The Legendary Logic Test Everyone Fails

Newsthink is produced and presented by Cindy Pom

Thank you to our Patrons, including:
John & Becki Johnston
Igli Laci

Image/video sources:

[Newsthink]

*What other videos would you like to watch?*

[fellopiantube]

this does't look like a logical puzzle. this looks like a psychological test to prove a thesis

[vishensivparsad]

Turns an extra card in the first puzzle to gaslight the rest of the puzzles😂

[coolParadigmes]

😇 There is a huge bias in the explanation because we can prove the assumption is wrong with only one card, only if we get a counter example and in that case the 1st card would have been just fine! You turn it and it's a "4", rule broken! You turn the 4th card and it's a "C" then you don't know anything more! So if you are unlucky you need to turn all cards that could break the rule, so the 1st and the 4th card! Furthermore the under age drinkers test is very different (we have more information), the 1st drinker is out automatically because he is old enough and we know that we have only 4 people in the bar so the choice is obvious! Again with cards we are used to have many more cards in games, so to establish a global rule on only 4 cards "in front of you" is a bit confusing and useful to destabilize the player ... what about hidden cards?

[petem]

The tendency to seek confirmation rather than disconfirmation also explains why most programmers struggle to find errors in their code. Generally, what makes software QA engineers effective is their ability to identify ways to find scenarios which will demonstrate software failures/incorrectness.

[zanshibumi]

This video should be called, "The Legendary Logic Test Everyone Fails, and then, unable to admit their own limitations, desperately flail for an excuse as to how it's not their fault for various reasons."

[Burgers_Pizza]

I think all my experience with playing Backgammon finally amounted to something positive. This was hardly puzzling.

[stephanf]

A lot of people are salty that they didn't follow the rule as stated and think the rule needs to be more complex but that is simply not correct. As stated it is very obvious what to do there are no reasons to add parameters and shit because than it is to simple cause it would explain the result already.

[mr.k]

So apparently I’m of the 10%, ok, but honestly I don’t think that being overly precise is always an advantage in every day life. To the contrary, “overthinking” can be a disadvantage in many cases as well. For example if time is a bigger factor etc.
It’s simply always more dependent on the context. It is by no means the case that system 2 always wins.

[FreeSpeech]

You flip card 19 over to see if the bar is breaking the law. However, based on that reasoning, you can flip the card with a beer on it to see if the bar is breaking the law, too. If it shows anything less than 21, then the bar breaks the law. That aside, your instructions should state specifically that ONLY ONE card can be flipped.

[joshuaadamstithakayoutubel]

I did better with the “harder” puzzle. I knew to check A and to ignore 2, but neglected to turn over 3. When it was converted into a real life scenario, I got distracted from the simple task.

[mutantryeff]

Study 1st Order Predicte Calculus. Write some rules in Prolog or Datalog.

[mr.boomguy]

Should've commented before the solution, but I thought that A and 2 you already knew, 3 and B you should test to see.

Edit: so... I failed the next with thr shapes because I thought it was the same. Check the rest, not the obvious

[secondsrule]

I don’t get it. I’d still flip the 1st card as this is only 100% confirmation. It’s different when you can’t touch it. Then 4 is logical choice.

[MrWonszBoa]

The 'card' rules in the first example are not clear. Because of the simple fact that this games uses cards. Everyone is accustomed to the fact that cards have a specific logic behind them, that dictates what particular front matches what particular back side. So it is natural and justifiable to assume in this case that the rule given to us "if A then 2" should work also the other way round. Not because of logic but because of the nature of most card decks we know from real life. The problem giver then feels and acts as if he or she is smarter than the presumably 90% of people who don't solve the puzzle 'correctly'. And then they continue to give a false interpretation of the discrepancy between the 10% and 75% of people who solve the 'beer' problem. It's not because of the slow and fast thinking but because the nature behind the beer problem matches the presumptions you naturally make while hearing the question. And it's not the case with the first puzzle where one's natural assumptions mismatch with the ones in the puzzle. All of this could have been avoided by making the first puzzle more abstract by saying something like: aliens left us four stones with these strange symbols....

[hamburglar]

Confirmation bias…….so true to why this world has so many issues.

[zanshibumi]

I love this video. It made me feel smarter in 20 seconds flat. \o/

[AnnemieM]

If the 3 card has a B on it, then it proves nothing in my mind. We still would not know if the A cart has a two on the other side. So I really don't get it. Am I wrong? If yes, please someone tell me.

[tigerlord]

I suck at system and like system 2 but I have a different interpretation for why the success rate increases so dramatically in the first experiment. I think most people are better at concrete/practical problems than they are at abstract ones. I like abstract arguments and get into trouble with a lot of people who refuse to think beyond the concrete. Who are we talking about rather than the question of principle?

[henrybottomsworth]

this video sucks

My initial thought was that all cards needed to be turned over. But then I repeated the rule and realized that turning over 2 is redundant if A had already been turned over and vice versa. So turning over 2 and getting A doesn't further "prove" the statement if A had been turned over already (although it may add further evidences) and getting a non A bears no relevance to the bare minimum proof of the statement after A had been turned over. You then turn over all non-A and non-2 faced cards in an attempt to disprove that the back has an A. This is what is necessary to entirely map out the proof within the scope of all cards presented. It is also to be said that turning over A contributes a guaranteed proof while from the perspective of turning over a 2, this only has one scenario that provides a proof and other scenarios don't contribute. Presuming the absence of a card with a visible A side, it would not be unreasonable at all to attempt to turn over 2 in search of that one scenario given there isn't that surefire A card to turn over.
Its a bad question though because it doesn't clarify what "prove" means (and most people don't know anyways exactly either) nor does it clarify the scope of proof. Its also totally unclear that there is distinction of perspective of what defines "one side" and "the other side". Some people may turn both A which provides a proof and additionally 2 in order to seek additional evidences which isn't necessarily supplementing a "proof" in the context that the video desires but in legal terms we tend to think of proof with the addendum "beyond a reasonable doubt" and in the courts additional evidences help beyond just 1 that is locally scoped.