5. Games and Specific Learning Behaviors
Now it is time to examine the specific learning behaviors exhibited while playing such a game.
The first learning behavior is the most obvious: gaining and interpreting information. Throughout the game, the player needs always to be aware of surroundings and to gather information about the back story of the game. This involves doing research, talking to other players (when possible) and using the interface to learn information. The learner also learns about alternative sources of information and may even join an online community to share such knowledge. This is a bit of a paradox, because as I point out later, most computer games are closed systems, while the gaming community tends to form in guilds, collecting knowledge and assembling it. It could be argued that many of the games have so many improbable escapes and impossible traps that a player could not play them without acquiring this specialized knowledge.
Within the game itself, gaining information involves two subtasks: filtering the available information and constructing queries that produce the right information.
The main problem with computer games is foreground and background. Important information should be available, but it should not be too visible or there would be no challenge in finding it. On the other hand, unless there are redundant clues pointing to such a location, a piece of information that is hidden in the background becomes as impossible to find as a needle in a haystack. A player can basically touch any object or place or person with a mouse to try to find more information. A successful game should teach a student to filter information and to recognize the most effective way to find information. To find a financial record for a bankrupt company, a player could touch every object in an office building, but that is not an efficient strategy. Over the duration of the game, the player might learn which rooms or file cabinets might be more likely to yield this information and not have to search through everything.
The other problem lies in the method to obtain information. For a lot of the role-playing games, the player might need to interact with the "non-player character" (NPC), a computer generated character that follows a preset script. Often it is impossible to converse with a NPC, but a conversation is usually just trying to give every conceivable response and walk through every little branch in the tree of possible conversation responses. This is a result of a programming constraint, but it really limits the ability of the player to learn while playing. To get the right answers you need to know the right questions to ask. But how do you know the right questions if you have no way to ask them?
In a typical NPC conversation (See Diagram: How a Player Obtains Information), the player initiates a conversation, and the NPC responds with an initial script. Perhaps the NPC holds important information—how does the player obtain it? Aside from walking through every conversation possibility, the the player can possess an object that triggers a a different script from the NPC. Or perhaps the player can perform an action that causes the NPC to respond in a certain way. NPC's might present information indirectly, perhaps with a rambling monologue full of local color and irrelevant details hiding vital clues. Indirect presentation at least taxes the player's critical thinking skills, but it is an inadequate solution to a crucial epistemological problem that lies at the heart of computer games: the inability to ask intelligent questions.
Or to put it in programming terms, the ability to query the computer game (represented by the NPC) for information. It is commonly assumed that game engines lack the ability to process natural language commands. But in the days of Ask Jeeves, the technology to make natural language queries is upon us.
It really isn't that hard. A linguistic interaction with an NPC would actually be like a database query (see the diagram, "Linguistic Interactions.)" If a player asks a question starting with the word, "when, " the computer knows that the question is really a query using X and Y filters where the result will be the Field "Time"?
If I ask "Where can I find water?", that question can be translated into this request:
"Return any data for the field "place" so that the agent "I" can perform the action "find" for the object "water." If the outcome is nil, then restate the question in the negative: "You cannot find water here."
"Does a wizard live in the village? "
The computer would act like this: Verify if the agent "wizard" performs the action "living" in the place "in the village." If the wizard exists which has a place "village" and has an associated behavior "Living," then answer with "Yes, one lives in the village on 4th street."
Obviously I am describing it in simple terms, and the increasing capabilities of natural language processing should make it easy game programs to establish a simple vocabulary and simple query structure for delivering information. The player could choose vocabulary items in a drop down box under each sentence part. If such a query engine could be incorporated into a game, then NPC's could assume a more central role in the game, and the emphasis of the game could shift from observation and following clues to constructing meaningful questions and interviewing. Once the player has the ability to formulate meaningful questions and receive meaningful replies from NPC's, the game developer no longer has to worry about exposition and can focus instead on constructing scenarios that involve investigation. Once this happens, computer games become by definition learning games.
A computer game, especially the more extensible systems, offer great creative possibilities. When a player needs to instantiate an object or a character, he or she needs to individualize it and to do the necessary research to give it verisimilitude. Games basically involve the interactions of 5 components: Actors, Objects, Contexts, Behaviors and History.
As gaming technology matures, a player may gain more ability to construct the virtual world. In MOO's, players already have the ability to create and manipulate objects and contexts. The construction of these elements involve the mastery of other related domains. For example, creating a context means mastering the domain of geography, architecture, navigation and even the social setting for a particular time period. If a game takes place in Renaissance England, then the player will need to become familiar with the use of space within buildings, plumbing, architectural constraints and the décor for the surroundings, as well as the ambient environment, the weather, the sanitary conditions and navigational aids.
