker or a portrait artist that does play poker. This interpretation means there is an additional condition in either case, (A Not B) or (A B. If Not B is thought to be less likely than B, then it's reasonable to argue that (A Not B) is less likely than (A B, albeit not guaranteed.
reply
It actually depends on how you interpret the problem. When you ask are they more likely to be a portrait artist or a portrait artist that also plays poker, it can be reasonably interpreted as are they more likely to be a portrait artist that doesn't play poker or a portrait artist that does play poker. This interpretation means there is an additional condition in either case, (A Not B) or (A B. If Not B is thought to be less likely than B, then it's reasonable to argue that (A Not B) is less likely than (A B, albeit not guaranteed.
reply
blobbo
i choose the portrait artist because it's only one thing she has to do, and doing one thing is easier,
for example you would probably find more people that play tennis than people who play tennis, basketball, soccer, volleyball, gamble, trade in the stock market, own a supermarket chain, made a commercially successful game series, and streams playing a block game on twitch then also uploads the same thing to youtube.
also it would kinda be weird if someone was playing poker at a casino while drawing on a canvas.
reply
i choose the portrait artist because it's only one thing she has to do, and doing one thing is easier,
for example you would probably find more people that play tennis than people who play tennis, basketball, soccer, volleyball, gamble, trade in the stock market, own a supermarket chain, made a commercially successful game series, and streams playing a block game on twitch then also uploads the same thing to youtube.
also it would kinda be weird if someone was playing poker at a casino while drawing on a canvas.
reply
Redstone
I instantly understood the probability of her being an artist is higher, because I instinctively broke it down to a mere number problem
I'm not sure why, but I considered What percentage of artists like playing poker?
And suddenly I didn't even care about the prior information - I just figured, statistically speaking, if she must be an artist, she probably doesn't play poker
I don't know what encouraged me to think this way - maybe it's that these are always trick questions
reply
I instantly understood the probability of her being an artist is higher, because I instinctively broke it down to a mere number problem
I'm not sure why, but I considered What percentage of artists like playing poker?
And suddenly I didn't even care about the prior information - I just figured, statistically speaking, if she must be an artist, she probably doesn't play poker
I don't know what encouraged me to think this way - maybe it's that these are always trick questions
reply
Ivan
I think theres something wrong whith this reasoning.
Since both options presented show the possibility that the subject is a portrait artist, this possibility suddenly becomes a given truth, since you cannot choose otherwise. Lucy must be a portrait artist.
Therefore the poker option is forcefully taken into account on its own when performing the choice, as if the question was presented as: is Lucy a poker player or not?
reply
I think theres something wrong whith this reasoning.
Since both options presented show the possibility that the subject is a portrait artist, this possibility suddenly becomes a given truth, since you cannot choose otherwise. Lucy must be a portrait artist.
Therefore the poker option is forcefully taken into account on its own when performing the choice, as if the question was presented as: is Lucy a poker player or not?
reply
Mark
So, we're supposed to make a determination between 2 situations which are not even partially confirmed with a pretty much unrelated background as our only lever to make this choice. We have no way of knowing WHAT Lucy really does without real evidence. We are mislead by assumed absolutes. Just because I know person X drinks diet soda gives me NO basis to claim they eat sweet relish on donuts.
reply
So, we're supposed to make a determination between 2 situations which are not even partially confirmed with a pretty much unrelated background as our only lever to make this choice. We have no way of knowing WHAT Lucy really does without real evidence. We are mislead by assumed absolutes. Just because I know person X drinks diet soda gives me NO basis to claim they eat sweet relish on donuts.
reply
Add a review, comment
Other channel videos
