CHAPTER

Building Empathy,
Compassion, and Achievement
in the Jigsaw Classroom
ELLIOT ARONSON
Distinguisfwd Visiting Professor, Stanford University, Stanford, California

In their remarkable little book, The Children are Watching, educators Ted and
Nancy Sizer suggest that the classroom curriculum constitutes only a small part
of what a youngster actually learns in school. Students pay close attention to
and learn from just about everything that happens in and around their school.
For example, if the building is drafty, has broken windows, unswept floors,
grimy walls, cracked linoleum, and leaky toilets, the students get the message
that the adult community does not care a lot about their education. I would
add that, by the same token, students may learn as much from the style or
process in which information is conveyed as they do from the information
itself.
! hasten to explain. It goes without saying that there are a great many ways
to convey information to students. Let us say we want the students to learn
about World War II. The teacher can create two or three lectures about the
causes and consequences of that war. Alternatively, the students can simply
read the basic facts about World War II in a textbook. As a third alternative, the
teacher can assign students to do their own research in the library, or have
students interview people who served in the military or lived through the war
period in the United States, Europe, and Asia. The teacher might require
students to work individually or in groups. To demonstrate what they have
learned, students might take a test, write a term paper, or give an oral presen~
ImprovingAcademicAchievement
Copyright 2002, Elsevier Science (USA). All rights reserved. 209
210 Elliot Aronson

tation to the entire class. The teacher might also run the class as if it were a quiz
show where he or she asks questions and the students show their quickness
and mastery of the subject by raising their hands as soon as they know the
answer.
Each of these methods of conveying and retrieving information sends its
own special message to students. Some of these messages may be unin~
tended. Teachers who lecture send the message that they are an expert source
of information and that the things that they know are the important things to
learn. Teachers who dispatch students to the library send the message that it is
useful for students to become skillful researchers as well as to learn about the
topic at hand. Teachers who require students to interview a war veteran convey
the implicit message that not all important information is contained in books
or conveyed by people in the teaching profession. Teachers who run their class
like a quiz show or contest indicate that quickness, assertiveness, and com~
petitiveness are important aspects of the learning endeavor--and perhaps of
life itself.
The point is that the learning students derive from the process of their edu~
cational experience is powerful indeed. In classes where students are expected
to raise their hands as soon as they know the answer and take tests graded on a
curve, the process encourages students to compete individually against each
other. The implicit message is that the other students are competitors for scarce
resources. It would not be surprising, then, if this process were to create tension
among peers and tended to discourage trust or friendship among youngsters
who were not already friends beforehand.
In American schools, this kind of competitive process is the predominant
method employed in most classrooms most of the time. High school students
who have gone through several years of participation in such a competitive
process are likely to view the world (both inside and outside of school) as one
gigantic game of musical chairs, as a dog~eat~dog place where the prizes go to
those who are quickest, or strongest, or most aggressive, or most charming, or
most athletic. Such a process may implicitly encourage students to look for
weaknesses or flaws in their peers--to find reasons for excluding or taunting
those who falter, or seem different, or socially awkward and to think of them as
weird or as losers.
This may be a major reason why research reveals that most American high
schools are clique~driven and exclusionary. Even casual observation of the
social climate of the typical high school reveals a clear hierarchy of who is in
and who is out that is remarkably similar in all parts of the country. In most high
schools, the youngsters at the top of the social pyramid are athletes (especially
football players), cheerleaders, class officers, attractive women, and "regular
guys." Near the bottom of the social pyramid are those youngsters who are the
"wrong" race or the "wrong" ethnic group; dress differently; are too short, too
tall, too thin, too fat, too "nerdy", or just do not fit in easily. Interviews with
high school students in all parts of the country indicate that just about every
l O. Building Empathy, Compassion, and Achievement in the Jigsaw Classroom 211

student in a given school can name the hierarchy of in-groups and out-groups
and can identify where each of their classmates can be placed in that hierarchy
(Aronson, 2000; Gibbs & Roche, October 1999; Lewin, 1999; Townsend, 1999).
Those youngsters in the relatively small in~group want to differentiate
themselves from the losers. They do not associate with them, sometimes they
tease or taunt them; occasionally they bully them. In most cases, those near
the bottom of the social pyramid suffer in silence, retreating further and further
from the mainstream. The more they are ignored, excluded, or taunted, the
further away they drift. For a great many youngsters, the high school atmos~
phere is extremely unpleasant. For some, it is a living hell. Given this kind of
social atmosphere, and given the fact that teenagers spend almost half their
waking hours embedded in that atmosphere, it should not be surprising that
occasionally, some of these students go over the edge--doing serious damage
to themselves or others. The most recent statistics are chilling: One of five
teenagers has seriously contemplated suicide; one of ten has made an attempt
at suicide (Goldberg & Connelly, 1999).
In recent years, when we think of students going over the edge and doing
serious damage to themselves or others, we are immediately reminded of
Columbine High School in Littleton, Colorado. You will recall that, in April of
1999, two students, Eric Harris and Dylan Klebold, armed with an arsenal
of guns and explosives, went on a rampage, killing a teacher and 12 of their
fellow students before turning their guns on themselves. While tragic events of
this sort are not everyday occurrences, they are not as rare as we would like to
believe. At the turn of the century, there were nine rampage killings in high
schools and junior high schools in a period of 18 months.
What is to be done? How does a society protect its students from these
senseless killings? In the aftermath of a horrendous event like the Columbine
massacre, we are tempted to look for instant solutions before we fully under~
stand the cause of the problem. Thus, following the Columbine massacre,
Congress voted to tack onto the crime bill an amendment giving states the
right to allow the display of the Ten Commandments on school walls and
bulletin boards. In the interest of making schools safer, there was a rush to
install metal detectors. School administrators have also asked students to
report other students who threaten violence or who even seem different
(dress strangely, keep to themselves, and so on). Some school officials have
ordered personality tests to be administered to all students--tests aimed at
profiling those students who might be most apt to go on a murderous rampage.
Some of these interventions are silly. Others are harmful. Still others, while
harmless, miss the point by a wide margin. Here, ! would like to make a
distinction between root cause solutions and what I like to call "pump handle"
solutions. Here is what I regard as the defining example of this distinction. In
1854, Dr. John Snow, England's leading epidemiologist, was charged with the
task of stanching a cholera epidemic that was devestating London. Dr. Snow
needed to act quickly. He first determined that most of the cases of cholera
212 Elliot Aronson

clustered around a particular well near the center of the city. But he made no
attempt to educate the local residents or to convince them that it might be
dangerous to drink from that well. He simply removed the pump handle so that
water could no longer be drawn from it. It worked. In many respects it was a
perfect solution--for that particular epidemic, in that particular place, at that
particular time.
But Dr. Snow did not stop with this measure. After ending the immediate
threat, he proceeded to do some systematic sleuthing, trying to get at the root
cause of the epidemic. That is, he raised the scientific question: Why was that
particular well contaminated and not others? Within a few months, he was able
to conclude with a high degree of certainty that the contamination was caused
by fecal matter from nearby latrines leaking into the water supply being tapped
by the well. This discovery led to legislation requiring latrines to be built at a
reasonable distance from wells. By getting at the root cause of the problem, Dr.
Snow succeeded in preventing future epidemics not only in London but
throughout England and, eventually, throughout the entire civilized world.
With this in mind, let us take a close look at the solutions to the school
shootings proposed by policymakers:

1. Posting the Ten Commandments. This intervention may be politically

expedient, because it creates the illusion that Congress is doing something;
but, as public policy, it is pathetically feeble. Needless to say, the Ten
Commandments is a magnificent code of ethics; but posting it on a school's
bulletin board will have very little impact on the behavior of any students hell~
bent on murdering their classmates. Every high school student knows about
the Ten Commandments. I am reasonably certain that most could even recite
the most relevant items on the list. They know they are not supposed to kill
or steal. They know they are supposed to honor their fathers and their
mothers. They probably even know they are not supposed to mention God's
name in vain, although my guess is that few are certain what that one means.
But being able to recite the Ten Commandments and living by the Ten
Commandments are two very different things. Those students who already
have a moral compass have no need of posters to remind them that one
should not kill or steal. Those who do not have a moral compass will not
be prevented from killing or stealing by seeing a piece of paper tacked to the
wall.
2. Metal Detectors. There is no doubt in my mind that installing metal
detectors in every entrance to every school in the country will reduce killings,
but it is far from ideal. The downside of this action is that it makes the school
seem like a scary place and suggests that, as a society, we cannot trust our
own kids not to kill each other. Moreover, installing metal detectors is far from
a perfect pump-handle solution. It would not make the schools perfectly safe.
If teenagers with grievances and guns are motivated to shoot their fellow
10. Building Empathy, Compassion, and Achievement in the JigsawClassroom 213

students, they can easily accomplish their mission without even entering the
school building. For example, in Jonesboro, Arkansas, the shooters fired on
their classmates from behind trees surrounding the school after first luring
their victims out of the building by setting off the school's fire alarm. Metal
detectors do not get anywhere near the root of the problem; needless to say,
the cause of the recent rash of school shootings is not the lack of metal
detectors in schools.
3. Profiling Potential Troublemakers. Identifying the youngsters who are
loners, different, or awkward either by asking other students to point them
out and turn them in or by forcing them to take some sort of personality
test will not solve the problem. Indeed, singling out these youngsters--the
very youngsters who are most hurt by having been excluded--is likely to
exacerbate the root problem by increasing their level of exclusion. More~
over, personality tests simply are not as accurate or as precise as the
general public thinks they are. Using such tests in an attempt to identify
troubled youngsters will undoubtedly target a great many who are not
troubled at all, while allowing some seriously troubled individuals to slip by
undetected.

TAKING A CLOSE LOOK AT THE

CLASSROOM ATMOSPHERE

I should point out one obvious fact: Young mass murderers are not shooting
people in their neighborhood, or in the local video arcade, or in the fast food
outlet. They kill their classmates and teachers, and sometimes themselves, in or
around the school building itself. Looking for root causes in individual path~
ology is an approach that seems sensible on the surface but it does not get to
the root of the problem. What is it about the atmosphere in the schools them~
selves that makes these young people so desperate, diabolical, and callous?
Why do they seek revenge, or a twisted notion of turning humiliation into pride,
by shooting their classmates? In what ways have they felt rejected or, humili~
ated at school? Are schools doing the best they can to develop students'
characters as well as their intellects? Can schools do better at creating inclu~
sive, caring communities with positive role models for students?
As suggested earlier, my observations of the typical classroom atmosphere
suggest that it is highly likely that the perpetrators were reacting in an extreme
and pathological manner to a general school atmosphere--an atmosphere of
exclusion--that many if not most of the student body find unpleasant, dis~
tasteful, difficult, and even humiliating (Aronson, 2000).
My observations receive strong support from the way high school students
from all parts of the country reacted to the tragic events in Littleton. Shortly
after the Columbine massacre, a search of the Internet revealed powerful
214 Elliot Aronson

feelings being expressed by large numbers of teenagers. The overwhelming

majority fairly crackled with expressions of anguish and unhappiness, describ-
ing how awful it feels to be rejected and taunted by their more popular
classmates. Several weeks before the contents of the Harris/Klebold video-
tapes were released, many of the writers were convinced that Harris and
Klebold must have had similar experiences of rejection and exclusion. I hasten
to add that none of these teenagers condoned the shootings; yet their Internet
postings revealed a surprisingly high degree of understanding and empathy for
the suffering that Harris and Klebold must have endured. While none of the
students posting their comments on the Internet had any intention of following
the lead of Harris and Klebold, some admitted that they had had fantasies of
doing similar things. That should make us sit up and take notice, not so that we
can track down those kids and hospitalize them, but rather, to try to figure out
how to improve the atmosphere of the school so that youngsters might
become prone to have such violent fantasies.

CHANGING THE CLASSROOM ATMOSPHERE

Our policymakers framed the question in the context of the Columbine

massacre: How can we prevent such events in the future? But, as implied
above, the question is a far broader one. How can we change the process of
the typical classroom so that our schools can transform themselves into
more humane social environments for all students? The general answer is
to change the dynamic of the classroom from dog-eat-dog competition to a
more cooperative, more caring one. Easier said than done. Many schools
have actively attempted to counteract the negative influences of excessive
competition by asking kids to try to cooperate with one another. It would be
hard to find a preschool or elementary school that did not actively encourage
children to share, to work harmoniously with their peers, and to behave
respectfully and cooperatively with one another. Many elementary schools
now have students sit in small groups at tables, rather than in rows of
individual desks.
But simply assigning students to work together in groups to produce a joint
report does not guarantee true cooperation. Unstructured attempts to encour-
age cooperation in the classroom usually fail to accomplish their ultimate goal
and might even backfire if not carefully designed. Most often the group
dynamics of an unstructured "cooperative" situation mirror the larger com-
petitive classroom dynamic. The one or two most able or most motivated
students put themselves forward to do most of the work while simultaneously
resenting the fact that they are carrying the load for the entire group. And the
less able or less motivated students end up doing little, learning little, and
feeling inadequate. These so-called cooperative groups are cooperative in
name only.
10. Building Empathy, Compassion, and Achievement in the Jigsaw Classroom 215

THE JIGSAW CLASSROOM

The problem with cooperative learning assignments is not that they do not
work. It is that they need to be carefully designed and rigorously structured for
them have the intended effect. One successful model, with a three-decade
track record, is the jigsaw classroom. Jigsaw is a specific type of group learning
experience wherein each student must cooperate with his or her peers to
achieve his or her individual goals. Just as in a jigsaw puzzle, each piece--each
student's part--is essential for the production and full understanding of the
final product. If each student's part is essential, then each student is essential;
and that is precisely what makes this strategy so effective.
Here is how it works: The students in a history class, for example, are divided
into small groups of five or six students each. Suppose their task is to learn
about World War II. In one jigsaw group let us say that Sara is responsible for
researching Hitler's rise to power in pre~war Germany. Another member of the
group, Steven, is assigned to cover concentration camps; Pedro is assigned
Britain's role in the war; Melody is to research the contribution of the Soviet
Union; Willie will handle Japan's entry into the war; Clara will read about the
development of the atom bomb.
Eventually each student will come back to his or her jigsaw group and will try
to present a vivid, interesting, well-organized report to the group. The situation
is specifically structured so that the only access any member has to the other
five assignments is by listening intently to the report of the person reciting.
Thus, suppose Steven does not like Pedro and thinks Sara is a nerd. If he
heckles them, or tunes out while they are reporting, he cannot possibly do well
on the exam that follows.
To increase the probability that each report will be factual and accurate, the
students doing the research do not immediately take it back to their jigsaw
group. After doing their research, they must first meet with the other students
(one from each of the jigsaw groups)who had the identical assignment. For
example, those students assigned to the atom bomb topic will meet together
to work as a team of specialists, gathering information, discussing ideas,
becoming experts on their topic, and rehearsing their presentations. We call
this the "expert" group. It is particularly useful for those students who might
have initial difficulty learning or organizing their part of the assignment, for it
allows them to benefit from paying attention to and rehearsing with other
"experts," to pick up strategies of presentation, and generally to bring them~
selves up to speed.
After this meeting, when each presenter is up to speed, the jigsaw groups
reconvene in their initial heterogeneous configuration. The atom bomb expert
in each group teaches the other group members what she has learned about
the development of the atom bomb. Each student in each group educates the
whole group about his or her specialty. Students are then tested on what they
have learned from their fellow group members about World War II.
216 Elliot Aronson

What is the benefit of the jigsaw classroom? First and foremost, it is a

remarkably efficient way to learn the material. Our research shows that elem~
entary school students learn the material faster and perform significantly better
on objective exams than a control condition of students learning the same
material in more traditional classrooms (Aronson, Stephan, Sikes, Blaney, &
Snapp, 1978; Aronson & Patnoe, 1997; Lucker, Rosenfield, Sikes, & Aronson,
1977). But even more important, in terms of the present discussion, the jigsaw
process encourages listening, engagement, and empathy by giving each
member of the group an essential part to play in the academic activity. Group
members must work together as a team to accomplish a common goal; each
person depends on all the others. No student can achieve his or her individual
goal (learning the material, getting a good grade) unless everyone works
together well as a team. Group goals and individual goals complement and
bolster each other. This "cooperation by design" facilitates interaction among
all students in the class, leading them to come to value each other as contribu~
tors to their common task (Aronson, 2000; Aronson & Patnoe, 1997).
My graduate students and I invented the jigsaw strategy in Austin, Texas, in
1971. We invented jigsaw as a matter of absolute necessity to help defuse a
highly explosive situation. The city's schools had recently been desegregated
and, because Austin had always been residentially segregated, white young~
sters, African~American youngsters, and Hispanic youngsters found themselves
in the same classrooms for the first time in their lives. Within a few weeks, long~
standing suspicion, fear, distrust, and antipathy between groups produced an
atmosphere of turmoil and hostility, exploding into interethnic fistfights in
corridors and schoolyards across the city. The school superintendent called
me in to see if my students and I could possibly do something to help students
learn to get along with one another. After observing what was going on in
classrooms for a few days, we concluded that intergroup hostility was being
exacerbated by the competitive environment of the classroom.
In every single classroom, the students worked individually and competed
against each other for grades. Here is a description of a typical fifth grade
classroom we observed:
The teacher stands in front of the class, asks a question, and waits for the children to
indicate that they know the answer. Most frequently, six to ten youngsters raise their
hands. But they do not simply raise their hands; they lift themselves a few inches off
their chairs and stretch their arms as high as they can in an attempt to attract the
teacher's attention. To say they are eager to be called on is an incredible understate-
ment. Several other students sit quietly with their eyes averted, as if trying to make
themselves invisible. These are the ones who don't know the answer. Understandably,
they are trying to avoid eye contact with the teacher because they do not want to be
called on.
When the teacher calls on one of the eagerstudents, there are looks of disappointment,
dismay, and unhappinesson the facesof the other students who were avidlyraisingtheir
hands but werenot calledon. If the selectedstudent comes up with the right answer,the
teacher smiles, nods approvingly,and goeson to the next question. This is a great reward
10. Building Empathy, Compassion, and Achievement in the JigsawClassroom 217

for the child who happens to be called on. At the same time that the fortunate student is
coming up with the right answer and being smiled upon by the teacher, an audible groan
can be heard coming from the children who were striving to be called on but were
ignored. It is obvious they are disappointed because they missed an opportunity to
show the teacher how smart and quick they are. Perhaps they will get an opportunity next
time. In the meantime, the students who didn't know the answer breathe a sigh of relief.
They have escaped being humiliated this time.

