The book is called Mind At Play: The Psychology Of Video Games by GEOFFREY R. LOFTUS & ELIZABETH F. LOFTUS Basic Books, Inc., Publishers New York

Preface
[SPOILER:a5d5496eb9] PREFACE
Our aim is to shed light on the intriguing phenomenon of video games.
Along the way, we'll introduce some of the most farranging ideas in
modern psychology and also provide an entrée to the world of
computers, for the appeal of the games is largely psychological and
video games owe their very existence to the computer revolution.

When we set out to write a book on the psychology of video games, we
tried to adopt a relatively neutral stance. We read everything we
could about the games. We went to video arcades to play the games and
to talk to the owners, players, and onlookers. We talked to parents
and critics. At the same time, we explored the contributions that
research in psychology might provide for an understanding of the
games.

After we had poked around for a while, some themes began to emerge. A
major one was the computer theme: video games are fundamentally
different from all other games in history because of the computer
technology that underlies them. The marriage of games and computers
has produced both costs and benefits. It enables, for example, the
design of games that are extremely compelling to play. Critics would
call the games addictive. Proponents would call them great fun.

A second theme involves ability. Playing a video game requires
intricately tuned skills. How are these skills acquired? What are the
mental components that go into them?

A final theme revolves around education. We believe that the games
combine two ingredients—intrinsic motivation and computer-based
interaction—that make them potentially the most powerful educational
tools ever invented. We have discovered, much to our delight, a number
of research projects that are striving to harness this educational
power. Some are succeeding. More will succeed in the coming years.

While writing this book, we've had help from a variety of people who
deserve special thanks. Craig Raglund provided a number of perceptive
suggestions about the potential uses of video games in education. Hank
Samson and Jim Diaz, who are much better players than we've yet
become, engaged us in lively discussions about reinforcement. Ellen
Markman, Delia Gerhardt, and Brian Wandell read and provided useful
comments on early versions of several chapters. And, finally, there's
no way to adequately thank Judy Greissman, our editor at Basic Books,
who initiated the whole idea and who did a magnificent job shepherding
it through all stages from start to finish.[/SPOILER:a5d5496eb9]
Chapter 1: Videomania
[SPOILER:a5d5496eb9] CHAPTER 1: VIDEOMANIA
Venturing into a video arcade, you find a decidedly mixed crowd. To be
sure, most players are "typical teenagers," who play the video games
for at least a few hours every week. 1 But a not uncommon sight is the
corporation executive, the housewife, the construction worker.
According to one survey, about half the game players (in arcades and
elsewhere) are over the age of twenty-six. 2

The economics of the video game craze are staggering. Each year more
than $5 billion is spent in the video arcades alone. 3 And while the
video parlor operators are busily collecting their quarters,
microcomputer manufacturers are expected to make similarly large sums
selling both home computers and the software to go with them.
Advertisements for home computers in traditional publications describe
the virtues of keeping the checkbook balanced, maintaining Christmas
card lists, and teaching the children to program. However, by far the
major use of home computers is for video games, and indeed the
potential home video game market provided a major incentive for the
development of many home computers in the first place. Six or seven
years ago hardly any video games existed. But today arcade and home
video games comprise an industry that has reached over $7 billion.

While questioning people in the course of preparing this book, we
uncovered a wide range of feelings about the games, most of them quite
passionate. A twenty-three-year-old computer engineer, David, was
playing portable video games non‐ stop on a flight we shared with him.
"What do you like about these games?" we asked. His answer was quite
definite: "I think they're entertaining. They fascinate me. I can't
believe I can hold something almost as small as a credit card that can
play a game I haven't mastered. They're a challenge. What's most
intriguing is that I know because of my work that there is a pattern
to these games. And I haven't yet figured it out. But I keep getting
closer. I keep getting better."

On the other hand, Glen, a twenty-five-year-old property manager,
hates video games. He says just as definitely: "I get no satisfaction
out of beating a machine!" And Jane, a thirty‐ eight-year-old
management consultant, sees them as a soporific for teenagers, an
aesthetic nightmare, and is adamant that they are no good at all for
anything whatsoever. The opinions of public figures reflect this
controversy. The U.S. Surgeon General, Everett Koop, decries video
games, while Isaac Asimov, one of the most respected science writers
in the United States, extolls their educational benefits. 4 Given
these extreme differences of opinion, we find the job of trying to
understand the video game explosion even more challenging.Figure 1.1
shows the "family tree" of video games. Their immediate parents were
the digital computer and the arcade game. The computer side of this
parentage will be traced in chapter 6; the arcade side, in chapter
4.Most games involve competition of one sort or another. But somewhere
along the line, solitary games evolved, in which competition, if it
even exists, is with yourself (for example, trying to top your
previous best score) or with some abstract entity such as a deck of
cards or a machine. Most video games are, or can be, solitary games.
You play chiefly against the machine.Three conceptual ingredients
enter into the immediate background of video games:
1. Sound and fury. Flashing lights, bizarre noises, and continuously
displayed, astronomical scores were incorporated in pinball machines.
Often associated with sleazy bars and arcades and thought to be
controlled by organized crime, nonetheless pinball machines managed to
build up a mystique. They were colorful and gaudy. Presumably in an
effort to give the illusion of variety, different games in an arcade
represented an enormous variety of concepts, ranging from the Vietnam
War to the Indianapolis 500 to the Playboy penthouse. However, all
these games were virtually identical in terms of how they were played
and what the goals were.

FIGURE 11 The family tree of video games.

2. Death and destruction. In the 1960s a new kind of game began to
compete with pinball for arcade space. These games were usually
automated in some fairly sophisticated way and usually involved
violence of one sort or another. In Bomber Pilot, for example, the
player, after inserting a quarter, was seemingly placed at the
controls of a bomb-laden jet plane and presented with varying terrain
passing below. The goal was to drop bombs on targets that would appear
for a few seconds beneath the aircraft and then vanish. Points were
awarded for successful hits, with the highest numbers of points
being awarded for the destruction of high-density population areas,
such as large cities, and strategic targets, such as enemy missile
bases. The player was constantly under threat of enemy antiaircraft
fire and therefore had to worry about taking evasive action as well as
aiming the bombs. Like pinball, these arcade games were supplemented
by exotic flashing lights, violent noises, and rapidly increasing
scores, which were prominently displayed.
3. Computer control. In the 1970s another game arrived, unobtrusively,
on the scene. This newcomer, Pong, differed from its predecessors in
several ways. First, and most important, it was entirely under the
control of a computer, and except for the player's joysticks, there
were no moving parts. Everything was electronic. In a major way, Pong
heralded the dawn of a new era.
Pong's second distinction was that it somehow acquired an immediate,
broad social acceptance. It suddenly appeared in all sorts of
places—in cocktail lounges, train stations, airliners— where no one
would dream of putting either pinball or the death and destruction
games. Although the reasons for this broad social acceptance are not
entirely clear, it is interesting to speculate. First, size doubtless
played a part. The older games, which used heavy mechanical parts,
were large and difficult to transport (and were certainly not welcome
in places like airplanes where size and weight are at a premium).
Pong, with a computer at its heart, was much more mobile. Second, in
the years following its introduction, Pong's price—along with the
prices of all other computer-based goods—fell rapidly, and thus the
game became widely available. In fact, in the mid-1970s versions of
Pong—primitive by today's standards, but revolutionary then—began to
find their way into individual households. And, finally, Pong's
central theme was not the violence and kitsch of the previous arcade
games. Instead, it mimicked the then-genteel racquet games such as
tennis and squash. This feature may well have provided the lubrication
necessary to ease the game into polite society.

For whatever reasons, Pong managed to escape from the smoky, seedy
atmosphere of its pinball arcade predecessors, and it set the stage
for the widespread status currently enjoyed by today's video games. As
we have indicated, the computer basis of Pong, with its attendant
implications for cost and mobility, was a critical ingredient of this
transition. In chapter 6 we shall summarize the computer revolution
and its critical role in the psychology of video games.

Throughout this book, we are going to take the theories and
experiments of psychologists and use them to understand the video game
phenomenon that has sent many children into video arcades and many
parents into fits of nervousness. When the surgeon general marches
through the country crying, in essence, "Warning. Video games may be
hazardous to your children's health," should we believe him? Dr. Koop
has argued that there is nothing constructive about the games and that
in fact they may be teaching children to kill and destroy since that's
what most of the games are about. In this book, however, we'll take
the position that his fear may be completely unwarranted. Video games,
at least in some form, are going to be with us for quite some time,
and it is important to analyze dispassionately their psychological
costs and their benefits. We should not ban video games without a deep
and thoughtful analysis, any more than we should ban hopscotch or
Monopoly.

When people ask "What good are these games, anyhow?" the suggestion is
often heard that they have a direct benefit of increasing some skill
like eye-hand coordination. But so do many activities, such as
baseball and sewing. What are not usually considered are the indirect
benefits that video games can and do yield. These can be quite
unexpected and enormously powerful. We refer to such benefits as the
creation of an intense interest in computers, which has led many of
the game players of the early 1980s to jobs as computer programmers
with major corporations. We interviewed one such man, Greg, who at the
age of twenty-three had landed a programming job with a growing
software company just south of San Francisco. Greg spends his days
writing computer programs and claims he's happier than he has ever
been in his life. Five years earlier, Greg's parents worried that he
was spending too much time playing video games. They thought he might
be "addicted" to the games the way other kids seemed to become
addicted to drugs or alcohol. Now—five years later—they take
tremendous pride in their son's work. They have come to realize that
the games were the start of his intense interest in computers that led
to his career.

What was it about video games that Greg found so appealing? Why was he
willing to forgo sporting events and trips to the beach to spend time
in video arcades? To address such questions, we now draw upon the
field of psychology.[/SPOILER:a5d5496eb9]

Chapter 2: Why Video Games Are Fun
[SPOILER:a5d5496eb9] CHAPTER 2: WHY VIDED GAMES ARE FUN
Syndicated columnist Ellen Goodman has described her own initiation
into video games. One cloudy day she was waiting for an airplane in
the Detroit airport. She had time on her hands and thought she would
try a quick game of Pac-Man. Before she knew it, she was hooked;
Pac-Man took her for every last quarter she had. It began innocently
enough. She put in her first quarter but, not yet having a feel for
the game, she shoved Pac-Man into the arms of the nearest monster.
Undiscouraged by defeat, she tried again. She did a little better this
time. Something about the game made her think she could win. So she
kept at it. Fortunately for Goodman, she was able to break the habit
before it broke her. Once she gained some distance —away from the
clutches of Pac-Man—she thought about it
clearly. "Pac-Man hooks only those people who confuse victory with
slow defeat." 1

Why do people find the games so compelling? In this chapter we will
illustrate how the psychological concepts of reinforcement, cognitive
dissonance, and regret help explain the process of video game
addiction. Although very few psychological studies deal directly with
the issue of why video games are fun, we found one that does. At the
end of this chapter, we'll describe it.

Pac-Man, by Way of Example
In describing how experimental psychologists might view and explain
video game behavior, it's useful to have one example to rely on
throughout. Because it has been one of the most popular games, we've
chosen Ellen Goodman's nemesis, Pac‐ Man. For the benefit of any
readers who have been living somewhere besides Earth for the past few
years, we'll provide a brief description of Pac-Man here.

The game Pac-Man gets its name from the Japanese term paku paku, which
means "gobble gobble." The character Pac‐ Man is a little yellow
creature who looks like he's smiling. He is set in a somewhat complex
maze that's initially filled with yellow dots. The player controls
Pac-Man's movements with a four-directional joystick or control knob,
thereby allowing him to move left, right, up, or down. His jaw faces
the direction in which he's moving.

As he glides around through the maze, Pac-Man gobbles up the dots.
Simultaneously, however, he's vigorously pursued by four monsters
(each a different color) named Inky, Blinky, Pinky, and Clyde. If one
of the monsters catches him, Pac‐ Man slowly folds up and wilts away
while the machine provides sympathetic noises. Pac-Man can be eaten
only three times before the game ends, the score reverts to zero, and
another quarter is required for further play.

Pac-Man has a variety of ways of combatting and outwitting the
monsters. First, by adept maneuvering, he can keep away from them.
Second, at each of the four corners of the board is a glowing,
extra-strength dot called an energizer. Whenever Pac-Man eats an
energizer, some, or all, of the monsters turn blue. When a monster is
blue, contact between it and Pac-Man results in the monster's demise
rather than in Pac-Man's. However, monsters remain blue only for brief
periods following Pac-Man's consumption of an energizer; moreover, a
monster destroyed by Pac-Man doesn't stay destroyed, but instead
returns to action following a short interlude in the penalty box.

If Pac-Man eats all the dots in the maze, the player has completed a
"board" and a new maze with fresh dots appears. This provision of new
boards can continue indefinitely; however, with each successive board,
things become more difficult. The monsters move faster and are blue
for shorter periods of time; Pac-Man moves slower; and so on. In
between each board, a short, amusing skit occurs on the screen.

When playing Pac-Man, the player is rewarded in many ways. For
example, points are awarded for gobbling up the dots and for
destroying monsters. Additionally, there is some symbol in the center
of the board. What the symbol is depends on how many boards the player
has managed to get through. For instance, a cherry appears on the
first board, a strawberry on the second, until finally, a key appears
on the twelfth board and all subsequent boards. Naturally, Pac-Man
aficionados have memorized the sequence of symbols, and the symbols
that signify that many boards have been accomplished are the most
prestigious. The symbols themselves can be eaten by Pac-Man for
progressively increasing numbers of points. And finally, additional
sources of reinforcement are the amusing skits that occur between
boards.

