[BLANK_AUDIO] >> Sometimes we make mistakes against logic. We just had four examples in the former video for example where there was a logic and we made a mistake. This is the world of deduction but of course, the other world induction the world where I live, the world of creativity is the world of another kind of mistakes. You cannot be wrong because truth doesn't exist in this world, but the danger is, maybe to go too fast, maybe not to think enough, and I'm going to demonstrate that with three exercises. If I ask you, what's this? Most probably you will answer this is letter T, fine, now I ask you, what's this? [BLANK_AUDIO] What came in your mind? Most probably, two T's. But now, imagine I didn't show the first one, the T alone. Immediately show this one, the double one. Probably, you could have other answers. This is a table with a hole or this is like a balance sheet with liabilities and assets et cetera, et cetera. So, you see, you influence, in the second question, by the answer of the first question. This is an example of cognitive bias and it happens everywhere. One of the most incredible examples I, I've seen is this one. You have a group of people, half of the people are asked, when is Genghis Khan born. Okay? The other group had the same question but before they receive another question. Which are the three first digits of your phone number? The second group puts the burst of Genghis Khan more often before the year 2000, 981 or something like that. Just as if this souvneir of three digit, of the phone number left something like we need an answer in three digits, isn't that incredible? It is, but it happens everywhere. It's a good example of a committive by us, you have hundreds of them, and many, many books describes exactly the danger of this type of thinking. Let's go to another example with this one, look at the following set of companies Goldman Sachs, Deutsche Bank, American Express, and Pfizer. And I ask you, which one doesn't belong to the group? Most probably, you will think, oh, Pfizer, because the first three has a name with two names, like Goldman Sachs, or maybe you think, hey, they're all in the financial world. Pfizer is pharmaceutical, so you drop Pfizer. Now let's move to the next one. Consider this set, and I can express Pfizer, Goldman Sachs, and Deutsche Bank and the question is the same, which one is a bit odd? Which one doesn't belong to the group? Maybe you think it's Deutsche Bank. Why? Because you move with a three first step American Express, Pfizer oh, I'm in the United States Pfizer. Okay, I'm sure. Deutsche Bank is Europe so you drop Deutsche Bank. So you have the same set and probably you will drop two different companies because you frame the problem differently. You remember in lecture number three I think when we build this H. According to the present mental model, you're going to judge the next one. It is a beautiful example too a last example to show, show it's not a mistake because you are not right. You are not wrong. You can remove. I don't care. But, I just want to show that you don't control 100% what happens in your mind and that's the world of in, induction of course. Let's have a last example in the same category of problems as this one. Suppose there is a rule. A rule somewhere that produces set of three numbers. And I give you an information, 2, 4, 6 is produced by the rule and I, I want you to find the rule. But the only way for you to find it, you have to propose sets of numbers, of three numbers and I will tell you yes or no, it fits or it doesn't fit the rule. Are you ready? Maybe the first set you want to propose is something like 10, 12, 14, and I'll say yes, it fits the rule. Maybe you come with 20, 22, 24, and I'll say yes again, it fits the rule. But is this the most efficient way to find the rule? I don't think so. What happened? You have a hypothesis. Three even numbers in ascending order, and what you do you try to prove you right. Try to propose a set of numbers that doesn't fit the rule. It's going to be more efficient, you'll see. So you'll try for example, two, four, a billion I see yes, it fits the rule. Then you can say, three, two, one. I will tell you no, it doesn't fit the rule and slowly you will find the rule. The rule produces sets of three numbers in ascending order. Not necessary even number, not necessary in a sequence and it shows something very important. When we are asked something immediately we have an hypothesis, and we will try to confirm our hypothesis. It's much more efficient to go against, to prove this is not true. [BLANK_AUDIO]