On interviewing several of the teachers we learned that virtually all of them

started the school year with a determination to treat every student equally and
encourage all to do their best, but the students quickly sorted themselves into
different groups. The "winners" were the bright, eager, highly competitive stu~
dents who fervently raised their hands, participated in discussions, and did well
on tests. Understandably, the teacher felt gratified that these students re~
sponded to her teaching. She praised and encouraged them, continued to
call on them, and depended on them to keep the class going at a high level
and at a reasonable pace.
Then there were the "losers." At the beginning, the teacher called on them
occasionally, but they almost invariably did not know the answer, or were too
shy to speak, or could not speak English well. They seemed embarrassed to be in
the spotlight; some of the other students made snide comments--sometimes
under their breath, occasionally out loud. Because the schools in the poorer
section of town were substandard, the African~American and Mexican~Ameri~
can youngsters had received a poorer education prior to desegregation. Conse~
quently, in Austin, it was frequently these students who were among the losers.
This tended unfairly to confirm the unflattering stereotypes that the white kids
had about minorities. They considered them stupid or lazy. The minority stu~
dents also had preconceived notions about white kids: they were pushy show~
offs and teacher's pets. These stereotypes were also confirmed by the way most
of the white students behaved in the competitive classroom.
After a while, the typical classroom teacher became discouraged in trying to
engage the students who were not doing well. She also felt it was kinder not to
call on them and expose them to ridicule by the other students. In effect, she
made a silent pact with the "losers"; she would leave them alone as long as
they were not disruptive. Without really meaning to, she gave up on these
students, and so did the rest of the class. Without really meaning to, the
teacher contributed to the difficulty the students were experiencing. After a
while, these students tended to give up on themselves as well--perhaps
believing that they were stupid, because they sure weren't getting it.
It required only a few days of intensive observation and interviews for us to
have a pretty good idea of what was going on in these classrooms. We realized
that we needed to do something drastic to shift the emphasis from a relent~
lessly competitive atmosphere to a more cooperative one. It was in this context
that we invented the jigsaw strategy. One of our first interventions was with fifth
graders. First we helped several fifth grade teachers devise a cooperative jigsaw
2 18 Elliot Aronson

structure for the students to learn about the life of Eleanor Roosevelt. We
divided the students into small groups, diversified in terms of race, ethnicity,
and gender, making each student responsible for a certain portion of Roose-
velt's biography. Needless to say, at least one or two of the students in each
group were already viewed as "losers" by their classmates.
Carlos was one such student. Carlos was very shy and feeling insecure in his
new surroundings. English was his second language. He spoke it quite well, but
with a slight accent. Try to imagine his experience: After attending an inad-
equately funded, substandard neighborhood school consisting entirely of
Hispanic students, like himself, through the fourth grade, he was suddenly
bussed across town to the middle class area of the city and catapulted into a
class with Anglo students who spoke English fluently, seemed to know much
more than he did about all the subjects taught in the school, and were not
reluctant to let him know it.
When we restructured the classroom so that students were now working
together in small groups, initially, this was terrifying to Carlos. For now, he
should no longer slink down in his chair and hide in the back of the room. The
jigsaw structure made it necessary for him to speak up when it was his turn to
recite. Although he had gained a little added confidence by rehearsing together
with others who were also studying Eleanor Roosevelt's work with the United
Nations, he was understandably reticent to speak when it was his turn to teach
the students in his jigsaw group. He blushed, stammered, and had difficulty
articulating the material that he had learned. Skilled in the ways of the com-
petitive classroom, the other students were quick to pounce on Carlos' weak-
ness and began to ridicule him.
One of my research assistants was observing that group and heard some
members of Carlos' group make comments such as, "Aw, you don't know it,
you' re dumb, you' re stupid. You don't know what you' re doing. You can't even
speak English." Instead of admonishing them to "be nice" or "try to cooper-
ate," she made one simple but powerful statement. It went something like this:
"Talking like that to Carlos might be fun for you to do, but it's not going to help
you learn anything about what Eleanor Roosevelt accomplished at the United
Nations--and the exam will be given in about 15 minutes." What my assistant
was doing was reminding the students that the situation had changed. The
same behavior that might have been useful to them individually in the past,
when they were competing against each other, was now going to cost them
something very important: the chance to do well on the upcoming exam.
Needless to say, old, dysfunctional habits do not die easily. But they do die.
Within a few days of working with jigsaw, Carlos' groupmates gradually realized
that they needed to change their tactics. It was no longer in their own best
interest to rattle Carlos; he was not the enemy, he was on their team. They
needed him to perform well to do well themselves. Instead of taunting him and
putting him down, they started to gently ask him questions. But how? What kind
of questions? In effect, they had to put themselves in Carlos' shoes to find a way
10. Building Empathy, Compassion, and Achievement in the JigsawClassroom 2]9

to ask questions that did not threaten him but could facilitate his reciting in a
clear and understandable manner. After a week or two, most of Carlos's group~
mates developed into skillful interviewers, asking him relevant questions to
elicit the vital information from him. They became more patient, figured out the
most effective way to work with him, helped him out, and encouraged him. The
more they encouraged Carlos, the more he was able to relax; the more he was
able to relax, the quicker and more articulate he became. Carlos's groupmates
began to see him in a new light. He became transformed in their minds from
a "know-nothing loser who can't even speak English" to someone they could
work with, someone they could appreciate, maybe even someone they could like.
Moreover, Carlos began to see himself in a new light, as a competent, contrib~
uting member of the class who could work with others from different ethnic
groups. His self~esteem grew and as it grew, his performance improved even
more; and as his performance continued to improve, his groupmates continued
to view him in a more and more favorable light.
Within a few weeks, the success of the jigsaw was obvious to the classroom
teachers. They spontaneously told us of their great satisfaction with the way
the atmosphere of their classrooms had been transformed. Adjunct visitors
(such as music teachers and the like) were little short of amazed at the
dramatically changed atmosphere in the classrooms. Needless to say, this
was exciting to my graduate students and me. But, as scientists, we were not
totally satisfied with these testimonials; we were seeking firmer, more objective
evidence--and we got it. Because we had randomly introduced the jigsaw
intervention into some classrooms and not others, we were able to compare
the progress of the jigsaw students with that of the students in traditional
classrooms in a precise, scientific manner. After only 8 weeks there were clear
differences, even though students spent only a small portion of their class time
in jigsaw groups. When tested objectively, jigsaw students expressed signifi~
cantly less prejudice and negative stereotyping, were more self~confident, and
reported that they liked school better then children in traditional classrooms.
Moreover, this self-report was bolstered by hard behavioral data: For example,
the children in jigsaw classes were absent less often than students in traditional
classrooms. Finally, as mentioned earlier, on objective exams, the students in
jigsaw classrooms performed significantly better than students learning the
same material in traditional classrooms. Close inspection of the data revealed
that the differences in objective exam performance were due primarily to
improvements in the scores of minority students; the Anglo students per-
formed equally well in jigsaw as in traditional classrooms (Lucker et al., 1977).

COOPERATION: JIGSAW AND BASKETBALL

You might have noticed a rough similarity between the kind of cooperation that
goes on in a jigsaw group and the kind of cooperation that is necessary for the
2 20 Elliot Aronson

smooth functioning of an athletic team. Take a basketball team, for example. If

the team is to be successful, each player must play his or her role in a coopera~
tive manner. If each player is hell bent on being the highest scorer on the team,
then each would shoot whenever the opportunity arose. In contrast, on a
cooperative team, the idea is to pass the ball crisply until one player manages
to break clear for a relatively easy shot. If I pass the ball to Sam, and Sam whips
a pass to Harry, and Harry passes to Tony who breaks free for an easy lay~up, I
am elated even though I did not receive credit for either a field goal or an assist.
This is true cooperation.
As a result of this cooperation, athletic teams frequently build a cohesive~
ness that extends to their relationship off the court. They become friends
because they have learned to count on one another. There is one difference
between the outcome of a typical jigsaw group and that of a typical high school
basketball team, however, and it is a crucial difference. In high school, athletes
tend to hang out with each other and frequently exclude nonathletes from their
circle of close friends. In short, the internal cohesiveness of an athletic team
often goes along with the exclusion of everyone else.
In the jigsaw classroom, we circumvented this problem by the simple device
of shuffling groups every 8 weeks. Once a group of students was functioning
well together, once the barriers had been broken down and the students
showed a great deal of liking and empathy for one another, we would re~form
the groupings. At first, the students would resist this re~forming of groups.
Picture the scene: Debbie, Carlos, Tim, Patty and Jacob have just gotten to
know and appreciate one another and they are doing incredibly good work as a
team. Why should they want to leave this warm, efficient and cozy group to join
a group of relative strangers?
Why, indeed? After spending a few weeks in the new group, the students
invariably discover that the new people are just about as interesting, friendly,
and wonderful as their former group. The new group is working well together
and new friendships form. Then the students move on to their third group, and
the same thing begins to happen. As they near the end of their time in the third
group, it begins to dawn on most students that it is not the case that they
happened to luck out and land in groups with four or five terrific people. Rather,
they realize that just about everyone they work with is a worthy human being; all
they need to do is pay attention to each person, to try to understand him or her
and good things will emerge. That is a lesson well worth learning.

LONG-TERM EFFECTS?

Jigsaw works. Moreover, jigsaw is compatible with other teaching methods.

If jigsaw is used for as little as one hour per day, it has been shown to have
positive effects. How permanent are the effects? If students participated in the
jigsaw classroom in the fifth or sixth grade, would the positive impact remain
10. Building Empathy, Compassion, and Achievement in the Jigsaw Classroom 2 21

even if they never experienced jigsaw again? Unfortunately, we do not have a

definitive answer to this question. We do have some tangential evidence that
the effects of jigsaw may become a permanent part of the individual's way of
looking at the world. In a clever experiment, Diane Bridgeman showed that the
empathy required by jigsaw takes on the form of a more or less permanent
ability that generalizes and is used outside the confines of the classroom.
In Bridgeman's experiment, she worked with fifth graders, half whom had
spent 2 months participating in jigsaw classes; the others had spent that time
in traditional classrooms. Bridgeman showed them a series of stick-figure
cartoons about a young boy their own age. In the first panel, the boy is looking
sad as he waves good~bye to his father at the airport. In the next panel, a letter
carrier delivers a package to the boy. In the final panel, the boy opens the
package and finds a toy airplane inside and bursts into tears. Bridgeman asked
the children why they thought the boy had burst into tears at the sight of the
airplane. Nearly all of the children could answer correctly, because the toy
airplane reminded him of how much he missed his father and that made him
sad. Then Bridgeman asked the crucial question: "What did the letter carrier
think when he saw the boy open the package and start to cry?"
Most children of this age make a consistent error; they assume that every~
one knows what they know. Thus, the youngsters in the control group thought
that the letter carrier would know the boy was sad because the gift reminded
him of his father leaving. But the children who had participated in the jigsaw
classroom responded differently. Because they were better able to take the
perspective of the letter carrier--to put themselves in his shoes; they realized
that he would be confused at seeing the boy cry over receiving a nice present
because the letter carrier had not witnessed the farewell scene at the airport.
Offhand, this might not seem very important. After all, who cares whether
kids have the ability to figure out what is in the letter carrier's mind? In point of
fact, we should all care--a great deal. The extent to which children can develop
the ability to see the world from the perspective of another human being has
profound implications for empathy, prejudice, aggression, and interpersonal
relations in general. When you can feel another person's pain, when you can
develop the ability to understand what that person is going through, it in~
creases the probability that your heart will open to that person, and it becomes
difficult to harm him or taunt him. Moreover, because Bridgeman's data suggest
that empathy, is a skill--not unlike riding a bike--that can be used in a variety
of situations, the implication is that the major impact of the jigsaw classroom
might have longqasting effects.
I do have some additional evidence but I am afraid that it is merely anec~
dotal. Nevertheless, I will mention it for what it may be worth. Over the past 20
years, I have received unsolicited letters from young men and young women
who, many years earlier, had undergone such a transformation. To give you
some of the flavor of this experience, I would like to share one such letter with
you.
222 Elliot Aronson

Dear Professor Aronson:

I am a senior at University. Today I got a letter admitting me to the Harvard Law

School. This may not seem odd to you but, let me tell you something. I am the 6th of 7
children my parents had--and I am the only one who ever went to college, let alone
graduate, or go to law school.

By now, you are probably wondering why this stranger is writing to you and bragging to
you about his achievements. Actually, I'm not a stranger although we never met. You
see, last year I was taking a course in social psychology and we were using a book you
wrote called The Social Animal, and when I read about prejudice and jigsaw it all
sounded very familiar--and then, I realized that ! was in that very first class you ever
did jigsaw in--when I was in the 5th grade in Austin. And as ! read on, it dawned on me
that I was the boy that you called Carlos. And then I remembered you when you first
came to our classroom and how I was scared and how I hated school and how I was so
stupid and didn't know anything. And you came in--it all came back to me when I read
your book--you were very tall--about 6 ~ feet--and you had a big black beard and you
were funny and made us all laugh.

And, most important, when we started to do work in jigsaw groups, I began to realize that
I wasn't really that stupid. And the kids I thought were cruel and hostile became my
friends and the teacher acted friendly and nice to me and I actually began to love school,
and I began to love to learn things and now I'm about to go to Harvard Law School.

You must get a lot of letters like this but I decided to write anyway because let me tell you
something. My mother tells me that when I was born I almost died. I was born at home
and the cord was wrapped around my neck and the midwife gave me mouth to mouth
and saved my life. If she was still alive, I would write to her too, to tell her that I grew up
smart and good and I'm going to law school. But she died a few years ago. I'm writing to
you because, no less than her, you saved my life too.

Sincerely,
XXXX XXX

As you m i g h t imagine, I was deeply t o u c h e d by this letter. It is just a b o u t the

m o s t m o v i n g letter I have ever received. But w h e n I read the signature I was
startled to discover t h a t it did n o t b e l o n g to the b o y t h a t I had in m i n d - - t h e
b o y w h o in my previous writings I had referred to as " C a r l o s . " The y o u n g m a n
w h o w r o t e me t h a t lovely letter was mistaken.
I have a clear m e m o r y of sitting there w i t h the letter in my h a n d t h i n k i n g
a b o u t t h a t y o u n g man and h o w w r o n g he was. A f t e r a few minutes, I fell into a
reverie in w h i c h I began to realize t h a t p e r h a p s t h a t y o u n g m a n was n o t
mistaken after all. That is, a l t h o u g h I had a specific fifth grader in m i n d w h e n
I w r o t e a b o u t Carlos, there are a great m a n y children w h o c o m e p r e t t y close to
fitting t h a t description. In my reverie I began to grasp the i m p l i c a t i o n s of the
possibility that, all over America, there are t h o u s a n d s of youngsters w h o t h i n k
they are Carlos. And, in the d e e p e s t possible way, t h e y are all Carlos. Carlos is
any child w h o has been the u n h a p p y recipient of p u t - d o w n s , taunting, and
rejection at the h a n d s of his or her peers, leading to a d i m i n u t i o n of his or her
self~esteem, and w h o has m a n a g e d to turn t h a t a r o u n d because the structure
of the classroom changed, paving the way for a different set of responses. To
i0. BuildingEmpathy,Compassion,andAchievementin the JigsawClassroom 2 23

the child involved, it feels like a miracle. To the social psychologist it is still
another vivid example of the power of the situation: how what looks like a
small, simple change in the structure of a social environment can have an
enormous impact on the experience of the people in that environment. This is
an experience that can last a lifetime. 1

A POSTSCRIPT

Following the Columbine massacre, there was a lot of negative publicity about
the atmosphere at Columbine High School and how that atmosphere might
have contributed to the tragedy. Some of it mentioned how athletes domin-
ated the school and how unpopular students were taunted and excluded. The
criticism, although not inaccurate, was unfair in the sense that the atmosphere
at Columbine was no different than in almost every high school in the country.
In response to the criticism, some of the Columbine students attempted to
justify their exclusion of Harris and Klebold. Typical of these remarks was
comment by a member of the Columbine football team:
Columbine is a good clean place except for those rejects. Most kids didn't want them
there. Theywere into witchcraft. Theywere into voodoo. Surewe teased them. But what
do you expectwith kids who come to school with weird hairdos and horns on their hats?
If you want to get rid of someone, usually you tease 'era. (Quoted by Gibbs & Roche,
December, 1999)

It is my belief that if the jigsaw technique had been widely used in the
Littleton school system a few years earlier, the young man and his friends might
have developed some additional compassion and empathy as well as a greater
tolerance for diversity. If so, they would undoubtedly have been delighted
rather than angered by the diversity represented by the kids "who come to
school with weird hairdos." I may be wrong, but I am fairly certain that, if this
had been the case, the Columbine massacre would never have occurred. Again,
I may be wrong; but it certainly is worth thinking about.

Teachers' Questions and Answers

Q: As a teacher, I often use the jigsaw method with my students and it
works extremely well. But, often there are some students, usually main~
streamed special education students, who have tremendous difficulty relay~
ing the content specifics to their groups even after prepping in their their
mastery group. How can the jigsaw be adapted to accommodate these stu~

1I should mention in passing that during the past two decades, educational researchers have
developed and tested several alternative cooperative classroom methods. Although these alterna~
tive methods differ from jigsawin severalminor respects, they have in common a basic cooperative
structure and produce similar positive effects. For a summary of these techniques, see Slavin,
Aharan, Kagan, Hertz-Lazarowitz,Webb, and Schmuck (1985).
2 24 Elliot Aronson

dents so that they can fully participate without their group missing essential
information?
A: ! have found that, in situations like this, it is helpful to assign another
student in the jigsaw group to the special~ed student, someone to serve as a
coach. This would need to be a very bright, mature student. In addition to
learning her own paragraph, she would learn enough about the special-ed
student's paragraph to ask probing questions or to fill in where the special-
ed student might have left something out. This is a special role that most
bright, mature students take to like a duck to water. (If no such student exists in
a particular jigsaw group, this is a role that the teacher might want to take on.)
(3: I have found that we do not always have enough time to complete all of
the parts of the jigsaw within one period. What are some ways you would
recommend to divide up the process over 2 days? Or is there a way to shorten
the process so as to fit it into one period?
A: Needless to say, it works best if it can all be squeezed into one period. If
not, I would do the learning of the material and the expert group in one period
and save the final jigsaw group for the next period.
Q: It's very clear to me that the jigsaw can be used effectively if the
curriculum covers, say, the life of Eleanor Roosevelt. But I'm having trouble
seeing how one could use the jigsaw method for teaching math. Is there a way
to do so? Does it work as well as it does for more humanities-style curricula?
A: You are absolutely right. It is easier to work with social studies than with
a progressive subject like math, where it is difficult for students to grasp step 3
without having gone through steps 1 and 2. At the same time, I am happy to
report that some math teachers have had very good success using jigsaw to
teach arithmetic, mathematics, and statistics. Just to take one stunning, recent
example, David Perkins and Renee Saris at Ball State University have used
jigsaw, to great effect, in an undergraduate statistics course (see Journal of
Teaching of Psychology, 2001, Vol. 28, pp. 111-113).

References
Aronson, E. (1978). Tile jigsaw classroom. Beverly Hills, CA: Sage.
Aronson, E. (1999). The social animal (Chap. 4). New York: Worth/Freeman.
Aronson, E. (2000). Nobody left to fiate: Teacfiing compassion after Columbine. New York: Worth/Freeman.
Aronson, E., & Bridgeman, D. (1979). Jigsaw groups and the desegregated classroom: In pursuit of
common goals. Personality and Social Psychology Bulletin, 5, 438-446.
Aronson, E., & Patnoe, S., (1997). Cooperation in tile classroom: The jigsaw method. New York: Longman.
Bridgeman, D. (1981). Enhanced role~taking through cooperative interdependence: A field study.
Child Development 52, 123 I-1238.
Gibbs, N. (1999, October 25). A week in the life of a high school, Webster Groves. Time Magazine.
Gibbs, N., & Roche, C. (1999, October 25). Columbine. Time Magazine.
Gibbs, N., & Roche, T. (I 999, December 20). The Columbine tapes. Time Magazine.
10. Building Empathy, Compassion, and Achievement in the Jigsaw Classroom 22 5

Goldberg, W., & Connelly, M. (1999, October 20). New York Times.
Lewin, T. (1999, May 2). Terror in Littleton: The teenage culture; Arizona high school provides
glimpse into cliques. New York Times.
Lucker, W., Rosenfield, D., Sikes, J., & Aronson, E. (1977). Performance in the interdependent
classroom: A field study. American Educational Research Journal, 13, 115-123.
Sizer, T., & Sizer, N. (1999). The students are watching. Boston: Beacon Press.
Slavin, R., Aharan, S. Kagan, S., Hertz~Lazarowitz, R., Webb, C., & Schmuck, R. (Eds.) (1985).
Learning to cooperate, cooperating to learn: New York: Plenum.
Townsend, P. (1999, May 23). High school cliques. Santa Cruz Sentinel, p. A9.
 CHAPTER

II
Intelligence Is Not Just Inside
the Head: The Theory of
Successful Intelligence
ROBERT J. STERNBERG
Yale University, New Haven, Connecticut

Many people believe that intelligence is just inside the head. According to
this view intelligence can be defined in terms of a fairly narrow set of cognitive
skills of the kinds measured by conventional intelligence tests--answering
vocabulary questions, solving mathematical reasoning problems, manipulating
images of geometric objects in the head, and so forth. In this chapter ! argue that
the notion of intelligence as inside the head and as operationalized in narrowly
based intelligence tests is incomplete. I argue for a concept of successful
intelligence, according to which intelligence is the ability to achieve success in
life, given one's personal standards, within one's sociocultural context. One's
ability to achieve success depends on one's capitalizing on one's strengths and
correcting or compensating for one's weaknesses through a balance of analyt-
ical, creative, and practical abilities in order to adapt to, shape, and select

Preparation of this article was supported under the Javits Act Program (Grant No. R206R00001 ) as
administered by the Office of Educational Research and Improvement, U.S. Department of Educa-
tion. Grantees undertaking such projects are encouraged to express freely their professional
judgment. This article, therefore, does not necessarily represent the position or policies of the
Office of Educational Research and Improvement or the U.S. Department of Education, and no
official endorsement should be inferred.
Requests for reprints should be sent to Robert J. Sternberg, Department of Psychology, Yale
University, P.O. Box 208205, New Haven, CT 06520-8205.

