Unmasking the Face by Paul Ekman

Prévia do material em texto

unmaskingthe face
Other booksby the authors:
EMOTION IN THE HUMAN FACE:
Guidelines for Researchand an Integration of Findings
DARWIN AND FACIAL EXPRESSION:
A Century of Researchin Review
Paul Ekman is Professorof Psychologyat the University of
California at San Francisco.He is a fellow of the American
PsychologicalAssociation and the American Association for
the Advancementof Science.
Wallace V. Friesen is Lecturer in Psychology and Re-
search Psychologistat the University of California at San
Francisco,where he has worked as co-investigatorwith Paul
Ekman since 1965.
UNMASKI
THE ACE
A guide to
ｾ ｾ ｾ
recognizingemotions
from facial clues
PAUL EKMAN and WALLACE V. FRIESEN
PRENTICE-HALL, INC., ENGLEWOOD CLIFFS, NEW JERSEY
Library of CongressCatalogingin PublicationData
Ekman,Paul.
Unmaskingthe face.
(A SpectrumBook)
Includesbibliographies.
1. Facial expression. I. Friesen,WallaceV.,
joint author. II. Title.
BF637.C45E38 152.4'2 74-14544
ISBN 0-13-938183-X
ISBN 0-13-938175-9(pbk.)
© 1975by PRENTICE-HALL, INC., EnglewoodCliffs, New Jersey
A SPECTRUM BOOK
All rights reserved.
No part of this book may be reproduced
in any form or by any means
without permissionin writing from thepublisher.
10 9 8 7 6 5 4 3 2
Printedin the United Statesof America
Prentice-Hall International, Inc. (London)
Prentice-Hall of Australia Ply., Ltd. (Sydney)
Prentice-Hallof Canada,Ltd. (Toronto)
Prentice-Hallof India Private Limited (New Delhi)
Prentice-Hallof Japan,Inc. (Tokyo)
For Patricia and Myriam
contents
acknowledgments xi
1
introduction 1
2
why mistakes are made
in understandingfacial expressions
of emotion 10
3
researchon facial expressions
of emotion 21
4
surprise 34
5
fear 47
6
disgust 66
ix
x contents
7
anger 78
8
happiness 99
9
sadness 114
10
practice faces 129
11
facial deceit 135
12
checking your own facial expression 154
13
conclusion 167
appendix
the facial blueprint photographs 169
appendix II
practice photos for chapter 10 173
appendix III
log and judge sheets 203
index 209
acknowledgments
We are grateful to the National Institute of Mental Health (NIMH)
for making it possible for us to study facial expressionand body movement
during the last eighteenyears. Paul Ekman was able to start our research
when he was awardeda predoctoralresearchfellowship from NIMH from
1955 to 1957. During military service from 1958 to 1960 Ekman and
Friesenbecameresearchassociates,a relationshipwhich was later formalized
when Friesenjoined the project in 1965. A postdoctoralresearchfellowship
from NIMH madeit possiblefor Ekman to pursuethe researchfrom 1960
to 1963. Later, when the pressureof teachingseemedlikely to curtail re-
search,a Career DevelopmentAward from NIMH to Ekman allowed the
team to continue the researchfrom 1966 to 1972. During all these years,
at eachcritical juncture, the late Bert Boothe, director of the ResearchFel-
lowship Branch, provided invaluable help, interest, and adviCe. The Clinical
ResearchBranchof NIMH providedcontinuoussupportfor our researchon
facial expressionandbody movementfrom 1963 until now (MH 11976-09).
Its supportallowed us to study mentalpatientsand also has madeit possible
for us to work togethersince 1965.
We arealsogratefulto the AdvancedResearchProjectAgency (ARPA)
of the Departmentof Defensefor supportingour studiesfrom 1966 to 1970.