The first time you drop a quarter into a Pac-Man game, you might get a
score of 1,000, if you're lucky. Less than a minute will pass, and
your Pac-Man will be eaten three times in disconcertingly rapid
succession. But chances are you'll play again. And again. And again.
You might get through a couple of boards, and your score might get up
to 5,000 or so. But what will puzzle you most is this: the top score
for the day will be posted on the machine. It might be 56,000. Or
102,000. It could be over 500,000.

How could anyone get such a score? In the next chapter we'll focus on
the mental and physical skills that go into video game facility. For
the moment, however, let's concern ourselves with the question of how
people become so motivated that they'll play video games hour after
hour, day after day. To do so, we'll consider the phenomenon of
reinforcement.

Mechanics of Reinforcement
On a clear Vermont night, a young boy sits methodically scanning the
sky in search of shooting stars. In a sleazy Las Vegas casino, a
glassy-eyed old woman mechanically deposits nickel after nickel in a
slot machine. And in a posh Chicago suburb, a teenager spends hours
playing Gallaxian at the video parlor every afternoon when school lets
out.

These three situations have a common element: in each one, behavior is
dictated by what psychologists refer to as the partial reinforcement
effect. To understand this effect—which is a critical psychological
ingredient of video game addiction—it will be useful to provide a
thumbnail sketch of reinforcement itself and the role it plays in
shaping behavior. Video games are designed to take your money, and
they have an uncanny way of doing so. They play on the ordinary
person's weaknesses for reinforcement.

BASIC CONCEPTS
Reinforcement is the provision for you of something that you like. In
each of the three preceding examples, reinforcement of some sort is
involved. For the aspiring astronomer, seeing a shooting star is a
reinforcement. For the Vegas gambler, the slot machine's payoff is a
reinforcement. And for the video game player, beating a previous high
score or winning a free game or shooting down enemy spaceships is a
reinforcement. There are a variety of psychological theories designed
to explain the role of reinforcement in behavior. Central to all of
them, however, is the idea that any behavior that is followed by
reinforcement will increase in frequency. In short, video games that
do something to make a player feel good will be played again and
again.

Certain elementary kinds of human behavior can be analyzed nicely by
referring to well-studied principles of reinforcement. To get the
complete picture, of course, we will need to go beyond reinforcement
and examine other important topics such as motivation and social
pressure. For the moment, though, we consider only the subject of
rewards.

From observations made with rats, pigeons, monkeys, and other
organisms—and some studies with humans—we have come to know a great
deal about how reinforcement hooks people and gets them to behave in
certain predictable ways.

Experiments with rats are the easiest to do, and certain laws of
reinforcement emerge in their simplest form in a rat study.

A typical experiment to investigate reinforcement involves a white
laboratory rat placed in what is called a "Skinner box," named after
Harvard psychologist B. F. Skinner, who invented it. A Skinner box is
a cage containing a protruding lever that the rat can push and a small
container into which food can be dispensed by the experimenter. When a
novice rat is initially placed into the box, it wanders around,
performing the sorts of behaviors that rats typically perform:
exploring, sniffing, grooming. Eventually—probably out of sheer
boredom—it presses the lever, whereupon a rat pellet appears in the
container. This is a reinforcing event, which undoubtedly makes the
rat very happy. The event, like any reinforcement, leads to an
increase in the behavior that just preceded it—in this case,
lever-pressing. Eventually the rat will be pressing the lever at a
rapid clip and eating rat pellets to its heart's content. Already you
may be sensing a correspondence between the rat in the Skinner box and
the human in front of the "video parlor box." Both are continually
performing actions that lead to reinforcement. The rat gets crunchy
food, while the video game player gets higher scores and free games.
While food would be reinforcing for the video gamer, there are
undoubtedly times when given the choice between (1) a slice of pizza
and (2) a chance to play Space Invaders and perhaps achieve the high
score for the day, the game would be a more powerful reinforcer.

SCHEDULES OF REINFORCEMENT
Game designers confront many decisions when trying to create a game
that people will like. One question is: How often should a player be
reinforced? Is it a good idea to make sure that players never leave
their first game without some form of reinforcement? Or should the
games be created so that they are sufficiently difficult and several
plays are necessary before a single rewarding event occurs? As we
shall see, game designers have apparently stumbled on the optimal
strategy for reinforcing people so they (like the rat that keeps
pressing) will continue dropping quarters at a rapid clip.

To see this, let's return to the rat. In the scenario we have just
sketched, the rat was reinforced after each lever press. This is
called continuous reinforcement—the rat gets a reward each time it
presses the lever. This schedule is usually necessary to get the rat
started.

However, continuous reinforcement is just one of many possible
schedules of reinforcement. Suppose that after the rat starts pressing
away, we stop providing food pellets after each press and instead
provide them only now and then. What will the rat do? It will continue
pressing, at least for a while. At some point, however, the rat will
again be reinforced, since eventually another pellet will follow a
lever press. This schedule of reinforcement is called partial
reinforcement—reinforcement is intermittent rather than continuous.
Partial reinforcement is a powerful way of hooking both rats and
people. They keep responding in the absence of reinforcement because
they are hoping that another reward is just around the corner. The
gambler keeps pulling the slot machine lever even though he has lost
ten times in a row because he hopes that the next time around he'll
win big. This would never happen if reinforcement had been continuous.
If it were, then as soon as the money stopped, the gambler would
quickly decide that the machine was no good anymore and turn his
attention elsewhere.

EXTINCTION
We knew a woman we'll call Ruth who went to Las Vegas at least twice a
month. Each time she stayed for three days and spent most of her
waking hours diligently pumping nickels into a slot machine. If you
happened to wake up as the sun was rising, you'd find Ruth glued to
the machines. Her case was classic. When she was in her early
twenties, Ruth went to Las Vegas for the first time to celebrate her
best friend's twenty‐ first birthday. While there, she experienced a
"big win" of $850, which was more than her monthly salary. This
experience made a lasting impression on her and from then on she was
hooked.

What if reinforcement were cut off altogether? As you might expect,
Ruth and the other players would not stop playing immediately. Rather,
they would doggedly continue to play for some time before giving up in
exasperation. This decline and eventual cessation of behavior (lever
pumping) in the absence of reinforcement (money) is referred to as
extinction, and the length of time it takes for the behavior to cease,
or extinguish, is referred to as the extinction period.

How long will it take Ruth to stop playing these machines? This
depends heavily on the schedule of reinforcement that she had been
exposed to earlier. If Ruth was used to winning each time she dropped
a coin, then she would extinguish very quickly. However, this
continuous schedule of reinforcement would have been very unlikely in
Vegas. It is far more likely that she had been reinforced
intermittently—that she had been on a partial reinforcement
schedule—and in this case the extinction period is considerably
longer.

But which partial reinforcement schedule leads to the longest
extinction periods? It turns out that the "variable" schedules (either
variable ratio—say, one in five turns, on the average—or variable
interval—say, every minute, on the average) are the most powerful ones
because they lead to the longest extinction periods. More precisely, a
variable schedule with moderately long intervals between
reinforcements is a good idea for game designers since it leads people
to continue to play the longest in the face of nonreward. However, if
the variable schedule is too long, a person might actually extinguish
when the game designer had not meant this to happen. With a very long
interval between reinforcements, the person has no way of being sure
that the reinforcement drought is ever going to end. He or she may
actually give up playing rather than drop the next quarter that would
lead to a reward.

In sum, we know a great deal about reinforcement and how it affects
people. A partial reinforcement schedule leads to behavior that (1)
occurs more rapidly and (2) is more resilient to extinction than does
a continuous reinforcement schedule. The dependence of extinction on
the prior schedule of reinforcement has been dubbed the partial
reinforcement effect. Taken together, these two effects of partial
reinforcement produce what looks very much like "addictive behavior."

The question of why the partial reinforcement effect occurs has long
been of interest to psychologists, and a variety of experiments have
been carried out to test various theories. For our purposes, however,
the important thing is that the phenomenon happens at all. Knowing how
rates of responding and resilience to extinction are affected by
reinforcement schedules should—in principle, anyway—allow us to
account for the seemingly addictive behavior engendered by video
games. Furthermore, we can explain why some video games are more
addictive than others.

With these concepts in hand, it is easy to see the underlying
principle that governs behavior not only in the case of Ruth, the
compulsive gambler, but also in the case of the young, aspiring
astronomer and the teenager in the video parlor. In all cases, the
person is under a partial reinforcement schedule: slot-machine payoffs
occur only infrequently, as do shooting stars and video game wins.
Moreover, the schedules are of the variable sort (slot-machine
payoffs, shooting stars, and video game wins do not occur in a fixed,
systematic way) and therefore produce the most powerful resistance to
extinction. According to the principles of partial reinforcement,
therefore, the behaviors involved should be highly resistant to
extinction, which indeed they are. In all cases, the person is willing
to pursue the behavior for lengthy periods of time, even in the
absence of reinforcement.

REINFORCEMENT AND VIDEO GAME DESIGN
Given an understanding of these principles, the task of a person who
designs and manufactures video games is more focused. The designer's
goal, of course, is to make money on the game. This goal is achieved
by ensuring that the eventual players will insert quarters into the
game as rapidly as possible. It's to the designer's advantage to
design a game that reinforces the player on the most addictive
schedule possible. And this usually turns out to be a variable-ratio
or a variable-interval schedule.

What this means in the world of video games is that reinforcement will
be somewhat unpredictable. A reward might come on the average of once
every ten times a player plays. For example, the player might achieve
three complete boards in Pac-Man only once every ten times, or so,
that he or she plays. This is an example of a variable-ratio schedule.
If a reinforcement came on average once every ten minutes, with the
actual times ranging randomly from once every ten seconds to once
every half hour, our player would be on a variable-interval schedule.
So, for example, if the Space Invader's mother ship appeared and was
destroyed on this sort of schedule, we would say that the player was
on a variable-interval schedule. These irregular schedules of
reinforcement are, in part, what cause video games to be so compelling
and irresistible.

In trying to implement these reinforcement schedules, an interesting
problem arises for the video game designer. To understand the nature
of this problem, it is useful to consider not a video game but,
rather, pinball. As you probably know, pinball is played with a steel
ball, initially ejected via a spring mechanism, into the playing area.
While in the playing area, the ball can strike various knobs, springs,
and other assorted paraphernalia, all of which cause the score to
increase. If, however, the ball rolls down to where it originated,
that ball is eliminated and a new ball must be ejected. The player has
three kinds of control over what is going on. First, the initial
ejection of the ball can range from soft to powerful, depending on how
far back the player draws the spring-loaded plunger. Second, near the
player is a small gate, partially guarded by flippers that the player
can manipulate. These flippers, if used properly, eject the ball back
into the field before it can roll out of play. And finally, the
player, by using his or her whole body, can tilt the entire machine
ever so slightly, in order to influence the path of the ball. The tilt
can't be too much, however, or "tilt" will register on the scoreboard
and the game will end.

Any game—pinball included—can't be too easy, or it will provide
continuous reinforcement for practiced players, which, as we've seen,
doesn't really lead to much of an addiction to the game. On the other
hand, the game can't be too difficult —that is, reinforcement can't be
too intermittent—because then most novice players will never get
enough reinforcement to become addicted to the game in the first
place. Just as the rat in the Skinner box never really begins pressing
the lever unless initial reinforcement is more or less continuous, so
the would-be game addict needs some early reinforcement in order to
get interested in the game. This means the pinball game designer is
forced to an intermediate stance—the game is made moderately
difficult. This solution has two difficulties. First, and probably
most serious, many potential players won't ever start playing pinball,
because they don't get reinforced enough when they first start playing
and can't play very well. Second, a really expert player will be
reinforced continuously, which, as we've seen, doesn't produce much
addiction. A colleague of ours named Graham, who teaches at the
University of Aberdeen, tried a game one evening and got a score of
zero. He said he never wanted to play again. A half hour later,
another player —fifteen-year-old Dennis—tried the same game and, much
to Graham's dismay, gave up after ten minutes because it was "much too
easy."

Enter video games. The feature that sets them apart from all other
games is the extremely flexible nature of the digital computer that
controls them. We'll talk more about computers in chapter 6. For now,
it's sufficient to realize that the computer can be programmed to make
the games easy to begin with and progressively more difficult. A good
example of this is seen in Pac-Man. The major determinants of
difficulty in that game are such things as the speed of Pac-Man
himself, the speed of the monsters, the period of time that the
monsters remain edible by Pac-Man, and so on. These factors change
from board to board such that the game becomes progressively more
difficult as play continues. A novice player is usually able to get
through one complete board after only a few trys. However, only a very
few experts—who have played literally thousands of games—are able to
make it up to the highest level of difficulty. The same sort of
strategy is seen in the design of other popular games such as Space
Invaders, where the invaders move faster, shelters disappear, and the
player's life becomes generally more difficult and harrowing as the
game progresses.

There is another advantage of having the computer control the
reinforcement schedules. Suppose we turn into a nation of Pac-Man
experts. Suppose, that is, virtually everyone practiced Pac-Man enough
to be able to play it perfectly. Wouldn't reinforcement then become
continuous, with the resultant lack of addiction? No problem. Another
salient feature of a computer program is that it is very easy to
modify. It would be a trivial job to make the monsters go even faster
or make Pac‐ Man go even slower. Less trivial, but still not
especially difficult, would be the insertion of features that are
altogether new —for example, a new monster could be created that is
even more adept at devouring Pac-Man and generally creating havoc than
are the current ones. In contrast, the capability of easily changing
the rules of the game is not present with a precomputer game such as
pinball, where all such changes would involve difficult-to-modify
mechanical devices rather than simple-to-modify computer programs.