ImprovingAcademicAchievement
Copyright2002,ElsevierScience(USA).All rightsreserved. 227
228 RobertJ.Sternberg

environments (Sternberg, 1997, 1999). If we adopt this notion of intelligence, I

argue, we immediately will have tools available to improve children's perform~
ance in school and to enhance their achievements in life. This view requires us to
view intelligence not just from a psychometric perspective or even a cognitive
perspective, but from a social perspective as well.
I divide my argument into three main parts. First I argue that conventional
notions of intelligence are incomplete. Second ! suggest an alternative notion
of successful intelligence that expands on conventional notions of intelligence.
Finally I draw some conclusions about the nature of intelligence and its
implications for schooling.

INADEQUACY OF CONVENTIONAL NOTIONS

OF INTELLIGENCE

Conventional notions of intelligence are incomplete. To the extent that one

wishes to define intelligence very narrowly, perhaps they work. But in the
modern world, where so many skills are needed to thrive and even to survive,
perhaps narrow notions of intelligence are no longer adequate. Moreover, they
are based on a doubtful premise. This premise is of intelligence as a unitary
construct, a view that dates back to a proposal by British psychologist Charles
Spearman (1904) and a view that is most likely not entirely correct. So contem-
porary theories of intelligence based on this notion, such as those proposed by
psychologists Arthur Jensen (1998) and John B. Carroll (1993), cannot be
entirely correct either.
There now has accumulated a substantial body of evidence suggesting that,
contrary to conventional notions, intelligence is not a unitary construct. This
evidence is of a variety of different kinds. One of the main kinds of evidence
traditionally adduced to support the unitary notion is the pattern of all positive
correlations that is frequently observed among ability tests. The assumption
has been that this pattern of all positive correlations among tests reflects an
underlying common ability being measured by all the tests. This ability is often
referred to as g, or general ability. But I believe there is good evidence to
suggest that the general factor is not a function of some inherent structure of
intellect. Rather, it reflects limitations in the kinds of individuals tested, the
kinds of tests used in the testing, and the situations in which the individuals are
tested.
One kind of evidence suggests the power of situational contexts in testing.
For example, David Carraher, Terezinha Nufies, and their colleagues have
studied a group of children that is especially relevant for assessing intelligence
as adaptation to the environment (see Nufies, 1994). The group comprised
Brazilian street children. Brazilian street children are under great contextual
pressure to form a successful street business. If they do not, they risk death at
the hands of so~called "death squads," which may murder children who,
11. Intelligence Is Not Just Inside the Head 229

unable to earn money, resort to robbing stores (or who are suspected of
resorting to robbing stores). The researchers found that the same children
who are able to do the mathematics needed to run their street business
areoften little able or unable to do school mathematics. In fact, the more
abstract and removed from real~world contexts the problems are in their form
of presentation, the worse the children do on the problems. These results
suggest that differences in social context can have a powerful effect on
performance.
Such differences are not limited to Brazilian street children. Jean Lave (1997)
showed that Berkeley housewives who successfully could do the mathematics
needed for comparison shopping in the supermarket were unable to do the
same mathematics when they were placed in a classroom and given compar~
able problems presented in an abstract form. In other words, their problem was
not at the level of mental processes but at the level of applying the processes in
specific environmental contexts.
Steve Ceci and Steve Liker (1986) showed that, given tasks relevant to their
lives, men would show the patterns of effects similar to those shown by women
in the Lave studies. These investigators studied men who successfully handi~
capped horse races. The complexity of their implicit mathematical formulas
was unrelated to their IQ. Moreover, despite the complexity of these formulas,
the mean IQ among these men was only at roughly the population average or
slightly below. Ceci and Liker also subsequently found that the skills were really
quite specific: The same men did not successfully apply their skills to compu-
tations involving securities in the stock market. Again, context matters.
In our own research, we have found results consistent with those described
above. These results have emanated from studies we have done around the
world.
For example, in a study in Usenge, Kenya, near the town of Kisumu, we
were interested in school-aged children's ability to adapt to their indigenous
environment. In collaboration with others (Sternberg et al., 2001), I was
involved in studies of practical intelligence for adaptation to the environ-
ment. We devised a test that measured children's informal tacit knowledge for
natural herbal medicines that the villagers believe can be used to fight various
types of illnesses. We do not know if all or any of these medicines are actually
effective. But from the standpoint of our study, the important thing is that
the villagers think they are and therefore that knowledge about them is worth
possessing.
We measured the children's ability to identify the medicines and what they
are used for. Based on work we had done elsewhere, we expected that scores
on this test would not correlate with scores on conventional tests of intelli~
gence. To test this hypothesis, we also administered to the children the Raven
Coloured Progressive Matrices Test, which is a measure of fluid or abstract~
reasoning~based abilities, as well as the Mill Hill Vocabulary Scale, which is a
measure of crystallized or formal~knowledge~based abilities. In addition, we
230 Robert J. Sternberg

gave the children a comparable test of vocabulary in their own Dholuo lan-
guage. The Dholuo language is spoken in the home, English in the schools.
We did indeed find no correlation between the test of indigenous tacit
knowledge and scores on the fluid-ability tests. But to our surprise, we found
statistically significant correlations of the tacit-knowledge tests with the tests of
crystallized abilities. The correlations, however, were negative. In other words,
the higher the children scored on the test of tacit knowledge, the lower they
scored, on average, on the tests of crystallized abilities. This surprising result
can be interpreted in various ways. But based on the ethnographic observations
of the cultural anthropologists on our team, Geissler and Prince, we concluded
that a plausible scenario takes into account the expectations of families for
their children and the resultant ways in which the children are socialized.
Most families in the village do not particularly value formal Western school-
ing. There is no reason they should, as their children will for the most part
spend their lives farming or engaged in other occupations that make little or no
use of Western schooling. These families emphasize teaching their children the
indigenous informal knowledge that will lead to successful adaptation in the
environments in which they will live. At the same time, there are some families
in the village that have different expectations for their children. They hope that
their children eventually may be able to leave the village and to go to a
university, perhaps the University of Nairobi. These families tend to emphasize
the value of Western education and to devalue indigenous informal knowledge.
Thus the families typically value and emphasize one or the other kind of
knowledge but not both.
The Kenya study suggests that the identification of a general factor of
human intelligence may tell us more about patterns of schooling and espe-
cially Western patterns of schooling than it does about the structure of human
abilities. In Western schooling, children typically study a variety of subject
matters from an early age and thus develop skills in a variety of skill areas. This
kind of schooling prepares the children to take a test of intelligence, which
typically measures skills in a variety of areas. Often intelligence tests measure
skills that children were expected to acquire a few years before taking the
intelligence test. But as Barbara Rogoff (1990) and others have noted, this
pattern of schooling is not universal and has not even been common for much
of the history of humankind. Throughout history and in many places still,
schooling, especially for boys, takes the form of apprenticeships in which
children learn a craft from an early age. They learn what they will need to know
to succeed in a trade, but not a lot more. They are not simultaneously engaged
in tasks that require the development of the particular blend of skills meas-
ured by conventional intelligence tests. Hence it is less likely that one would
observe a general factor in their scores, much as we discovered in Kenya. The
developing world provides a particularly interesting laboratory for testing
theories of intelligence because many of the assumptions that are held as
dear in the developed world simply do not apply. In particular, children receive
11. Intelligence Is Not Just Inside the Head 231

very different socialization from the socialization they receive in the United
States.
Even in the United States, however, children can receive very diverse kinds
of socialization. Lynn Okagaki and I did a study in San Jose, California, of
parents' and teachers' conceptions of what it means to have an intelligent
child (Okagaki & Sternberg, 1991, 1993). We found that parents of children of
different ethnic groups have somewhat different conceptions (sometimes
called implicit theories) of what it means to have a smart child. In particular,
Latino parents more emphasized the importance of social~competence skills,
whereas Anglo and Asian parents more emphasized the importance of cogni~
tive~competence skills. Teachers, however, had conceptions of smart children
that were more similar to the conceptions of the Anglo and Asian parents than
they were to the conceptions of the Latino parents. We found that the more the
parents' conceptions of intelligence matched the conceptions of the teachers,
the better the children performed in school. In other words, teachers are not
evaluating children just on some "true" basis of what it means to be intelligent,
but rather on their own conception of what it means to be intelligent. And of
course, the teachers' conceptions of intelligence are likely to reflect the way
they were socialized.
Another kind of socialization effect stems from familiarity with testing in the
first place, a point that Patricia Greenfield (1997) has very much emphasized in
her writings. We agree with Greenfield. A study done in Tanzania in which I
collaborated with Elena Grigorenko points out the risks of giving tests, scoring
them, and interpreting the results as measures of some fixed underlying intel~
lectual ability or abilities (Sternberg, Grigorenko, Ngorosho, Tuntufye, Mbise,
Nokes, Jukes, & Bundy, in press). We administered to young school children in
Bagamoyo, Tanzania, tests such as a form~board test and a Twenty Questions
Test, which measure the kinds of skills required on conventional tests of intelli~
gence. Of course, we obtained scores that we could analyze and evaluate,
ranking the children in terms of their supposed general or other abilities.
However, we administered the tests dynamically rather than statically, loosely
following a procedure first set out by Russian psychologist Lev Vygotsky (1978)
and later carried out as well by Israeli psychologist Reuven Feuerstein (1979).
Dynamic testing is like conventional static testing in that individuals are tested
and inferences about their abilities are then made. But dynamic tests differ from
conventional static tests in that children are given some kind of feedback to
help them improve their scores. Vygotsky suggested that the children's ability
to profit from the guided instruction the children received during the testing
session could serve as a measure of children's zone of proximal development
(ZPD), or the difference between their developed abilities and their latent or
underlying capacities. In other words, testing and instruction are treated as
being of one piece rather than as being distinct processes.
In our assessments, children were first given the ability tests. Then they were
given a brief period of instruction in which they were able to learn skills that
232 Robert J. Sternberg

would potentially enable them to improve their scores. Then they were tested
again. Because the instruction for each test lasted only about 15 minutes, one
would not expect dramatic gains. Yet, on average, the gains were statistically
significant. More importantly, scores on the pretest showed only weak al~
though significant correlations with scores on the posttest. These correlations,
at about the .3 level, suggested that when tests are administered statically to
children in developing countries, they may be rather unstable and easily
subject to influences of training. The reason, of course, is that the children
are not accustomed to taking Western~style tests and so profit quickly even
from small amounts of instruction as to what is expected from them. Of course,
the more important question is not whether the scores changed or even
correlated with each other, but rather how they correlated with other cognitive
measures. In other words, which test was a better predictor of transfer to other
cognitive performance, the pretest score or the gain from the pretest score to
the posttest score? We found the gain score to be the better predictor, by a
factor of 4. In other words, any general-factor score or, really, any other factor
score obtained from the pretest, which was equivalent to a typical statically
administered test, would be of substantially lower validity than would be a gain
score measuring learning at the time of test as obtained from a dynamically
administered test.
If intelligence is not just a single thing that can be measured by a conven~
tional static test of intelligence, what is it? I argue that it comprises three
things, each of which is a different aspect: of intelligence.

THREE ASPECTS OF SUCCESSFUL INTELLIGENCE

The intelligence one needs to attain success in life comprises analytical,

creative, and practical aspects (Sternberg, 1985). According to my theory of
successful intelligence, a common set of processes underlies these three
aspects of intelligence. Metacomponents,or executive processes, plan what to
do, monitor things as they are being done, and evaluate things after they are
done. Examples of metacomponents are recognizing the existence of a prob~
lem, defining the nature of the problem, deciding on a strategy for solving the
problem, monitoring the solution of the problem, and evaluating the solution
after the problem is solved. In writing a term paper on patterns of child rearing,
for example, a student has to recognize the need to write the paper, define a
topic on which to write the paper, decide on a strategy for getting the paper
done, monitor how the paper is going while he or she is writing it, and then
evaluate it to make sure it is ready to be handed in. Performance components
execute the instructions of the metacomponents. For example, inference is
used to compare and contrast different approaches, say, to child rearing.
Knowledge~acquisition components are used to learn how to solve problems
or simply to acquire knowledge in the first place. For example, selective
11. Intelligence Is Not Just Inside the Head 233

encoding is used to decide what information one is reading is relevant in the

context of one's writing the term paper on which one is working
Although the same processes are used for all three aspects of intelligence,
these processes are applied to different kinds of tasks and situations depending
on whether a given problem requires analytical thinking, creative thinking,
practical thinking, or a combination of these kinds of thinking.

Analytical Intelligence
Analytical intelligence is involved when the components of intelligence are
applied to analyze, evaluate, judge, or compare and contrast. It typically is
involved when components are applied to relatively familiar kinds of problems
where the judgments to be made are of an abstract nature.
In some of my early work, I showed how analytical kinds of problems, such
as analogies and syllogisms, can be analyzed in terms of the core component
processes underlying performance on these problems. In this research, I gave
people problems such as testlike analogies or series problems, and collected
their response times or error rates so that I could determine the underlying
information~processing components people used when they solved these
problems. The goal of this research was to understand the information~pro~
cessing origins of individual differences in (the analytical aspect of) human
intelligence.
Research on the components of human intelligence yielded some interest-
ing results. For example, in a study of the development of figural analogical
reasoning in second, fourth, and sixth graders, Bathsheva Rifkin and I found
that although children generally became quicker in information processing with
age, not all components were executed more rapidly with age (Sternberg &
Rifkin, 1979). The encoding component, representing the time it takes to figure
out what is in a stimulus, first showed a decrease in component time with age
and then an increase. Apparently, older children realized that their best strat~
egy was to spend more time in encoding the terms of a problem so that they
would later be able to spend less time in operating on these encodings. A
related finding was that better reasoners tend to spend relatively more time
than do poorer reasoners in global, up-front metacomponential planning,
when they solve difficult reasoning problems. Poorer reasoners, on the other
hand, tend to spend relatively more time in local planning. Presumably, the
better reasoners recognize that it is better to invest more time up front so as to
be able to process a problem more efficiently later on.
The study with Rifkin also yielded another strange and totally unexpected
finding. A substantial number of second graders received scores of zero on the
analogical reasoning test. In other words, they got absolutely no problems
correct. This was a puzzling finding indeed. Because it was hard to believe that
the children could be so lacking in analogical reasoning skills, we decided to
look closely at their data. We discovered that the children who got no problems
234 RobertJ.Sternberg

correct, instead of circling one of the two answer options printed at the right of
each page, had circled one of the two givens of the problem, printed at the left
of each page. What were they doing?
It turned out that the children, who were students in a Hebrew day school,
were accustomed to reading English in the morning and Hebrew in the after~
noon. English is read left to right, but Hebrew is read right to left. Unfortunately
for us, we had tested in the afternoon. So the children did what they were
accustomed to doing in the afternoon, that is, they read from right to left. Note
how even in what appears to be a wholly cognitive test, socialization matters.
The children were responding in a way that made sense in terms of their
socialization in the school.
Some of our studies concentrated on knowledge~acquisition components
rather than performance components or metacomponents. For example, in
one set of studies, Janet Powell and I (Sternberg, 1987b; Sternberg & Powell,
1983) were interested in sources of individual differences in vocabulary. We
were not content just to write these off as individual differences in knowledge,
because we wanted to understand why it was that some people acquired this
knowledge and others did not. What we found is that there were multiple
sources of individual and developmental differences. The three main sources
were in knowledge~acquisition components, use of context clues, and use of
mediating variables. For example, in the sentence, "The blen rises in the east
and sets in the west," the knowledge-acquisition component of selective
comparison is used to relate prior knowledge about a known concept, the
sun, to the unknown word (neologism) in the sentence, "blen." Several context
cues appear in the sentence, such as the fact that a blen rises, the fact that it
sets, and the information about where it rises and sets. A mediating variable is
that the information can occur after the presentation of the unknown word.
We did research such as that described above because we believed that
conventional psychometric research sometimes incorrectly attributed individ~
ual and developmental differences. For example, a verbal analogies test that
might appear on its surface to measure verbal reasoning might in fact measure
primarily vocabulary and general information. The analogy MITIGATE: EXACER~
BATE :: AMELIORATE :(a. worsen, b. improve, c. extend, d. contract), for
example, is more likely to measure indMdual differences in vocabulary than
individual differences in reasoning, at least for most populations. In fact, in
some populations, reasoning might hardly be a source of individual or devel-
opmental differences at all. And if we then look at the sources of the individual
differences in vocabulary, we would need to understand that the differences in
knowledge did not come from nowhere: Some children had much more fre~
quent and better opportunities to learn word meanings than did others. For
example, they may have grown up in a house with more books or more
opportunities to be exposed to print in the English language. Note, therefore,
how socialization matters even when one tests something as cognitive as
vocabulary.
11. Intelligence Is Not Just Inside the Head 23 5

The kinds of analytical skills we studied in this research can be taught. For
example, in one study, we tested whether it is possible to teach people better to
figure out meanings of unknown words presented in context (Sternberg, 1987a).
In one study, we gave participants in five conditions a pretest on their ability to
figure out word meanings. Then the participants were divided into five condi~
tions, two of which were control conditions that lacked formal instruction. In
one condition, participants were not given any instructional treatment. They
were merely asked later to take a posttest. In a second condition, they were
given practice as an instructional condition, but there was no formal instruc~
tion, per se. In a third condition, they were taught knowledge~acquisition
component processes that could be used to figure out word meanings. In a
fourth condition, they were taught to use context cues. In a fifth condition, they
were taught to use mediating variables. Participants in all three of the theory~
based formal~instructional conditions outperformed participants in the two
control conditions, whose performance did not differ. In other words, theory~
based instruction was better than no instruction and better than practice
without formal instruction.
Research on the componential bases of intelligence was useful in under~
standing individual differences in performance on conventional tests of intelli-
gence. But it became increasingly clear to me that this research basically served
to partition the variation on conventional tests in a different way, rather than
serving to uncover previously untapped sources of variation. Children develop
intellectually in ways beyond just what conventional psychometric intelligence
tests or even Piagetian tests based on the theory of Jean Piaget (1972) measure.
So what might be some of these other sources of variation? Creative intelli~
gence seemed to be one such source of variation, a source that is almost
wholly untapped by conventional tests.