Lee Hough, former director of ARPA, convinced us of the importanceof
studying facial expressionand gesture in different cultures. He helped us
overcomeour reluctanceto try to resolve the argumentover the possible
universalityof facial expressionand gesture.When we launchedour research
in a remoteareaof New Guinea,RowenaSwanson,the monitor of the grant,
found ways to solve the administrativeand bureaucraticobstacles.
We are grateful to Silvan S. Tomkins for his contagiousexcitement
about facial expressionof emotion. He encouragedus to learn how to read
xi
xii acknowledgments
faces and to teachothersto do so. For the past ten years,PatsyGarlan has
been an invaluable help as we have approachedthat point in each experi-
ment where the resultsof our work are conveyedto others.She has always
had a keen understandingof our research,has worked to make our writing
lucid, and hascritically examinedour ideasand searchedfor ambiguitiesand
contradictions.We arealso grateful to our friends, colleagues,and employees
who have been enthusiasticabout our studiesof the face and our attempts
to teach others what we have learned.Randall Harrison, John Bear, Allen
Dittmann, and Stuart Miller gave many helpful suggestionsabout how to
present this material in an understandableway. Harriett Lukes not only
typed the manuscript,but was an enthusiasticfirst reader.Nina Honbo helped
invaluably in keeping us organizedand encouragingthe completion of the
materials.We cannotthank by namethe many peoplewho have worked on
the researchreportedin this book; we are grateful for their fine work and for
their extraefforts which gaveus the time to write this book.
Our special thanksgo to our friends, students,and colleagueswho let
us showtheir facesin this book. We hopethat you, our readers,get to know
them well.
introduction
What This Book Is About
This book is about faces and feelings-your own and those of the
peoplearoundyou. The first focus is on what the feelings look like, in other
people'sfaces and in your own. Photographsshow the facial blueprints of
the major emotions-howsurprise,fear, anger, disgust, sadness,and happi-
ness are registeredby changesin the forehead, eyebrows, eyelids, cheeks,
nose, lips, and chin. Common confusionsthat plague the recognition of ex-
pressionsof emotions are clarified by pictures highlighting the differences
betweensurpriseand fear, anger and disgust, sadnessand fear. The subtle-
ties of facial expressionsof emotion are revealedin pictures that show the
family of expressionsfor each feeling. Surprise,for example, is an emotion
with a big family. Thereis not onesurprisefacial expression,but many-ques-
tioning surprise,dumbfoundedsurprise,dazedsurprise,slight, moderate,and
extremesurprise.The complexitiesof facial expressionsare shown in photo-
graphsof how different emotionscan blend into a single facial expressionto
show sad-angry expressions,angry-afraid expressions,surprise-fearful ex-
pressions,and so forth.
You can use this information about the blueprintsof facial expression
to better understandthe feelings of others,even when they are trying not to
reveal their feelings (Chapter 11 on "Facial Deceit"). Or you can use the
knowledge of the blueprints of facial expressionto learn about your own
face, to becomemore awareof what your face is telling you about how you
feel and what your face is telling others.Chapter12, "Checking Your Own
Facial Expressions,"describeshow you can determinewhether your facial
expressionsare characterizedby a particular style. For example,are you a
facial withholder (never showing anything on your face), an unwitting ex-
2 introduction
pressor(not knowing you are showinga feeling when you do so), or a sub-
stitute expressor(thinking you are showing an angry face when, in fact, you
look sad)?
The blueprints of facial expression,whether used to understandothers
or yourself, is then the first focus of this book. The secondfocus is the feel-
ings themselves.Although everyoneusesthe terms anger, fear, sadness,etc.,
few people really understandfully their own experienceof these emotions.