Other Aspects of Reinforcement
Knowing about the partial reinforcement effect gives any video game
designer an edge in designing a particularly appealing game. But there
is more that the designer needs to know. For example, Pac-Man gobbles
yellow dots that are worth 10 points each. Why 10 points? Is this the
best number of points to award a player for each dot devoured? These
questions raise the important issue of how big the reward ought to be.
Another principle of reinforcement is necessary for understanding
behavior, and that concerns the size, or magnitude, of reinforcement.

MAGNITUDE OF REINFORCEMENT
There is no question that behavior is related to the size of the
reinforcing event. Rats, for example, will run faster and more
frequently if they are rewarded with more food rather than less food.
People will work harder and play longer on a slot machine if they have
a chance of winning $1,000 than if they have a chance of winning only
$100. Intuitively, of course, this doesn't seem surprising.

When it comes to video games, however, the issue of reinforcement
magnitude becomes a bit more interesting. You may have noticed that
the number of points one accumulates in video games always seems to be
very large, even if the player is just a novice. For example, in
Pac-Man, a player acquires 10 points for devouring each dot, 200 to
1,600 points for devouring the monsters, and so on. Thus even on a
very first Pac-Man effort, a player can generally score in excess of a
few hundred points.

Why is this? Why not, for example, just one point per dot and 20 to
160 points for the monsters? At the very least, there'd be less space
needed on the screen to display the score. When we look at things in
terms of reinforcement principles, the reason is clear: large rewards
lead to faster responding and greater resistance to extinction—in
short, to more addiction— than do smaller rewards. From the point of
view of the video game manufacturer, of course, points are free—the
cost of manufacturing and programming the game is the same whether
small or large numbers of points are awarded to the game's eventual
players.

Given these considerations, you might ask why the game designers
stopped where they did in terms of point magnitudes. Why only 10
points per dot in Pac-Man? Why not 100 or 1,000? The answer is
twofold. First, the point magnitudes, after all, have to be something,
and whatever they're made to be they could always be higher. So the
actual values chosen by the designer are somewhat arbitrary. Second,
however, at some point people stop having an intuitive grasp of what
some magnitude means—that is, above some magnitude, any amount is
psychologically pretty much equal to any other similarly high amount.
To get a feeling for this phenomenon, imagine that you are a
participant in a TV game show and you are given the following choice:
(1) you can either have $1 for sure, or (2) a coin will be tossed, and
you will receive $10 if the coin comes up heads but nothing if the
coin comes up tails. Almost invariably people choose the latter
alternative, figuring that $10 is worth so much more than $1 that the
chance of winning the $10 is worth the risk of losing the coin toss
and forsaking everything. But now imagine a new version of the choice:
either you get a sure $1 million or a coin is tossed and you get $10
million if the coin comes up heads but nothing if the coin comes up
tails. Now we find that people almost invariably choose the first
alternative. For most people, $1 million and $10 million are
psychologically pretty much the same thing—they're both "very large
amounts of money." Thus it makes perfect sense, psychologically, to
opt for the sure thing—the million dollars—rather than the choice that
involves a 50 percent chance of getting nothing.

Bearing this example in mind, we see why video game scores —despite
principles of reinforcement—can't be too large. In Pac-Man, for
example, the designer wants the accomplishment of eating a monster to
be psychologically much greater than the accomplishment of eating a
dot. This is done by awarding differential numbers of points, but, as
in the money example, the absolute value of the points can't be too
large. A million points for eating a dot would, psychologically, be
not very dissimilar from 20 million points for eating a monster; they
are both just "huge numbers of points."

DELAY OF REINFORCEMENT
We have pointed out that any behavior will increase in frequency if
that behavior is followed by reinforcement. It turns out that the
delay between the behavior and the reinforcement is, in most cases,
very important: the shorter the delay, the quicker will the behavior
increase in frequency. In other words, short delays lead to more
powerful reinforcement effects.

In many real-life situations, delay of reinforcement is very long. For
example, if we save money in a savings account, it's a long time
before we begin to see the interest accumulate or are able to withdraw
the saved money in order to make some large purchase. Because of this
delay of reinforcement, the behavior of saving money isn't as frequent
as it otherwise would be—in everyday language, we say that saving
money is difficult. So, many people don't save money; instead they
spend it as soon as they get it, and reinforcement is immediate.

In the case of video games, however, at least some sort of
reinforcement is always provided immediately. In most cases, a score
of some sort is prominently posted somewhere in the display, and the
score changes the instant we shoot down an enemy ship or eat a
monster. It is, in part, this instant reinforcement that makes the
behavior of playing video games so satisfying and therefore so
prevalent.

MULTIPLE REINFORCEMENTS
One aspect of video games that sets them apart from most other games
is that they can be, and usually are, much more complicated than
arcade-type games have traditionally been. Pac-Man, for example, has
many different ways to reinforce you. You can eat dots; you can eat
monsters; you can avoid monsters; you can eat the symbols; you can get
through boards; you can see between-board skits; hear music; and so
on. Other games, invented more recently than Pac-Man, are even more
complicated.

From the standpoint of what makes games fun, these multiple
reinforcements are important because different people enjoy different
things. By using a "kitchen-sink" approach— that is, by inserting into
a game a wide variety of things that might be reinforcing—the designer
winds up with a game that appeals to a wide variety of people and
will, accordingly, be widely played. This flexibility of video games
can be contrasted with that of pinball. Pinball, for all its bells and
whistles, really provides only limited types of reinforcement—you see
the ball move, you hear the sound effects, you see your score
increasing, and that's about it. The reason for this contrast, once
again, is that the computer program that underlies a video game is
itself infinitely flexible, whereas pinball, being mechanical, has to
be kept relatively simple or else it will become prohibitively costly.

If video games are reinforcing in a variety of ways, at least some of
the reinforcement is no doubt extrinsic, taking the form of praise and
admiration from peers and other onlookers. But it's perfectly possible
to play a video game by yourself and feel gratified when you do well
or when you improve your performance. This kind of reinforcement is
called intrinsic reinforcement. Video games can provide very powerful
intrinsic reinforcement, which is probably a very important reason for
their success. The fundamental source of intrinsic reinforcement—the
very person receiving the reinforcement—is perpetually present.

Cognitive Dissonance
As we have seen, video games have a variety of ways of reinforcing
players. But, at least for the games played in video arcades, there is
another side of the picture: you have to pay for them. You might
expect that the reinforcement obtained from the games themselves might
in some sense be countered by the punishment that stems from having to
insert quarter after quarter. Interestingly enough, however, a large
body of social psychological research suggests that the opposite may
be true: games may be more reinforcing, not less, if you have to pay
for them.

During the 1950s and 60s, a group of psychologists (led by Leon
Festinger and his colleagues) developed a theory called cognitive
dissonance to account for some seemingly paradoxical types of
behavior. The paradox is that people sometimes seem to enjoy things
that are less reinforcing over other things that are more reinforcing.
Consider, for example, an experiment reported by Festinger and
Carlsmith. 2 In this experiment, a group of people performed a
repetitious, tedious, and thoroughly boring task. After completing the
task, the group was asked by the experimenter to lie to a new group of
people —to tell them that the task was more fun than it actually was.
One group was offered $20 to lie, whereas the other group was offered
only $1. Finally, after the lies had been told, the people were asked
to rate how much they enjoyed the original task. It turned out that,
contrary to what you might expect on the basis of reinforcement
effects, the $1 group claimed to like the task much better than did
the $20 group.

Why did this happen? Cognitive dissonance theory assumes that when a
person performs acts or holds beliefs that are in conflict with one
another, the person will act so as to reduce the conflict. In the
$1/$20 experiment, the conflict was between the people's knowledge
that they were performing a boring task and their knowledge that they
had told someone else that the task was fun. Why did they lie? People
who were paid $20 had adequate justification—they were hired guns,
paid to lie. The $1 group didn't have this handy justification, and
their only recourse was to change their attitude about the task. By
believing that the task was more interesting, they created a
justification for the positive report that they made about it.

We can recast this sort of finding into a statement about extrinsic
versus intrinsic reinforcement. Given that a person was performing
some act to begin with, it's necessary to have some kind of
reinforcement to account for it. If there's extrinsic reinforcement,
as there was for the $20 group, that's fine. But if there's
insufficient extrinsic reinforcement, as was true for the $1 group,
then intrinsic reinforcement had to be generated—the subjects had to
decide that the task was more intrinsically fulfulling. This sort of
effect can be seen more directly in an experiment by Lepper, Greene,
and Nisbett. 3 Here, nursery school children were given the choice of
playing or not playing with marking pens. One group of children was
given a reward for playing with the pens, whereas another group was
given no reward. It turned out that the reward group played with the
pens less than did the no-reward group. Again, this appears to be the
opposite of what you would expect from reinforcement theory; and
again, it can be explained if you assume that, in the absence of
external reward, the pens developed a powerful, intrinsically
reinforcing quality of their own. As an aside, it's interesting to
note that this experiment has close analogues in real life, since
parents will often pay their children for academic success. This
practice is probably unwise, since such extrinsic reward may remove
the intrinsic motivation that produces optimal and most satisfying
academic performance.

WHAT IF VIDEO GAMES WERE FREE?
Video games have the interesting quality that they take your money but
provide you with no tangible, extrinsic rewards that you can put in
your pocket and take home. By anyone's definition, having your money
taken away from you is not reinforcing —on the contrary, it's
punishing. This means that video games must develop qualities that
provide powerful intrinsic reward. Since people are standing there
having their money taken away, they must develop the attitude that
whatever they're doing is a lot of fun. In other words, if games were
free, people would probably like them less. To our knowledge, no
research has ever been done that has appropriately compared free games
with money-devouring ones in terms of how enjoyable the games are
perceived to be. But a large body of psychological research indicates
that games requiring at least some minimal amount of money (like a
quarter) would be perceived as more enjoyable than free games.

Does this mean that arcade operators should be advised to make the
games more expensive? Not necessarily. It's important to keep track of
the distinction between how much a person enjoys the game on the one
hand and how much the person plays the game on the other. A game that
costs a dollar might be perceived as more enjoyable than a game that
costs a quarter. But there comes a point at which, enjoyable or not,
cost would be prohibitive and the game wouldn't be played. Thus
there's a tradeoff between enjoyment and cost, which implies that some
intermediate game cost is the optimal one. Any scientist would shriek
in agony at this unjustifiable conclusion.

Regret and Alternative Worlds
So far we've talked about reinforcement in terms of being rewarded for
things that you have done—like getting high scores. There's another
side of this motivational coin, however, which is regret over things
that you haven't managed to accomplish. In most situations regret is
something that you just have to live with. But that's not true with
video games. Often when playing a video game, the game ends because
you've made a mistake, and you immediately know exactly what you've
done wrong. "If only I hadn't eaten the energizer in this game before
trying to grab that cherry," you say to yourself. "I knew it was the
wrong thing to do, and I did it anyway." But now you don't have to
just sit there being annoyed and frustrated. Instead you can play the
game again and correct that mistake. So in goes another quarter. But
in the process of playing again, you make another mistake. And spend
another quarter to correct it. And so it goes.

Two psychologists, Daniel Kahneman and Amos Tversky, 4 have recently
been studying the phenomenon of regret. To give you a flavor for the
kind of things they've discovered, consider the following question:

Mr. Smith and Mr. Jones both have to catch planes. They're on
different flights, but since both flights leave at 9:00 A.M., they
decide to take a cab together. Owing to a combination of unfortunate
circumstances, the cab is late and doesn't arrive at the airport until
9:30. On consulting with the airline agent, the two men discover that,
whereas Mr. Smith's flight left on time at 9:00, Mr. Jones's flight
was delayed and left at 9:28, only two minutes ago. Who is more upset,
Mr. Smith or Mr. Jones?

Given this question, people invariably and immediately report that Mr.
Jones was more upset. Why is this? After all, both men missed their
flights and, objectively, they're both in equal difficulty. Kahneman
and Tversky offer an explanation in terms of alternative worlds that
can be constructed in the mind. They propose that, when some
unfortunate event occurs, the victim constructs an alternate reality
in which the unfortunate event didn't occur. The less this alternative
world differs from reality, the worse the victim then feels. In the
example at hand, it's very easy for Mr. Jones to construct an
alternative world in which he caught his flight. "If we had just gone
through that yellow light instead of stopping for it," he might say to
himself, "then I would have made my flight." Mr. Smith, on the other
hand, would have a much more difficult time constructing an
appropriate alternative world. In order for him to have made his
flight, they would have had to have gone through the yellow light and
have not gotten stuck behind that trailor truck, and have not had to
wait so long for the cab to pick them up in the first place. Since Mr.
Smith's alternative world would differ so substantially from the real
world, he would wind up with a lot less regret than would Mr. Jones.

Regret in video game play fits quite nicely into this framework. The
mistake that (in general) ends the game is the last thing, or close to
the last thing, you did prior to the game ending (eating the energizer
before trying to get to the other side of the maze in the example we
gave before). Therefore, the alternate world in which the mistake was
not made is extremely close to the real world in which the mistake was
made—and that's just the situation that produces maximal regret.
Naturally, given the opportunity to make that alternate world a
reality and eliminate all your regret, you'll avail yourself of the
opportunity. You play again.

An even more striking example of the alternative world phenomenon as
it operates in computer games is seen in Adventure. Adventure games
are very similar to the precomputer game of Dungeons and Dragons. In
them, the player is placed into some kind of hypothetical mazelike
environment, where both danger and excitement abound. In one of them,
for example, you (the player) are in a nuclear power plant and your
mission is to find and defuse a time bomb that has been placed
somewhere in the plant by a wicked saboteur. Carrying out this mission
requires a complex series of actions, not the easiest of which is
figuring out your way around the plant to begin with: learning not
only where various rooms, nooks, and crannies are relative to one
another, but also which actions—complying with the automatic security
locks and so on—are necessary in order to cross from one place to
another.

