How Logical Are You? (Psychology of Reasoning)

How Logical Are You? (Psychology of Reasoning)

The psychology of reasoning is a branch of
cognitive psychology that studies that way in which humans use logic to come to conclusions
based on available information. But not all reasoning is valid, and people may use invalid reasoning to come to illogical conclusions. So just how logical are we as humans? And for that matter – how logical are you? A logic test developed by Peter Wason in 1963 revolutionised the field of reasoning. This classic selection task features cards with
letters on one side and numbers on the other side. Subjects are then presented with 4 cards
and a rule: If a card has a vowel on one side, it must have an even number on the other side. And the task is this: which card (or cards)
must be turned over in order to determine whether or not the rule has been followed. So basically, you must turn over cards which can guarantee whether this rule is true or false I’ll give you a few seconds if you want to
pause the video and think about it, before revealing the answer… OK… the correct answer is…the cards which must be turned over are: A and 7 These are the two cards which must be turned over to
assess whether or not the rule has been followed. Various studies has shown extremely poor results
for tasks like this, in which only 4% of participants get the correct answer. The most common wrong
answer being A and 2. This can possibly be explained by confirmation
bias. The rule mentions both a vowel and an even number, so it may seem logical to choose
the vowel and even number cards. While this may seem logical, as I will demonstrate, it’s actually fundamentally illogical. I’ll now go through each card and explain
why you do or do NOT need to turn over each card over. The key to this task is you need to understand that you must try to falsify the rule, not confirm it So the first card, A, this is the most obvious,
and virtually all participants correctly conclude that this card must be turned over. Because
of course, if there’s an odd number on the opposite side, the rule has been broken. Now the next card, the K card, this is basically an irrelevant card, which cannot give us any information. If we turn it over and it’s an
odd number, that’s fine. If we turn it over and it’s even number… that’s also fine.
So if neither outcome breaks the rule, there’s no need to turn the card over. The 2 card is where things get tricky… and
where most participants slip-up. If we read the rule again, “If a card has vowel on one
side, it must have an even number on the other side”. Now, it’s important to realise is that
this rule only goes one way. Therefore we cannot conclude that if there’s an even number on
one side, there must be a vowel on the other side. This is not the case. If we turn the 2 card over, and there’s a
consonant on the other side… that’s fine. This does NOT break the rule. And obviously,
if we turn it over and there’s a vowel, this also doesn’t break the rule. So again, if
neither outcome can break the rule, then we cannot obtain any relevant information from
this card, and therefore there is no need to turn it over. Finally, the 7 card. This card must be turned
over. Why? To make sure that there is NOT a vowel on the other side. If there is, this
breaks the rule because the card has a vowel on one side and an odd number on the other side. So looking at both potential outcome of all
4 cards, we can see that there are only two ways in which the rule can be broken, and
therefore these two, and only these two, cards that must be turned over to determine whether
or not the rule has been followed. This rule is basically just an “If, then”
statement. Or it can be written as “P therefore Q” in which P is the antecedent, and Q is
the consequent. Each of the 4 cards represent the 4 possible premises: P, not P, Q, not
Q. So ‘P’ would be a vowel. ‘Not P’ would
be a consonant. ‘Q’ would be an even number. And ‘Not Q’ would be an odd number. Now when presented with each of these premises,
we can make inferences. These can be divided into two types of reasoning – inductive reasoning,
and deductive reasoning. Inductive reasoning is invalid, while deductive reasoning is valid. The first and most obvious is when presented
with P (in this example, the A card), since the rule is P therefore Q, given P, we can
obviously deduce that Q must follow. This is known as affirming the antecedent, or modus ponens, which is a valid, deductive form of reasoning Next when presented with ‘not P’, in this
example, the card K, we could infer ‘not Q’. This is known as denying the antecedent,
which uses inductive reasoning and is therefore invalid. Just because you’re presented with
‘not P’ does NOT necessarily mean that ‘not Q’ must follow. When presented with Q, people often infer
P. But like I said, the rule only goes one way. So just because the rule is “P therefore
Q”, does not mean we can turn it around and say “Q therefore P”. This is known as affirming the consequent, which is another invalid form of reasoning Now, finally, when presented with ‘not Q’,
we can infer ‘not P’. This is often missed by participants of tasks like this, despite
being a valid, logical form of reasoning – known as denying the consequent, or modus tollens.
This is valid because if we have ‘not Q’, the only logical conclusion would be ‘not P’. If we apply this to a real-world example,