For example,what is it really like to be afraid?What doesit feel like in your
body?What situationsmakeyou afraid?Can you alwaysanticipatewhen you
will be afraid? Can you be both afraid and angry at the sametime? When
you are afraid, do you get aggressive,withdrawn, or thoughtful? Do you
laugh fear off, or do you breakout in hives?Do you ever enjoy being afraid
-watching a horror movie, for example?Do other people react the same
way you do when they are afraid? Does the same thing happen to their
breathing?Do the samesituationsthat make you afraid make othersafraid?Or do you sometimesthink or say, "1 can't understandwhy he was afraid;
that wouldn't bother me" or "1 can't understandwhy she wasn't afraid; 1
was terrified."
Most people would have a difficult time answeringsuch questions,at
least aboutsomeof the emotions.This may happenwhen you are, to quote
the cliche, out of touch with your feelings. This can also happenwhen you
are in touch with your feelings but don't realize the extent to which your
experienceof an emotionis unique.Thereis usually oneemotion, and maybe
more, that you do not shareopenly, describingthe feeling to others.It might
be an emotionyou in someway dreadexperiencing,or can'tcontrol, or habi-
tually never think about.Or it may be an emotionyou experiencekeenly but
completelyprivately. You may discover, for example,after extendedcontact
with an intimate, that what distressesyou does not distresshim. After the
courtshipis over, a marriagemay founder when the matesdiscover they not
only feel and expressangerdifferently but that the one who explodescan't
tolerate the one who holds it in, or the one with the long fuse can't accept
the one with the short fuse.
The secondfocus of the book is upon just these issues,describingin
detail the experienceof eachof the emotions.It providesas much as we know
abouthow theseemotionscan be experiencedin daily life. You can use this
information to learn about your own emotional experiences,the things you
sharewith others and the way in which you differ. You may find out about
experiencesyou are missing. And you may discover the basis for some of
your feelings which you haveneverfully understood. You can use this infor-
mation also to understandother people'sfeelings, to have a basisother than
your own emotionalexperiencefor comprehendingwhat an emotionmay be
like for anotherhumanbeing.
Who This Book Is For
thIs book is for psychotherapIsts,mInIsters, physIcIansand nurses,
trial lawyers, personnelmanagers,salesmen,teachers,actors. . ..
The psychotherapistmust know how people experienceemotions. He
must be alert to what the face may tell him about his patient'sfeelings. He
can't rely solely upon the patient'swords, for sometimesthe patient can't
describehis feelings. The patient'sface may show the emotionbeing experi-
encedevenwhen it is too upsettingto put into words, when he doesn'thave
the words to describehis feelings, or when he doesn'tknow how he feels.
The minister, or indeed anyone who counselspeople, has the same
needsas the psychotherapist.
The physician and the nurse also need to understandemotions and
facial expressions.Peoplehave an emotional reaction to their illness or the
threatof illness which may be crucial in its outcome.The physicianand the
nursemust understandthe different ways peopleexperiencefear, for this is
a commonemotionalreactionto the possibility of illness and treatment,and
may heightenpain, preventearly detectionof illness, interfere with treatment
plans,etc. And not everyonewho is afraid of cancer,or surgery,will experi-
encethat fear in the sameway. Recognizingsadness,which is often a reac-
tion to loss, and helping a patient to deal with it can be an important factor
in the patient's recovery when there is lingering or permanentdisability. If
many of the theories about psychosomaticdisorder are correct, then the
experienceof anger also should be understood.Patientswith a physical ill-
nessare often reluctantor embarrassedto mention their feelings about being
sick, their fear or sadness,their feelings of self-disgust,and so forth. Physi-
cians and nursesmust, therefore,learn to interpret accuratelyfacial expres-
sions and facial signs that emotionsare being controlled.
The trial lawyer often can't trust the words of a witness or client. He
needsanothersource, such as the face, to tell him how the person really
feels. Trial lawyers emphasizethe importanceof readingemotionalreactions
in picking jurors and in assessinghow a jury, once selected,reacts to dif-
ferent lines of argument.
