Ambiguity and Domain Knowledge – Two Pillars of Puzzle Design

By Clara Fernández-Vara
Share this article

Puzzles tease us constantly. They point out to a gap, a mystery, something that we do not know. Most of us cannot stand the uncertainty of not knowing – we must fill that gap, solve the mystery, learn the information that is eluding us. That moment when we achieve and solve the mystery is the moment of insight, as I explained in my previous blogpost.

The levels of teasing also have to do with the essential paradox that puzzles are built on. Puzzles hide information at the same time that they demand it, as Helene Hovanec explains in The Puzzler's Paradise.

Let us look at this riddle: “What type of cheese is made backwards?”

Think about it.

I will wait for you.

Do you have an answer?

The answer is “Edam.”

Which is “made” backwards, as the text of the riddle was asking.

Many puzzles are relatively easy; in this case, the answer to the question was in the text itself. You may have come up with the solution right away and this may have not been a challenge at all, because you are already a well-trained mind gamer. If you did not come up with the answer, do not feel bad about it – it was not meant to be obvious. You fell in the trap of the riddle, probably because you were expecting more difficult, as it would fit a website that is filled with challenging and mind-bending puzzles. Perhaps you thought that you may have to know something about how to elaborate cheeses. Turns out that you only had to know a bit about different types of cheeses from all over the world, or at least know the name of this specific cheese from The Netherlands.

That is the essential paradox of puzzles. They are based on a tug-of-war between the person posing it and the person who has to solve it. But unlike a tug-of-war, a successful puzzle does not involve one side winning over the other; rather, a good puzzle involves a well-designed problem, which will entice the player and keep them thinking until they arrive at the solution. And when they do, as I described in the previous post, they will be happy, they will feel clever because they just learned a new thing. The puzzle designer, the riddler, does not necessarily win by presenting a puzzle with an impossible solution that nobody can figure out. How can we know that the puzzle actually has a solution? A designer who defies their players with an impossible challenge is cheating, because they are robbing players of their moment of insight, of their reward after a lot of brain-unravelling. Same with puzzles where only the puzzlewright knows the solution, because it does not make the challenger more intelligent or accomplished than the person trying to solve the puzzle. There has to be a solution, even if it is convoluted and it hurts your brain—the relief comes when you solve the puzzle.

This means that a well-crafted puzzle has to be a balanced tug-of-war, where the solution breaks the rope and everyone falls gracefully to the ground and laughs. Finding that balance is an art in itself, because when we devise a puzzle, we have to think about what kind of player we are creating it for. In my previous blogpost I explained that in the process of solving a puzzle what we do is connecting information. Where that information comes from is a different story. In the case of the riddle above, we have all the information in the text in front of us – we just have to spell “made” backwards and that is the answer. But we also have to know that there is a type of cheese called Edam. If one is not much of a cheese connoisseur, the solution may not click because we are missing the key piece of information.

The kinds of information that are needed to solve a puzzle are called domain knowledge, a term that defines the related concepts within a specific field or topic. The domain knowledge of cheeses, for example, involves knowing their types, how they are made, what kinds of milk are used to make them, how they taste, where they come from, or how they pair with other foods. Some domains are general and shared by many people within a culture, such as how to behave in a public space, or how to buy things in a store. Others are very specialized, such as how to fix electronic appliances or trading in the stock exchange.

If the knowledge the player needs to solve the puzzle is easily accessible or teachable, the puzzle can start being very easy. That is the case of sudoku puzzles, where knowing the different numbers and the rules of the puzzle is enough to get us started. Magic squares can be more complicated because they involve some basic – or advanced – algebra in order to solve them, which is a more specific knowledge.

Some puzzles teach us a new domain knowledge in order to solve them – the videogame Portal involves opening inter-dimensional puzzles with a portal gun in order to navigate a space and getting from point A to point B. Designers could not expect players to know about impossible physics in order to tackle the challenges on each level, so they devised tutorials that taught players how to open and close portals and how to use them to their advantage. That does not mean that the puzzles are easier, it is just that the player is equipped with the knowledge to solve them – making the connections is still up to them.

Designing a puzzle then requires being aware of what information the player needs to know in order to solve it, and deciding how much of that information we want to provide them with and what they should figure out on their own. It is okay to create a puzzle where finding out that information is part of the challenge – but then game designers need to anticipate whether their players are up for it. Players participating in a competition are ready to take on the challenge, whereas people playing casually may walk away from puzzles that feel like homework rather than an entertaining brain teaser. It is not a matter of intelligence or studies, but what kinds of domains the player is familiar with.

About the Author

Clara Fernández-Vara is the co-founder of Fiction Control, a narrative design company, and Associate Arts Professor at the NYU Game Center. She's a game designer and writer as well as an academic, so her work combines scholarship with the creation of narrative games both for research and in the commercial sphere.

Share this article