we could say something like “All tigers have stripes” as our premise. Then we could make inferences using both deductive and inductive reasoning for example: “If it’s a tiger, it has stripes” “If it’s not a tiger, it doesn’t have stripes” “If it has stripes, its a tiger” “If it doesn’t have stripes, it’s not a tiger” Looking at the results of the card selection
task, one might quickly jump to the conclusion that humans are not very good and reasoning,
and are therefore illogical. But one important thing to note is that content is crucial to
a task like this, and that participants perform significantly better when presented with cards that have real-life applicable examples. This can be seen in a different form of the
card selection task. This time, each card represents a person. On one side of the card is the drink in which that person is drinking, and on the other side is their age. Participants are presented with 4 cards, and the following rule: “If a person is drinking alcohol, they must be 18 years or older”. The task is once again to turn over cards
in order to determine whether or not the rule (or in this case, the law) is being followed. In this case, 72% of people correctly predict that the cards ‘Beer’ and ’16’ must be turned over The ‘Beer’ card has to be turned over, to
make sure the person drinking that is old enough. The ‘Water’ card is irrelevant,
since it doesn’t matter what age the person is. The ‘25’ card is also irrelevant since
it doesn’t matter whether they’re drinking alcohol or not, since they’re over 18. And
the ‘16’ card must be turned over, to make sure that person is not drinking alcohol. It’s much easier for us to reason in tasks
like this because we can actually apply it to real-life scenarios. In terms of actual variables, this test is
actually quantitatively identical to the first. The statement “If a person is drinking alcohol, they must be 18 years or older”, is just another “If P, then Q” statement, where drinking alcohol is P and being 18 year or older is Q. The 4 cards also represent the same 4 premises
as before – P, not P, Q, not Q. This also helps highlight why certain inferences are illogical. The most common wrong answer in the first task was the A and 2 cards. But
as stated earlier, turning over the 2 card is illogical. The 2 card, is representing
the premise Q. In this real-world example, representing the Q premise is the ‘age 25’
card. So it becomes more obvious why this is illogical.
Just because a person is over 18, does not necessarily mean they are drinking alcohol. So no matter what they’re drinking, they’re not breaking the law, and therefore there
is no need to turn over that card. So the rule P therefore Q cannot be simply
turned around to Q therefore P. “If a person is drinking alcohol, they must be at least
18 years or older” cannot be turned around to “If a person is 18 years or older, they
must be drinking alcohol”. This is inductive reason and is therefore invalid. It’s worth pointing out though, it’s possible
to be wrong by using deductive reasoning, and vice versa, it’s possible to arrive
at a correct conclusion using inductive reasoning. Deductive reasoning is only as correct as
its premise. So given an incorrect premise like “all birds can fly”, we could use
modus tollens to logically deduce that “if it can’t fly, it’s not a bird”. This
is obviously false because a penguin is a species of bird that can’t fly. And if we return to our tiger example, if
we see a tiger, and say “It has stripes, therefore it’s a tiger”. This is an invalid
form of reasoning, yet we have arrived at a correct conclusion. In fact, inductive reasoning is actually incredibly
useful, and we use it on a daily basis with incredible accuracy. Some of the most obvious
statements and assumptions we make, are actually using inductive reasoning. Things as obvious as “the sun will rise
tomorrow” or “if I drop this coin, it will fall to the ground”. These statements,
though obvious, are actually using inductive reasoning How do we really know the sun will rise tomorrow?
Because it’s risen every day of our lives so far? But nothing that has happened in the
past can guarantee what will happen in the future, we can only make predictions, and
increase the certainty of our predictions, but we can never guarantee outcomes. In fact, everything we know about the universe,
we know through inductive reasoning. How do you think we get these premises in the first
place? How do we know that all tigers have stripes? Through inductive reasoning. But
there’s no way to know for sure that all tigers have stripes. Maybe there’s an undiscovered
species of tiger in a jungle somewhere with no stripes. It’s just that every tiger anyone
has ever seen has had stripes, so we can say with a high degree of confidence that all
tigers have stripes. But it’s impossible to know for sure. In fact, according to the strictest rules
of logic, it’s impossible to know ANYTHING for sure. Through scientific research, we
can only increase the likelihood that something is true, but can never actually confirm it. A famous philosophical quote states that “Nothing
can be known, not even this” But just because something can’t be known
with 100% certainty, when all the scientific evidence points to something, any rational
person would accept it as fact. So when it comes to reasoning, it’s not
just about logic, but also about common sense, and rationality too. Thanks for watching.