The personnelmanagerand the salesmanmay be in the samespot as
the trial lawyer. In conductingan employmentinterview, the personnelman-
agerwants to recognizesignsthat the applicantis controlling his feelings. For
example, is his apparentconfidencegenuine, or is it covering nervousness
abouthis capability. Is he really as interestedin this job as he sayshe is. The
face can allow him to checkon how the applicantreally feels about the job.
The salesmanknows that the feelings which motivate a decisionto buy may
never be put into words, or the words may not be trustworthy.
The teacherneedsto know whetheror not the studentsunderstandwhat
3
4 introduction
he is saying. Interest, concentration,and perplexity are shown on the face
The actor must understandthe complexity of emotional experiencein
attemptingto convey an emotion in performance.The discussionof decep-
tion may help him preventhis personalfeelings from leaking through in his
performance.When the actor is feeling his way into the emotionalexperience
of a character,he needsto be sure that his expressionof that emotion is
commonly understood.And he should find the facial blueprints useful in
understandingand perfectinghis own ability to show emotions.
All theseprofessionals-psychotherapists,physicians,nurses,trial law-
yers, personnelmanagers,salesmen,teachers-needalso to becomeaware,
like the actor, of the impressionsthey give in their own facial expressionsto
their various "audiences."
this book is also for job applicants,loan seekers,customers,
voters, jurymen. . ..
The job applicant and the loan seekerneed to know what impression
they give by their own facial expressionsto the personnelmanageror bank
officer, for certainly they are under scrutiny. They may want also to watch
the face of the personnelmanageror bank officer to gaugehis reaction to
them.
Customersmay be concernedabout the honestyof the salesman-has
that car really beendriven by just one little old lady?
The voter is often concernedwith the man as well as his politics-how
trustworthyhe is in what he says,whetheror not his campaignpromisescan
be believed. In thesedays of television politics, this becomeseven more of
an issue.
The juryman can't assumethat the witness or defendantis telling the
truth, or knows the truth. Understandingfacial expressionsmay help him
distinguish what the person testifying actually feels from what he wants
people to think he feels. The juryman must also understandthe experience
of emotions if he is to comprehendfully the motivations behind certain
criminal acts. Whether or not there were mitigating circumstancesmay de-
pendupon the emotionalstateof the defendant.The reliability of a witness's
account may dependupon understandinghis emotional experienceat the
time of the crime as well as at the time he is giving his account.
this book is for friends, spouses,parents,lovers, relatives. . ..
The information in this book is relevant as well to any relationships
that are not mechanical,perfunctory, and businesslike.Everyone has rela-
tionshipsin which there is little or no emotionalinvestment.Feelingsare not
shared;neither personmakes an attempt to know the other's feelings, and
introduction 5
to do so would be an affront. Everyonealso has relationshipsin which the
intimatesharingof feelings is the main core. It is no accidentthat in intimacy
facesmove closer together.Intimatesalso look more into eachothers' faces.
Peoplekeep in sight picturesof the faces of thosewith whom they feel inti-
mate. Thougha telephonecall is better than a letter, if you expectan impor-
tant emotional experienceor want to describe one that has just occurred
(wedding, divorce, death, job promotion, etc.), you want to see the look on
the other person'sface and you want your face to be seen.
Wanting to sharefeelings does not necessarilymake this an easy thing
to do. Intimatesmay find it hard to understandor acceptthe differencesin
how they experiencean emotion. Intimacy may not survive such differences
precisely becausethey are not understood."You can'tbe angry about that;
I don't believe it!" "If you were afraid, why didn't you tell me?" It is very
hard to understandthat people you care about, people you love, don't ex-
periencefeelings the sameway you do. Facial expressionsthat show feelings
may be misinterpretedor missedentirely. If you don't fully understandthe
different ways a feeling may be experienced,how another'sway may differ
from your way of experiencingthat feeling-if you don't know the various
ways the face may show that samefeeling-the chancesfor misunderstand-
ing, for seeminginconsideration,multiply. This book is no panaceafor the
problemsof intimacy, for not all those problemsare due to misunderstand-
ing, and misunderstandingscan't be resolved by reading a book. But the
descriptionsof the varieties of emotional experienceand the blueprints of
facial expressionsshould help.