Creative Intelligence
Intelligence tests contain a range of problems, some of them more novel than
others. In some of our work we have shown that when one goes beyond the
range of unconventionality of the tests, one starts to tap sources of individual
differences measured little or not at all by the tests. According to the theory of
successful intelligence, (creative) intelligence is particularly well measured by
problems assessing how well an individual can cope with relative novelty. Thus
it is important to include in a battery of tests problems that are relatively novel.
These problems can be either convergent or divergent in nature.
In work with convergent problems, some of it done in collaboration with
Sheldon Tetewsky (Sternberg, 1982; Tetewsky & Sternberg, 1986), we pre~
sented individuals with novel kinds of reasoning problems that had a single
best answer. For example, they might be told that some objects are green and
others blue; but still other objects might be grue, meaning green until the year
2000 and blue thereafter, or bleen, meaning blue until the year 2000 and green
236 Robert J. Sternberg

thereafter. Or they might be told of four kinds of people on the planet Kyron:
blens, who are born young and die young; kwefs, who are born old and die old;
halts, who are born young and die old; and prosses, who are born old and die
young. Their task was to predict future states from past states, given incom~
plete information. In another set of studies with Joyce Gastel (Sternberg &
Gastel, 1989a, 1989b), people were given more conventional kinds of inductive
reasoning problems, such as analogies, series completions, and classifications,
but were told to solve them. But the problems had premises preceding them
that were either conventional (dancers wear shoes) or novel (dancers eat
shoes). The participants had to solve the problems as though the counter~
factuals were true.
In these studies, we found that correlations with conventional kinds of tests
depended on how novel or nonentrenched the conventional tests were. The
more novel the items, the higher the correlations of our tests with scores on the
conventional tests. We also found that when response times on the relatively
novel problems were componentially analyzed, some components better
measured the creative aspect of intelligence than did others. For example, in
the "grue-bleen" task mentioned above, the information~processing compon~
ent requiring people to switch from conventional green-blue thinking to
grue-bleen thinking and then back to green-blue thinking again was a particu~
larly good measure of the ability to cope with novelty.
In collaborative work with Todd Lubart (Sternberg & Lubart, 1995) using
divergent reasoning problems having no one best answer, we asked people to
create various kinds of products where an infinite variety of responses were
possible. Individuals were asked to create products in the realms of writing, art,
advertising, and science. In writing, they would be asked to write very short
stories for which we would give them a choice of titles, such as "Beyond the
Edge" and "The Octopus' Sneakers." In art, they were asked to produce art
compositions with titles such as "The Beginning of Time" and "Earth from an
Insect's Point of View." In advertising, they were asked to produce advertiser
ments for products such as a brand of bow tie and a brand of doorknob. In
science, they were asked to solve problems such as one asking them how
people might detect extraterrestrial aliens among us who are seeking to escape
detection. Participants created two products in each domain.
We found that creativity is relatively although not wholly domain~specific.
Correlations of ratings of the creative quality of the products across domains
were lower than correlations of ratings and generally were at about the .4 level.
Thus, there was some degree of relation across domains, at the same time that
there was plenty of room for someone to be strong in one or more domains but
not in others. More importantly, perhaps, we found, as we had for the conver~
gent problems, a range of correlations with conventional tests of abilities. As
was the case for the correlations obtained with convergent problems, correl~
ations were higher to the extent that problems on the conventional tests were
nonentrenched. For example, correlations were higher with fluid than with
11. Intelligence Is Not Just Inside the Head 237

crystallized ability tests, and correlations were higher, the more novel the fluid
test was. Even the highest correlations, however, were only at the .5 level,
suggesting that tests of creative intelligence tap skills beyond those measured
even by relatively novel kinds of items on conventional tests of intelligence.
Creative~thinking skills can be taught, and in a collaborative endeavor with
Wendy Williams, we have devised a program for teaching them. Insight skills, a
form of creative~thinking skill, can also be taught. In collaborative work with
Janet Davidson, we divided gifted and nongifted fourth grade children into
experimental and control groups. All children took pretests on insightful
thinking. Then some of the children received their regular school instruction
whereas others received instruction on insight skills. After the instruction of
whichever kind, all children took a posttest on insight skills. We found that
children taught how to solve the insight problems using knowledge~acquisition
components gained more from pretest to posttest than did students who were
not so taught.
Tests of creative intelligence go beyond tests of analytical intelligence in
measuring performance on tasks that require individuals to deal with relatively
novel situations. But how about situations that are relatively familiar, but in a
practical rather than an academic domain? Can we measure intelligence in the
practical domain, and if so, what is the relation of practical intelligence to
intelligence in more academic kinds of domains?

Practical Intelligence
Practical intelligence involves individuals' applying their abilities to the kinds of
problems that confront them in daily life, such as on the job or in the home
(Sternberg, Forsythe, Hedlund, Horvath, Snook, Williams, Wagner, & Grigorenko,
2000). Practical intelligence involves applying the components of intelligence to
experience so as to (a) adapt to, (b) shape, and (c) select environments. Adapta~
tion is involved when one changes oneself to suit the environment. Shaping is
involved when one changes the environment to suit oneself. And selection
is involved when one decides to seek out another environment that is a better
match to one's needs, abilities, and desires. People differ in their balance of
adaptation, shaping, and selection, and in the competence with which they
balance among the three possible courses of action.
Much of our work on practical intelligence has centered on the concept of
tacit knowledge. We define this construct, for our purposes, as what one needs
to know to work effectively in an environment that one is not explicitly taught
and that often is not even verbalized. We represent tacit knowledge in the form
of production systems or sequences of "if-then" statements that describe
procedures one follows in various kinds of everyday situations.
We typically have measured tacit knowledge using work~related problems
that present problems one might encounter on the job. We have measured
tacit knowledge for both children and adults and, among adults, for people in
238 Robert J. Sternberg

various occupations such as management, sales, academia, and the military. In

a typical tacit-knowledge problem, people are asked to read a story about a
problem someone faces and to rate, for each statement in a set of statements,
how adequate a solution the statement represents. For example, in a paper-
and-pencil measure of tacit knowledge for sales, one of the problems deals
with sales of photocopy machines. A relatively inexpensive machine is not
moving out of the showroom and has become overstocked. The examinee is
asked to rate the quality of various solutions for moving the particular model
out of the showroom. In a performance-based measure for sales people, the
test taker makes a phone call to a supposed customer, who is actually
the examiner. The test taker tries to sell advertising space over the phone. The
examiner raises various objections to buying the advertising space. The test
taker is evaluated for the quality, rapidity, and fluency of the responses on the
telephone.
In collaborative studies with Richard Wagner, we found that practical intelli~
gence as embodied in tacit knowledge increases with experience, but it is
profiting from experience, rather than experience per se, that results in in-
creases in scores (Wagner & Sternberg, 1986). Some people can have been in a
job for years and still have acquired relatively little tacit knowledge. We also
have found that subscores on tests of tacit knowledgemsuch as for managing
oneself, managing others, and managing tasksmcorrelate significantly with
each other. Moreover, scores on various tests of tacit knowledge, such as for
academics and managers, are also correlated fairly substantially (at about the
.5 level). However, scores on tacit-knowledge tests do not correlate with scores
on conventional tests of intelligence, whether the measures used are single-
score measures or multiple-ability batteries. Despite their lack of correlation
with conventional measures, the scores on tacit-knowledge tests predict per-
formance on the job as well as or better than do conventional psychometric
intelligence tests. In one study done at a management training center, the
Center for Creative Leadership, we further found that scores on our tests of
tacit knowledge for management were the best single predictor of performance
on a managerial simulation. In a complex statistical procedure, scores on
conventional tests of intelligence, personality, styles, and interpersonal orien-
tation were considered first and scores on the test of tacit knowledge were
considered last in predicting success on two simulations of managerial per-
formance. Scores on the test of tacit knowledge were the single best predictor
of managerial simulation score. Moreover, they also contributed significantly to
the prediction even after everything else was entered first into the equation. In
recent work on military leadership with Jennifer Hedlund, Joseph Horvath,
Colonel George Forsythe, Wendy Williams, and others, we found that scores
on a test of tacit knowledge for military leadership predicted ratings of leader-
ship effectiveness, whereas scores on a conventional test of intelligence and
on our tacit-knowledge test for managers did not significantly predict the
ratings of effectiveness (Sternberg et al., 2000).
11. Intelligence Is Not Just Inside the Head 2 39

Craig Smith, Michael Barnes, and I also have done studies of social intelli~
gence, which is viewed in the theory of successful intelligence as a part of
practical intelligence (Sternberg & Smith, 1988; Sternberg & Barnes, 1988). In
these studies, individuals were presented with photos and were asked to make
judgments about the photos. In one kind of photo, they were asked to evaluate
whether a male-female couple was a genuine couple (i.e., really involved in a
romantic relationship) or a phony couple posed by the experimenters. In
another kind of photo, they were asked to indicate which of two individuals
was the other's supervisor. We found females to be superior to males on these
tasks. Scores on the two tasks did not correlate with scores on conventional
ability tests, nor did they correlate with each other, suggesting a substantial
degree of domain specificity in the task.
Practicalqntelligence skills can be taught. Williams, Blythe, White, Li, Gard~
ner, and I have developed a program for teaching practical intellectual skills,
aimed at middle-school students, that explicitly teaches students "practical
intelligence for school" in the contexts of doing homework, taking tests,
reading, and writing (Williams et al., 1996). We have evaluated the program in
a variety of settings and have found that students taught via the program
outperform students in control groups who did not receive the instruction.

Combining Analytical, Creative, and

Practical Intelligence
The studies described above looked at analytical, creative, and practical intelli~
gence separately. But a full validation of the theory of successful intelligence
would require research that looks at all three aspects of intelligence in con-
junction. To date, we have done two such sets of studies.
In one set of studies in collaboration with Michel Ferrari, Pam Clinkenbeard,
and Elena Grigorenko, we explored the question of whether conventional edu~
cation in school systematically discriminates against children with creative and
practical strengths. Motivating this work was the belief that the systems in
schools strongly tend to favor children with strengths in memory and analytical
abilities (Sternberg, Ferrari, Clinkenbeard, & Grigorenko, 1996; Sternberg, Gri~
gorenko, Ferrari, & Clinkenbeard, 1999).
We devised a test for high school students of analytical, creative, and prac~
tical abilities that consisted of both multiple~choice and essay items. The
multiple~choice items required the three kinds of thinking in three content
domains: verbal, quantitative, and figural. Thus there were nine multiple~choice
and three essay subtests. The test was administered to 324 children around the
United States and in some other countries who were identified by their schools
as gifted by any standard whatsoever. Children were selected for a summer
program in (college~level) psychology if they fell into one of five ability groupings:
high analytical, high creative, high practical, high balanced (high in all three
abilities), or low balanced (low in all three abilities). Students who came to Yale
240 Robert J. Sternberg

were then divided into four instructional groups. Students in all four instruc~
tional groups used the same introductory psychology textbook [a preliminary
version of Sternberg (1995)] and listened to the same psychology lectures. What
differed among them was the type of afternoon discussion section to which they
were assigned. They were assigned to an instructional condition that empha~
sized either memory, analytical, creative, or practical instruction. For example,
in the memory condition, they might be asked to describe the main tenets of a
major theory of depression. In the analytical condition, they might be asked to
compare and contrast two theories of depression. In the creative condition, they
might be asked to formulate their own theory of depression. In the practical
condition, they might be asked how they could use what they had learned about
depression to help a friend who was depressed.
Students in all four instructional conditions were evaluated in terms of their
performance on homework, a midterm exam, a final exam, and an independent
project. Each type of work was evaluated for memory, analytical, creative, and
practical quality. Thus, all students were evaluated in exactly the same way.
Our results suggested the utility of the theory of successful intelligence. First,
we observed when the students arrived at Yale that the students in the high
creative and high practical groups were much more diverse in terms of racial,
ethnic, socioeconomic, and educational backgrounds than were the students in
the high-analytical group. In other words, just by expanding the range of abilities
we measured, we discovered more intellectual strengths than would have been
apparent through a conventional test. Moreover, the kinds of students identified
as strong differed in terms of populations from which they were drawn in
comparison with students identified as strong solely by analytical measures.
We found the general factor of intelligence to be very weak, suggesting that
the general factor is probably relevant only when a fairly narrow range of
abilities is measured, as is typically the case with conventional tests. We
found that testing format had a large effect on results: Multiple~choice tests
tend to correlate with other multiple~choice tests, almost without regard to
what they measure. Essay tests show only weak correlations with multiple
choice, however. We further found that after we controlled for modality of
testing (multiple~choice versus essay), the correlations between the analytical,
creative, and practical sections were very weak and generally nonsignificant,
supporting the relative independence of the various abilities. We found that all
three ability tests--analytical, creative, and practical~significantly predicted
course performance. When multiple-regression analysis was used, at least two
of these ability measures contributed significantly to the prediction of each of
the measures of achievement. Perhaps as a reflection of the difficulty of deem~
phasizing the analytical way of teaching, one of the significant predictors was
always the analytical score. (However, in a replication of our study with low-
income African~American students from New York, Deborah Coates of the City
University of New York found a different pattern of results. Her data indicated
that the practical tests were better predictors of course performance than were
11. Intelligence Is Not Just Inside the Head 24 ]

the analytical measures, suggesting that what ability test predicts what criter-
ion depends on population as well as mode of teaching.) Most importantly,
there was an aptitude-treatment interaction whereby students who were
placed in instructional conditions that better matched their pattern of abilities
outperformed students who were mismatched. In other words, when students
are taught in a way that fits how they think, they do better in school. Children
with creative and practical abilities, who are almost never taught or assessed in
a way that matches their pattern of abilities, may be at a disadvantage in
course after course, year after year.
In a follow-up study with Bruce Torff and Elena Grigorenko, we looked at
learning of social studies and science by third graders and eighth graders
(Sternberg, Torff, & Grigorenko, 1998a, 1998b). The third graders were students
in a very low income neighborhood in Raleigh, North Carolina. The eighth
graders were students who were largely middle to upper-middle class studying
in Baltimore, Maryland, and Fresno, California. In this study, students were
assigned to one of three instructional conditions. In the first condition, they
were taught the course that basically they would have learned had we not
intervened. The emphasis in the course was on memory. In a second condition,
they were taught in a way that emphasized critical (analytical) thinking. In the
third condition, they were taught in a way that emphasized analytical, creative,
and practical thinking. All students' performance was assessed for memory
learning (through multiple~choice assessments) as well as for analytical, cre~
ative, and practical learning (through performance assessments).
As expected, we found that students in the successfulqntelligence (analyt~
ical, creative, practical) condition outperformed the other students in terms of
the performance assessments. One could argue that this result merely reflected
the way they were taught. Nevertheless, the result suggested that teaching for
these kinds of thinking succeeded. More important, however, was the result
that children in the successfulqntelligence condition outperformed the other
children even on the multiple~choice memory tests. In other words, to the
extent that one's goal is just to maximize children's memory for information,
teaching for successful intelligence is still superior. It enables children to
capitalize on their strengths and to correct or to compensate for their weak~
nesses, and it allows children to encode material in a variety of interesting ways.
Thus the results of two sets of studies suggest that the theory of successful
intelligence is valid not just in its parts but also as a whole. Moreover, the
results suggest that the theory can make a difference not only in laboratory
tests, but in school classrooms as well.

CONCLUSION

The time has come to move beyond conventional theories of intelligence and
its development. In this chapter I have provided data suggesting that conven~
242 Robert J. Sternberg

tional theories and tests of intelligence are incomplete. The general factor of
intelligence may be an artifact of limitations in populations of individuals
tested, types of materials with which they are tested, and types of methods
used in testing. Indeed, our studies show that even when one wants to predict
school performance, the conventional tests are fairly limited in their predictive
validity. I have proposed a theory of successful intelligence that fares well in
construct validations, whether one tests in the laboratory, in schools, or in the
workplace. The greatest obstacle to our moving on is in vested interests in a
way of thinking that is no longer working, if it ever worked at all. If we want to
maximize individuals' achievement in school and in the workplace, new notions
are needed. We now have ways to move beyond conventional notions of
intelligence; we need only the will.

Teachers' Questions and Answers

Q: As an elementary school teacher, how could I enhance my students'
analytical, practical, and creative skills? What kinds of interventions and exer-
cises would you recommend?
A: Fortunately, we have written a book that addresses this question. The
book is Teaching for Successful Intelligence (Sternberg & Grigorenko, 2000).
In the book, we show the kinds of strategies that can be used. To develop
analytical skills, children are encouraged to analyze, compare and contrast,
judge, critique, and evaluate. To develop creative skills, children are encour-
aged to create, invent, explore, discover, and imagine. To develop practical
skills, children are encouraged to use, apply, implement, and put into prac~
rice.
Q: Often, especially in highly competitive colleges and universities, the
importance given to SAT scores far outweighs any other parameter used in the
admissions process. In my experience, high scores on these standardized tests
do not necessarily translate to outstanding performance in college. Can you
comment on the validity of this observation? In the light of your findings do you
think that universities ought to expand their selection criteria to include
dimensions of successful intelligence? Do you see any evidence that colleges
and universities may be moving in this direction?
A: Yes to everything! We are currently working with the College Board to
experiment with new tests that we have developed that measure creative and
practical in addition to analytical abilities. We hope that someday these tests
will supplement measures such as the SAT. Ultimately, we hope to see a test
battery that measures the complete set of skills proposed in the theory of
successful intelligence.
11. Intelligence Is Not Just Inside the Head 243

References
Carroll, J. B. (1993). Human cognitive abilities: A survey of factor.~analytic studies. New York: Cambridge
University Press.
Ceci, S. J., & Liker, J. (1986). Academic and nonacademic intelligence: An experimental separation.
In R. J. Sternberg & R. K. Wagner, (Eds.), Practical intelligence..Nature and origins of competencein the
everyday world (pp. 119-142). New York: Cambridge University Press.
Feuerstein, R. (1979). Tile dynamic assessmentof retarded performers..The learning potential assessmentdevice
theory, instruments, and tecfiniques. Baltimore, MD: University Park Press.
Greenfield, P. M. (1997). You can't take it with you: Why abilities assessments don't cross cultures.
American Psychologist, 52, 1115-1124.
Jensen, A. R. (1998). The g factor.. Tile science of mental ability. Westport, CT: Praeger/Greenwoood.
Lave, J. (1997). The culture of acquisition and the practice of understanding. In D. I. Kirshner & J. A.
Whitson (Eds.), Situated cognition: Social, semiotic, and psychological perspectives(pp. 17-35). Mahwah,
NJ: Erlbaum.
Nufies, T. (1994). Street intelligence. In R. J. Sternberg (Ed.), Encyclopedia of human intelligence (Vol. 2,
pp. 1045-1049). New York: Macmillan.
Okagaki, L., & Sternberg, R. J. (1991). Cultural and parental influences on cognitive development. In
L. Okagaki & R. J. Sternberg (Eds.), Directors of development:Influences on the developmentof children's
thinking (pp. 101-120). Hillsdale, NJ: Erlbaum.
Okagaki, L., & Sternberg, R. J. (1993). Parental beliefs and children's school performance. Cfiild
Development, 64, 36-56.
Piaget, J. (1972). The psychology of intelligence. Totowa, NJ: Littlefield Adams.
Rogoff, B. (1990). Apprenticeship in thinking: Cognitive development in social context. New York: Oxford
University Press.
Spearman, C. (1904). 'General intelligence,' objectively determined and measured. American ]ournal
of Psychology, 15, 201-293.
Sternberg, R. J. (1982). Nonentrenchment in the assessment of intellectual giftedness. Gifted Child
Quarterly, 26, 63-67.
Sternberg, R. J. (1985). Beyond/Q: A triarchic theory of fiuman intelligence. New York: Cambridge
University Press.
Sternberg, R. J. (1987a). Most vocabulary is learned from context. In M. G. McKeown & M. E. Curtis
(Eds.), Tile nature of vocabulary acquisition (pp. 89-105). Hillsdale, NJ: Erlbaum.
Sternberg, R. J. (1987b). The psychology of verbal comprehension. In R. Glaser (Ed.), Advances in
instructional psychology (Vol. 3, pp. 97-151). Hillsdale, NJ: Erlbaum.
Sternberg, R. J. (1997). Successful intelligence. New York: Plume.
Sternberg, R. J. (1999). Successful intelligence: Finding a balance. Trends in Cognitive Sciences, 3,
436-442.
Sternberg, R. J., & Barnes, M. (Eds.) (1988). The psychology of love. New Haven, CT: Yale University
Press.
Sternberg, R. J., Ferrari, M., Clinkenbeard, P. R., & Grigorenko, E. L. (1996). Identification, instruc-
tion, and assessment of gifted children: A construct validation of a triarchic model. Gifted Child
Quarterly, 40, 129-137.
Sternberg, R. J., Forsythe, G. B., Hedlund, J., Horvath, J., Snook, S., Williams, W. M., Wagner, R. K., &
Grigorenko, E. L. (2000). Practical intelligence in everyday life. New York: Cambridge University Press.
Sternberg, R. J., & Gastel, J. (1989a). Coping with novelty in human intelligence: An empirical
investigation. Intelligence, 13, 187-197.
Sternberg, R. J., & Gastel, J. (1989b). If dancers ate their shoes: Inductive reasoning with factual
and counterfactual premises. Memory and Cognition, 17, 1-10.
Sternberg, R. J., Grigorenko, E. L., Ferrari, M., & Clinkenbeard, P. (1999). A triarchic analysis of an
aptitude-treatment interaction. European/ournal of Psychological Assessment, I5, 1-11.
244 Robert J. Sternberg

