Puzzle sheds light on government policy, corporate America and why no one likes to be wrong


[Read the post]


It took me a while to catch on:


I arrived at the same conclusion, in about the same number of trials. I was interested to see that you tested negative numbers. I assumed the puzzle was limited to the Natural numbers.

Now I wonder if it will accept non-integer values.


I would have found this much more interesting if I hadn’t already seen this trick in a Youtube video some weeks ago. So it goes.


I’d say that “being” wrong is a philosophical problem. Providing a correct or incorrect answer does not indicate or change who you “are”. But that people commonly refer to their problem-solving experiences with intense feelings of identification seem to me a huge obstacle to learning. As this example shows, numerous guesses can be used to understand the underlying rule. Arriving at the correct answer on the first try would simply be a lucky guess. One can test as many as they choose.

I agree that people generally prefer to not solve problems incorrectly, but the value of realizing that errors have been made would seem obvious. No amount of sunken cost is going to make poorly audited thinking any more functional.


I kind of liked the article that accompanied the test.
Confirmation bias. I would also like to point out that the New York Times used another rhetorical trick that the media use a long. False equivalencies.

Liberals think conservatives are ignoring climate change data.

But conservatives were right in the olden days when the scientists said “that population growth would create widespread food shortages.”

But the difference is that people DID pay attention to the issue of food shortages because the population growth would explode and DID explode. We actually DID something about it.

We killed a lot of people. *Thanks to multiple horrible plans, policies and wars
We limited population growth. * Thanks birth control
Still didn’t stop population growth enough SO

We figured out ways to increase crop yields. *Thanks researchers

If we had NOT listened to reality and made changes we would NOT be where we are now.


My first tries were almost the same, too! To my surprise, it does accept non-integers and the rule applies to them just the same. Fun experiment with fascinating results and interesting background in the explanation.

(I fooled around quite a lot after these first guesses, trying out non-integers and very large numbers, but I was fairly certain after 0 1 2 went through and 0 0 0 and -1 -2 -4 didn’t).

EDIT: Whoops, noticed that 0.5 0.25 and 0.0125 shouldn’t work anyway; typo’d one too many zeroes!


The example is a great way to lead you into confirmation bias. I was on that path, then tested -1 (and flubbed it) and 0, and realized that pattern didn’t fit and then poked around to determine the pattern was really simple. Nice article, this one was interesting.


I guessed it in four sequences:

1 2 4
3 6 12
0 0 0
1 2 3

…but then I’m a bit of a math geek and took a math class involving pattern-solving in high school (our teacher was writing a college level textbook and used us as guinea pigs).

For me this didn’t really work to show my confirmation bias - I was actually intentionally trying to get negatives, since those can often give as much information as positives when problem solving. (At least mathematical problem solving, depending on the problem.)


If I were making this quiz, I’d use the rule “sequence number is not a prime number”. After the first few guesses, people would start to believe that they were getting it.


At the NYT it’s generally OK to go with quick, probably premature judgments


I suck at concisely describing patterns of numbers.

But failed responses were quite valuable in my search.


damn-- forgot this point.

Numbers cannot be complex. Numbers cannot be irrational. I don’t think I even tested non integers less than zero. What a fool I’ve been.


whoa, lots of guesses, mark. didn’t think to capture the screen, but I did
16,32,64 - yes
10,12,14 - yes
1,1,1 - no
10,11,12 - yes

and that gave me the answer


IAAMathematician. I did about 15 guesses, maybe 9 Yeses and 6 Nos, before guessing (correctly). I cared more about being correct than minimizing guesses.


What’s powering this quiz? I couldn’t get it to run in either of my browsers.


I was part of the unwashed masses that just doubled everything, and when it to worked just gave my answer (double the previous number) only to be revealed to be a jumping-to-conclusions fool. (But,on the bright side, took me less than a minute to be done with the thing.)


I had almost the exact same progression. While it was the “No” to the 0,0,0 that threw up the first red flag for me, it was actually the semi-unexpected “Yes” for 1,2,3 that confirmed that my initial supposition was incorrect.

I did do a few more examples after that sequence, though, to confirm my new hypothesis.


I broke it with i, 2i, 3i.


when i saw this yesterday, i should have actually read it. saw the 2^n pattern and just assumed that was it. didn’t feel the need to actually try any numbers.