Edited by Richard Walker, Wednesday 18 March 2026 at 22:49
Google 'Married Puzzle' and you'll get around 7,000,000, so it is quite famous. If you've not seen it, it usually goes like this
Jack, who is married, is looks at Anne, but Anne is looking at George, who is unmarried. Is a married person looking at an unmarried person?
Yes
No
Not enough information
It's often captioned '80% Get It Wrong' and indeed Alex Besos wrote about it in the Guardian and ran an online survey that attracted 200,000 responses of which just over 70% were wrong.
I first saw the puzzle some time ago but encountering it again today reminded me what a well constructed question it is, skilfully written to make it seem more puzzling than it actually is. Here's a different question
I flipped a coin three times. The first time it came up H and the third time it came up T. Did the pair HT appear in the sequence?
I hope it's not too hard to see the answer must be 'Yes'. Given the description we might think about the sequence of coin flips and the only possibilities are HHT and HTT, both of which contain the pair HT.
But the two questions have exactly the same underlying structure; we have just replaced 'married' with H , 'unmarried' with T, and 'is looking at' with 'was followed by'.
However we've lost the distracting baggage the original came with, a storyline that hints at the intriguing possibility of a love triangle (often the question includes a picture of the three people) and perhaps makes us feel vaguely that being married or not might have something to do with the solution. I've set my version in a context that nudges us towards think in a more analytical way and makes it easier to see the answer.
The Guardian article drew on an article in Scientific American that mention the original source of the puzzle as being Hector Levesque, a well-known researcher in artificial intelligence at the University of Toronto, who devised it for a research project. I've tried to find the original publication but haven't to date succeeded in identifying which of his publications it appears in.
In some ways it reminds me of the equally famous Linda Problem from the book Thinking, Fast and Slow, by Daniel Kahneman, which contrasts two Systems of thought: System 1, which is fast, intuitive and economises on mental computation, versus System 2, which is slow, methodical and computationally intensive.
System 1 excels when something is routine or demands rapid and instinctive response, whereas System 2 is needed in more complicated situations that require slow and careful logical thought to arrive at a correct solution.
Here is the Linda Problem
Linda is 31 years old, single, outspoken, and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice, and also participated in anti-nuclear demonstrations.
Which is more probable?
Linda is a bank teller.
Linda is a bank teller and is active in the feminist movement.
Most people choose Option 2 but that cannot be correct. There are more bank tellers than there are back tellers who are active in the feminist movement, just like there are more people who live in the UK than there are people who live in the UK and collect stamps. So the probability of 2 being true is less than the probability of 2 being true.
I love this example! The amazing thing to me about it is that when I first saw it I got the right answer but it felt wrong and for all the times I have explained it to someone or written about it, it still feels that 2 is correct. It stands as an example of how wrong an instinctive answer can sometimes be.
I think the link between the two problems, the Married Puzzle and the Linda Problem, is that the majority (and wrong) response in both cases is the System 1 response, but both scenarios require a System 2 approach if we are to solve them correctly.
The Famous Married Puzzle
Google 'Married Puzzle' and you'll get around 7,000,000, so it is quite famous. If you've not seen it, it usually goes like this
It's often captioned '80% Get It Wrong' and indeed Alex Besos wrote about it in the Guardian and ran an online survey that attracted 200,000 responses of which just over 70% were wrong.
I first saw the puzzle some time ago but encountering it again today reminded me what a well constructed question it is, skilfully written to make it seem more puzzling than it actually is. Here's a different question
I hope it's not too hard to see the answer must be 'Yes'. Given the description we might think about the sequence of coin flips and the only possibilities are HHT and HTT, both of which contain the pair HT.
But the two questions have exactly the same underlying structure; we have just replaced 'married' with H , 'unmarried' with T, and 'is looking at' with 'was followed by'.
However we've lost the distracting baggage the original came with, a storyline that hints at the intriguing possibility of a love triangle (often the question includes a picture of the three people) and perhaps makes us feel vaguely that being married or not might have something to do with the solution. I've set my version in a context that nudges us towards think in a more analytical way and makes it easier to see the answer.
The Guardian article drew on an article in Scientific American that mention the original source of the puzzle as being Hector Levesque, a well-known researcher in artificial intelligence at the University of Toronto, who devised it for a research project. I've tried to find the original publication but haven't to date succeeded in identifying which of his publications it appears in.
In some ways it reminds me of the equally famous Linda Problem from the book Thinking, Fast and Slow, by Daniel Kahneman, which contrasts two Systems of thought: System 1, which is fast, intuitive and economises on mental computation, versus System 2, which is slow, methodical and computationally intensive.
System 1 excels when something is routine or demands rapid and instinctive response, whereas System 2 is needed in more complicated situations that require slow and careful logical thought to arrive at a correct solution.
Here is the Linda Problem
Most people choose Option 2 but that cannot be correct. There are more bank tellers than there are back tellers who are active in the feminist movement, just like there are more people who live in the UK than there are people who live in the UK and collect stamps. So the probability of 2 being true is less than the probability of 2 being true.
I love this example! The amazing thing to me about it is that when I first saw it I got the right answer but it felt wrong and for all the times I have explained it to someone or written about it, it still feels that 2 is correct. It stands as an example of how wrong an instinctive answer can sometimes be.
I think the link between the two problems, the Married Puzzle and the Linda Problem, is that the majority (and wrong) response in both cases is the System 1 response, but both scenarios require a System 2 approach if we are to solve them correctly.