this book is for you alone
Understandingemotional experienceapplies not just to your relation-
shipswith othersbut also to your relationshipwith yourself. It can help you
understandthe most private, personal,unique part of your self. This is a
part of your self which hasenormouspower over your life. Your work, your
life, and even your death can be determinedby your feelings. Sexual needs
may not be satisfied,hunger not met, work not completedbecauseof feel-
ings that interfere. Feelingscan motivate the taking of your own life or the
life of another. Strugglesof an extraordinarynature may be endured,awe-
some feats accomplishedbecauseof feelings. Yet we know less about our
feelings than we do about our teeth, our car, or our neighbor'sescapades.
This is not a self-help book, but it may lead you to a better under-
standingof emotionsand of your own emotional life. Although learning the
facial blueprints will chiefly improve your ability to spot emotion in others,
it can also teach you to be keenly aware of what your own facial muscles
are telling you-aboutyou.
6 introduction
What You Can Get From This Book
You canreadthis book, or you canstudy it. It dependsupon your goal.
Do you want simply to increaseyour knowledgeof emotion?Or do you also
want to mastera skill? Both are possible, but the skills take longer to ac-
quire than the knowledge.You can readthis book in a matter of hours, con-
centratingon the words and looking at the pictures. You will learn a good
deal aboutemotionalexperiences,your own and other people's-information
that should be helpful in all the ways describedabove. But this approach
won't teachyou how to spot emotionin people'sfaces much better than you
are already able to do. You won't be any better in spotting slight signs of
emotion, blendsof emotions,signs of emotionalcontrol, the leakageof emo-
tions, and so forth. To be able to do thesethings requires the extra invest-
mentof learningthe facial blueprintsso well that you canusethe information
without thinking aboutit. It mustbecomea skill.
You may feel you don't need muchimprovementin interpretingfaces,
or you may needit for certain emotionsbut not for others,or you may find
you are pretty far off in telling how people feel from their faces. In the sec-
tions on facial appearancein Chapters4, 5, 6, 7, 8, and 9 are practice in-
structions.Each of those chaptersalso tells you how to make certain faces,
constructingchangesin expressionso you can better learn how the face
works. When you can accuratelyrecognize the faces shown in Chapters4
through 9, Chapter10 showsyou how to use a new set of faces to practice
further and improve your skills.
How This Book Relates to Popular
Books about Body Movement and
Nonverbal Communication
Thereis little overlapin information betweenthis book and the popular
books about body movementor nonverbal communicationwhich have ap-
pearedin the last few years. Those books said almost nothing about facial
expression,chiefly becausethe information about the face has not beenavail-
able before. Our two previous books on facial expression(Emotion in the
Human Face and Darwin and Facial Expression) are intended for the re-
searcherwho is seekinginformation on how to study facial expression,and
for the scholaror studentwho wants to know what researchhas beendone
on facial expression.They addresssuch questionsas whether facial expres-
sions of emotion are universal, and whether people can accurately tell an
emotion from the face. But they don't show what universal expressionsof
emotionlook like or showthe readerhow to makeaccuratejudgmentsof emo-
introduction 7
tion. This is our first attemptto providethis information,and to our knowledge
it is the first attemptanyonehasmadeto do this.