You actually play this game by issuing a sequence of instructions to
the computer. In return, the instructions are carried out to the best
of the program's ability; and, also, brief descriptions are provided
of objects that can be seen and of actions that occur.

A couple of key features heighten the game's interest. First, there
are lots of ways that you can go wrong and kill yourself. For
instance, you can accidentally blow up the bomb, you can fall off a
ledge, you can die of radiation poisoning, and so on. But fortunately,
you can save the game at any stage, so if you do make a mistake, you
can go back to the point at which the game had been saved—that is,
prior to when the mistake had been made. Again we see a classic case
of an alternative world. "If only I had put on the radiation suit,"
you say to yourself, "I wouldn't have died that horrible death in the
radiation chamber." And since the alternative world in which you put
on the radiation suit is very close to the "actual" world in which you
didn't, regret is very high. But since you saved the game, you can go
back and create that alternative world, thereby eliminating the
regret. So you do. Computer games provide the ultimate chance to
eliminate regret; all alternative worlds are available.

Research on Video Games

Since video games are a relatively new phenomenon, psychologists have,
thus far anyway, performed relatively little research on the games
themselves. One piece of research that has been done, however, is a
Stanford University Ph.D. dissertation by Thomas Malone.5 Malone was
primarily concerned with educational techniques, and his dissertation
was aimed at finding ways of making classroom learning more fun.
Noting the mass appeal of video games and the intrinsic reinforcement
that they provide, he ventured to suggest that these games may prove
to be superb teaching devices. This educational theme recurs
throughout Malone's dissertation; however, the research that he
reports was concerned primarily with the features of certain games
that made them fun to play.

Malone studied school children (kindergarten kids through eighth
graders). All the children had been playing with computer games in a
weekly class when the study began in 1979. The survey involved
relatively nonstandard games, but it is still highly suggestive.

Malone asked the children to rank a variety of computer games they had
played on a simple four-point scale. He then analyzed the features of
the games and concluded that incorporation of a specific goal was the
single most important feature in making a game enjoyable. Other
popular features were score‐keeping; audio and visual effects; the
degree to which players had to react quickly; and randomness
(unpredictable games are preferred). The games ranked highest
incorporate these features; those ranked lowest don't.

Malone also asked children what they liked about the games. Almost 40
percent of the reasons the students gave dealt with fantasy. For
example: "I like it because it's just like Star Wars." "I like it
because it has bombs." "I like blasting holes in the other snake." Of
course, children also provided other reasons. Some liked a game for
its challenge, others because it was easy to do well. One student
liked Petball because "you can get a high score very fast.... It's
easy to get bonuses. I like to win; I'm a sore loser." Interestingly,
while the children talked a lot about fantasy, the games they rated
most favorably were notably low in fantasy: perhaps children are no
better than adults at reporting their own motivations.

The results of Malone's survey provide interesting but inconclusive
suggestions about what makes the games fun to play. His next step was
to create new versions of particular games in which some of the
hypothesized important features were missing. He investigated two
different games, Breakout and Darts. Breakout requires primarily
muscle skills, whereas Darts requires primarily thinking, or
intellectual, skills.

Breakout is a derivative of Pong, the first video game, which was a
simulation of table tennis. Figure 2.1 shows a typical screen display
in Breakout. In this game, the player manipulates a knob that controls
the vertical motion of the paddle, shown on the left side of the
screen. The paddle is used to hit a ball that bounces against the wall
on the right-hand side of the screen. The wall is made up of multiple
layers of bricks, and each time the ball strikes the wall, it knocks
out one brick. The ultimate goal is to knock out all the bricks from
the wall. In addition, however, points are awarded for each brick that
is knocked out, and the score is continuously displayed on the screen.
The player is allowed a total of three balls and uses up one ball each
time he misses the ball with the paddle.

Based on the results of his survey, Malone generated a list of
features that, he hypothesized, contributed to Breakout's popularity.
There are clear goals: increasing the score by knocking out bricks and
ultimately tearing down the whole wall. The game keeps score. It has
audio effects—tones sound when the ball bounces off a wall—and visual
effects—the ball moves, the bricks break from the wall. It has
fantasy—the player destroys the wall, perhaps imagining himself
escaping from prison or rescuing a hostage.

Malone then proceeded to create several versions of Breakout and to
compare them with the original. To see how important was the challenge
of getting a higher score, he created some variations in which the
computer did not keep track of the score. To see how important was the
visual stimulation of watching the bricks break out, he created some
variations in which the bricks did not actually break away—the ball
just bounced back and forth against the wall, with a point being
awarded for each bounce. To answer other questions, he created other
variations. Malone's subjects—who, by the way, were Stanford students
this time, rather than the younger children used in his initial
survey—played the various games and indicated which ones they liked
and did not like.

FIGURE 2.1 Breakout display. A player manipulates a knob that controls
the vertical motion of the paddle shown on the very left of the
screen. The paddle is used to hit a ball, which bounces against the
wall on the right-hand side of the screen—a wall made up of eight
layers of bricks. The score refers to the number of bricks in the wall
that the player has successfully knocked out. A ball is used up
whenever the player misses the ball with the paddle, and the number of
balls left before the game ends is shown underneath the score. From T.
W. Malone, What Makes Things Fun to Learn? A Study of Intrinsically
Motivating Computer Games (Palo Alto, CA: Xerox, 1980), p. 24. Used by
permission of the author and the publisher. Subsequently reproduced in
T. W. Malone, "Toward a Theory of Intrinsically Motivating
Instruction," Cognitive Science 4 (Ablex, 1981): 345, and used also
with permission of Ablex Publishing Co.

The results were clear. The most important feature in determining how
much the game was liked was the breaking out of a brick when the brick
was hit by the ball. The versions in which the wall remained intact
when struck by the ball were not liked nearly as well (even though the
score increased just as it had before). Two other features—the
computer's keeping score and the ball's bouncing off the paddle (as
opposed to being simply ejected from the paddle)—were also important,
but less so than the gradually deteriorating wall. Why is the breaking
out of the bricks so appealing? Although the experiment doesn't permit
a definite answer, there are various possibilities. Watching a
deteriorating wall of bricks provides visually compelling
entertainment, it provides a cumulative scorekeeping device, and it
shows you how far you are from reaching the ultimate goal of
destroying the wall. Any one or a combination of these effects could
be responsible.

Malone showed clearly that when both the score and the brick
destruction were removed from the game, people didn't like it at all.
Without these features, the game had very little purpose. In this
degenerate version of Breakout, the players might try to keep the ball
moving as long as possible but they have no easy way of knowing how
well they are doing. Without the goal the game is no fun.

In Breakout, people learn a sensorimotor skill—they learn how to move
the paddle in such a way as to successfully maneuver the ball. But
playing Breakout doesn't require any higher‐ level skills such as
thinking, remembering, or problem solving. For this reason, Malone
next turned his attention to a new game—Darts—that did teach a bona
fide academic skill, that of estimating magnitudes on a number line
and expressing them as mixed numbers. (A mixed number is an integer
plus a fraction, such as 1 3/8.) This experiment (in which fifth
graders were used as subjects) revealed, among other things, some
intriguing differences between the types of games preferred by boys
and girls.

In the game of Darts, a number line is presented with specified
numbers defining the ends of the line, as shown in figure 2.2. There
are three "balloons" protruding from the line, and the player's job is
to decide which numbers correspond to the positions of the balloons.
The player types a guess, and a dart (shown on the right) is moved to
the position indicated by the player and fired. If the number
corresponds to a balloon's position on the line, the balloon is burst.
In the example shown in figure 2.2, if the player were to type 3 3/16,
the lowest balloon would burst. If the number doesn't correspond to a
balloon's position, the dart remains stuck in the line and the
incorrect number that had been typed in is indicated, as shown in
figure 2.2. Here the player incorrectly typed 3 7/8. A total of three
darts is provided; thus a perfect player can burst all three balloons.

FIGURE 2.2 Darts display. Thomas Malone describes the set-up this way:
"Three balloons appear at random places on a number line on the
screen, and players try to guess the positions of the balloons. They
guess by typing in mixed numbers (whole numbers and/or fractions), and
after each guess an arrow shoots across the screen to the position
specified. If the guess is right, the arrow pops the balloon. If
wrong, the arrow remains on the screen, and the player gets to keep
shooting until all the balloons are popped" (p. 31). In this example,
the player has previously made an incorrect guess that 3 ⅞ is the
position of a balloon. But he's right this time by typing in 3 ½, and
the dart will move across the screen to that position, bursting the
middle balloon. From T. W. Malone, What Makes Things Fun to Learn? A
Study of Intrinsically Motivating Computer Games (Palo Alto, CA:
Xerox, 1980), p. 32. Used by permission of the author and the
publisher. Subsequently reproduced in T. W. Malone, "Toward a Theory
of Intrinsically Motivating Instruction," Cognitive Science 4 (Ablex,
1981): 349, and used also with permission of Ablex Publishing Co.

In addition to the obvious visual effects, there are abundant auditory
effects in this game. For example, circus music begins the game, and
to reward the player who pops all three balloons a short song is
played.

Malone again tried to find out what it was about the game that made it
fun and whether any new variations would make it more enjoyable. He
created a version in which, after each incorrect try, the player was
told in which direction and by how much the answer was wrong. In other
words, the player was given "constructive feedback" such as being told
"A little too high" or "Way too low." In other variations, the
balloons were broken, but not by the darts; rather, when the correct
position was typed in, a balloon over on the right side of the display
burst. Thus the visual display of bursting balloons was more or less
the same in the two versions; however, in one the player could
fantasize that the dart itself was bursting the balloon, whereas this
fantasy wasn't possible in the other version. Finally, some versions
of the game had the original music, whereas other versions had no
music.[/SPOILER:a5d5496eb9]

Chapter 2 Continued
[SPOILER:db06e850b3] Different players played the different versions of the game and
then indicated how much they liked it.

The most intriguing result to emerge from the experiment was that boys
and girls differed substantially in terms of which features determined
their preferences. For every feature that was examined, girls and boys
reacted in the opposite direction —if boys liked a particular feature,
girls disliked it, and vice versa. Some of the most striking
differences were the following: Girls liked music, whereas boys
disliked it. Girls liked (and boys disliked) being told (verbally) how
they were doing, whereas boys liked (and girls were relatively
indifferent to) having a visual, or graphic, representation of how
they were doing. Finally, boys liked having bursting balloons and
especially liked the version in which the balloons appeared to be
burst directly by the darts. Girls disliked both of these balloon
representations, and especially the latter.

In addition to these game characteristics, Malone also found that
certain characteristics of the group he studied influenced how well
the students liked the game. For example, those students who
considered themselves to be good in math liked the game better than
those who considered themselves to be poor in math. Students who
thought they did well in the game liked it better than those who
thought they did poorly in the game—although preference was unrelated
to how well students actually did, providing further evidence for a
distinction between what students think and what they do.

Malone worried that in reporting his results he risked perpetuating
stereotypes of human beings based upon their gender. There is ample
reason to believe, he urged, that the game preferences primarily
reflect differences in the ways boys and girls are socialized in our
culture. But whatever the basis for these preferences, it is important
to understand them. For example, if a mathematical game like Darts
happens to be designed in a way that appeals more to boys than to
girls, then a sex difference toward mathematics may be unwittingly
created.

The boys in this study liked the arrows and balloons fantasy, while
the girls did not. Why? One possibility is that destroying balloons
with arrows is aggressive, and the aggressiveness underlies the
difference in preference.

A major reason given for why video games are fun is that they are
responsive. In a world in which people are often too wrapped up in
themselves to give you the time of day, the games are just the
opposite. As a player, you get feedback all the time. The experiment
with Darts showed that fantasy was more important than feedback, but
as Malone has pointed out, the fantasy in these games is a unique form
of responsive fantasy. The fantasy is feedback.

When we watch a movie or read a book, we passively observe the
fantasies. When we play a computer game, we actively participate in
the fantasy world created by the game. For this reason alone, the
computer game might an ideal vehicle for learning. We'll return to
this educational theme in chapter 5, but for the moment, it's worth
noting that Malone felt he had the beginnings of a learning theory
that was "intrinsically motivating." By this, he meant a kind of
learning in which reinforcement comes from within the person rather
than from the outside world. For Malone, there are three major
ingredients inherent in the student-computer game experience that make
the games such ideal vehicles for learning. These three ingredients
are: challenge, fantasy, and curiosity.

Challenge comes into play because the games provide a goal to be
reached and an uncertain outcome. The ideal game, if it is to provide
the challenge necessary for true intrinsic motivation, undoubtedly
includes an element of chance—or at least something that seems like
chance to the learner. In card games, the cards are typically dealt
randomly to players, and uncertainty is thereby introduced. Similarly,
in a game like Darts the particular problem to be solved at any
moment, whether it is the location of 3 1/2 or 2 5/8, is more or less
randomly determined. Challenge is also achieved by having a "variable
difficulty level": as the player gets better at the game, the game
gets harder. The effort, skill, or knowledge required to reach some
subgoal may increase. This keeps the player from getting bored,
providing the requisite challenge.

