Category Archives: game studies

Game Studies: Rules, Emergence, and Information

Of all the games we played last week, I was most interested by Mastermind, as it felt the most systematic and logic-based. From the six base colors and the four possible slots for these codes, you had 6x6x6x6, or 1,296, outcomes initially. I found myself thinking through my succession of moves each round logically whittling down these potential outcomes, in a somewhat mathematical way.

In other words, with the code already made, I was free to work within my own head and not respond to the moves of the other player in an ongoing, emergent way, unlike Blockus and Connect Four. I could stick to my own system or strategy and not have to worry about the way other players shifted their strategy. Over a succession of rounds in Mastermind, this may change, but in the context of the night, it did not. It was almost like the game was completely inside my own head as a series of logical possibilities scaffolded over a series of decisions.

This in-my-head quality made the game feel more like what Salen and Zimmerman say about “decision trees,” where one could trace out a consistent set of approaches, including a winning strategy, through a flow chart. With the numbered layout of six colors and four slots, the game has a series of discrete decisions mapped along two possible axis of variability, unlike Chess with its range of varied moves. But this only describes the possible moves, which is where the role of information came in.

Again, in Blockus and Connect Four, my move changed based on the information of the ongoing moves done by the other player(s). As Chris encroached on my territory, for example, I started making a stronger effort to block him. But in Mastermind, my moves changed based on the information provided by my own past decision. In this way, strategy was still emergent in Mastermind as I got new information, but the components of the game–like the input of other players–affected that emergence less. I could create an arc of moves that, theoretically, could be almost algorithmic and unchanged as each guess progressed.

In the information theory approach, as Salen and Zimmerman define, “information is a measure of how certain you can be about the nature of a signal” (193). As they breakdown Mastermind, as the guesser gets more feedback about the nature of the code through the black and white begs, “the guesser narrows down the possible answers (decreasing uncertainty), carving out a single guess fro ma range of all possible guesses” (194-95). Initially, the uncertainty is total–1,296 possible outcomes–but each round reduces the possible answers, so that by my final guess, I was no longer guessing. I logically knew exactly what my answer was.

In the other games, I never experienced this level of certainty. There was no set “strategy,” just an emergent web of possible strategies based on the changing game board, particularly in Blockus. I could get a sense for where people may go, but I never “knew.” I got a sense, as Even points out in his post, that a more expansion-oriented strategy seemed effective, or that certain pieces could fit into key choke-points for other players. But chance, or uncertainty, remained high. Noah, for example, could deliberately not play the “right” move, as we often did in connect four to delay the victory. Or one of us would miss the right move. Or a new move would come up based on the third or fourth person. All of these variables would, in turn, affect my move, feeding back into the system and affecting theirs.

In this way, Blockus, and to a lesser extent Connect Four, felt more “complex” and more uncertain than Mastermind. Or perhaps, it felt more emergent. But importantly, as Salen and Zimmerman point out, these feelings may simply be feelings, and the formal backdrop, the “constituative rules,” of all the games had a similar logic that operationalized in different experiences.

ENG 730: “Werewolf” and Fundamentals of Play

I had played “Werewolf” a handful of times in the past, mostly at parties with larger groups of people. These past times also had different variations, like the inclusion of a witch who could silence a villager in the night and no dead goat/cow/chicken/corn to start the game.

This initial dead goat/chicken/cow/corn felt like one of the more significant parts of the game, as (unless you are the Seer and get lucky) you lack information, making the choice of the vote more random. Salen and Zimmerman’s concept of the “anatomy of choice ” highlights the character of this initial vote I think. As they break it down, the anatomy of choice includes the following five steps:

  1. What happened before the player was given the choice?

  2. How is the possibility of choice conveyed to the player? 

  3. How did the player make the choice?

  4. What is the result of the choice? How will it affect future choices?

  5. How is the result of the choice conveyed to the player?

If we take the “voting of a villager to exile” as our choice, then some of the answers to these questions would be the same in latter votes. For example, the possibility of choice is conveyed through the narrator announcing the daylight and the need to vote, and the immediate result would be the showing of the player’s card and their exile.

But other elements differ. For one, the dead goat offers no information, so in the initial vote, we are all suspect. Initially, we had no way to really back up a decision–no answer to the “how” in an experiential sense for making our choice. Later on, we could use evidence from the course of play, but the first vote feels more awkward, dangerous, and random.

I feel like that created our role-playing. With the addition of roles, it eased the difficulty of this initial vote because we could essentially make up narrative reasons, easing the sense of randomness. As time went on, however, we seemed to embrace more of the arbitrary nature of this vote with a “go for it” approach. The randomness also heightened our appeal to the gambler’s fallacy and other more probabilistic thinking. Later votes invited more meta-gaming, analysis, and psychology.

The result of the choice, in a larger sense, was also important. If the villages got the wolf in the first turn, winning was much easier, raising the stakes.

I think what makes Werewolf an interesting game is that the choice–while the same procedure through each round–changes so much as the in-game context changes. As Salen and Zimmerman note about design, “Design is the process by which a designer creates a context to be encountered by a participant, from which meaning emerges.” Almost like stages in a game-show, as contestants get eliminated, each new vote provides a new “level,” “round,” or context to be encountered by the players as participants. And as the context changes, the experience changes. Patterns and skills emerge, but each vote presents a new problem to solve, a new experience of “meaningful play.”

This creates an elegant scaffolding between the “macro” and the “micro” levels of choice and outcome, to use Salen and Zimmerman’s terms. Each round, the available information changes and the stakes change in terms of our immediate “quantifiable” goal, and while the larger strategy for either wolf or villager may be the same, the tactics for each round are flexible and rhetorically situated.

Early on, for example, the wolf may want to appear completely innocuous and invisible. As time goes on, the wolf may want to eliminate more villagers, especially in the penultimate round. Or, initially, the wolf may kill the other wolf, making them appear innocent for the rest of the game. The round helps determine the more productive choice, to which the player must respond and be sensitive to, while maintaining the larger goal in mind.

Works Cited:

Salen, Katie, and Eric Zimmerman. Rules of Play: Game Design Fundamentals. Cambridge, MA: MIT P2004.