In our final meeting for the month of March, the prototyping team continued to examine the concept of a game with enemies representing the five senses. Our prototype from the previous week had involved only one enemy and one sense (sight) and was a puzzle that required the player to alter his silhouette with various strange items to disguise himself. This week, we tried adding more enemies (and thus more senses) to the game.

We decided that we wanted to continue with the idea of having an inventory or tool set that the player could use to disguise themselves. This time, though, instead of providing physical items that the player could pick up, we made the protagonist into a shape-shifter, and their inventory was the different characteristics or attributes that they could take on--there were five possible characteristics for each sense, like "triangular" or "square" for sight, "rattly" and "shrill" for hearing, "sticky" or "rough" for touch, and so on. The main idea of the game was that the player has to move across the board while avoiding the five enemies or shape-shifting to disguise themselves from one or more enemies as they moved into range.

Our first iteration divided the game board up into a 4 x 3 grid in which each section of the grid had a set of five characteristics associated with it, one characteristic for each of the five senses. For example, a section might be designated to be "bitter, round, cold, squeaky, and sweet-smelling." Five enemies--each of which had only one sense available to them--moved in set paths about the board, and the player had to try to get from one side of the board to the other without being detected. The player and the enemies moved in turns, and at the end of the player's turn he had to choose three attributes from a list (each attribute had to be from a different sense), and think of an object to shift into that had all the desired attributes (if they were unable to think of an object with the desired attributes, they would have to choose a different set of attributes). For example, if the player's turn ended in the section of the board described above, they might choose to take on the traits "round", "cold", and "sweet-smelling", and declare that they were transforming into key lime pie.

If an enemy ran into the player, the enemy would check to see if the player's current shape matched the section of the board they were in. If the player's shape matched the monster's expectations, then the player would escape detection. However, if the monster detected a trait that didn't fit in the section, or detected no trait all for its particular sense that is, if the player didn't choose a characteristic that corresponded to the monster's sense), the player would be discovered. In the previous example, the player would not be detected by the smelling monster, since they had the "sweet smelling" attribute, but WOULD be detected by the tasting monster, because they don't have the "bitter" attribute. We decided that when the player was detected, they would have to choose one trait from their list to give up forever--that is, for the remainder of the game, they would be unable to transform into an object with that characteristic.

At this stage, the game was still very simple. Our first playtester enjoyed trying to think of objects with the required traits, and found the task both challenging and amusing. However, he felt that having to give up a trait upon discovery didn't seem like enough of a consequence, since it was fairly easy to figure out which traits would be necessary and which could be safely discarded. He also found navigating the board to be too simple, since the obvious strategy was to stick close to the walls to avoid the enemies and there was nothing to prevent the player from doing this.

To address these problems, later versions of the game had more complicated layouts that gave the players more choice or forced the player to backtrack. We also made the grid finer, allowing for more sections with their own distinct sets of associated characteristics, and changed the rules of enemy movement to try to make them less predictable. We hoped that these changes would make navigation less straightforward and would make the choice of which attributes to sacrifice more difficult. Further playtesting revealed that while we were (eventually) somewhat successful in creating a more interesting experience when it came to navigation, players still found it obvious which traits they should give up. In addition, making the enemies' movement patterns clearer meant that it was trivial for the player to decide which three attributes (of the five associated with each section) to take on at every turn, because they could easily predict which enemies would move into range.

On the other hand, all our playtesters liked the overall concept of enemies that represented the five senses, and many of them had a lot of fun trying to come up with objects that would match the set of three traits that they had selected (some of the objects that players came up with were hair gel, ninja stars, rusty drainpipes, and various kinds of pie). The task of future sessions, then, is to build on these aspects that clearly work while we try to solve the problems that our playtesting identified. Clearly we need to make significant changes to what happens after an enemy discovers you, but other possible directions to investigate include: changing the layout so that the player is wandering through rooms of a house rather than simply going from start to finish on a mostly linear path, maybe adding some exploration so that the player only finds out what the required traits of a room are when they enter; revising the shape-shifting system so that instead of giving players a list of characteristics to choose from, we give them a library of objects they can change into, and they have to figure out which ones have the attributes that are needed in a certain area; or adding difficulty to the game by putting in some randomness or adding some real-time sections.