Fantasies are the second ingredient for making a learning environment
more interesting and more educational. For Malone, fantasy-inducing
environments are those that evoke mental images—images of physical
objects such as balloons or images of social situations such as being
the ruler of a kingdom. Long before Malone's work, theorists such as
child psychologist Jean Piaget assigned a central role to make-believe
play in the development of skills in children. So Malone's idea about
the importance of fantasy is not completely new; his contribution,
rather, is to identify the important role it plays in making video
games an ideal vehicle for learning.

The final ingredient for making a learning environment more
interesting is the evocation of the learner's curiosity. For this, one
needs to provide an optimal level of informational complexity. By
optimal level, Malone means that the environment should be neither too
complicated nor too simple with respect to how much the learner
already knows. The world of learning should be novel and surprising;
at the same time, it cannot be incomprehensible. Finding the optimally
complex environment constitutes the fundamental challenge for the
designers of the educational video games of tomorrow.[/SPOILER:db06e850b3]

Chapter 3: Games And The Cognitive System
[SPOILER:db06e850b3]CHAPTER 3: GAMES AND THE COGNITIVE SYSTEM
Ability is the focus of this chapter. What aspects of mind figure in
the performance of an act requiring complex skills, such as playing
video games? Psychologists refer to the mind as the "cognitive system"
because the "mind" isn't really a unitary entity. Rather it is an
elegant system of delicately intertwined and finely tuned components.
The means by which these components are combined into an ability—such
as the ability to play a video game—is called a strategy. In the pages
to come, we'll describe both the components themselves and the ways in
which they can be combined into strategies.

A major theme of the chapter is that quite different strategies can be
used for accomplishing the same mental goal—the goal, for example, of
being good at a particular video game. Which strategy is appropriate
for a particular person depends on which of his or her mental
components are good. Some people have very fast reaction times, while
others are good at memorizing things.

A second, related theme is that of time and how long it takes to do
things. A typical person has a reaction time of about a fifth of a
second. We'll see that the quality of many mental components is
measured in terms of the amount of time that a component takes to do
something. Most video games are designed so that if you're faster than
the game at something, you win; if you're slower, you lose.

We have referred to the cognitive "system" and its "components." To
illustrate these concepts, we'll use a familiar example: a stereo
system. A sophisticated system might consist of a turntable/cartridge;
amplifier; tuner; reel-to-reel and cassette tape deck; several pairs
of speakers, any combination of which can be in operation at any given
time; and two sets of stereo headphones. To evaluate the system, we
would have to consider the quality of each component and then the
degree to which the user is adept at combining the components to make
the system capable of carrying out a variety of functions— producing
music for a party, providing a soothing background, masking the sound
of outside traffic, making tapes for the car stereo, and so forth.
Since the system is so complex, it's capable of doing each of these
things in a variety of ways. It's the user's job to figure out which
is the best way for any given task and to configure the system
accordingly.

The Mind as a System
The cognitive system, too, can be conceptualized as consisting of
components, and a particular combination of these mental components,
designed to accomplish some particular goal,
is referred to as a strategy. Later we shall discuss how specific
strategies are appropriate for specific people playing specific video
games. But first, let us introduce the mental components themselves. 1

SENSORY MEMORY
At any given moment, our five senses are being bombarded by a
tremendous amount of incoming information from the environment. When,
for example, you're standing in a video game parlor playing Donkey
Kong, visual information originating from the game screen, as well as
from much of the rest of the video parlor, is entering the cognitive
system via your eyes. Auditory information in the form of honks and
beeps from your game and others, along with the cries, whispers, and
conversation of the denizens of the parlor, is entering the cognitive
system through your ears. You're receiving tactile information from
the feel of the buttons and levers of the game through the skin of
your fingers, olfactory information about the hot dog being consumed
by the person standing next to you, and gustatory information about
the soft drink that you're sipping in between button pushes.

All information that enters the system through the sense organs is
initially placed into a sensory memory. One sensory memory corresponds
to each sensory modality, thus there are five sensory memories in all.
Each sensory memory has a very large capacity for holding
information—indeed, experimental evidence suggests that a sensory
memory may hold all the information that initially enters the system
from the environment. But information in it doesn't stay around very
long. In the case of the visual modality, for example, information
remains in the sensory buffer for only about a quarter of a second (or
250 milliseconds). Within this short time it is transferred to the
next storage area of the cognitive system or decays away and is lost
forever.

ATTENTION
If, as in our example, you were at the video parlor playing Donkey
Kong, you would need some but by no means all of the information
entering the system through your eyes, only a very small amount of the
information entering through your ears and skin, and probably none of
the information entering through your nose or tongue. Not only do you
not need this excess information, but it you were to hold onto it, it
would probably hinder you in your attempt to play the game
efficiently.

However, some portion of the incoming information is critical for you.
You have to be able to see what barrels are rolling toward you, for
example, or you're most certainly going to be hit by them. So the
question is: How do we filter out the information we don't need, while
at the same time retaining the information that we do need?

This filtering process is what is referred to as attention (or
selective attention), and people generally filter information very
efficiently. Psychologists have shown this efficiency in studies of
the "cocktail party phenomenon." 2 Imagine that you're sitting on a
couch at a crowded cocktail party in which a number of conversations
are occurring simultaneously. Bill and Jim are talking on your left
and, at the same time, Sue and Jane are talking on your right. If you
attend to Bill and Jim's conversation, then you'll find that you're
completely unaware of Sue and Jane's conversation. However, it's
perfectly possible to switch your attention to the right-hand
conversation, at which point you'll stop being aware of the left-hand
conversation. This switch of attention doesn't require moving a muscle
—it is something that occurs completely within your mind. All the
conversations from the entire party, including the two in question,
have been entering your sensory memory, but you have been attending
to, and thus have been aware of, only one conversation at any given
time. All the others have been eliminated—filtered out of sensory
memory and quickly lost from the cognitive system.

The cocktail party example involved sound—information coming in
through the auditory modality. There are analogous instances of such
attentional effects in the visual modality. Suppose, for example, that
you're playing the game of Sabotage. Sabotage works as follows: you,
the player, are in charge of a large cannon that sits on the ground.
As the game progresses, you're attacked by a variety of flying
objects, chiefly helicopters, and paratroopers that are dropped by the
helicopters. You can use your cannon to shoot down both the
helicopters and the paratroopers. You are charged one point per shot,
but you earn various numbers of points for everything that you shoot
down. More points are awarded for destroying helicopters than for
destroying paratroopers. However, if four paratroopers manage to land
unscathed, they will team up to sabotage you, thereby resulting in the
destruction of your cannon and the termination of the game.

In this game you tend to concentrate on destroying helicopters until a
disturbing number of paratroopers are in the air, at which point you
concentrate on the paratroopers. Thus you attend to different sets of
incoming information. While attending to the helicopters, for example,
you're quite unaware of the paratroopers—indeed, you have to
periodically shift attention away from the helicopters just to make
sure that no paratroopers have slipped in unnoticed. Likewise, while
concentrating on shooting down the paratroopers, you almost completely
lose track of the helicopters. Again we see that all environmental
stimuli—both the helicopters and the paratroopers—are perpetually
registered by the cognitive system in the sense that they all enter
the sensory memory. 3 However, your attentional abilities allow you to
attend to only one set of stimuli or the other.

Since performance in video games depends, in large part, on the speed
at which you're able to do things, the question of how fast you can
shift your attention from one set of information to another is quite
important. In Sabotage, for example, if you waited too long to notice
the paratroopers, your game would quickly be over.

Some shifts of attention involve eye movements. What are eye
movements? An explanation requires a short digression here. The entire
area that we can see at all is called the total visual field. The area
that's directly in the center of the visual field is called the
central field and the rest is called the visual periphery.

Because of the way our eyes are built, there's much of the visual
field that we can't see very well at any given instant. Rather, we can
make out only fine details in the central field, which is quite
small—less than 1 percent of the total visual field. To demonstrate
this, try focusing your eyes on one word of text in this book. If you
keep your eyes steady, you'll find that only about one word is really
readable. Words on either side of the one you're focusing on—as well
as words above and below it—are fuzzy and indistinct.

What about the periphery? We can see objects in the visual periphery,
but we can't see them very well. Nonetheless, the visual periphery is
very useful. For instance, we're able to detect when something new
appears in the periphery, or when something moves or changes color.
Detection of such changes in the periphery is often a sign that
something interesting or important is happening there. Some event in
the visual periphery often signifies that you should shift your gaze
to the area where the event is occurring, in order to assess what's
happening. Thus we need to make eye movements in order to keep
ourselves updated on what's happening in the world. When playing
Sabotage, a quick eye movement to that fuzzy object in the periphery
can tell you that a bomber is on its way and that immediate action is
necessary.

Not all eye movements are alike. One common type is called a saccade
(French for "jerk" or "jolt") which is a quick jump of the eye from
one place to another. In between saccades are periods during which the
eye is relatively stationary; these are called fixations. It is during
these fixations that information gets into the mind; nothing gets in
while the eye is making a saccade. 4 When doing something like video
game playing, where things are happening at a rapid clip, making
saccadic eye movements turns out to be time-consuming. Saccades
themselves take place quite rapidly—most take less than a thirtieth of
a second to complete. But a bottleneck arises because once the eye
arrives somewhere, it is forced to stay there for a minimum of about a
fifth of a second before it can move again. That is, fixations last a
minimum of about 200 milliseconds.

This can cause problems if, for example, you move your eye to some
particular place, quickly assess what's going on there, and then
notice via your peripheral vision that something else important is
happening elsewhere. Because of the inherent physiology of our visual
system, you're stuck where you are for about a fifth of a second
before you can switch your gaze to investigate this new development. A
fifth of a second may not seem like much in the grand scheme of
things, but in a video game events are taking place so fast that the
difference between being able to do something in, say, a tenth of a
second instead of a fifth may make a big difference.

In addition to switching attention via eye movements, it's also
possible to switch attention without moving our eyes if we're
switching attention between things that are very close together.
Again, it's easy to demonstrate this to yourself. Stare again at a
word of text; don't move your eyes. Notice that you can switch
attention back and forth between two adjacent letters in the word.
This type of attention shift takes about a twentieth of a second (50
milliseconds) to carry out.

SHORT-TERM MEMORY
Via the practice of selective attention, only certain information from
sensory memory actually gets noticed. But what actually happens to the
objects that we attend to? Attended information is transferred—that
is, copied from—sensory memory to a new component of the cognitive
system referred to as short-term memory.

Short-term memory has several salient characteristics. First, it is
generally identified with consciousness. That is, whatever we're
currently aware of, or conscious of, is exactly that information
currently in our short-term memory. Second, short-term memory has a
relatively small capacity. In contrast to sensory memory, which
appears to be of virtually unlimited capacity, short-term memory can
hold only about seven items—it's large enough to hold a seven-digit
telephone number, for example. We can access the contents of
short-term memory very quickly —if, for example, you're holding a
string of digits (such as a telephone number) in your short-term
memory, you can scan through them at the rate of about thirty digits a
second (roughly one digit every 33 milliseconds). We lose information
from short-term memory moderately quickly; information in it will
generally be forgotten after fifteen to twenty seconds. So if you had
just looked up a telephone number and someone interrupted you to ask a
question, the number would probably be forgotten. However, this
forgetting process can be prevented by rehearsal: by repeating the
contents of short-term memory over and over to ourselves, forgetting
will be prevented. By rehearsing information, we can keep it in
short-term memory indefinitely. Finally, short-term memory is also our
"working memory." It's where information is manipulated when we plan
things, figure things out, and so on. This is important, because if
we're maintaining a lot of information in short-term memory via
rehearsal, we'll have less short-term memory capacity left to do other
things, such as planning strategies and focusing attention. Suppose
that you're playing Defender. While you're playing, you have a good
deal of planning to do. You have to be constantly thinking about where
you'll be aiming, whether you might want to escape into hyperspace,
and so on. In order to carry out all of this, it is important to have
your short-term memory clear. Short-term memory is like the amplifier
in the stereo system; it's the heart of the system, and it's important
to learn to use it as efficiently as possible.

LONG-TERM MEMORY
The next major component of the cognitive system is long‐ term
memory—our repository of general knowledge. It contains such things as
our name, our ability to speak the language, things that we've learned
at work or in school, and so on.

The storage capacity of long-term memory is virtually unlimited.
Further, while information can be forgotten, such forgetting is
relatively slow. Whereas information is lost from sensory memory in
less than a second, and from short-term memory in less than a minute,
information will remain in long-term memory for days, months, years,
or even decades. How long it will remain depends on how well it was
originally stored there. Since information makes its way into
long-term memory via short-term memory, it's necessary to keep
information in short‐ term memory for some period of time in order to
get it into long-term memory. This makes intuitive sense. If you're
told a person's name and don't attend to it at all—or even if you
attend to it but then immediately forget it—you'll be unable to
remember the name later on.

In general, you'll find that if you just maintain information in
short-term memory by rehearsing it, then the longer you maintain it,
the better it will be entered into long-term memory. The efficiency of
entering information into long-term memory can be improved by
so-called elaboration methods. They include such tricks as forming
mental images of whatever it is you're trying to remember, associating
the to-be-remembered information to things that you already know, or
making up rhymes such as "Thirty days hath September "

When you learn a new video game, you have to remember many things
about how to play the game and what the consequences are of various
actions. Under what circumstances is it useful to turn tail and run
instead of taking an offensive stance? How long will your armored
shield last before becoming useless? How many points does it cost you
for each shot? And so on.

