Why the Linda problem matters
The Linda problem, or the Conjunction fallacy, is designed to lead you to a cognitive error.
It's a really common error, and there's no shame in making it! I did.
Here's what it does:
- Give a bunch of information to prime your monkey brain to think about one thing
- Ask a question that has nothing to do with the information provided.
Here's why this is important:
-- Thought experiments give practice in a safe space to make the error.
-- You are then free to practice introspection, or to make another common mistake and blame the question.
-- Life is filled (FILLED) with times you are given information that has nothing to do with a decision you need to make. In the Real World, this is often done sneakily and tricky to manipulate you to making decisions that are good for someone else, not for you.
-- When done on behalf of companies, there's another special word for this: marketing. When we're tricked by companies, we (and others) lose money.
-- When done on behalf of governments, we have a special word for this: propaganda. When we're tricked by propaganda, people die.
-- By thinking about these errors and by knowing that we are likely to suffer them, we can do that hardest of cognitive tasks: belief revision.
In summary: problems like the Linda problem let you practice hard cognitive tasks.
Subscribe to:
Post Comments (Atom)
.
ReplyDeleteThis was a very interesting exercise. What do you think about alternative readings of the Linda problem, as discussed here? psychologytoday.com - Linda The Bank Teller Case Revisited
ReplyDeleteJohn Jainschigg : Wow... that article literally commits the conjunction fallacy during its response: “We are told that Linda was a devoted feminist during college.” We were? Or... is the author interpreting the first paragraph of the question in conjunction with the later question? (Hint: I don’t really think that’s a question, as they obviously are)
ReplyDeleteOh, man, I messed this one up because I was thinking about which one was probable (a judgement of likelihood) rather than more probable (a judgement between two possibilities). My answer was "Neither because we have insufficient information to judge whether Linda is a feminist or a bank teller."
ReplyDeleteBut yes, based purely on probability, it is more likely that she is a bank teller than a bank teller AND something else, unless something else is perfectly correlated with bank teller (of which we have no information).
Yes, Tony Lower-Basch -- obviously, the author is doing what you say. The setup of the Linda problem is such that it encourages us to do so. The point (or one point) the article is making is that this is actually a useful skill: being able to infer what a questioner means to inquire after, rather than the question they ask in the austere and purely-literal sense.
ReplyDeleteNotes that the solution when it comes to both marketing and propaganda is to ultimately reject and expose the question as a trick. So yay?
ReplyDeleteJohn Jainschigg : I grasp that point (having watched nearly a dozen people use variations of it in the past day).
ReplyDeleteI worry about whether the people checking (and rechecking) that we all understand that point also understand that in the vast majority of Linda-question problems there is no questioner apart from the person making the judgment. People bridge the gap between the question reality asks them, and the question they think needs to be answered anyway.
Rabbit Stoddard Or, as "we" were asked during Watergate: Follow the money. Who benefits from me believing as I am being asked to do?
ReplyDeleteContinuing to ask that question would be unquestionably good for the world.
Yes it is a useful skill in conversation between two equal parities, to bridge the conversation gap. The problem arises when marketing/propaganda machines can use your propensity to make leaps of faith towards their point of view while they do not have to explicitly state the point they are trying to make. They get to have clean hands during the misinformation campaign. It is our job as good citizens to inform ourselves, and always be vigilant against it.
ReplyDeleteKimberley Lam What I love that you did there was, basically:
ReplyDelete1. Make a cognitive mistake.
2. Realize it. Look into why.
3. Solve it.
That's the goal! That we make mistakes is irrelevant. It's what we do with it that matters, and fixing the problem is awesome.
Thanks!
Ha! The real test is if I fall for it again.
ReplyDelete