Sternberg, R. J., Grigorenko, E. L., Ngorosho, D., Tuntufye, E., Mbise, A., Nokes, C., Jukes, M., &
Bundy, D. A. (in press). Assessing intellectual potential in rural Tanzania school children.
Intelligence.
Sternberg, R. J., & Lubart, T. I. (1995). Defying tile crowd: Cultivating creativity in a culture of conformity.
New York: Free Press.
Sternberg, R. J., Nokes, K., Geissler, P. W., Prince, R., Okatcha, F., Bundy, D. A., & Grigorenko, E. L.
(2001). The relationship between academic and practical intelligence: A case study in Kenya.
/ntelligence, 29, 40 I-418.
Sternberg, R. J., & Powell, J. S. (1983). Comprehending verbal comprehension.American Psychologist,
38, 878-893.
Sternberg, R. J., & Rifkin, B. (1979). The development of analogical reasoning processes, lournal of
Experimental Cfiild Psychology, 27, 195-232.
Sternberg, R. J., & Smith, E. E. (Eds.) (1988). The psychology of human thougfit. New York: Cambridge
University Press.
Sternberg, R. J., Torff, B., & Grigorenko, E. L. (1998a). Teaching for successful intelligence raises
school achievement. Pfii Delta Kappan, 79, 667-669.
Sternberg, R. J., Torff, B., & Grigorenko, E. L. (1998b). Teaching triarchically improves school
achievement, lournal of Educational Psychology, 90, 374-384.
Tetewsky, S. J., & Sternberg, R. J. (1986). Conceptual and lexical determinants of nonentrenched
thinking. Journal of Memory and Language, 25, 202--225.
Vygotsky, L. S. (I 978). Mind in society: The developmentof higher psychological processes.Cambridge, MA:
Harvard University Press.
Wagner, R. K., & Sternberg, R. J. (1986). Tacit knowledge and intelligence in the everyday world. In
R. J. Sternberg & R. K. Wagner (Eds.), Practical intelligence: Nature and origins of competence in the
everyday world (pp. 51-83). New York: Cambridge University Press.
Williams, W. M., Blythe, T., White, N., Li, J., Sternberg, R. J., & Gardner, H. I. (1996). Practical
intelligence for school..A handbookfor teacfiers of grades 5-8. New York: HarperCollins.

Suggested Reading
Ceci, S. J. (1996). On intelligence...more or less (expanded edition). Cambridge, MA: Harvard University
Press.
Gardner, H. (1983). Frames of mind: The theory of multiple intelligences. New York: Basic.
Sternberg, R. J. (1997). Successful intelligence. New York: Plume.
 CHAPTER

12
Being and Becoming a Good
Person: The Role of Emotional
Intelligence in Moral
Development and Behavior
DAVID A. PIZARRO AND PETER SALOVEY
Department of Psychology, Yale University, New Haven, Connecticut

There can be little doubt that one of the primary concerns of parents and
educators is that children under their care grow to develop a strong sense of
moral responsibility. Within the past few years especially, a fear that something
can go wrong--that the process by which children acquire this sense could
failmhas led to increased attention to the formation of moral character.
Nationwide, programs have been instituted that implement a standardized
curriculum of moral and values education in classrooms. This has been due
in part to the media exposure of shocking incidents involving children (in many
cases very young children) acting violently in schools across the nation, but can
also be understood as an attempt to apply what researchers have learned
about moral development over the past decades to the home and the class~
room. The topic itself has fascinated students of psychology for more than a
century, and has inspired a great deal of research within the field. Indeed, it can
be argued that moral development is one of the most important processes for
psychologists to study, as one of the most critical conditions for the survival of
society is that its members learn the differences between right and wrong.

Please address correspondence to Peter Salovey, Department of Psychology, P.O. Box 208205,
New Haven, CT 06520-8205. E~mail. [email protected].

ImprovingAcademicAchievement
Copyright2002,ElsevierScience(USA).All rightsreserved. 247
248 David A. Pizarro and Peter Salovey

The study of emotion in psychology, on the other hand, has ebbed and
flowed, enjoying periods of flourishing theory and research, yet also spending
time on the "backburner" of psychological thinking. Despite its cyclical nature,
the study of emotion has yielded valuable information to our understanding of
human thinking and behavior. In recent years, psychology has experienced a
surge of interest concerning the role of emotions in various psychological
processes, including their role in judgment and decision making, and motiw
ation. Part of this rebirth in interest about emotions has been due to a public
fascination with popular theories of emotional intelligence and emotional
competency--theories that describe a set of emotional skills that are import~
ant for social functioning. The surge of public interest in emotional intelligence
has been accompanied by a plethora of books and articles on the topic
(reviewed by Mayer, Salovey, & Caruso, 2000). Emotional intelligence has
been implicated by many as an important skill or set of skills necessary for
social adjustment and happiness, yet until present times relatively ignored.
Some have gone as far as to equate emotional intelligence with moral charac~
ter, making the terms "emotionally intelligent" and "moral" nearly synonym~
ous (e.g., Goleman, 1995; Saarni, 1999). This interest has raised intriguing
questions concerning the role emotions and emotion~related skills play in
our moral lives. This chapter seeks to shed light on some of these important
issues. Among the questions we address are, What role do various emotions
play in the moral development and education of children, as well as in the
moral decision-making of adults? And, is it necessary to be emotionally intelli~
gent to be a moral person? Before we do so, however, let us first take a look at
what we mean by "emotional intelligence."

WHAT IS EMOTIONAL INTELLIGENCE?

Although emotions play a large role in the daily life of all people, it is clear that
there are large individual differences in the way people deal with emotions;
some people seem to be more adept emotional managers than others, for
example. For instance, receiving a bad grade on a project might incite one child
to intense anger, causing the child to act out in class, while another child,
although angered at first, might use his or her feelings to motivate working
harder on the next project. The various emotion~related skills that are
employed by individuals have been grouped together and labeled emotional
intelligence (Mayer & Salovey, 1997; Salovey & Mayer, 1990). The emotional
intelligence framework was proposed as an attempt to organize the growing
body of research on emotions and their influence on cognition and behavior
into a single theory that highlighted the way individuals use these skills in
their everyday lives. Emotional intelligence has been divided into four main
branches, each focusing on a different set of emotional skills: (a) the ability
12. Being and Becoming a Good Person 249

to perceive, appraise, and express emotion; (b) the ability to use feelings in
cognitive activities; (c) the ability to understand emotion and emotional know~
ledge; and (d) the ability to regulate or manage emotions to promote emotional
and intellectual growth (Mayer & Salovey, 1997).
Many theorists, researchers, and journalists eagerly picked up on emo~
tional intelligence, and framed it as a skill that was of critical importance to be
a caring, moral, and otherwise well-adjusted person (e.g., Goleman, 1995).
This characterization is not at all surprising. After all, perceiving, generating,
and regulating emotions are generally things we do to maintain and improve
relationships with others. Very often we enlist our abilities at managing emo~
tions when we become angry with a best friend or have an argument with
our spouse, for instance. In these situations, the ability to understand how
the other person is feeling and to regulate our emotional reactions comes in
very handy, and serves to fulfill the goals we have of remaining friends or
of staying married for longer than a few months. It seems natural, then,
that understanding how emotion~related skills affect social relations should
be of primary importance to those interested in studying emotional intelli~
gence.
The relationship between emotional intelligence and moral character is,
however, not as clear-cut as might first appear. The same emotional skills
that make some individuals good, caring people can also be used to achieve
more nefarious goals. Criminals who are masters at deception or con artists
who are trained to manipulate others may in some ways be among those
highest in at least some of these emotional skills (Salovey & Mayer, 1990).
Any discussion of how emotions and emotional skills relate to moral develop~
ment and behavior must take this into account.

A (VERY) BRIEF HISTORY OF MORAL PSYCHOLOGY

AND EMOTION

Although research on moral development has grown enormously within the last
50 or so years, most of the interest has come from researchers within the
tradition known as cognitive.-developmental. Building on the work of Piaget (1932),
Kohlberg (1969), and others, researchers from this tradition have had the
strongest influence in the field of moral psychology. Unfortunately for those
of us interested in emotions, this tradition has largely ignored the role of feelings
in the way children grow to become moral individuals. Rather, it has focused on
the development of the child's cognitive abilities and the way these develop-
mental changes affect the child's moral world view. Kohlberg, for instance,
viewed the child as progressing from an early morality based on parental author~
ity to a fully autonomous morality, based on an understanding of universal
2 50 David A. Pizarro and Peter Salovey

moral principles. Although Kohlberg sometimes touched on the topic of emo~

tions, his theory never adequately sought to describe the role of emotions in the
development of a mature morality.
Within the last 20 years, however, social and developmental psychologists
have turned their attention to emotional processes in the social development
of children and in the judgments and behaviors of adults. Developmental
researchers specifically have sought a description of how emotions work in
fostering a child's moral sensibilities. There are at least two reasons why
theorists consider emotions important for the study of moral development
and behavior. First, emotions are powerful sources of motivation. In other
words, most emotions are associated with what has been labeled an action
tendency, or a propensity to engage in certain actions when experiencing a
specific emotion. For instance, anger is associated with the tendency to attack
and fright with the tendency to escape (Lazarus, 1991). Emotions such as
empathy and guilt motivate prosocial behaviors [constructive behaviors that
are also generally understood as moral (Eisenberg, 1986)]. Helping a person in
need and apologizing for hurting someone are examples of these kinds of
behaviors. Knowing in what circumstances emotions motivate us to do good
is therefore an important component of understanding moral behavior in
general. Even cognitive theories recognize this role of emotions as an energy
source for moral functioning.
Because emotions are powerful motivators of action, they are also import~
ant mechanisms that aid in the process of socialization (the process by which
children come to internalize the values, norms, and morals of their parents and
society at large). Emotions, as naturally occurring events in children, are
resources that can help in the transmission of norms and values. By inducing
emotions such as guilt, empathy, shame, and even disgust (see Rozin, Haidt, &
McCauley, 1993), parents can mold children's responses to behaviors, events,
and people. In time, these emotional reactions come to occur naturally in the
child in appropriate situations, and act as internal sources of motivation and
constraint. It is these two features of emotion, their motivational arousal and
their role in socialization, that have made the study of emotion an important
area of inquiry to researchers interested in moral development and prosocial
behavior.
When discussing the emotional intelligence of morality, we focus on those
emotions that seem to have a distinctly moral nature, such as empathy and
guilt. This is not to say that emotions such as happiness and sadness do not
affect our moral lives. Certainly, we become happy when we do good things for
others and sad when we ponder the misfortune of others. However, for the
purposes of our discussion, we focus on the emotions investigators have
generally lumped together as morally relevant. In our discussion, we organize
the functions of these various emotions using the four branches of the emo~
tional intelligence framework (Mayer & Salovey, 1997; Salovey, Woolery, &
Mayer, 2000).
12. Being and Becoming a Good Person 2 51

PERCEIVING AND APPRAISING EMOTIONS

When discussing moral rights and wrongs, there seems to be room for a lot of
disagreement. Issues such as abortion, the death penalty, and same~sex
unions, for example, seem to draw a moral dividing line across our nation.
Fortunately, however, there exists much more agreement about issues con~
cerning right and wrong than it might seem. For instance, causing innocent
individuals to suffer is rarely seen as anything but a morally reprehensible
action. Because moral events usually center around the presence of victimiza~
tion, encountering the suffering of innocent others in our daily lives is a strong
indication that a moral event is taking place. It is generally not the case,
however, that people wear signs around their neck that announce their pain,
such as "I just got dumped by my girlfriend" and "My favorite aunt just passed
away." Rather, there are subtle signals sent by those individuals that clue us in
to their distress. These signals act as efficient forms of communication. How~
ever, as may seem obvious, a signal must be perceived and understood for it to
be effective. The ability to perceive emotions accurately in others is thus a very
important emotional skill, arguably the most fundamental skill of all when it
comes to human relationships. This ability to know how others are feeling has
been labeled empathyby many researchers (Eisenberg & Miller, 1991; Feshbach
& Roe, 1968; Hoffman, 1987), and has been the most well-researched of all the
so-called moral emotions.
What is empathy? Some researchers have defined empathy strictly in terms
of the ability to take the perspective of others, a distinctly cognitive ability.
However, empathy can be seen as having both a cognitive component and an
emotional component, that of actually feeling an emotion that is more appro~
priate to the other person's situation (this is the definition offered by Hoffman,
1998). In fact, the word "empathy" literally means, "to feel oneself into"
(Wisp~, 1987). For our purposes, we restrict our definition of empathy to the
emotional arousal one feels when presented with the emotional experience of
another, particularly the distress of another. It is this empathic arousal that
allows us to feel suffering when others are suffering and, thus, motivates us to
help the individual in distress.
Most people are able to experience empathy, but some are better than
others. For instance, there is evidence that some antisocial youth suffer from
an inability to feel empathy, an "empathic dysfunction." As Gibbs (1987)
states, "Empathy is available in most [juvenile] offenders but is not readily
elicited and tends to be either an isolated impulse or a mawkish sentiment. In
either case, the empathy is superficial and erratic; when it lingers, it is readily
suppressed by self~centered motives or aggressive impulses. (p. 303)." It has
also been argued that psychopathic individuals suffer from a lack of the
capacity for empathic affect (Blair, 1995). This serves as further reason to
consider empathy a truly moral emotion: if you cannot feel empathy, chances
are you are not a very moral person.
2 52 David A. Pizarro and Peter Salovey

What influences the ability to experience empathy? It is clear that a capacity

for empathy is not all~or~nothing; individuals vary as to their degree of em~
pathic responsiveness. Various factors have been pointed to as important for
the development of empathy (for a review see Barnett, 1987; Davis, 1996, pp.
62-81). These include genetics (e.g., Matthews, Batson, Horn, & Rosenman,
1981), childhood temperament (e.g., Mehrabin, 1980), a stable and positive
relationship with a caregiver (e.g., Mussen & Eisenberg~Berg, 1977), and discip~
lining techniques that focus on inducing empathy in children (e.g., Hoffman &
Saltzstein, 1967). Summarizing the research on the development of empathy,
Barnett (1987, p. 156) concluded:

The development of empathy and related responses would appear to thrive in an

environment that (1) satisfies the child's own emotional needs, (2) encourages the
child to identify, experience, and express a broad range of emotions, and (3) provides
numerous opportunities for the child to observe and interact with others who, through
their words and actions, encourage emotional sensitivity and responsiveness to others.

Parents and educators should therefore strive to create an environment in

which children are encouraged to take the perspective of others, to imagine
what the other person is feeling, and to be active in speaking to their children
about emotions.
Before we conclude that the capacity for empathy is sufficient to be a moral
individual, we must discuss its limitations. As was mentioned above, empathic
arousal motivates us to alleviate the suffering of others. It seems as if the
distress we feel when in the presence of distressed others can be alleviated
only by helping the individual in need. In fact, individuals tend to help even if
there is an easier escape from the empathic distress, for instance, by leaving
the situation (e.g., Batson, Dyck, Brandt, Batson, Powell, McMaster, & Griffitt,
1991). People high in dispositional empathy (people with an "empathic per~
sonality") are also more likely to engage in helping behavior (Davis, Mitchell,
Hall, Lothert, Snapp, & Meyer, 1999). However, being "high" in empathy is not
a surefire qualification for being a moral individual, nor is it a guarantee that a
person will always do the right thing. One of the interesting features of empathy
is that it is more easily elicited for people that we perceive as similar to us (e.g.,
Feshbach & Roe, 1968) and that we view as innocent (Betancourt, 1990).
Conversely, the more different we perceive others to be, and the more at
fault we think they are, the less likely we are to experience empathy for them
and, thus, the less likely we are to help them if they are in need. This "empathic
bias," as Hoffman (1987) has labeled it, is one reason we cannot always rely on
our emotional reactions as a reliable guide to moral truth. Sometimes, we have
to try very hard to feel empathy for others, by imagining ourselves in their
position and by focusing on similarities rather than differences. Indeed, one of
the primary tasks of parents and educators should be to make the empathic
response available in children regardless of perceived differences between
themselves and the victim.
12. Being and Becoming a Good Person 2 53

Another problem with empathy is that sometimes we feel empathy for

individuals who we know do not deserve it. For instance, we may feel sorry
for a criminal who had a rough childhood, only to find out that he committed
numerous brutal murders. In this situation, felt empathy must be "squashed"
so that the motivational consequences (helping the murderer) dissipate.
When speaking of empathy as a moral emotion, then, one must be careful
not to assume that the capability to feel what others are feeling is the same as
making mature moral decisions. Indeed, the ability to perceive and appraise
emotions in others is an ability that may even be used to manipulate others.