In playing video games, speed is of the essence—particularly the speed
with which you can retrieve information from long‐ term memory.
Psychological experiments have revealed that when you're confronted
with a very familiar symbol, such as a letter, it takes you about a
tenth of a second to retrieve, or to recognize, the name of that
symbol. 5 This fact is important in a game such as Asteroids, in which
various types of objects (for example, large asteroids, small
asteroids, UFOs, and so on) appear at random times and in random
places, and it's your job to identify them as soon as possible so you
can take appropriate action. Since it takes about a tenth of a second
to determine what each one is, a limit is placed on how fast you can
deal with these objects as they appear. A game designer could thwart
the efforts of most people to play a game by designing the game so
that players are required to recognize objects in only a twentieth of
a second.

When you sit down to play a new video game, you will find that the
games you played earlier in the day can influence how well you do on
the new game. The earlier games can actually interfere with your
ability to learn the new one. Interference more generally is an
important characteristic of long-term memory; it refers to the problem
you have remembering one thing as a result of learning some other,
related thing.

Interference can work in two directions—forward and backward. So,
games you learned earlier can influence a new game you are currently
learning. But the game you are currently learning can also influence
the ease with which you will learn future games. This is especially
true if the games are similar to one another. Knowing the problems
that interference can create, some choices of what games to play in
succession are wiser than others.

Suppose you've learned to play Asteroids and you then become intrigued
with Defender, which is similar to Asteroids but also has some
important differences. You may concentrate on Defender for a while and
become quite good at it. However, if you then go back to playing
Asteroids, you may discover that your game has deteriorated and that
you're now making responses that are appropriate to Defender—the game
you just learned—but not appropriate to Asteroids, the game you
originally learned. Learning Defender would have created retroactive,
or backward, interference with respect to playing Asteroids.

Similarly, suppose that you have learned a whole series of
"shoot-'em-down" type games such as Astro Blaster, Space Invaders,
Gallaxian, and so on. Now you're getting a little bored and want to
learn a new game. If the new game is another shoot-'em-down type—for
example, Phoenix—you'll find that it will be hard to learn;
interference from the games you already know will cause inappropriate
responses. This would be an instance of proactive, or forward,
interference. Chances are that you would have an easier time learning
an entirely new kind of game, such as Pac-Man or Donkey Kong.

One final aspect of long-term memory is pertinent to the learning of
video games. You may find that when you start learning a new game,
you'll play continuously for hours and hours. Not only will this tend
to deplete your supply of quarters, but, it turns out, it's not the
optimal way to learn. For obvious reasons, this kind of learning
strategy is referred to as massed practice. Massed practice has been
found to be inefficient relative to spaced practice, in which you take
numerous breaks between games. You may have noticed that if you play
many games in a short period of time, you eventually seem to be
getting worse rather than better. Moreover, if you take a break and
return the next day, let's say, then on your very first try you may do
the best you've ever done. This is known as reminiscence. It's
probably the most dramatic example of the advantages of spaced
practice.

EXPECTANCY
Suppose the game you are playing requires you to press a button the
moment you notice that an enemy saucer has materialized out of
hyperspace onto your screen. This is an example of one of the most
fundamental tasks the cognitive system has to do—it has to respond as
soon as possible after some event occurs in the visual field. Earlier
we mentioned that it takes about a fifth of a second to react to such
a stimulus. However, this figure is highly dependent on the degree to
which you expect the event to occur. If you're not expecting
something, it takes longer to react; if you are expecting something,
it takes less time to react.

When you're learning to play a video game, therefore, it's important
to know as accurately as possible when things are likely to occur so
that you can anticipate them and react as quickly as possible. The
difference between a reaction time of a fourth of a second (250
milliseconds) and a fifth of a second (200 milliseconds) can easily be
the difference between shooting down the enemy and getting shot down
yourself. In any event, being able to anticipate is a matter of
learning contingencies among various events. In other words, given
that some particular event has occurred—say, the appearance of an
enemy ship in Space Invaders—what is most likely to happen next?

In fact, a variety of things could happen next, and what will happen
depends on the goals of the people who originally designed the game.
If they wanted to make things very difficult for you, they could
design things to happen completely randomly, in which case no event
will be predictive of any other event and you will never be able to
put expectancy to use. However, most games (and real life) do not work
this way. Usually, the occurrence of a particular event provides you
with information about what will happen in the immediate future: some
events have an increased probability of occurring, whereas others have
a decreased probability of occurring. It is an important task of
long-term memory to store these event dependencies, and good players
concentrate on doing just that. In other words, when learning to play
a game, they concentrate on what events are likely to follow—or not to
follow—what other events. This way, they are able to use this
information in the future and set up appropriate expectancies for what
is about to happen. The major benefit of this strategy is that these
players are able to respond faster, and that is one of the major
reasons that they are good players.

THE VERBAL/VISUAL DISTINCTION
The graphic designs, the funny bleeping sounds, and the brief verbal
messages are some of the most enticing qualities of video games.
Occasionally the mind is strained while it is forced to deal with all
of this incoming information at once. Coping with visual information
(such as the designs), auditory information (such as the bleeps), and
verbal information (such as the messages) simultaneously can, however,
be a lot easier for a person than coping with multiple visual,
multiple auditory, or multiple verbal inputs at one time.

To simplify the discussion, let's consider the visual versus verbal
comparison. It's fairly clear that we have two separate mental
subsystems to handle these two separate types of inputs. Further, it
appears that the two mechanisms can operate independent of one
another. To see what we mean by this, let's return to the stereo
system example. Suppose you wanted to record from a record and from a
radio at the same time. You would be able to do this by recording from
the record on your cassette recorder at the same time that you
recorded from the radio on your reel-to-reel recorder. Like the
handling of verbal and visual information by the cognitive system,
these two operations could be carried out simultaneously and
independently by the stereo system.

To get a feeling for the presence of both your verbal and visual
subsystems, try the following demonstration. First, imagine the block
letter E. Now imagine yourself going around the letter, identifying
each corner as an "in" corner or an "out" corner. The speed at which
you can perform this task depends very strongly on the manner in which
you make the actual identification of each corner. Try it in two
different ways. First, just say (out loud) either "in" or "out" as you
mentally arrive at each corner. Now try it again, but this time, point
to either your left or your right to signify "in" or "out." You'll
find that the pointing method will take you much longer than the
speaking method. It's usually a very powerful and dramatic effect. 6
Why does this effect occur? The reason is that the task of imagining
the block letter and determining whether a particular corner is an in
or an out corner is a visual task. Pointing is another visual task,
whereas speaking is a verbal task. Thus, when you're imagining the
corners and pointing at the same time, you're doing two visual tasks
at the same time, which overloads the visual mechanism. However, when
you're imagining the corners and speaking at the same time, you're
doing one visual task and one verbal task. Your visual and verbal
mechanisms don't interfere with one another; they have no trouble
operating at the same time.

This independence of visual and verbal mechanisms manifests itself in
a variety of ways when video games are being played. Practiced video
players are perfectly capable, for example, of holding a
conversation—with colleagues, with themselves, or with the machine
itself—without impairing their ability to execute the visual/motor
activities needed to play the game. Such players can also execute
these abilities at the same time as they are verbally working out a
strategy for the seconds to come ("Let's see, I'll pick up that
energizer in the upper left-hand corner, then zoom to the middle of
the board for the cherry, then get all the dots in the lower right,
but leave the energizer intact . . ." a player might say to herself as
she deftly gobbles up the dots and avoids the monsters). However, the
same player would be ill advised to imagine one path of Pac‐ Man—a
visual activity—while at the same time engaging in the other visual
activity of actually guiding Pac-Man around the maze.

From the standpoint of video game playing, one important facility that
is associated with the visual subsystem is that of mental
transformations. In general, a mental transformation is the process of
taking some visual stimulus and imagining it to be in some physical
state other than the one it's in. For example, you could look at an
object in the room, such as a chair, and mentally shrink it or expand
it, or place it somewhere else in the room, or rotate it to another
position. To get a feeling for what a mental transformation is,
suppose you are driving a car, headed south. Suppose also that you
must make a complex series of turns to get where you are going and you
must consult a map. But a problem arises: if you hold the map in its
normal way, with north facing upward, since you are driving south, the
directions on the map won't correspond to the directions in which you
must go. There are two common solutions to this problem. Some people
will keep rotating the map, so that "up" on the map will always be the
same as the direction in which the car is traveling. Other people,
however, have the ability to "mentally rotate" the map so that they
can always imagine it as being oriented in the same direction as the
car. This latter solution involves a particular kind of mental
transformation known as mental rotation.

It's easy to see how an ability to perform mental rotations could help
your video game playing. In Asteroids, for example, you must mentally
move a target asteroid to where it's going to be in a few seconds and,
at the same time, mentally rotate your cannon to see if you're going
to be in the correct position to shoot it down. Likewise, when objects
move off the screen, you must be able to mentally calculate where
they're going to reappear if you're going to keep an edge on the game.

The map-reading example illustrates that people differ in their
ability to perform mental transformations. Some people are able to
rotate the map mentally, whereas others must rotate it physically in
order to understand where they're going. Using an ingenious procedure
developed by Roger Shepard 7 of Stanford University in which people
are timed while they mentally rotate objects, it has been found that
people who are good at visualizing things are faster to mentally
rotate objects than are people who are poor at visualizing. Moreover,
children and elderly adults are slower to mentally rotate than are
middle-age adults. Men are occasionally faster than women but
sometimes the sexes perform equally quickly. This observation enables
us to explain why a person can perform exceptionally well on one video
game but not on another. If the second game requires especially fast
mental rotation and the person happens not to be an especially fast
mental rotater, he or she may never be able to master it.

People also differ quite substantially in their ability to process
information visually versus verbally. For example, males tend to do
better than females on those spatial tasks that require the
visualization or manipulation of objects in space. However, the
advantage that males have over females is rather slight, and most
probably arises from different learning experiences rather than from
any innate sex distinction. More generally, it is clear that some
individuals—regardless of their sex— do better at one thing relative
to another.

We have already suggested that good visualizers have an edge in video
game playing relative to poor visualizers. The reason for this, of
course, is that video games, by their very nature, require visual
thinking. The visually represented objects on the screen are
constantly changing, and a person who is able to mentally track these
changes, and who can imagine what the configuration of objects will be
several seconds hence, is in a better position to plot the appropriate
actions than is the person who doesn't have these abilities. We can
speculate that these individual differences in proclivity to use
visual versus verbal strategies are, in part anyway, what makes some
people seem inherently good at playing video games whereas others seem
inherently not so good.

We have already described how different strategies may be used to
accomplish the same goal. The distinction between visual and verbal
thinking provides an apt example of how different strategies may be
put to use. As we have mentioned, most video games emphasize the use
of visual skills. Where does this leave a person who isn't so good at
visual thinking? Probably the best solution for playing the games is
to work out novel strategies that emphasize verbal skills instead. In
Pac‐ Man, for example, progress can be made in various ways. One way
is to just rely on your instincts, judging which way the monsters and
you are going to be headed and trying generally to aim Pac-Man so that
he and they won't converge when they're not blue but will converge
when they are. This kind of "seat of the pants" strategy basically
makes use of visual skills.

But you could also use more rational, logical, verbal strategies. For
example, you could memorize and plan out various routes that you have
established as being relatively safe. Or you could devise an
intermediate strategy of, say, planning the order in which you're
going to eat the energizers and plan to avoid the monsters as best you
can in between.

How do you tell whether or not you are a good visualizer? If someone
looks at a watch and tells you that it is 8:37, can you easily conjure
up a mental picture of a clock reading 8:37? Or do you have to
struggle to mentally create this image, slowly picturing the small
hand set at 8 and then, while trying to keep the small hand glued to
where it belongs, picturing the large hand pointing to the lower
left-hand corner? There are several psychological tests that have been
used to measure how good at visualizing a person is, some of which
have been used by Canadian psychologist Allan Paivio. 8 For example,
in one test Paivio asked subjects to think of a cube of a certain size
and color that is sliced up into many smaller cubes. Next subjects
were asked how many of the smaller cubes have two colored surfaces,
how many have three colored surfaces, and so on. Based on the results
of this test, as well as others, subjects could be characterized as
being good or poor at visualization.

There is another test that can assist you in determining whether you
are a good visualizer. In Figure 3.1 you will see a list of pairs of
states with their shapes shown in the right-hand column. Look only at
the names on the lefthand side (covering the shapes on the right), and
place the six pairs in order so that the pair whose shapes are most
similar are at the top of the list and the pair that is least similar
is at the bottom. Now repeat the process while looking only at the
shapes.

FIGURE 3.1 How good a visualizer are you? See the text for instructions.

From M. Matlin, Cognition (New York: Holt, Rinehart & Winston, 1983),
p. 106, based on R. N. Shepard and S Chipman, "Second-order
Isomorphism of Internal Representations: Shapes of States," Cognitive
Psychology 1, no. 1 (1970): 1-17. Redrawn and used by permission of
Holt, Rinehart & Winston, Inc., Academic Press, Inc., M. Matlin, R. N.
Shepard, and S. Chipman.

Are your two lists similar to each other? If you put pair B
(Colorado—Oregon) near the top of both lists and pair C (Oregon—West
Virginia) near the bottom, you may have fairly good visual imagery.
Note that this is not a good test for distinguishing exceptional
visualizers, since most people looking only at the names make
judgments that are fairly similar to their judgments when looking only
at the shapes. One exception to this consistency in judgment is
Nevada; many people from the eastern United States are under the
erroneous impression that most of the Western states are square, and
this distorts their ability to judge the similarity of shapes when
given only the names.

Tests such as these can be used to identify people who have a facility
with visualization and consequently those who are likely to be good at
video games that require a visualization skill.

Motor Performance
We have concentrated so far on how information is gotten from the
environment and is then manipulated within the cognitive system.
Operating somewhat independently of the cognitive system is the motor
system, the part of the mind responsible for initiating muscle
movements. The sort of skilled movement required for video games is
called motor performance.

