لطفا منتظر باشید ...
Rewards and Schedules
Operant conditioning not only affects the kinds of choices players make during the course of a game, but also their general motivation to continue playing. More than just shaping good and bad behaviors, rewards and punishments shape a player's sense of pleasure and overall play experience. Game designers Neal and Jana Hallford point out this design challenge: It's surprising how many developers forget that it's the victories and the treasures-not the obstacles-that make people interested in playing in the first place. If you stop giving out the carrots that will keep players excited, or even worse, if you start punishing them for their curiosity, you're only going to drive away the very people who want to enjoy your game.[17]
Keeping players engrossed in your game as they play is the second of the two seductions of game design. Hallford and Hallford are absolutely correct that players need to be rewarded, that they need to accomplish tasks and feel satisfaction as they play. Although punishments are important, on balance a play experience needs to be pleasurable. Otherwise, nobody is having any fun.
What kinds of rewards can games offer players? There are as many kinds of rewards as there are forms of play. Hallford and Hallford list four general types. Although these categories were written about computer role-playing games, they suggest the kinds of rewards other kinds of games might contain.
Rewards of Glory. Glory rewards are all the things you're going to give to the player that have absolutely no impact on the game play itself but will be things they end up taking away from the experi-ence. This includes winning the game by getting all the way to the end, completing a particularly difficult side quest, or defeating the plots of evil monsters.
Rewards of Sustenance. Rewards of this nature are given so the player can maintain their avatar's status quo and keep all the things they've gained in the game so far. This might include health packs that heal injuries, mana potions that increase a player's magical abilities, high-tech armor that shields a player from e-mag radiation, robots that remove curses or diseases, or even storage boxes or beasts of burden that allow a player's avatar to carry more resources along with them.
Rewards of Access. Rewards of access have three critical features: they allow a player access to new locations or resources that were previously inaccessible, they are generally used only once, and they have no other value to the player once they've been used. Keys, picklocks, and passwords are typical examples of this kind of reward.
Rewards of Facility. Rewards of facility enable a player's avatar to do things they couldn't do before or enhance abilities they already possess. When well handled, they should increase the number of strategies and options that player will have for playing the game. A good example of a facility reward might be a magic orb that lets an avatar walk through a stone wall or a cybernetic software up-grade that lets them shut down enemy gun turrets from a distance.[18]
Punishments, negative reinforcement, and positive reinforcement are important game design tools. They not only teach players what actions to take and not to take in a game, but also craft larger structures of pleasure. These structures assure that players are properly rewarded for spending the time to take part in the experience designed for them. But using reinforcement successfully in a game means more than just knowing what kinds of pleasures to provide. It is equally important to know how to integrate rewards and punishments into the experiential structures of a game. How often does reinforcement occur? How powerful is the reward or punishment? Do reinforcement factors change over time or remain the same? Behavior theory has devoted much study to reinforcement schedules. A reinforcement schedule refers to the rate a subject is given reinforcement over time. These reinforcement patterns, along with a network of integrated rewards and punishments, help shape the fabric of any game experience. There are two basic kinds of reinforcement: fixed and variable.
Fixed reinforcement means that rewards or punishments are occurring at a steady, continuous rate. A fixed ratio means that the outcome occurs a set number of times that the behavior is performed, such as a player getting a chevron for every five waves of aliens defeated. A fixed interval refers to a regular amount of time between reinforcements, as when a power-up appears in a game every 30 seconds as long as a player can stay alive.
With variable reinforcement, the rewards and punishments are coming at irregular intervals. Variable ratio means that the outcome happens after an irregular number of intervals, like slot machine payoffs that occur at a random rate. With a variable interval, the reward or punishment occurs at random time intervals, as in mechanical children's games like Don't Wake Daddy, in which daddy will wake up (with negative consequences) after a random amount of time.