  • Mia Tann

    Is the host eating gum while broadcasting?

  • Thomas Baranowski

    Damn. At first my answer was just A but when he revealed the answer (being A and 7) I immediately realized my mistake.

  • Shinigami North

    lol, i choosed to turn no cards.
    I think it's better for to jump from a bridge XD

  • Blue eyed Eurasian

    I think this is not that difficult. Just don't rush into the answer

  • Kelvin Price

    As far as falsification being the only true evidence Karl Popper once said " there arw two types of theories, theories that have been disproven and theories yet to be disproven". Understanding logic tests and how people rarely use logic make one afraid to go in front of a jury, or even most judges.

  • Rei Zero

    Bad teacher

  • Tries Everything

    I take issue with the idea that the vowel : even number relationship is only one way. The nature of cards is such that each individual card has only two outcome sets, represented by two sides. Now, these cards being “cards” indicate that they will have no more than or less than two sides (to suggest that the rule doesn’t state that cards have to be two sided is just unnecessarily annoyingly literal). Therefore, given these 2 potential outcome sets, we are left with 4 qualifiers – vowel, consonant, even, odd. By indicating that for a rule to be true a card with a vowel on one side must have an even number on the other side is effectively saying that the vowel qualifier directly yields the even qualifier, which, because there are only 2 outcome sets per card, indicates that this rule is then reciprocal, meaning if you flip over the “2” and get a B, you have disproven a rule. That might be our if scope of this because it’s technically a proof by induction… but still

  • FV N • 55 years ago

    I got the ak27 correct and the beer and 16 thing wrong lol

  • Ace Adventura

    Isn't there a more interesting test?
    Letters and numbers are boring af

  • Man in ushanka

    By the way, the premise that ''deductive reasoning is always correct'' always annoyed me in Sherlock Holmes. Because deductive reasoning CAN lead to incorrect results.
    Thanks for clarifying this point in the video.

  • ashley miles

    Instinctively on the first I knew turning over A & 2 would be wrong but I didn't know why so I had to keep watching to know why it was wrong 😅

  • Johan Jonasson

    Surely you must turn over the K as well. What's to say there's not a vowel on the other side? Or are we only testing from the perspective of this side of the cards…?

  • Graydiation

    How long did 5.1k-people pause video?

  • Onion skin 1,00

    Holy moly I got a lightbulb flipped

  • Säm Pie

    no Logical at all

  • bumbujahe

    lol i didn't even understand what the question meant

  • Renat Avetisyan

    If you have good logical thinking. You don't want to do bad things to society.

  • tanveen kaur

    your voice is annoying

  • Marine Trisinscius

    You can know you have conscious experiences. You can make that into a premise onto which you can deductively conclude you conscious has substance in some form, which means your conscious experience and the medium it functions in have the property of existence.
    Even if everything is deterministic and we lack all free will, you can still use this reasoning to have SOME true knowledge.

  • Your email address will not be published. Required fields are marked *