To successfully create and manipulate objects, a player would need to learn a little about mechanics, functionality, chemical composition and physical characteristics of the object being constructed. In addition to providing a link to learning domains like mechanics and physics and chemistry and archeology, a game with these sort of objects can call into question the role of ownership, commodification and consumerism in our society. That may sound like a stretch, but in the case of the "Sims," the accoutrements for the house (the stereo, the swimming pool, the cabinet) become the focal point of the game. A player won't be able to help wondering after only a few hours of game play whether acquisition of Sims objects actually furthers one goals and whether this modern American emphasis on material possessions is just plain silly. Is the player playing to make friends or to make money or to increase the overall number of material possessions?
"Behaviors" are another component in the gaming world. Behaviors are associated with objects, actors or contexts. Objects can be associated with behaviors which can be performed by actors. Actors are allowed to perform certain behaviors at any given moment (defined by objects, contexts and history). How does one explore the permissible behaviors for an object or agent? The main way is by trial-and-error, experimentation and perhaps inductive reasoning. When a player witnesses an unusual or inexplicable behavior, he or she needs to find what object or actor caused it and why. Often, a player may need to make a connection between the behavior of a door (for example) and the button which is pressed to open it.
The main way that a player can understand a behavior is to replicate it. In a similar way, a scientist tries to understand a natural occuring phenomenon by trying to replicate it. One challenge to a game is guessing the function of an object by its outward appearance. This is no different from how an archeologist analyzes a find and tries to reconstruct its original purpose.
History is another component in gaming. It refers to a chronology of events. The history of a actor can be equated to that actor's personal experience and the acquisition of objects or navigation through certain contexts. The history of an object might affect the ability of an object to work properly, or it may refer to the previous owners of the object or the previous ways in which the object could have been used. History might affect the permissible behaviors for actors, objects and locales. If a magic wand is used too many times, it may no longer be able to cast spells.
Game history might also tie into the actual study of history. It might involve the act of interpreting runes to find a relic or in understanding how a previous chain of events had an impact on the gaming world today. The ability to create realistic game components depends on placing them in a historical context or at least in recognizing how and why change occurs.
The final and perhaps most crucial game component is the actor. This refers to the character the player builds as well as any NPC he creates. At the moment, almost no games allow players to create additional NPC's, but this is not outside the realm of possibility. In the original Dungeons' and Dragons, the Dungeon Master player had this capability, and today for games to really take off, the job of creation needs to be taken off the backs of the gaming companies and put on the backs of willing volunteers.
For the time being though, the player's main task will be to create a virtual character for the game. This alter ego will be based upon a template provided by the game, but it is up to the player to personalize it. For Everquest and other games, the player can create gender, race, dress even religion. As avatars become more common, creating an appearance, a style of dress and talk seem necessary for playing the character well.
Role playing involves a certain amount of empathy and understanding a person with values that may differ from our own. Playing an alter ego who bears no resemblance to who the player plays in real life allows the player to play with different identities and also to reveal aspects of personality that might not normally come out in normal life. A chemistry geek can imagine herself a ballet dancer; a jock may imagine himself a successful businessman, and for once they are not shackled by societal expectations. (The veneer of anonymity helps a lot).
When portraying a character, a player needs to deal with the issue of stereotyping and of playing a role in a way that does not reinforce a stereotype. In fact, often players do this anyway, but usually it is the form of external characteristics. But when the player runs out of physical or vocal mannerisms, one has to worry about creating a plausible character, a requirement that is even more important in the multiplayer game. A player who cannot generate an interesting character may be ignored by other players.
The game allows the player to see the dichotomies between actual role and fantasy role of players. Part of role-playing is developing an empathy for people with unusual characteristics and of ignoring the physical attributes that normally obsess people in real life. Since it is so easy for a player to choose a nice-looking avatar, the player learns to look beneath the surface and to realize that appearances are not always what count in real life. The ability to create a virtual character certainly affects gender attitudes. Knowing that a male can play a female and vice versa is a reminder that identity is a society construct, not only in the game universe but in real life as well.
Games like these educate players in starting and maintaining virtual relationships. They aren't really that different from real relationships, and maintaining one means understanding the mutual need for privacy and acceptance of the limited role one plays. A person may develop dozens, even hundreds of extremely casual acquaintances online, and a game can be an environment for how to manage them successfully. It is the modern day dance floor or water cooler. Perhaps one doesn't play an online game to chat, just to play the game. But it is soon clear that another player's background or race and gender has little to do with that player’s ability to play the game.
With online role playing games, there is ample opportunity for cooperation and helping others towards a common solution and convincing others who might disagree with you. That is the active part. Playing any game involves setting a strategy and comparing alternatives. Especially when other players are involved, a good part of the game involves making predictions about future events. If my character hides behind this boulder, will the troll be able to find me? If my character tries to fight a group of trolls, will other trolls come? If my character tries to cross the bridge, will it collapse from the weight? If my character acts like a peasant, will other players be willing to share information about the sunken treasure?