EMOTIONS THAT HELP US THINK

The idea that emotions are forces that act contrary to reason has plagued
Western thinking since the days of the earliest Greek philosophers. Plato (1988)
characterized emotions as being akin to wild horses that need to be controlled
by the "rational" rider. Freud (1977) also viewed most emotions as strong
instinctual forces that must be conquered by the Superego, that portion of the
mind that was in charge of matters of conscience. The notion that emotions
disrupt cognitive activities persists even today in conceptions of emotional
processes (Mandler, 1975; Simon, 1981). For example, labeling someone as
being "too emotional" is synonymous with calling him or her irrational. Simi~
larly, crimes of passion are punished less severely than cold, calculated acts
because emotions are seen as temporarily seizing the will of the individual,
rendering him or her unable to make informed decisions in the planning of
actions.
Although there are some investigators who continue to maintain a strong
position concerning the divide between reason and passion (e.g., Metcalfe &
Mischel, 1999), most researchers agree that emotions often serve to facilitate
reasoning, rather than hinder it (e.g., Damasio, 1994; Salovey & Mayer, 1990).
Emotions prioritize events, pointing the individual toward problems in the
environment that are of immediate importance (Easterbrook, 1959; Leeper,
1948). In this sense emotions serve to aid judgment, by steering thoughts in the
right direction. For instance, negative moods encourage careful, deliberate
ways of thinking, causing people to elaborate more on problems than they
would in a more positive mood. The presence of happy moods, on the other
hand, encourages a more creative style of thinking, leading some to listen to
happy, upbeat music to facilitate creative thoughts (Isen, 1993; Palfai & Salo~
vey, 1993; Schwarz, 1990).
Stated simply, emotional reactions focus our cognitive resources on the
problem at hand. It is no different with moral emotions. Moral emotions
prioritize thinking about our moral principles and beliefs, motivate appropriate
moral judgments, and prepare us to take moral action. When we become
distressed at the sight of another individual suffering, the negative arousal
2 54 David A. Pizarro and Peter Salovey

mobilizes our mental resources and facilitates thinking concerning moral ques-
tions such as why the individual is suffering, whose fault is it that she is
suffering, and what course of action should be taken to help her. For example,
when, on a winter day, we come across a man who is obviously cold and
hungry, and who appears to be homeless, the empathy we feel turns our
thoughts toward the plight of the homeless and the inequalities of society.
Or, it may cause us to pronounce harsh moral judgments on the individual,
attributing his or her position to some flaw of character. In either case, the
immediate felt empathy is what served to stimulate thinking about the moral
implications of the situation.
If empathy generally motivates us to help, guilt is what motivates us to make
amends, either by seeking to repair the damage to a valued relationship, as
when we ask forgiveness for offending a friend, or by correcting our behavior to
be consistent with our principles concerning how one should act. Guilt is a
negative evaluation of a specific behavior, and usually occurs when we feel as if
we have violated one of our moral principles, for example, by hurting someone
else or by otherwise acting in a manner unbecoming of how we think we should
act. In contrast to shame, which usually causes us to focus on ourselves, guilt
shifts the focus to the transgression and is associated with a desire to undo
what has been done (Tangney, 1999). For instance, a young child who feels
guilty for hitting his best friend will most likely find it hard to spend too much
time organizing his baseball cards. The guilt he is experiencing will turn his
thoughts toward how he hurt his friend and to what he should do to make
things better. Similarly, if we have hurt the feelings of a good friend, we are
easily distracted if we try to work, because our thoughts are constantly turning
to the damage we have done. Although one can have maladaptive levels of
guilt, in normal individuals guilt is an incredibly adaptive emotion, because it
maintains relationship health by motivating individuals to repair any damage
done to the relationship.
Once an emotion such as empathy is aroused in an individual, and thoughts
turn to matters of a moral nature, one will naturally draw conclusions regarding
the situation (Hoffman, 1998). Moral judgments, the conclusions drawn by
individuals concerning the moral rightness or wrongness of actions or events,
often influence the presence of subsequent emotions. If we feel empathy in the
presence of a distressed other and realize that her or his distress is due to the
unjust actions of some third party, our empathy is likely to turn into "empathic
anger." For example, when viewing footage of police brutality directed toward
an innocent African American man, the empathic distress we feel may turn into
anger at such a violation of basic rights. If, on the other hand, we feel empathy
for the distress of another, but realize that we are the cause of the others'
distress, empathy transforms into guilt. For example, the distress aroused
when seeing our younger siblings crying uncontrollably quickly turns into guilt
when we realize that they are crying because of something we said. Empathy
combines with attributions of blame and other moral judgments, and it is the
12. Beingand Becominga Good Person 255

motivational power of the emotion in combination with our judgments that

informs our subsequent actions.
The knowledge that moral emotions will mobilize our thinking concerning
moral issues (such as the plight of the homeless or the importance of not
hurting those we love) is knowledge that can be used to serve our individual
moral goals. By taking the perspective of other people, for instance, we can
make ourselves feel empathy for someone with whom we may not have
otherwise concerned ourselves. A pragmatic use of these emotional skills is
therefore an advantage, in that the emotions encourage the critical thinking
necessary to work through moral situations and moral dilemmas, and they
harness the full motivational force of the emotion.

EMPLOYING EMOTIONAL KNOWLEDGE

Human interactions are full of complex emotional information. The ability to

understand and discern this emotional information adds quality and depth to
our own lives, and allows us to understand better the lives of others. Know~
ledge such as what emotion an artist is trying to convey through her work or of
the complex combination of emotions that are making us feel a certain way is
considered a sign of a healthy emotional life. Knowledge concerning how
emotions work and are communicated and the way that people employ this
knowledge is organized under this third branch of emotional intelligence.
Among the skills are the ability to define emotions, the ability to understand
complex blends of emotions, and an accurate understanding of the likely
transitions between emotions (Mayer & Salovey, 1997).
How is this emotional knowledge used in moral life? Thus far we have
discussed how emotions work as motivation through their immediate action,
as in the empathy we feel when we see someone in distress or the guilt we feel
immediately after we hurt someone. But emotions also motivate us from a
distance. In other words, merely anticipating that we might feel an emotion is
sometimes enough to affect our present behavior. A child who is thinking
about cheating on an exam might be motivated not to do so because she
knows that she would feel guilty immediately following the act. In this case,
knowledge of the emotional consequences of an act becomes an important
determinant for whether or not a person will be motivated to avoid performing
an "immoral" action.
This type of emotional knowledge, although crucial when it comes to
behaving morally, takes time to develop. One of the most interesting findings
concerning children's knowledge of moral emotions is the so~called "happy
victimizer" effect (Arsenio & Kramer, 1992). Very young children expect that a
wrongdoer will feel good after having committed a moral transgression. There is
a clear age trend in this phenomenon; as children develop (usually between the
ages of 6 and 10) they come to believe that a wrongdoer will feel badly after
2 56 David A. Pizarro and Peter Salovey

having committed a transgression. The development of this ability is critical; if

a child is unaware of the emotional consequences of his or her act, there will be
little motivation to avoid it (or to perform it). Pointing to the emotional
consequences of an act can serve to strengthen the connections between
transgressions and the feelings of guilt. In fact, there is evidence that disciplin-
ing children by pointing out how they feel after an act (what is called inductive
discipline) is the most efficient form of discipline, because it pairs emotional
consequences with certain acts.
A second type of emotional knowledge, which takes time to develop as well, is
the knowledge of complex combinations of emotional states. This knowledge is
one of the most important feats of mature emotional and cognitive develop~
ment. When, for instance, we see an Olympic athlete in a track race fall, we are
aware not only of the physical pain he is experiencing, but also of the disappoint~
ment he must feel at achieving so much and failing at such a critical moment, and
also of the possible wound to his pride caused by falling in front of an audience of
millions. Such an understanding of the complex emotions experienced by the
individual is important in informing our subsequent actions. Should we help
ease his physical pain? What types of things should we say to him to ease his
emotional pain? Should we remain quiet rather than speak to him? An inability to
answer these complex questions concerning the individual involved renders our
helping abilities rather useless. If we were unable to figure out some answers to
these questions, our helping behaviors would be similar to those of young
children, who often offer a safety blanket or a favorite toy (decidedly not the
kind of help that most adults would want). Adequately helping others means
knowing how they may be feeling in the larger context of their life experiences.

EFFECTIVELY REGULATING EMOTIONS IN

OURSELVES AND IN OTHERS

Regulating Emotions in Ourselves

Emotional regulation is perhaps one of the most important features of emo-
tional intelligence when it comes to moral judgments and behaviors. Emotional
reactions sometimes need to be guided in the right direction, lest they steer us
into the wrong one. This is obvious for negative emotions such as anger; if
anger is not regulated, it can motivate us to act inappropriately. It is less
obvious why we would need to regulate emotions such as empathy. After all,
empathy is a good thing, is it not?
This discussion should be prefaced with a point concerning emotions that is
especially useful when discussing moral emotions. Sometimes emotions are
elicited almost automatically (see Hodges & Wegner, 1997). When survival is
threatened, we react immediately with fear. In the same manner, when the desire
for social approval is threatened (e.g., by someone who made us look foolish in
12. Being and Becoming a Good Person 2 57

public), we immediately feel angry. We also have immediate and automatic

reactions of empathy, disgust, jealousy, and nearly every other emotion. The
bright side is that human beings are not mindless animals condemned to act on
our every impulse, so we are constantly able to choose the emotions that are
appropriate and those that are not by stepping outside the emotion and decid~
ing whether or not it is appropriate, then regulating it accordingly (see Gross,
1999). If we have an emotion that we believe we should not be having, or at least
that we think would be wrong to act on, we can enlist a higher-order desire to
regulate that emotion. If we have an immediate emotional reaction, such as
anger, and with it comes the desire to act on that emotion (attack our offender),
we can step outside the emotion and act as judges of it. We can have desires
about desires or emotions about emotions (Gottman, Katz, & Hooven, 1997). In
the case of the anger, it is the greater desire to do the right thing that motivates
us to regulate our emotional reaction.
Sometimes, it is the case that the immediate emotional reaction and the
greater desires match up with each other nicely. For instance, the motivation
brought on by empathy (the desire to help the person in need) goes along very
well with our greater desire to "do good," and the end result is that we perform
the action. In the cases where they do not match up, however, our skills in
regulating our emotions are called into play, and we take on the role of
emotional managers. When we become angry with a boss, for instance, we
know that we cannot slap him in the face. Thinking about our anger and turning
our thoughts toward constructive ways of dealing with the problem are strat~
egies that are often effective in the regulation of the emotion.
Not all people have mastered this skill, as one might guess. Oftentimes,
individuals allow inappropriate emotions to exert their full motivational force,
with the end result sometimes being disastrous. The recently coined "road
rage" phenomenon, where drivers become so angry that they stop at nothing
until they satisfy their revenge on other drivers, certainly attests to the unfor-
tunate consequences of poor emotional regulation.
This discussion of emotional regulation should not be taken as evidence
that emotions are bad and that by regulating them we necessarily mean
eliminating their effects. On the contrary, emotional reactions that are chan~
neled constructively can act as excellent sources of motivation. Anger at the
presence of societal injustices, when effectively regulated (which may mean
letting ourselves experience the anger fully rather than suppressing it), can
motivate individuals to great moral achievements, for instance. In fact, there
are some instances in which regulating an emotion, by not allowing ourselves
to fully experience it, may have drastic consequences for ourselves and others.
For instance, suppressing anger is thought to affect various physiological
mechanisms that are vital to our health (Petrie, Booth, Pennebaker, & Davison,
1995). Similarly, suppressing an emotion such as empathy might have disas~
trous moral consequences, allowing people to do things they would never
otherwise do (such as harm innocent people).
2 58 David A. Pizarro and Peter Salovey

Regulating Emotions in Others

So far we have talked about the importance of regulating emotions in our-

selves, but what about regulating emotions in other people? At first thought,
this might seem rather manipulative. However, in everyday life we know indi~
viduals who are skilled at manipulating emotions in other people in ways that
are considered constructive. For instance, the friend that everyone turns to
when they are feeling down is often sought out because of his or her ability to
"raise spirits" and make people feel better. Motivational speakers and
preachers are also good regulators of emotion in others, and are applauded
for these skills. So, while one can certainly see the manipulation that might
occur by regulating the emotions of others, by and large people use these skills
for the achievement of noble goals.
The ability to regulate the emotions of others is a critical skill when it comes
to the socialization of children. As we have mentioned, effective disciplining is
often achieved by using children's natural emotional reactions as sources of
motivation. Parents can capitalize on these emotional reactions by generating
them in children when appropriate. The induction of emotions or moods is
something that psychologists interested in emotions often do in an experi~
mental setting. For instance, in our emotions laboratory we often induce
moods by asking people to watch sad movie clips or listen to happy music
(e.g., Palfai & Salovey, 1993; Salovey, 1992; Salovey & Birnbaum, 1989). Simi~
larly, when disciplining a child, caretakers can take advantage of the ease with
which children are likely to experience emotions such as empathy and guilt,
and use it to motivate appropriate moral behaviors.
The children of parents that tend to induce emotions such as empathy and
guilt when a moral situation arises are more likely to internalize moral norms
efficiently (Hoffman & Saltzein, 1967). For instance, pointing out the conse~
quences of stealing, thus inducing empathy for the victim of the theft, is an
effective way of teaching children not to steal. By repeatedly inducing empathy
in similar situations, children come to associate the act of stealing with em~
pathy for the victim, and this emotional energy provides an internal source of
motivation in the children. This type of discipline stands in sharp contrast to
disciplinary tactics in which caretakers merely exert their authority over chil~
dren, threatening them with punishment if they do not act in a moral manner.
In any future moral situations, children who were disciplined through the use
of inductive methods will continue to act morally even in the absence of
external authority or threat, as compared with children who were disciplined
merely through an exertion of parental authority. Recent research on children's
development of "conscience" has supported these ideas (Kochanska, 1995,
1997).
Empathy and guilt are not the only emotions recruited in the transmission of
moral norms. Emotions such as shame and disgust are also implicated. Some
authors have pointed to the power of feelings of disgust when it comes to
12. Being and Becoming a Good Person 259

certain moral practices. For instance, vegetarians (who are vegetarian for moral
reasons) are more likely to find meat disgusting than vegetarians who become
vegetarian for health reasons (Rozin, Markwith, & Stoess, 1997). Inducing
disgust over certain practices may be one of the most powerful ways to get
children (and adults, as some religious leaders can attest) to stop performing
certain behaviors.
Inductive discipline works not merely as simple conditioning, i.e., the pairing
of emotions with situations. Through time, caretakers elaborate on the moral
principles involved in situations that arouse emotions, focusing on the similar~
ities across situations such as hitting others, stealing, or lying, and teaching the
child the appropriate principles involved. By linking moral principles to emo~
tions like this, thinking about moral principles becomes an emotionqaden task,
lending the principles greater motivational power. Every time there is a co~
occurrence of moral principles with empathic affect, the association causes a
bond between the two. Moral principles thus acquire a motivating power that
they would not have acquired without the effective pairing of the empathic
response. Moral principles come to elicit empathy and guilt, and conversely,
empathy and guilt elicit thinking about moral principles, leading Hoffman
(1987) to refer to them as "hot cognitions." This may help to explain why emo~
tions become so intense when individuals disagree about their moral beliefs.

A WORD ABOUT MORAL PRINCIPLES

In our discussion of emotions, we have briefly mentioned the importance of

moral principles in guiding the regulation of emotions and in elaborating on
them when disciplining children. As mentioned before, many psychologists
studying morality have largely ignored moral emotions, choosing to focus on
the development of cognitive abilities instead. They have focused on how
maturing cognitive abilities affect children's thinking about moral principles,
and how understanding these principles affects their judgments concerning
right and wrong. However, moral theorists who do focus on emotions have
been criticized for ignoring the role that an understanding of moral principles
plays in the moral development of children, choosing rather to focus on how
emotions act as rewards for doing good or punishments for doing wrong (e.g.,
Blasi, 1999).
It is our belief that any discussion of morality should ignore neither the role
of moral principles nor the role of emotions. In the moral lives of individuals, it
makes little sense to separate the two. We feel guilt when we violate what we
believe to be a moral principle. Guilt does not exist without previous iudgments
that certain acts are wrong. In the same manner, were it not for the emotions of
guilt and empathy we would have little motivation to act on our moral prin-
ciples. Because the two cannot be separated in real life, we do not think they
should be separated in our theoretical frameworks either.
260 David A. Pizarro and Peter Salovey

MORAL EDUCATION

Having covered a few key points concerning moral emotions and emotional
intelligence, we are now in the position to take a closer look at the implications
for moral and emotional education. There is a large push for the implementa~
tion of programs in schools across the nation that focus on the education of
character, values, and morals. Although this is not intended to be a review of
socio-moral~emotional education programs, there are a few points to be made
concerning the broad approaches that are currently in favor. In his article "How
Not to Teach Values," Kohn (1997) takes a critical look at many character
education programs, guiding his criticism by asking five questions he considers
vital. Among these are "What is the view of human nature [espoused by the
program]?" and "What is the theory of learning [espoused by the program]?"
Many character education programs adhere to an underlying assumption that
children are intrinsically evil and that their natural impulses must be curbed.
This bleak view of human nature, according to Kohn, leads to efforts at controll~
ing behavior by "breaking the will" of the child, and by offering the child
rewards for their good behavior. As Kohn correctly points out, this approach
is directly contrary to psychological research on motivation; one way to extin~
guish behavior is to encourage it with extrinsic rewards (Lepper, Green, &
Nisbett, 1973) . The rewarding of behavior (e.g., by giving tokens to children
when they are "caught" performing a good behavior) may undermine intrinsic
motivation.
Attempts to stop misbehavior by external punishments are just as ineffect~
ive. If character education is to work, it must foster internal motivation to do
good, and not depend on the presence of external rewards and punishment. As
discussed above, moral emotions are, by their very nature, internal sources of
motivation and constraint. The easy solution seems to be just to "teach"
emotional skills, focusing especially on moral emotions. In fact, it is strange
that more moral education programs do not pay special attention to emotional
education. But the goal sounds easier than it may actually be to accomplish it.
Berkowitz (1995) lists some reasons why this may be the case. First, develop~
mental evidence points to an early emergence (within the first 2 years of life) of
empathy, making its presence dependent on factors that occur before children
even reach school. Second, there is a general lack of research on how to
educate moral emotions. As Berkowitz states, "[T]he role of the school is to
direct the child to care for the good and abhor the bad; e.g., empathize with
victims and despise injustice. Unfortunately it is quite unclear how this is done. The
literature on moral education pays little attention to this issue" (p.25, em~
phasis added).
A more general approach to moral learning has also been popular in schools
across the nation--programs that focus specifically on issues such as conflict
resolution, emotional learning, and social development in children. Although in
most cases more broadly focused than Character education programs, these
12. Being and Becoming a Good Person 261

programs were also initiated because of the desire to minimize behaviors such
as interpersonal violence, drug abuse, suicide, and lack of civility among
students. There is only limited evidence at this point regarding the effective~
ness of many of these programs (see Lopes & Salovey, in press, for a review).
However, this is most likely due to a lack of controlled research (i.e., adequate
comparisons between programs) rather than an inability of these programs to
foster change in students.
Lest the state of affairs seem beyond remedy, it must be made clear that
there are strategies that can promote moral emotions in children. We have
already mentioned some strategies for effectively inducing emotions in others.
For instance, pointing to the consequences of a child's actions is an effective
method of promoting an empathic response in the child, and capitalizes on the
child's natural tendency to feel for others. Below we present further strategies
that may promote the education of moral character through the use of emo~
tions:

1. Build an environment that encourages the expression and discussion of

emotions. There is no substitute for having good models of emotional skills.
The way in which parents and educators treat and talk about emotions has
been shown to be an important part of the child's ability to adjust (Gottman et
al., 1997).
2. Be an effective regulator of emotions in children, especially when con~
fronted with moral situations. For instance, induce empathy for innocent
victims of crime, or guilt when the child has hurt someone. The builtqn
motivation provided by these emotions will continue to exert an influence
even in the absence of caretakers.
3. One strategy for inducing empathy in children is to point to the similar~
ities between them and the victim, framing victims in ways that allow children to
fully experience empathy. In contrast, framing victims as different preempts
feelings of empathy. By fostering a universal respect for humanity, as opposed
to drawing boundary lines across races, religions, and nationalities, parents and
educators can ensure that children will not fail to experience emotions when
presented with the victimization of others.
4. Although there may be a heritable component to the tendency to
experience empathy, it is most certainly the case that we can improve this
ability in ourselves and in children. Encourage children to be constant "per~
spective takers," to learn to see the world through the eyes of others.
Encourage conversation about how others must be thinking or feeling. This
is especially important in situations where there is a conflict between two
parties. Encouraging both sides to take the perspective of others will help
children not only to feel what the others may be feeling, but is also an
important exercise in respecting the opinions of others. Make it a habit to
verbalize your empathic feelings when presented with the suffering or victim~
ization of others.
262 David A. Pizarro and Peter Salovey

5. Discuss important moral principles, and link them to moral emotions.

Discussions about justice and fairness will come naturally to the developing
child (e.g., when having to share toys with other children, or when having to
take the blame for mishaps). Seize these opportunities to engage children in a
discussion of justice and fairness. Say, for instance, why feeling anger at the
sight of unfair practices is okay (i.e., because the principle of justice has been
violated). If a child is effective at reasoning about moral issues, it is usually the
case that appropriate moral emotions will follow.
There is little reason to think that we are helpless when it comes to the
emotional and moral education of children. The truth of the matter is that we
can be systematic and effective in fostering the moral and emotional develop~
ment of children.