Let's invent a fictional game called Unlocker to illustrate several points about classical conditioning. Unlocker is a straightforward 2D computer game where the player controls an avatar seen from a top-down point of view. Moving through a series of rooms, the player must avoid traps, collect weapons, fight pursuing enemies, and collect keys that unlock doors to additional rooms and levels. Although combat can occur, it is not the intended focus of the game. The goal of Unlocker is to unlock as many doors as possible and earn the most points before dying. Even this relatively simple game contains many kinds of objects and events: open and locked doors, hidden keys, mobile enemies, movement, combat, manipulation of an inven-tory,and so on. Because many players will not read the instructions (and just as many will forget them soon after reading), how do you teach players what they are supposed to do in the game? Rewards and punishments are one means of shaping their behavior. The overall trajectory of the game is to open doors and move on to new rooms and new game levels. So you want your players to open those doors. Imagine the first time a player finds a key and uses it on a locked door. A sprite animation plays and the player sees the door swing open in the game space. The problem is that there is nothing in the game to let a player know that unlocking the door is a valuable action that brings the player closer to a positive outcome. The solution? Reward the player! Give the player bonus points for unlocking a door and make sure to add a Ka-ching! sound to emphasize the event. Make a gold star appear in the interface when a player unlocks a door-maybe five gold stars earn an extra life. Or flash a message on the screen that congratulates the player and shows a map to the next locked door. Each of these possible solutions represents a different way to reward the player for the action of unlocking. All of them combine internal, system-based rewards (points, extra lives, information about the next door location) with external, audiovisual rewards (sound effects, gold stars, congratulatory text). If you can craft the proper reward for your player, you will create a desire to achieve that satisfying reward event again, and the player's actions will follow suit. The same is true of negative reinforcements and punishments. By providing unpleasant feedback, you can teach your player what not to do in your game. Let's say your intention
Fixed schedules are best at shaping behavior if the subject is being punished: sending a child to his room every time he performs an undesired behavior is much more effective than sending him to his room only some of the time. On the other hand, for many kinds of reinforcements, especially positive ones, variable schedules are more effective. In gambling, players are usually rewarded at variable ratios. The repeat play of gamblers is strong evidence of the power of variable reinforcement. One game designer known for integrating ideas of operant conditioning and reinforcement schedules into his work is Gabe Newell, lead designer on the computer game Half-Life. During a panel discussion at an MIT conference on gaming, Newell discussed the way that Half-Life's design integrates these concepts: The rewards in Half-Life are getting to see new monsters, the plot is moving forward, getting to have a new fun weapon, getting to see something really cool…. You want to make sure that throughout the course of the game that they're getting rewards….You want to look at it from the point of view of a reinforcement schedule and say OK, that makes sense. I mean there were points in the game before we shipped where there were long lulls where basically all you were doing was stuff that you'd already done before, which in our view didn't represent a reward. So we said we've got to put more fun stuff in here or eliminate that section from the game.[19]
Newell makes a number of important points. He begins by listing some of the chief rewards that Half-Life provides for players, which include what Hallford and Hallford call rewards of facility (new fun weapons), access (meet new monsters and experience plot twists), and glory ("getting to see something really cool"). He also emphasizes the fact that it is rewards that sustain players through the course of a game.When his design team felt like players were not consistently receiving substantive rewards, they either increased rewards in that section or removed it entirely from the game. Half-Life utilizes principles of operant conditioning in other ways as well. Much of the success of Half-Life has been attributed to the way that it shapes player experience, creating a thriller-like tension while drawing the player slowly into its dangerous and mysterious spaces. For example, the intensely combative moments in the game are interspersed with uneventful stretches. As opposed to a typical "enemy lurking behind every door" structure, Half-Life creates uncertainty about when and where the terrifying mutant monsters of the game will appear. In some game levels, most doors do not open up to an opponent but instead to empty space. In these levels, Half-Life uses variable ratio punishments. Sometimes the player will be attacked when he opens a door or rounds a corner, but usually he is not. The experiential result of this design strategy is that in Half-Life, deadly threats seem to lurk in every dark shadow and beyond every closed doorway. Deploying enemies with restraint, creating a sparse pattern of unexpected, horrifying encounters, results in a more powerful experience through the use of fewer game elements.
The question arises: if rewards and pleasure are the keys to keeping players involved in a game, why don't tension-inducing punishments such as variable monster attacks drive players away? How can seemingly negative emotions create seductively positive play? We could answer by referring to similar pleasures in other media: the frightening ghost story or the gripping sci-fi cinema thriller. But there is a deeper principle at work. To play a game is to experience its pleasure. At the same time, we know from the lusory attitude that part of playing a game is to take on artificial challenges, inefficiencies adopted for no logical reason except that they make play possible. It is surely challenging to get a golf ball into a tiny little hole so many yards away on the green, but that hardly stops golfers from playing, just as the anxiety of Half-Life doesn't cause players to exit the program. On the contrary: challenge and frustration are essential to game pleasure. Without them, there would be no game conflict to struggle against and no pleasure would emerge from the process of overcoming adversity.
is that the game play of Unlocker is more about finding keys and unlocking doors than about combating enemies. In this case, you may need to find ways to punish players that go against the grain of the game. If combat is too rewarding, either in terms of structure (such as earning points) or experience (such as cool combat effects), your players will keep on fighting because the game intrinsically encourages them to do so. What are some solutions to this dilemma? You could make the enemies tougher, but that might simply encourage players to rise to the challenge.There are better "punishments" that can steer your players away from fighting. For example, maybe players earn no points from killing enemies, only from finding objects and unlocking doors. Or perhaps when they die as a result of combat, they are sent back to an earlier level. Or when they die, they lose points and game resources. Obviously, you do not want to punish the player too harshly for fighting or dying, but you do want to nudge them in the direction that you designed for them. Of course, if you continue to observe your Unlocker playtesters engaging in too much combat and not enough door unlocking, it might be telling you something else. Maybe something about the game's core mechanic makes combat more compelling than the hunt-and-unlock activities you intended as the game's focus. Perhaps you should turn Unlocker into a combat game. On the other hand, you could always remove the combat component entirely. It's your design. You decide.
Reward and punishment are two sides of a coin, both of them necessary to craft the structure of meaningful experience for players. Finding that elusive balance between positive and negative experience-between anxiety and pleasure-is one of the deepest challenges of game design. In the following section, we engage directly with these important questions. [17]Neal Hallford with Jana Hallford, Swords and Circuitry: A Designer's Guide to Computer Role Playing Games ( Roseville, CA: Prima Publishing, 2001), [18]Ibid. p. 157-160 [19]Gabe Newell, "Computer and Video Games Come of Age. A National Conference to Explore the Current State of an Entertainment Medium." February 10-11, 2000. Comparative Media Studies Department, MIT. Transcripts. Henry Jenkins.
