Today Jim Cummings sent me a link to a Youtube video that attempted to define the difference between puzzles and choices.

Here it is: Game Design Corner

Its a bit old, but Jim C. commented that this video had some interesting food for thought, and I agree with him. In it the James Portnow and Daniel Floyd attempts to differentiate between what he considers two distinct elements of game design. The first is **the problem** and the second is **the choice**. In the video Portnow claims that the two are different in that puzzles have solutions and choices don't. His ultimate argument is that choices and problems are two distinct tools and that game designers should be able to differentiate between them. If they can't then game developers can't hope to develop "art".

Now, what I am going to do is throw out the entire argument that games can be/are art. I won't address this here. Instead what I want to do is make a quick point about the distinction between choices and problems. Basically, I don't think they can be distinguished the way Portnow claims. I think he points out an interesting problem, but the theoretical distinction he is making doesn't take into account a few very important concepts.

What is a problem? What is a choice?

According to Portnow a problem has an explicit reward. The problem has a solution. The solution is discoverable. So what happens when the problem space is large enough that discovering the optimal solutions is difficult or impossible. This is actually the crux of what Jim C. wrote about in his last post (Jim C. and I have also published a paper on this that is currently in press). The problem of too much choice results from two things. There may be too many choices in the problem space (in fact the problem space may not even be definable past a very short period of time) or the information environment might be too complex or unfriendly enough that the optimal choice becomes incredibly difficult to determine.

Now, according to my interpretation of Portnow these problems would qualify as choices. He claims that problems have a discoverable solution and choices do not. This is not something that I can agree with. I believe that the difference between what Portnow calls problems and choices boils down to more than that. Here are some situations in which players would not be able to, or perhaps not want to spend the time, searching for an optimal outcome. Yet they are still faced with something that has an optimal outcome, or at least some outcomes that are "better" than others. Let's explore some of these:

1: The problem can have too many choices - There are too many branches in the choice tree for the optimal solution to be easily discovered. Determining the next move in a game of chess for a human is this type of problem. Yes, there maybe an optimal solution. There maybe many solutions that are optimal. In fact, since the next move depends on the probabilistic function of the opponents move there is no way to determine the "best" solution. In order to do that you have to have full information about your opponent something you can't have.

2: The problem can have a unfriendly choice environment - When a player is presented with a list of choices and does not have the expertise to differentiate the optimal strategy then the solution to the problem may not be discoverable. There maybe choices where the attributes are negatively correlated. Perhaps the choices are both useful, just in different contexts the choosing the best becomes difficult.

3: The problem can have a stochastic element - As I discussed above if a problem has a random element to it - playing against another player, a random number generator, etc. - then it becomes difficult to find an optimal outcome. In game theory (no different than games for fun) your outcome in a multiplayer game depends on the choices of other players. The outcome may even have multiple or context dependent Nash equilibrium meaning that the solution to the problem depends on the trajectories of many different players and small shifts in early conditions might change your optimal strategy. Think about paper, rock, scissors as an example. The best choice for me depends on the choice of my opponent. If I have to play my choice for a number of rounds then the best choice is the one that beats the largest percentage of other players. When it becomes time to choose again after playing a few rounds then the best choice is the choice that will beat the choice that I think the largest percentage of players will choose.

4: The player may have intrinsic motivations, not understand their own motivations, or there maybe individual differences in motivation. Not all games are transparent. It is not immediately obvious to players what choices are the best. In fact, in a sufficiently complex/changing environment the value of extrinsic motivations can consistently change (see number 3). Intrinsic motivations can also change in that players may not know exactly what they are motivated by. My own personal intrinsic motivations can change based on my life experiences, or how hungry I am. Portnow suggests that looking for a career is a choice; however, I think it is a problem that is simply driven by a variety of extrinsic and intrinsic motivations.

Basically, what I am saying here is that there are many different types of problems. Not all problems have an explicitly locked in optimal solution. Some of them have too many variables to determine the optimal solution, and some change over time. Having randomness, dependence on other players, or incomplete information does not change a problem to a choice. I want to thank James Portnow and Daniel Floyd for the insightful video; however, I also believe that the definition that he presents is inaccurate and fails to describe the tools of problems and choices in sufficient detail to actually make them useful.

*Choices and Problems: Are they different?* by *Travis Ross*, unless otherwise expressly stated, is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License.

I think I'm a bit more forgiving of the video than you are, Travis. I think he's on to a pretty fundamental difference here, though I agree that it's rarely a clear distinction in real game environments.

One thing I wonder about how players actually handle choices in the strict sense (where there is no 'right' answer). The video touches on this, but I think it's important to ask if players EVER really think something is 'just' a choice. Apart from something like customizing the appearance of your character (and perhaps even then in some games), my intuition is that players always assume that there is a best option for any given in-game decision. This idea could use more elaboration, but that's my two cents for now...

Jared, I agree with you on one thing for certain. I am not sure about the distinction. And really I myself should continue to elaborate on the problem. I think that the video introduces some interesting concepts to think about I just don’t think they are sufficiently, or perhaps even correctly, explicated.