CONCLUSION

Emotions play an enormous role in the moral development, moral judgment,

and moral behavior of individuals, and have often been ignored by researchers
in moral psychology. For a full psychological understanding of morality, one
must take emotional processes into account. The emotional intelligence
framework provides a useful background by which to organize the various
ways emotions work in moral processes. The ability to be effective in dealing
with emotions--accurately perceiving them, using them to guide thinking,
being knowledgeable about complex emotional states, and being effective
regulators of emotions--comprises skills that come into play in being a
moral individual. In fact, these skills can be used to teach children right from
wrong more effectively. There is no doubt that people who are poor at dealing
with emotions and emotional events would find it hard to maintain their moral
character over time. As it is, moral judgments are nearly always affected by our
emotions, and being bad at dealing with emotions would seem to imply being
bad at dealing with moral situations.
In discussing emotions and emotional skills, however, we must be wary not
to transform emotional intelligence into something it is not. Emotional skills
are merely one subset of all human skills. Cognitive abilities, emotional abil~
ities, and various other skills and talents are important in making us complete
individuals. And, as we know, any human skill can be used to achieve destruc~
tive goals. Just as an individual who has the IQ of a genius could use her or his
intelligence to hurt others (the notorious Hannibal Lecter of The Silence of the
Lambs comes to mind), so can a person who is high in emotional intelligence
use his or her skills to manipulate and hurt others. Being good at knowing how
others feel, regulating the emotions of others, and controlling one's display of
emotions are all skills that are prerequisites for any great leader, whether she or
he chooses to lead people to do good things or evil things. Emotional intelli~
gence is therefore not a cure~all for the ills of society. If tomorrow everyone in
12. Being and Becoming a Good Person 263

the world became emotionally intelligent, the world still might not be a para~
dise. However, by understanding the role of various emotional processes in the
development of morality and in our everyday moral behavior, we are that much
closer to being effective moral agents and effective moral educators.

Teachers' Questions and Answers

Q: In the wake of the Columbine tragedy, what kinds of interventions can
schools implement to help violent or withdrawn students deal more effectively
with their emotions? How early should these interventions take place? Is there
a point where it is simply too late to effect change?
A: Recent violent incidents reported in the news media have made the
mental health of our youth very salient. Partly in response to these incidents,
schools across the nation have implemented programs in an attempt to pre~
empt any future tragedies [according to one count, more than 300 such
programs are in place in the United States alone (Cohen, 1999)]. Although
they often go by different names (character education, positive youth develop~
ment, emotional intelligence, emotional literacy, social~emotional learning),
they usually have as their main goal the teaching of skills surrounding the
effective management of emotions, the building of healthy social relationships,
and the achievement of positive social and personal goals.
It is too early to offer a critical evaluation of the success of these programs.
However, there have been some optimistic reports. For instance, one of the first
of such programs (instituted in the public schools of New Haven, CT), has con~
tributed to the reduction of school violence and feelings of hopelessness among
students (Shriver, Schwab~Stone, & DeFalco, 1999). A conflict~resolution pro~
gram in New York City (Resolving Conflict Creatively) has also contributed to a
reduction in aggressive behavior; children who received more conflict~resolution
lessons were less aggressive overall (Aber, Brown, & Henrich, 1999). These early
findings provide some assurance that we are headed in the right direction.
As far as the ideal age of implementation, the easiest answer is the earlier
the better. This is not to say that older children and adults cannot benefit from
such training. It seems as if old dogs can learn new tricks when it comes to
emotional skills. It is never too late to teach a child to take the perspective of
others, for instance, or to teach children to reappraise situations so as not to
feel overwhelmed with violent emotions.
If there is a take-home message, however, it is that there is still much to
learn about the motives of children such as those involved in the Columbine
and Jonesboro incidents. It would be a mistake to say that emotional intelli-
gence training could have prevented such a tragedy--we just do not know at
this point. However, the hope remains that by paying closer attention to the
social and emotional well-being of children at high risk for such behaviors, we
may be able to prevent such tragedies in the future.
264 David A. Pizarro and Peter Salovey

References
Aber, J. L., Brown, J. L., & Henrich, C. C. (1999). Teaching conflict resolution: An effective school-based
approach to violence prevention. New York: National Center for Children in Poverty, The Josephine L.
Mailman School of Public Health, Columbia University.
Arsenio, W. F., & Kramer, R. (1992). Victimizers and their victims: Children's conceptions of the
mixed emotional consequences of moral transgressions. Child Development, 63, 915-927.
Barnett, M. A. (1987). Empathy and related responses in children. In N. Eisenberg & J. Strayer
(Eds.), Empathy and its development.. Cambridge studies in social and emotional development (pp.
146-162). New York: Cambridge University Press.
Batson, C. D. (1991). Tile altruism question.. Toward a social-psychological answer. Hillsdale, NJ: Erlbaum.
Batson, C. D., Dyck, J. L, Brandt, J. R., Batson, J. G., Powell, A. L., McMaster, M. R., & Griffitt, C.
(1991). Five studies testing two new egoistic alternatives to the empathy-altruism hypothesis.
In B. Puka (Ed.), Reaching out: Caring, altruism, and prosocial behavior. Moral development:A compendium
(Vol. 7, pp. 76- I 01). New York: Garland.
Berkowitz, M. W. (1995). The education of the completemoral person. Aberdeen: Gordon Cook Foundation.
Betancourt, H. (1990). An attribution~empathy model of helping behavior: Behavioral intentions
and judgments of help-giving. Personality and Social Psychology Bulletin, 16, 573-591.
Blair, R. J. R. (1995). A cognitive developmental approach to morality: Investigating the psycho-
path. Cognition, 57, 1-29.
Blasi, A. (1999). Emotions and moral motivation. Journal for the Theory of Social Befiaviour, 29, 1-19.
Cohen, J. (1999). Social and emotional learning past and present: A psychoeducational dialogue.
In J. Cohen (Ed.), Educating minds and hearts.. Social emotional learning and the passage into adolescence
(pp. 2-23). New York: Teachers College Press.
Damasio, A. R. (1994). Descartes" error: Emotion, reason, and the human brain. New York: Avon Books.
Davis, (1996) Empathy.. A social psychological approach. Boulder, CO: Westview Press.
Davis, M. H., Mitchell, K. V., Hall, J. A., Lothert, J., Snapp, T., & Meyer, M. (1999). Empathy,
expectations, and situational preferences: Personality influences on the decision to participate
in volunteer helping behaviors. Journal of Personality, 67, 469-503.
Easterbrook, J. A. (1959). The effect of emotion on cue ultization and the organization of behavior.
Psychological Review, 66, 183-201.
Eisenberg, N. (1986). Altruistic emotion, cognition, and behavior. Hillsdale, NJ: Erlbaum.
Eisenberg, N., & Miller, P. (1987). Empathy and prosocial behavior. Psychological Bulletin, 17,
273-282.
Feshbach, N., & Roe, K. (1968). Empathy in six- and seven-year olds. Child Development, 39, 133-145.
Freud, S. (1977). Introductory lectures on psychoanalysis (J. Strachey, Trans.). New York: Norton.
(Original work published 1920).
Gibbs, J. C. (1987). Social processes in delinquency: The need to facilitate empathy as well as
sociomoral reasoning. In W. M. Kurtines & J. L.Gewirtz (Eds.), Moral development through social
interaction (pp. 301-321). New York: Wiley.
Goleman, D. (1995). Emotional intelligence.. Why it can matter more than/Q. New York: Bantam.
Gottman, J. M., Katz, L. F., & Hooven, C. (1997). Meta-emotion.. How families communicate emotionally.
Mahwah, NJ: Erlbaum.
Gross, J. J. (1999). The emerging field of emotion regulation: An integrative review. Review of General
Psychology, 2, 271-299.
Hodges, S. D., & Wegner, D. M. (1997). Automatic and controlled empathy. In W. Ickes (Ed.),
Empathic accuracy (pp. 311-339). New York: Guilford Press.
Hoffman, M. L. (I 987). The contribution of empathy to justice and moral judgment. In N. Eisenberg
& J. Strayer (Eds.), Empathy and its development. New York: Cambridge University Press.
Hoffman, M. L. (1998). Varieties of empathy-based guilt. In J. Bybee (Ed.), Guilt and children. Boston:
Academic Press.
Hoffman M. L., & Saltszein, H. D. (1967). Parental discipline and the child's moral development.
Journal of Personality and Social Psychology, 5, 45-57.
12. Beingand Becoming a Good Person 265

Isen, A. M. (1993). Positive affect and decision making. In M. Lewis & J. M Haviland (Eds.), The
handbook of emotions (pp. 261-277). New York: Guilford.
Kochanska, G. (1995). Children's temperament, mother's discipline, and security of attachment:
Multiple pathways to emerging internalization. Child Development, 66, 597-615.
Kochanska, G. (I 997). Multiple pathways to conscience for children with different temperaments:
From toddlerhood to age 5. Developmental Psychology, 33, 228-240.
Kohlberg, L. (1969). Stage and sequence: The cognitive-developmental approach to socialization.
In D. A. Goslin, (Ed.), Handbook of socialization theory and research (pp. 347-480). Chicago: Rand
McNally.
Kohn, A. (1997). The trouble with character education. In A. Molnar (Ed.), The construction of children's
character (pp. 154-162). Chicago: National Society for the Study of Education.
Lazarus, R. S. (1991). Emotion and adaptation. New York: Oxford University Press.
Leeper, R. W. (1948). A motivational theory of emotion to replace "emotion as diorganized
response." Psychological Review, 55, 5-21.
Lepper, M. R., Greene, D., & Nisbett, R. E. (1973). Undermining children's intrinsic interest with
extrinsic reward: A test of the "overjustification" hypothesis./ournal of Personality and Social
Psychology, 28, 129-137.
Lopes, P., & Salovey, P. (in press). Emotional intelligence and social~emotional learning. In H.J.
Walberg, M.C. Wang, R.P. Weissberg, & J.E. Zins (Eds.), Building school successon social and emotional
learning. New York: Teachers College Press.
Mandler, G. (1975). Mind and emotion. New York: Wiley.
Matthews, K. A., Batson, C. D., Horn, J. & Rosenman, R. H. (1981). "Principles in his nature which
interest him in the fortune of others": The heritability of empathic concern for others.//ournal of
Personality, 49, 237-247.
Mayer, J. D., & Salovey, P. (1997). What is emotional intelligence? In P. Salovey & D. Sluyter (Eds.),
Emotional development and emotional intelligence..Implications for educators (pp. 3-31). New York: Basic
Books.
Mayer, J. D., Salovey, P., & Caruso, D. (2000). Models of emotional intelligence. In R. J. Sternberg
(Ed.), Handbook of intelligence (pp. 396-420). New York: Cambridge University Press.
Mehrabin, A. (1980). Basic dimensions for a general psychological theory.. Implications for personality, social,
environmental, and developmental studies. Cambridge: Oelgeschlager, Gunn, & Hain.
Metcalfe, J., & Mischel, W. (1999). A hot/cool system analysis of delay of gratification: Dynamics of
willpower. Psychological Review, 106, 3-19.
Mussen, P., & Eisenberg~Berg, N. (1977). Roots of caring, sharing, and helping.. The development of pro~
social behavior in children. San Francisco: Freeman.
Palfai, T. P., & Salovey, P. (1993). The influence of depressed and elated mood on deductive and
inductive reasoning. Imagination, Cognition, and Personality, I3, 57-71.
Petrie, K. J., Booth, R. J., Pennebaker, J. W., & Davison, K. P. (1995). Disclosure of trauma and
immune response to a hepatitis B vaccination program. ]ournal of Consulting and Clinical
Psychology, 63, 787-792.
Piaget, J. (1932). The moral ~udgment of the child. New York: Harcourt, Brace.
Plato (1988). Phaedrus. (Trans C.J. Rowe.) Warminster, England: Aris & Phillips.
Rozin, P., Haidt, J., & McCauley, C. R. (1993). Disgust. In M. Lewis & J. Haviland (Eds.), Handbook of
emotions (pp. 575-594). New York: Guilford Press.
Rozin, P., Markwith, M., & Stoess, C. (1997). Moralization and becoming a vegetarian: The transform~
ation of preferences into values and the recruitment of disgust. Psychological Science, 8, 67-73.
Saarni, C. (1999). The development of emotional competence. New York: Guilford Press.
Salovey, P. (1992). Mood induced self-focused attention, lournaI of Personality and Social Psychology,
62, 699-707.
Salovey, P., & Birnbaum, D. (1989). Influence of mood on health~relevant cognitions. ]ournal of
Personality and Social Psychology, 57, 539-551.
Salovey, P., & Mayer, J. D. (1990). Emotional intelligence. Imagination, Cognition, and Personality, 9,
185-211.
266 David A. Pizarro and Peter Salovey

Salovey, P., Woolery, A., & Mayer, J.D. (2000). Emotional intelligence: Conceptualization and
measurement. In G. Fletcher & M. Clark (Eds.), The Blackwell handbook of social psychology:
Interpersonal processes(pp. 279-307). London: Blackwell.
Schwarz, N. (1990). Feelings as information: Informational and motivational functions of affective
states. In E. T. Higgins & R. Sorrentino (Eds.), Handbook of motivation and cognition: Foundations of
social behavior (Vol. 2, pp. 527-561). Elmsford, NY: Pergamon Press.
Shriver, T.P., Schwab-Stone, M., & DeFalco, K. (1999). Why SEL is the better way: The New Haven
Social Development Program. In J. Cohen (Ed.), Educating minds and hearts: Social emotional
learning and the passageinto adolescence(pp. 43-60). New York: Teachers College Press.
Simon, H. (1981). Comments. In M. S. Clark & S. T. Fiske (Eds.), Affect and cognition (pp. 333-342).
Hillsdale, NJ: Erlbaum.
Tangney, J. P. (1999). The self-conscious emotions: Shame, guilt, embarrassment and pride. In T.
Dalgleish & M. J Power (Eds.), Handbook of cognition and emotion (pp. 541-568). Chichester: Wiley.
Wisp~, L. (1987). History of the concept of empathy. In N. Eisenberg & J. Strayer (Eds.), Empathy and
its development:Cambridgestudies in social and emotionaldevelopment.New York: Cambridge University
Press.
 CHAPTER

Mozart and the Mind: Factual

and Fictional Effects of Musical
Enrichment
FRANCES H. RAUSCHER
Department of Psychology
University of Wisconsin, Osfikosfi
9 Wisconsin

The "Mozart effect," a term coined by the Los Angeles Times, refers to the
finding that college students who listened to the first 10 minutes of a Mozart
sonata (K.448) scored higher on a spatial~temporal reasoning task immediately
afterward--an effect that lasted approximately 10 minutes. The original re~
search report, first published by my colleagues and me in the journal Nature
(Rauscher, Shaw, & Ky, 1993), received a disproportionate amount of attention
from the popular press. To our horror, the finding has spawned a Mozart effect
industry which includes books, CDs, web sites, and all manner of hyperbole.
Articles with titles such as "Mozart Makes You Smarter" and "Mozart Makes
the Brain Hum" have led readers to believe that classical music in general, and
Mozart in particular, can improve babies' math scores later in life, improve
scores on the Scholastic Aptitude Test (SAT), and turn average healthy children
into Einsteins. Unfortunately, press reports of scientific findings are powerfully
seductive to parents, educators, and policymakers. In fact, Georgia Governor
Zell Miller, based on his understanding of these results, asked legislators to
purchase classical music CDs for every newborn baby in the state. "No one
doubts that listening to music, especially at an early age, affects spatial~

ImprovingAcademicAchievement
Copyright 2002, Elsevier Science (USA). All rights reserved. 267
268 Frances H. Rauscher

temporal reasoning that underlies math, engineering, and chess." Far from no
one doubting it, there is no evidence at all for the claim that listening to
classical music CDs improves children's spatial~temporal reasoning or any
other aspect of intelligence, for that matter. The scientific reports made no
claims about general intelligence, SAT scores, or babies.
Although the term Mozart effect initially referred to the transitory increase in
certain visuospatial task scores following listening to a particular Mozart
sonata, the phrase has generalized to include, for example, the effects of
music instruction on spatial~temporal task performance. In numerous studies
preschoolers, kindergartners, and second graders who received piano instruc~
tion scored higher on spatial~temporal tasks than control groups who received
other instruction or no training (see, for example, Rauscher, 1999). My goal in
this chapter is to share with you the most recent research in this area. Because
this is an area where there is considerable temptation to overstate the scien~
tific findings in the interest of a particular advocacy position, I caution the
reader to employ a conservative interpretation of the data presented here.
Although the research has strong implications for policy and practice, it is
important to keep in mind that these studies were designed with an eye toward
determining the parameters of a scientific effect rather than with an eye toward
application. Questions to be addressed include: (1) What have researchers
discovered about instrumental instruction and spatial~temporal reasoning?
(2) What is the best age to begin instrumental instruction for spatial~temporal
enhancement? (3) How long do the effects of instrumental instruction on
spatial~temporal reasoning persist? (4) How might musical experiences affect
cognitive development? (5) Does enhancing spatial~temporal reasoning im~
prove mathematics scores? (6) What are the implications of this research for
educators and public policy?
To understand the effects of instrumental instruction on children's spatial~
temporal reasoning, it is necessary to understand what is meant by "spatial~
temporal." Virtually every healthy human has some degree of spatial~temporal
intelligence. To maneuver an armchair through a doorway and around a corner,
for example, one needs to picture its shape and which way to turn it before one
lifts it. Even animals can reason spatially. One researcher in the field of spatial
cognition, for example, reports an event he noticed while watching a German
shepherd play fetch with its owner (Cooper & Shepard, 1990). The owner threw
a long stick over the backyard fence, and the dog pranced over and immedi-
ately placed his head through an opening where a board in the fence had fallen
off. The dog grabbed the stick horizontally, jerked backward and, one moment
before the stick would have rammed the fence, rotated his head 90 ~ to pull the
stick neatly through the hole in the vertical direction. One can never know
whether the dog's foresight was conscious but, writes the researcher, "Might
[there] not have been a preparatory mental rotation of the stick," a rotation, I
might add, similar to the one that lets you picture this whole episode in your
head as you read? Children show this form of intelligence as soon as they start
13. Mozartandthe Mind 269

building block towers or putting together puzzles, and later as they reason
about ratios and fractions. Individuals with highly developed spatial skills often
become architects, sculptors, engineers, graphic designers, painters, mathem~
aticians, physicists--and musicians.
Lois Hetland, a researcher from Harvard University, recently published a
statistical review (i.e., meta~analysis) of all the causal studies she could find
that explored the effects of instrumental instruction on spatial abilities (Het~
land, 2000). Although not all studies showed positive effects, overall the data
were convincing. Hetland concluded, "Active instruction in music does appear
to enhance spatial~temporal performance for preschool and elementary~aged
children, at least while instruction is occurring and at least up through two
years of instruction. The effect is...remarkably consistent across this popular
tion of studies .... It is a solid finding."
The typical study included in Hetland's analysis compared spatial~temporal
scores of two to four groups of children. One group received music instruction
and the other(s) received either no instruction or instruction in an alternative
activity to control for the Hawthorne effect. ~ Instruction was provided either
individually or in groups of approximately 10 children, and lasted 10 to 60
minutes for periods ranging from 6 weeks to 2 years. In most studies children
were taught the piano or xylophone, generally in conjunction with Orff or
Kod~ly techniques that included listening, singing, movement, and learning
to read music. Spatial~temporal reasoning was typically tested before and after
instruction began.
The primary outcome of Hetland's analysis was that active music instruction
led to dramatic improvement of children's spatial~temporal task scores. In
addition to this finding, Hetland made several other interesting discoveries. To
determine if the effect was stronger for younger than older children, Hetland
compared the posttest scores of children aged 3 to 5 years with those of children
aged 6 to 12. Her results suggest that the spatial abilities of younger children are
more enhanced by active music instruction than are those of older children. This
finding is consistent with the notion that the age at which music instruction
begins is related to structural changes in the brain. The 4~year study reported
below illustrates the importance of Hetland's finding.