SKILL
A skill is a precise, finely tuned sequence of muscle movements,
usually designed to achieve a very specific goal. In general, a skill
is carried out in conjunction with feedback from the sensory system.
For example, a golf pro would never have learned his or her skill
without being informed where the ball landed after each stroke.
Similarly, to become an expert at playing a video game, you need not
only to develop the correct muscle patterns but also to coordinate the
appropriate sequences with the appropriate input from the screen —that
is, you need to develop what is referred to as eye-hand coordination.
While playing Pac-Man, for instance, you need to be able to
appropriately manipulate the joystick (a muscle skill) in a way that
is dependent on such things as where Pac-Man is in the maze, where he
is relative to the monsters, and so on.

PRACTICE
By any measure of performance quality that we use—time to carry out
the response, correctness of the response, or whatever—performance
will get better the more practice you've had. Most of the improvement
occurs when you're just starting. Even if you're very poor when you
begin learning a game, you'll almost certainly improve rapidly—at
least at first. Then your "improvement curve" ( figure 3.2) begins to
flatten out: the longer you play, the slower your subsequent
improvement will be. The curve is (at least roughly) logarithmic:
every doubling of the number of practices leads to an equal increment
in performance. Thus the second practice will produce the same
improvement as the first. However, to then get the same increment
again requires two more practices for a total of four. To get it yet
again, you need four more for a total of eight. Then you need to
double your practices to sixteen, then to thirty-two, and so on. Small
wonder it is time-consuming to become a really expert player.

Actually, this logarithmic rule is pertinent to improvement in almost
anything. A speaker system that costs, say, $1,000 certainly doesn't
seem ten times as good as one that costs $100, because quality of the
speaker system is logarithmically related to the effort of making it.
You keep having to double the effort that goes into the system in
order to obtain each additional unit increase in the system's quality.
Since you pay according to the effort, this means that price will rise
much faster than quality.

FIGURE 3.2 The improvement curve.

The amount of effort (practice) that you have to put into a skill
will, by the same reasoning, increase faster than the quality of the
skill. But notice another feature of this curve—it keeps going up
forever. No matter how much you practice, you'll always keep getting
better.

Various experiments have demonstrated this assertion. In a study
completed over twenty years ago, workers in a Cuban
cigar-manufacturing company who had rolled as many as 10 million
cigars continued to increase their speed of rolling cigars. However
here, as in virtually all cases, the rate of improvement decreased. 9

That we can continue to improve is a feature of the human motor system
that is particularly felicitous when you are becoming skilled at a
video game since, as pointed out earlier, most video games are
programmed to keep getting harder and harder as you keep getting
better and better.

With practice, many motor skills become increasingly automatic. You
can drive a car and carry on a conversation at the same time because
both of these skills are highly practiced. When a motor skill becomes
automatic, it means that it can be done with a minimum of conscious
control. Since conscious control is not needed anymore for completion
of the motor skill, it can be used to concentrate on other features of
the environment. A skilled pianist, for example, can forget about the
specific motor movements and concentrate instead on interpreting the
mood of a concerto, and an ace tennis player can play a decent game of
tennis while carrying on a conversation. The similar "automaticity"
occurs with truly experienced video gamers. We watched a skilled
Pac-Man player effortlessly control the joystick while simultaneously
talking with a friend and periodically reaching with her other hand to
take a sip of beer. Clearly her impressive performance had reached a
level of smooth, autonomous mastery.

MOTOR/COGNITIVE INDEPENDENCE
This autonomy is one consequence of an independence that develops
between the cognitive and motor systems. Many years ago, one of the
authors (GL) broke a finger and couldn't drive his sports car.
Strapped into the passenger seat, he found him-self unable to tell the
substitute driver where reverse was in the gearshift configurations.
Instead he had to actually move into the driver's seat and (somewhat
painfully) go through the motions of putting the car into reverse. His
motor system knew perfectly well where reverse was, but his cognitive
system apparently didn't have a clue. (And the motor system wasn't
about to reveal the whereabouts of reverse to its cognitive
colleague.) In this episode the cognitive and motor systems apparently
functioned relatively independently.

When you learn a motor skill, it is not under control of the motor
system from the start. At first you spend a lot of time thinking about
what you're doing. As learning progresses, it gets taken over to a
greater and greater degree by the motor system. This phenomenon is
nicely illustrated when you learn to touch-type. After you have
memorized the keyboard and fingering, you still have to take cognitive
(conscious) steps of going to long-term memory to retrieve information
about both the location of the key you want to strike and the finger
responsible for it. Only then is the motor system summoned to perform
the action. Gradually control is transferred from the cognitive to the
motor system. In fact, the expert typist, unlike the beginner, is
typically unable to quickly and accurately reproduce the keyboard any
more. What the fingers have learned, the mind has forgotten.

In recent years, a pair of books entitled Inner Tennis and Inner
Skiing have appeared. 10 Their major message is that, when you're
trying to learn a motor skill such as tennis or skiing, it's highly
detrimental to think about what you're doing. Instead you should just
turn control over to the motor system and let it go. The authors of
these books depict the cognitive and motor systems as two "selves,"
the cognitive system being "Self 1" and the motor system being "Self
2." Indeed, a good strategy for something like skiing—which is almost
entirely a motor skill—would be to think about something else (do
arithmetic problems in your head, for example), thus disabling the
cognitive system and rendering it unable to do its mischief.

MORE ON EYE-HAND COORDINATION
Eye-hand coordination is essentially the ability to perform an
appropriate sequence of motor skills in response to a particular
sequence of information entering the visual system from the
environment. It isn't exactly that some particular pattern of muscle
movements gets connected to some specific sequence of visual input.
Rather, the relationship is mediated by some intervening, higher-level
goal. Suppose, for example, that you are driving down a highway. Your
hands are on the top of the steering wheel at the two o'clock and ten
o'clock positions. The connection between visual input and motor
action seems quite straightforward—if the road curves left, your hands
"automatically" move left. Road right means hands right. Suppose,
though, that you shift your hands to the bottom of the steering
wheel—to the five o'clock and seven o'clock positions. The appropriate
muscle movements for a particular visual input are now the exact
opposite of what they were when your hands were on top of the wheel.
Now when the road curves right, you must move your hands left, and
vice versa. It's not just that you've learned two visual/motor
associations, one for "hands on top of wheel" and the other for "hands
on the bottom of the wheel"; you perform the appropriate muscle
sequence effortlessly no matter where on the wheel your hands happen
to be. Thus the appropriate connection can't be between a particular
visual input and a particular muscle sequence. Rather, the connection
must be between visual input, the muscle sequence, and some
higher-level goal (in this case, keeping the car on the road). This is
an elegant and extremely efficient—but not very well understood—manner
of designing a system of eye-hand coordination.

The reliance of motor skills on higher-level goals is obviously
beneficial when video games are being played because it means that
once a particular skill has been learned, it will transfer to slightly
different physical configurations of the same game. We knew an expert
Pac-Man player, for example, who had learned the game at the video
arcades. She was introduced to a homecomputer version of the game in
which Pac-Man was directed not by a joystick but by certain keys on
the computer keyboard. It took her very little time to become just as
expert at this game as she had been at the original arcade version. In
this instance, the actual motor response—pressing the appropriate
configuration of keys—was entirely different from the original
response of manipulating the joystick. But the higher-level
goals—guiding Pac-Man to the correct areas of the board, avoiding the
monsters, and so on—had not changed, and it was these higher-level
goals at which she had become an expert.

Strategies
So far we have been primarily concerned with each component of the
cognitive system as it applies to video games. Thus we have seen how
focused attention can be useful or detrimental, how various types of
interference can cause deterioration of video game performance, and so
on. Now we want to talk a little more about how all the components
work in concert. A particular choice of which cognitive components
will be used and how they will get put together is termed a strategy.
Earlier we discussed how a stereo system's performance depended both
on the workings of the individual components (a factor over which one
has only limited control) and on strategy, how the user chooses to
arrange the components. For example, if you wanted to provide a
classical music background while you were working in your basement
workshop, you might set the FM tuner to a classical station and switch
on the speakers you've set up in the workshop. But if you wanted to
fill the house with your favorite rock 'n roll music, you might use
the record player rather than the tuner and use all the sets of
speakers.

STRATEGIES FOR PLAYING VIDEO GAMES
There are also appropriate cognitive strategies for playing video
games, from simple and obvious to complex and subtle. Consider, for
example, the simple game of Breakout, in which, you will recall, there
is a brick wall against which you hit a ball using a paddle. Each time
the ball hits the wall, a brick disappears and you gain some number of
points. Your goal is to eventually knock out all the bricks in the
wall.

This game is quite simple and thus requires a fairly simple cognitive
strategy. In large part, the game calls for focusing visual attention
on where the ball is relative to where the paddle is. There are
virtually no memory requirements. But consider, in contrast, a much
more complex game such as Pac-Man. You need to focus attention on
where you are, where the nearest escape route is, where the monsters
are, and whether they're blue or not (recall that a blue monster can
be eaten by Pac-Man rather than vice versa). You have to use your
short-term memory to remember such things as what board you're on, how
many Pac-Men you have left, how many energizers you've consumed, how
long it's been since the monsters have been blue, and so on. You need
to use your long-term memory in order to remember the configuration of
the maze, where the escape tunnels are, and the behavior of the
monsters in certain situations.

Given this complexity, there are various appropriate cognitive
strategies. For example, you could rely primarily on long‐ term memory
and memorize routes that work well in a variety of situations. But
such a strategy would have several costs. First, you would have to
memorize the strategies in the first place, which would require a lot
of time (and a lot of quarters). Second, you would have to devote some
of your processing capability to remembering where you are in a given
route and where the appropriate place to go next is. This, of course,
means less processing capability for such things as focusing and
switching attention. Another disadvantage is that video game makers
can easily change the routes of the monsters, thereby rendering your
carefully learned routes obsolete. Finally, one wrong turn causes your
route to become fouled up. A different strategy might be to forgo the
specific routes and concentrate instead on trying as hard as possible
to avoid the monsters, while still staying in the general vicinity of
the uneaten dots. This way, use of memory would be kept to a minimum.
You wouldn't care about exactly where you were in the maze at any
given time. The general idea would be that if you could keep avoiding
the monsters, you would eventually get all the dots. This strategy is
somewhat inelegant, as you keep fussing around, apparently aimlessly,
for quite some time. However, it avoids the pitfalls of the
memorization strategy.

Given that there are at least two (and probably more) appropriate
cognitive strategies to use, which one should you use? One expert, Ken
Uston, 11 is partial to the route strategy and, in fact, devotes most
of his book to describing and developing very sophisticated and
complex routes. Prior to writ-ing his Pac-Man book, Uston had already
achieved a good deal of fame for his development and popularization of
gambling strategies, notably for the game of blackjack. 12 Like his
Pac-Man strategies, his gambling strategies are based on very complex
memorization strategies. Uston became an expert at these schemes and
used them to make huge amounts of money in Las Vegas, Reno, and other
international gambling spots. Thus it is clear that Uston, by his
nature or through a great deal of practice, is an expert at memorizing
and, for him, mastering a new strategy based on memorization would be
natural and easy.

But if you are a poor memorizer, you might want to develop a strategy
that requires a minimum of memorization. If you are slow at retrieving
information from long-term memory, you'll want a strategy that
minimizes such retrieval, and so on. 13

More generally, video game players may be concerned with a "strategy
for developing strategies." Most current video games are complex,
requiring complex strategies. One way of combating this complexity,
which actually applies to problem solving in general, is to break the
required actions down into constituent parts. In figure 3.3 we have
done this for the game of Sabotage. This breakdown yields a
hierarchical, or treelike, structure, where points on the tree are
goals and subgoals that we wish to accomplish. At the top is our
overall goal of making as many points as possible. Two major subgoals
are used to achieve this overall goal—keeping from being bombed and
keeping paratroopers from accumulating on the ground. Each of the
subgoals is itself achieved by one or more subgoals that are nested
underneath it.

FIGURE 3.3 Sabotage: the hierarchy of its goals.

When you break things down this way, it becomes much easier to see
exactly what has to be accomplished. Notice also that at the bottom of
the tree are relatively simple motor skills that have to be learned.
Once you have identified these bottom-level goals—these specific
skills—you should, if at all possi-ble, practice each one in isolation
since that will optimally provide you with the action/feedback
sequences necessary for learning the skill.

To see the utility of using a hierarchical strategy, let us take an
example of not using it. Recall that the predominant action in
Sabotage is the appearance of helicopters that, if you don't destroy
them with your gun, drop paratroopers. Helicopters (being large) are
both easier to hit and worth more points than paratroopers (which are
smaller). Moreover, if you concentrate on shooting down each
helicopter as soon as it appears, it won't get much of a chance to
drop paratroopers anyway. Thus focusing attention on helicopters and
trying to hit each one as soon as it appears seems to be a reasonable
strategy.

However, this strategy works only up to a certain point, because
eventually helicopters start appearing and dropping paratroopers at
such a frenetic pace that the sky soon becomes filled with
paratroopers, and you're reduced to firing blindly and continually.
Since each shot costs a point, you start losing points faster than you
gain them. It's hopeless anyway, since even this desperate behavior
soon becomes inadequate to catch all the paratroopers. So you watch
helplessly as you're surrounded and eventually blown away.