SCHOOL DISTRICT OF KETTLE-MORAINE STUDY

Kindergarten
Children from four kindergarten classrooms at two Wisconsin public elemen-
tary schools in the school district of Kettle~Moraine participated. Some chil~
dren received piano keyboard instruction (keyboard group) and others received

lThe Hawthorneeffect is the phenomenonthat wheneverone introducessomethingnew into a

curriculum or program,it has an enhancingeffect on a varietyof behaviors.
2 70 Frances H. Rauscher

no special training (no music group). We began by pretesting all the children
using two spatial~temporal tasks, a puzzle~solving task and a block~building
task, and one pictorial memory task. Based on previous research, we predicted
improvement for the spatial~temporal tasks only.
Immediately following the pretesting, a music specialist visited each class~
room to provide the keyboard group with 20-minute piano lessons two times
per week for the remainder of the school year (8 months). Children were taught
in groups of 8 to 10. The instruction involved singing and moving to the
compositions of the current and subsequent weeks, rhythmic clapping and
solf~ge, ear training, music notation, improvisation, interval and dynamic
exercises, and keyboard performance. The children in the no music group
engaged in journaling activities with their kindergarten teacher during music
lessons. These children were not permitted access to the keyboards.
Children were posttested twice, once following 4 months of lessons and a
second time following 8 months. For both spatial~temporal tasks the children
who had received the keyboard lessons scored significantly better than the
children who had not. Although no differences in pretest scores were found
between the two groups of children, after only 4 months of instruction the
keyboard group's puzzle~solving scores were 38% higher than those of the no
music group. The keyboard group's scores had improved by 52%. These
enhancements were similar in magnitude to those found in similar studies
using preschoolers as participants, despite the chaotic setting of the kinder~
garten classroom and the participation of older (5~ to 6~year~old) children.
Furthermore, after 8 months of lessons the difference between the two groups'
scores had increased in magnitude. The keyboard group scored 46% higher
than the no music group, representing a 65% improvement. Similar results were
found for the other spatial-temporal task, block building. As predicted, scores
on the memory task did not differ significantly for the two groups following
lessons. Thus, in contrast to what the Mozart zealots would claim, music
training effects were limited to spatial-temporal skills; they did not generalize
to other skills.

First Grade
We returned to the schools the following year, at the end of the first grade, to
retest the children. The school district had partially expanded its kindergarten
keyboard program into the first grade. Therefore, some first grade children
received keyboard lessons and others did not, depending on the logistics of
classroom assignment. We thus had three groups of children to retest. Some
children received keyboard instruction for 1 year (in kindergarten) and did not
receive instruction the second year (in first grade). These children therefore had
a 1~yeargap in their instruction, after which they were retested. A second group
of children received music instruction for 2 years (in kindergarten and first
13. Mozart and the Mind 27 ]

grade) and were retested after each year of instruction. Finally, a third group
received no music instruction at all. All children were tested using the same
three tasks used earlier.
Results indicated that the children who received keyboard instruction only
in kindergarten scored 15% lower on the puzzle~solving task 1 year after their
lessons had ended. In fact, these children's scores were not significantly
different from the scores of the children who had never received lessons.
However, the scores of the children who continued lessons through the first
grade had increased by approximately 17% since kindergarten. Finally, the
children who received no lessons showed only the improvement one would
expect from age. The blocbbuilding task followed a similar trend. Again, no
effects were found for the memory task. These data suggest that I year only of
keyboard music instruction wilt not induce long.~term effects on spatial.~temporal task
performance. Either the instruction must continue indefinitely for the effects
to persist (use it or lose it), or some critical amount of training is required
to produce lasting effects on spatial cognition. Unfortunately, it is too early in
the research to determine which of these two explanations is correct.

Second Grade
The following year we returned to the schools to retest the children yet again.
After viewing the data, the district superintendent had decided to provide
keyboard lessons to all her elementary school children. All children were to
receive instruction every year. This decision provided us with three groups of
children to retest, all of whom had participated in our study in previous
years. One group had received keyboard instruction in kindergarten and
second grade only (not in first grade), a second group had received the
instruction in all three grades (kindergarten, first grade, and second grade),
and a third group had received instruction in the second grade only. We
administered the same three tasks as before, after the children had completed
the second grade.
The data show that the children who received lessons in kindergarten and
second grade, but not in first grade, improved by approximately 37% after their
lessons were reinitiated in the second grade. The children who received lessons
for all 3 years continued to improve, although the improvement from first to
second grade was not significant (14%). This may be due to the presence of a
ceiling effect. 2 Finally, children who received the lessons in the second grade
only did not improve significantly. Consistent with Hetland's (2000) analysis,
these data suggest that the effects of keyboard instruction on spatial~temporal
task performance are found primarily for those children who begin training at
the earliest ages.

2"Ceiling effect" refersto the failure to observeany improvementin performanceowingto the

fact that the participant is alreadyperformingat maximumcapacity.
272 Frances H. Rauscher

Third Grade
We collected additional data from these children the following year, after they
had completed the third grade. This time we used a more difficult version of the
puzzle~solving task. The block~building and memory tasks were not adminis~
tered.
As before, three groups of children were retested, some of whom received
lessons in kindergarten, second grade, and third grade, some of whom re~
ceived lessons from kindergarten through third grade, and some of whom
received lessons in second and third grades only. Because of the difference
in task difficulty between the test items we administered in third grade
and those administered earlier, we did not compare the data collected
following the third grade with those collected in previous years. However,
we were able to compare the scores of the three groups of third grade
children.
The data are compelling. The children who received keyboard instruction for
4 consecutive years (through the third grade) scored 30% higher on the task
than children who received instruction in kindergarten, second grade, and third
grade, and 52% higher than the children who began instruction in second
grade. This lends further support to the importance of beginning the instruc~
tion early.

Summary
Consistent with previous studies this longitudinal study found that young
children who were provided with music instruction scored higher on spatial~
temporal tasks compared with children who did not receive the instruction.
The effect was significant after 4 months of instruction. No enhancement was
found for a nonspatial task: pictorial memory. However, when the music
instruction was terminated the children's scores began to decrease. The
children who received instruction over the entire 4 years of the study con~
tinued to score higher on the spatiaMemporal tasks. Finally, scores of the
children who began instruction in the second grade did not improve signifi-
cantly, and these children continued to score lower than all other groups in
the third grade.
Lois Hetland's meta~analysis provides further information regarding several
variables of interest to researchers and educators. In addition to age~of-onset
differences, she found that one-on-one instruction may lead to stronger
spatial skills than group lessons, although group lessons, as demonstrated
above, do appear to be effective. Furthermore, Hetland's analysis revealed
that instruction on the keyboard, rather than another musical instrument,
may not be necessary for spatial enhancement, although she recommends
caution in interpreting this finding: Only five of the studies included in her
analysis did not include keyboard instruction. Also, the inclusion of mover
13. Mozart and the Mind 273

ment in the music instruction did not affect spatial skills. Programs that in-
cluded movement produced similar effect sizes as those that did not include
movement. Finally, learning to read music may play a role. Although learning
standard musical notation does not appear to be necessary for spatial
enhancement, programs of music instruction that included literacy resulted
in greater spatial-temporal enhancements than programs of instruction that
did not.

THEORETICAL INTERPRETATIONS

The effects of music instruction on spatial-temporal abilities have been ex-

plained by two types of theories. Neuroscientific theories assert that music
instruction induces physiological changes in brain structure that consequently
affect spatial-temporal processing (Leng & Shaw, 1991). Indeed, recent re-
search suggests that the brains of musicians are different from those of
nonmusicians. For example, two structural magnetic resonance imaging (MRI)
studies have found that musicians who began piano instruction prior to age
6 or 7 had larger corpus callosi and greater asymmetry of the planum tempor-
ale (the brain's sound signal processor) relative to nonmusicians (Schlaug,
Jancke, Huang, & Steinmetz, 1995; Schlaug, Jancke, Huang, Staiger, & Stein-
metz, 1995). Furthermore, violinists who began training prior to age 12 dis-
played greater cortical representation of the digits of the left hand than
nonmusicians (Elbert, Pantev, Wienbruch, Rockstroh, & Taub, 1995). Finally,
one study found that musicians who started playing before age 9 showed
greater auditory cortical representation than those who began instruction
after age 9 or nonmusicians (Pantev, Oostenveld, Engelien, Ross, Roberts, &
Mannfried, 1998). Again, there was a significant positive correlation between
effect size and the age at which subjects initiated instruction: musicians who
began instruction before age 9 displayed the largest effects. These differences
in the brains of musicians and nonmusicians may be related to findings of
improved spatial-temporal abilities in children who began music instruction at
an early age.
Transfer theories, on the other hand, suggest that playing a musical instru-
ment and performing a spatial-temporal task require similar cognitive skills,
and thus the skills involved in making music may transfer to spatial-temporal
task performance (Rauscher, 1999). One approach to examining the nature of
the relationship between music and spatial-temporal reasoning is to analyze
the cognitive requirements shared by these two domains. For example, several
of the musical elements described by Serafine (1988), including temporal
succession, nontemporal closure, transformation, and abstraction, may have
parallel elements in the visuospatial domain. Perhaps the cognitive skills
required to process this type of information are used in performing both
musical and spatial-temporal tasks.
2 74 Frances H. Rauscher

THE LINK BETWEEN MUSIC AND MATH

An important practical question remains: Will children who score higher on

spatial~temporal tasks as a function of music instruction also score higher
on mathematical tasks? Although significant correlations have been found
between spatial~temporal task performance and mathematical ability, only
two studies have addressed the hypothesis that music instruction affects
mathematical reasoning. The first study (Gardiner, Fox, Knowles, & Jeffrey,
1996) found that first and second grade children who received 7 months of
supplementary music and visual arts classes achieved higher standardized
mathematics scores than children who received the schools' typical music
and arts training. However, because the two treatments were initiated together
it is difficult to determine which intervention, music or art training, may have
been responsible for the improvement.
The second study (Graziano, Peterson, & Shaw, 1999) compared the mathe~
matical reasoning (in particular reasoning about ratios and fractions) of second
grade children assigned to four groups: (1) keyboard instruction coupled with
exposure to a computer game designed to develop spatial~temporal reasoning;
(2) English instruction coupled with the same spatial-temporal training; (3)
spatial-temporal training only; (4) no treatment. Results indicated that the
mathematical reasoning scores of the children whose treatment included the
music instruction were significantly higher than those of the children in the
other groups. It is unfortunate that the researchers did not include a fifth group
of students who received keyboard instruction only. However, this study does
suggest that music instruction may enhance reasoning related to certain
mathematical abilities, and confirms the role of spatial~temporal reasoning in
some mathematical operations.

IMPLICATIONS FOR PUBLIC POLICY

The research reported in this paper has public policy implications. It seems
clear that children derive measurable educational benefits from music training
beyond those directly related to music. I believe that the results of these
studies must be included in music education advocacy efforts. Arguments
that emphasize the extra-musical benefits of music instruction are effective
and have saved school music programs. Disadvantaged children, whose care~
givers can afford neither the time nor the money to provide music lessons,
stand to lose the most if school music programs are cut back or eliminated. I
suggest that music advocates use all available evidence to convince policy~
makers of the importance of a music education for all our children.
Nevertheless, I feel it is important to acknowledge the possible dangers
associated with an argument of music for math's sake. Care must be taken to
ensure that scientific goals do not displace developmentally appropriate in-
13. Mozart and the Mind 275

struction. Decisions regarding music education curricula should be based on

musical goals only. Consistent with recent recommendations of the National
Association for the Education of Young Children (Bredekamp & Copple, 1997),
a position statement containing guidelines for the establishment of age~ap~
propriate music curriculum has been published by the Music Educator's Na~
tional Conference (TheSchoolMusic Program, 1994). MENC recommends a focus
on singing, listening, movement, instrumental instruction, creativity, and music
literacy as well as the development of musical knowledge of melody, rhythm,
timbre, and form. Musical play is also highly recommended, as is the encour~
agement of individual creativity. Kenney (1997) outlines specific teaching
strategies relevant to these instructional goals for newborns to children aged
8. I encourage scientists and educators to attend carefully to these guidelines
when considering the application of these research findings.
John Bruer, president of the James S. McDonnell Foundation and a leader in
the funding of educational research, cautions us that "neither neuroscientists
nor behavioral scientists have the vaguest notion of how differences in brains
translate into differences in IQ or how a brain that can pass third grade differs
from one that cannot" (Bruer, 1994). He further comments that "I don't want
to discount [brain research] because eventually we will know much more. In
twenty years, it's conceivable we will understand the brain circuitry involved in
reading, for example, and how learning to read changes neural circuitry as the
skills mature." However, today's students and teachers cannot wait 20 years
for neuroscience to unequivocally demonstrate the nature of the link between
brain function and cognition. The current research suggests that music instruc~
tion improves children's spatial abilities, whether due to neurophysiological
mechanisms or not. I believe that to exclude this research from discussions
arguing for music in the schools is to do a disservice to the children whose lives
will be affected when music programs are eliminated. Yes, much more research
is needed to provide converging evidence, and no, music is not a panacea for
poor academic achievement. However, it seems clear that music has benefits
to intellectual development that transcend music itself.

Teachers' Questions and Answers

Q: You say that the original effect of listening to Mozart on test perform~
ance lasted only l O minutes. Wl~at does this mean, exactly? Does it mean that
10 minutes into the test, there was no difference between a group that listened
to Mozart and a group that did something else? What if there was music playing
constantly say before, during, and after a test, or piped into classrooms
constantly, could this limited effect be stretched out? If not, why would such
an effect be so ephemeral?

A: Our claim that the cognitive effects of listening to Mozart are short~
term came from the fact that a l O-minute delay between music exposure and
spatial~temporal testing eradicated the enhancement. In delay circumstances,
2 76 Frances H. Rauscher

participants' scores were essentially the same as when they had been exposed
to relaxation instructions or silence. The answer to your question regarding why
the effect is so ephemeral ultimately rests in determining the cause of the
Mozart effect. There are some data suggesting that the effect is due to arousal,
which would suggest that when the music~induced arousal ends, so does the
cognitive enhancement. I find this argument plausible, but it does not account
for all of the data. For example, my own research finds that other arousing
music, such as Mendelssohn, does not improve performance on spatial~tem-
poral tasks. Why would a Mozart sonata improve task performance while a
Mendelssohn symphony, which subjects reported as equally arousing, not
affect performance? What we need is an experiment in which actual physio~
logical measurements (e.g., heart rate and galvanic skin response) are made of
participants before, during, and after exposure to Mozart and other music and
also during a subsequent spatial-temporal test. Only then can we determine if
arousal is a viable explanation.
Q: You argue very passionately for the importance of music programs. Is
your assumption that because music can improve spatial-temporal reasoning,
we should use it? Is there actually an advantage to developing these skills
indirectly through music, as opposed to teaching children spatial-temporal
skills directly, for example, by giving them lots of puzzles to solve?
A: I strongly believe that music should be included in the core curriculum
for the beauty and joy it brings into our lives. The fact that music instruction
has also been shown to enhance children's spatial abilities is an added benefit.
I am often asked by my own music specialists if there is anything "special" they
should be doing to enhance the children's spatial abilities. I tell them that they
should teach the children using their best musical judgment, and the effects
will follow.
A study by Gordon Shaw and his colleagues at the University of California,
Irvine, directly pertains to the second portion of your question. Shaw com~
pared the spatial~temporal reasoning scores of several groups of children,
including one group who received spatial~temporal training alone and another
group who received the same spatial~temporal training coupled with piano key-
board instruction. The spatial~temporal training consisted of several months of
playing a computer game designed to train spatial skills. The children were then
tested using items similar to those used in the computer training. Although
both groups scored higher than a group of children who received language or
no special training, results indicated that the group who studied the piano
along with the spatial~temporal training scored significantly higher than the
group who received the spatial~temporal training alone. This is particularly
interesting because the children who received the spatial~temporal training
alone had twice as much direct spatial~temporal training (via the computer)
than the children in the keyboard group. (For example, if the spatial~temporal
alone group was given 40 minutes of spatial~temporal computer training, then
13. Mozart and the Mind 277

the spatial~temporal/piano group would be given 20 minutes of computer

training and 20 minutes of piano training.) It seems that adding the music
instruction to the training provided the children with an advantage that the
computer training did not. Perhaps it is the multisensory nature of the piano
instruction, which requires kinesthetic, auditory, and visual skills, that contrib~
utes to its effects on spatial learning. Only further research will tell.
Q: Does the Mozart effect work both ways? That is, is there any evidence
to suggest that practicing with puzzles or other spatial~temporal activities
improves one's ability to learn music?
A: No one has explored the possibility that spatial~temporal training can
improve music learning, although I think it is a feasible hypothesis. I have
thought about conducting such an experiment for quite some time, but have
encountered major difficulties in the design of the study. For example, what
would be a suitable measure of "music learning"? Those of us who study music
cognition recognize that learning to play a musical instrument requires an
integration of several types of knowledge, physical coordination, listening skills
(including aural discrimination for pitch, duration, and intensity), planning
skills, and so forth.
I am currently conducting a study with Head Start children to try to separate
some of the musical skills that may be affecting cognitive performance. We are
providing the children with different types of music instruction (e.g., rhythmic
training, singing instruction, and piano instruction) and are then testing a
broad range of spatial abilities. I am hoping that the information gained from
this study will lead to a better understanding of the components of musical
learning that may be contributing to the enhancement. If we can eventually
determine precisely which aspect(s) of music learning is affecting spatial~
temporal skills, perhaps then we can determine if the effect works both
ways.

References
Bredekamp, S., & Copple, C. (Eds.) (1997). Developmentally appropriate practice in early childhood
programs: Revised edition. Washington, DC: National Association for the Education of Young
Children.
Bruer, J. (1994). Schools for thought: A science of learning in the classroom. Bradford, UK: Bradford Books.
Cooper, L. A., & Shepard, R. N. (1990). Turning something over in the mind. In R. R. Llinas (Ed.), The
workings of the brain.. Development, memory, and perception. New York: Freeman.
Elbert, T., Pantev, C., Wienbruch, C., Rockstroh, B., & Taub, E. (1995). Increased cortical repre~
sentation of the fingers of the left hand in string players. Science, 270, 305-307.
Gardiner, M. F., Fox, A., Knowles, F., & Jeffrey, D. (1996). Learning improved by arts training. Nature,
38/, 254.
Graziano, A., Peterson, M., & Shaw, G. L. (1999). Enhanced learning of proportional math through
music training and spatial~temporal training. Neurological Research, 21, 139-152.
Hetland, L. (2000). Learning to make music enhances spatial reasoning. Journal of Aesthetic Education,
34, 179-238.
278 Frances H. Rauscher

Kenney, S. H. (1997). Music in the developmentally appropriate integrated curriculum. In C. H.

Hart, D. C. Burts, & R. Charlesworth (Eds.), Integrated curriculum and developmentally appropriate
practice. Albany, NY: SUNY Press.
Leng, X., & Shaw, G. L. (1991). Toward a neural theory of higher brain function using music as a
window. Concepts in Neuroscience, 2, 229-258.
Pantev, C., Ooostenveld, R., Engelien, A., Ross, B., Roberts, L. E., & Manfried, H. (1998). Increased
auditory cortical representation in musicians. Nature, 392, 811-813.
Rauscher, F. H. (1999). Music exposure and the development of spatial intelligence in children.
Bulletin of the Council for Research in Music Education, I42, 35-47.
Rauscher, F. H., Shaw, G. L., & Ky, K. N. (1993). Music and spatial talk performance. Nature, 365,
611.
Schlaug, G., Jancke, L., Huang, Y., & Steinmetz, H. (1995a). In vivo evidence of structural brain
asymmetry in musicians. Science, 267, 699-701.
Schlaug, G., Jancke, L., Huang, Y., Staiger, J. F., & Steinmetz, H. (1995b). Increased corpus callosum
size in musicians. Neuropsychologia, 33, 1047-1055.
The school music program: A new vision. (1994). MENC: Reston, VA: Music Educators National
Conference.
Serafine, M. L. (1988). Music as cognition: The development of thought in sound. New York: Columbia
University Press.

Suggested Reading
Hetland, L. (2000). Learning to make music enhances spatial reasoning. Iournal ofAesthetic Education,
34, 179-238.
Rauscher, F. H., & Zupan, M. (2000). Classroom keyboard instruction improves kindergarten
children's spatial~temporal performance: A field experiment. Early Childhood Research Ouarterly,
15, 215-228.
Shaw, G. L. (2000). Keeping Mozart in mind. San Diego, CA: Academic Press.