The strategic error in this line of play is in not developing the
subskill of shooting down paratroopers during the early phases of the
game when they appear only infrequently and you can practice at a
leisurely pace. Realizing this error, you would change strategies and,
in the early phase of the game, deliberately allow the helicopters to
survive and to drop their paratroopers. Using this gambit, you could
concentrate fully on picking off the paratroopers. When you finally
develop some proficiency at this skill, you can return to the
strategy—overall, more efficient—of shooting down the helicopters as
rapidly as possible. But at the same time, you will be secure in your
knowledge that when the number of paratroopers starts to increase, you
can accurately and calmly hit each one with a single shot.

Video Games as Problem Solving
Before we can master a game, we have to learn to play it. The process
of moving from novice to highly proficient player can be viewed as
problem solving, a process that has been studied extensively by
psychologists.

There are three major aspects of a problem-solving situation: (i) the
original state, (2) the goal state, and (3) the rules. For example,
imagine your goal is to become proficient at Pac‐ Man. The original
state, or the situation at the beginning of the problem-solving
process, might be, "I have played games before but never a video
game." The goal state, reached when the problem is solved, is to
become proficient at Pac-Man. You might even make a more specific goal
for yourself, such as beating the previous high score on the machine
at the video parlor on Saturday, April 18. The rules refer essentially
to restrictions that must be followed as you go from the original
state to the goal state. They might include your wanting to become an
expert on the arcade rather than the home version of Pac-Man, or to
accomplish your goal without consulting any of the books on the game.

Good problem solvers seldom strike out randomly. Rather, they plan.
Often they break the problem into smaller subproblems, then
concentrate on solving those subproblems. In our observations of
players in video parlors, we have occasionally come upon a person who
sits down at a new game, inserts a quarter, and begins moving the
joystick around without any idea whatsoever about the goal of the
game. This approach typically gets the person nowhere.

The video player who mindlessly plunges in is failing to appreciate
the importance of the first phase of problem solving —understanding
the problem. To understand a problem, you must pay attention to the
important aspects of it and ignore the rest. In Pac-Man, it is
obviously important to pay attention to Pac-Man, the monsters, and the
energizers. The elapsed time since an energizer was consumed is
important. Whether the monsters are blue or not is vital. However,
whether a monster happens to be green or pink is unimportant and
irrelevant.

Once you, the problem solver, figure out what is essential, the next
step is to try one or more different strategies for solving the
problem. If your initial goal in Pac-Man is to eat all the dots on a
board and win a new one, you might begin by randomly moving the
joystick in one of the four possible directions. If you did this
enough times, you might eventually stumble on a method for achieving
your goal. However, this random approach would be inefficient and
unsophisticated. A more creative approach would be to try selective
routes, routes that would be more likely to lead to the goal. These
selective approaches are called heuristics, and they are far more
efficient than random methods. The novice Pac-Man player might try
running Pac-Man around the edge of the screen and then attempting to
eat the dots in the middle areas. This particular heuristic might not
lead to the desired goal immediately, but it could lead to the
postulation of other heuristics that might then be tried and might
ultimately succeed. One problem with the "edge-then-middle" heuristic
is that all of the energizers would be consumed early in the game, and
none would be available later, when Pac-Man desperately needed them.

Clearly, a heuristic that does not deplete the supply of energizers is
called for.

During the course of solving the Pac-Man problem, new strategies will
be discovered. When the authors first started playing Pac-Man, we used
to immediately consume an energizer whenever one of the four monsters
was pursuing. We soon learned that it was far better to wait until the
monster was close at hand before consuming the energizer. This
increased our opportunity to contact and destroy the monster while
still in an energized state. When learning to play a video game, good
players use many heuristics of this sort.

The process of dividing a problem into a number of subproblems, or
smaller problems, is called means-ends analysis and is a feature of
many successful problem-solving strategies. The process has received
its name because it involves figuring out the "ends," or goals, that
you want to attain and then devising certain "means," or strategies,
for reaching them. In general, as you solve the subproblems, you
continually reduce the difference between your original state and your
goal state. Suppose your larger goal in playing Pac-Man is to beat the
previous high score. Rather than going for this largish goal all at
once, it would make much more sense to break up the problem into a
number of smaller problems. These might include (1) reaching the first
energizer and consuming a monster, then (2) completing a board and
receiving a new maze with fresh dots, then (3) capturing the
strawberry symbol. Mastering each of the subproblems gets you closer
and closer to your ultimate goal. Put another way, as you complete
each subproblem, you continue to reduce the difference between your
original state (being a novice at Pac-Man) and your goal state
(beating the previous score). You have used a means-ends analysis.

In everyday life we use means-ends analyses so often that we often
take them for granted. [/SPOILER:db06e850

Chapter 3 Continued
[SPOILER:0e72f55b18] If you must pick up a friend who is arriving
tomorrow at the airport and your car has broken down, you might divide
your problem into two subproblems: (1) borrowing a car and (2) getting
yourself to the airport. Once the first subproblem is solved, the
difference between your original state and your goal state is
substantially reduced. Again, you have used a means-ends analysis.

In some instances, means-ends analysis may not be the best approach,
or can even mislead you. This occurs when the solution to a problem
depends on temporarily increasing the difference between the original
state and the goal state. For example, assume that you have identified
the consuming of an energizer as a first step in your ultimate plan of
eventually beating the high score for the day. The consuming of an
energizer is your first subgoal, and it is natural to think that you
would want to move the joystick in the direction of the nearest
energizer. But it sometimes makes sense to first move the joystick
away from the energizer rather than toward it (for example, if there
is a monster between Pac-Man and the energizer). In this instance you
temporarily increase the difference between your original state and
the goal state. With some games this process actually enhances your
chances of winning. Discovering that a move away from the goal will
actually lead you to your ultimate goal involves truly creative
problem solving.

Expert Learning
What distinguishes an expert at some skill from a novice? We've seen
some of the components that go into becoming an expert: you have to be
a problem solver to figure out the rules of the game, you have to
devise strategies to optimally accomplish a variety of goals, and you
have to attach appropriate motor responses to stimuli. In addition,
particularly when the skill of interest is video game playing, you
must also learn to perceive game situations as complete units.

What's meant by this? Psychologists talk of perceiving and processing
"chunks." A chunk is anything stored in long-term memory as a unitary
whole. For instance, the letter string MGAE is perceived as four
separate letters—four chunks. But the same letters presented as GAME
are perceived as one word —one chunk. The fewer the chunks you have to
process in order to accomplish some task, the more efficiently the
task can be done.

Various studies have linked the acquisition of expertise in game
playing to the fusing of many small chunks into fewer larger ones.
Consider chess. In one experiment, various board positions were shown
either to chess experts or to chess novices. The board positions were
either random configurations of the chess pieces or they derived from
actual games. Later the subjects had to reproduce the board positions
they had seen. Neither the novices nor the experts could reproduce the
random board configurations very well. The novices couldn't reproduce
the actual game configurations very well either, but the experts
could.

The boards involved perhaps twenty pieces. Apparently, however, the
experts saw the game configuration boards as a small number of chunks,
because any configuration resulting from an actual game was bound to
be very similar to some configuration that the experts had seen many
times before. This wasn't true for the novices; hence for them twenty
pieces constituted about twenty separate chunks. The random board
configurations were unfamiliar to everyone and were thus perceived by
all as many chunks. The general principle is: The fewer the number of
chunks in some stimulus, the easier is that stimulus to deal with.

The same principle can be applied to the learning of a video game.
Take a complex game like Defender. The novice is overwhelmed by what
takes place on the screen. Each component—each humanoid, each mutant,
each stretch of terrain— constitutes a separate chunk. It takes a good
deal of time to analyze all these chunks, and thus the novice is slow
to respond and is quickly defeated. As you become expert, however, you
begin to fuse all these chunks into fewer, bigger chunks, and begin to
see not individual objects but all the objects together as situations.
Since a situation is only one chunk, it's easy to analyze, and you can
respond to it rapidly and easily.

Designing New Games
Most video game futurists do not seem to be taking into account the
human cognitive system. Instead they emphasize what characters are
likely to have the appeal of Pac-Man. In an article in Psychology
Today writer Dan Gutman speculates on what it is about Pac-Man that
made him the biggest cultural hero between John Lennon and E.T. 14 He
speculated that the cute, cuddly character, who was involved in an
essentially nonviolent game, was especially appealing to women, and he
easily chewed up millions of their quarters. Although there may be
relatively fewer women at the arcades, when they do go, they seem to
enjoy Pac-Man. As for candidates for a future Pac‐ Man, Gutman
proposed "Q*Bert," a mangy-looking, noselike creature who hops on a
pyramid made of cubes, trying to make them all the same color. Every
time Q*Bert hops on a new cube it changes color. Or, if not Q*Bert,
then "Domino Man," who weaves his trail of dominoes through the
congested shopping-center parking lot. His major role in life is to
protect the trail from the Bumbling Bag Lady and the Reckless Little
Boy and his hot-rod shopping cart. Or, if not Q*Bert or Domino Man,
then perhaps "Millipede," who must defend its homeland from hordes of
marauding insects. Perhaps the cute, cuddly characters are what draw
some people to the games, but it seems likely that other
considerations would be far more important.

Just as video game players may well wish to think about which
cognitive abilities are required when devising a strategy, video game
designers may wish to consider what cognitive strategies will be
involved in any new game they might consider designing. Most probably
they would consider an existing game and realize that there's some
cognitive component—or set of components—that the playing of this game
doesn't really require. Thus, if some modification of the game could
be designed that would require those missing components, the game
would be that much more complicated and that much more interesting.
Let's look at an example of how this might work.

GROUND-LEVEL PAC-MAN
Certain of the games—Pac-Man and Defender, for example —are especially
exciting and fast-moving, requiring fast reactions and fine-tuned
eye-hand coordination. Others—for example, some of the "adventure,"
maze-running games—are slower-moving but more intellectually
appealing, making much greater demands on memory (remembering where
you are, where various rooms are, where you've left various objects,
how to go about achieving some goal, and so on). Further, some of the
adventure games stretch the imagination, allowing you to fantasize
yourself into a "real-life" situation. Imagine a game in which you
were in the Pac-Man maze, instead of looking down at it. You would be
swept down the corridors gobbling up dots wherever you found them,
evading the monsters, and, in general, doing what Pac-Man usually does
in a Pac-Man game. From your point of view, of course, many things
would have changed relative to the normal Pac-Man situation. Lacking
the bird's-eye view of the maze usually enjoyed by Pac-Man players,
you wouldn't know where the monsters were unless they happened to
appear in the corridor; thus monsters would unexpectedly leap out from
behind a corner, or would be lying in wait at the next turn. Moreover,
you would forget pretty quickly where you were in the maze since you
couldn't see yourself from the outside. As might be expected, this
uncertainty would lead to problems—for instance, once you had eaten a
row (or, as it appeared to you, a corridor) of dots, you wouldn't
quite remember where the rest of the unconsumed dots were. You
wouldn't have the traditional luxury of being able to glance around
and see where the energizers were and how many were left. Finally, you
rather than your little surrogate face would be the one in danger of
being obliterated at any moment.

This hypothetical invention, "Ground-level Pac-Man," might become a
reality; someone will take the concept and program it because,
technically, such three-dimensionality is entirely feasible. In fact,
someone has more or less thought of this idea. In Disney's box-office
hit Tron, the central character is a man called Flynn, who is an
expert computer programmer as well as a world-class video game player.
During most of the movie, Flynn is trapped inside a video game trying
to get out. As he zips through corridors, enemies continually try to
attack him. In the end—of course—he frees himself.[/SPOILER:0e72f55b18]

Psychology Terms Used In Book:
[SPOILER:0e72f55b18]Reinforcement-the provision for you of something that you like.
Schedule of reinforcement (aka partial reinforcement)—reinforcement is intermittent rather than continuous.
Extinction-decline and eventual cessation of behavior in the absence of reinforcement
Extinction period-the length of time it takes for the behavior to cease or extinguish.
Magnitude of Reinforcement-reward
Delay of Reinforcement
Multiple Reinforcements
Intrinsic Reinforcement
Cognitive Dissonance-paradoxical types of behavior.
Cognitive Dissonance Theory-theory assumes that when a person performs acts or holds beliefs that are in conflict with one another, the person will act so as to reduce the conflict.
Regret
Challenge
Fantasy
Curiosity
Sensory Memory
Attention
Filtering Process
Saccade-(French for "jerk" or "jolt") which is a quick jump of the eye from one place to another.
Fixations-periods in between saccades during which the eye is relatively stationary.
Short-term Memory
Long-term Memory
Expectancy
Verbal/Visual Distinction
Motor Performance-the motor system, the part of the mind responsible for initiating muscle movements. The sort of skilled movement required for video games is called motor performance.
Skill-a precise, finely tuned sequence of muscle movements, usually designed to achieve a very specific goal. In general, a skill is carried out in conjunction with feedback from the sensory system.
Practice
Eye-hand coordination-the ability to perform an appropriate sequence of motor skills in response to a particular sequence of information entering the visual system from the environment.
Strategies
There are three major aspects of a problem-solving situation: (i) the original state, (2) the goal state, and (3) the rules.
Chunk-anything stored in long-term memory as a unitary whole. For instance, the letter string MGAE is perceived as four separate letters—four chunks. But the same letters presented as GAME are perceived as one word —one chunk. The fewer the chunks you have to process in order to accomplish some task, the more efficiently the task can be done.[/SPOILER:0e72f55b18]

I'm bumping this thread hoping people are able to understand that 80s videogames still have the same psychological hooks as games of the 21st century.

There are more articles on the net on psychology of videogames, but they aren't as thorough as the book excerpts I posted.

Other articles:

http://www.gamasutra.com/view/news/29910/Analysis_The_Psychology_of_Immersion_in_Video_Games.php

http://www.gamasutra.com/view/feature/2289/the_psychology_behind_games.php