This book differs from the popular nonverbal-communicationbooks,
too, in that it is about the face, and therefore about emotions, while those
books are primarily about the body. Our studiesof the body,·publishedin
professionaljournals, have exploredthe differencesin what the face and the
body tell us. Emotionsare shownprimarily in the face, not in the body. The
body insteadshows how people are coping with emotion. There is no spe-
cific body movementpatternthat always signals angeror fear, but there are
facial patternsspecificto eachemotion.If someoneis angry,his bodymayshow
how he is coping with the anger. He may be tense and constrained(tight
muscular tension in arms and legs, stiff posture). He may be withdrawing
(a particular retreatedposition). There may be verbal attack (certain types
of hand movementswith his words), or the likelihood of physical attack
(posture,orientation, hand movements).However, any of thesemovements
can occur just as well when a personis afraid as when he is angry. Body
movementreveals,of course,not just how someonecopeswith emotion, but
also what a person's attitudes, interpersonalorientations, etc., might be.
Some of this has been explained in the popular "body" books, some has
regrettablybeen explainedbadly or erroneously,and much has just not been
discussed.In a few yearswe plan to publish anotherbook showinghow body
movement,facial expression,voice tone, and words fit together.They are all
important in understandingpeople.But the face is the key for understanding
people'semotional expression,and it is sufficiently important, complicated,
and subtle to requirea book to itself.
Why There Is Need for a Book
about Facial Expression
Although everyoneagreesthat understandingemotionsis crucial to per-
sonal well-being, to intimate relationships,and to successin many profes-
sions, no one teachesyou how to do this unlessyou get into serioustrouble.
And then, the vafious techniquesof psychotherapyusually focus only on
those few emotions which their theories consider. Though there is strong
evidencenow that the face is the primary signal system for showing the
emotions, no one taught you how to read those signals. And there is every
reasonto believe that you were not born with the knowledge.You have to
pick it up.
Much of what you know about emotions and facial expressionswas
shapedby your parentsand other membersof your family. The facesof your
8 introduction
parents,of your siblings, and of other people who took care of you were
the first you saw. Your family membersmay have beenvery expressiveor
very guardedin their facial expression.They may have shown you the full
repertoireof emotional expressions,or only a few. Perhapsyou never saw
an angry face, or a fearful face. People in your family may have shown
emotion in their faces as most other peopledo, or one of them might have
developeda strangeor peculiarway of looking disgustedor afraid, and that
may still influence your recognition of that emotion. As a child, you may
have been specifically instructed not to look at the facial expressionsof
others,or at leastcertain facial expressions.For example,somechildren are
told never to look at someonewho is crying.
As an adult, you may be more sensitive to some emotions than to
others.What you learnedabout readingemotionsin your own family might
have great applicability to understandingothers or relatively little, at least
for someemotions.Throughwatchingtelevision,ormovies,or a closefriend,
you may have improved upon and addedto your knowledge of facial ex-
pressions.Although almosteveryonecorrectly readssomefacial expressions,
few peoplerealizewhen they makemistakesor why they makethem.
The rules for translating a particular set of facial wrinkles into the
judgment that a personis angry, afraid, etc. would be very hard for most
peopleto describe.When you follow theserules you do so automatically,on
the basisof habits establishedso long ago that usually you don't know how
they operate,or even when theyoperate.In this sense,understandingfacial
expressionsof emotionis like driving a car. You don't think aboutwhat you
are doing when you do it. Unlike driving a car, with facial expressionthere'
never was an earlier periodin which you were specifically taught the skills.
There is no manualin which you can check how to correctmistakes.There
are no equivalentsto the traffic cop telling you when you missed or mis-
interpreteda signal.
Often you do not know that a facial expressionwas the basisfor your
hunch or intuition about someone.You may only sensesomethingabout a
personwithout being able to trace the sourceof your impression.And if you
don't know what led to your judgmentyou can't correct yourself if it turns
out to have been wrong. Sometimesyou are puzzled by someone'sfacial
expression;you can't figure out what he meant.Or you can figure out what
he meant by the look on his face, but you can't decide whether or not to
trust it. It is hard to check impressionswith others,becausethere just isn't
much of a vocabularyfor describingthe face itself. There are a lot of words
for the messagesyou get from the face (afraid, terrified, horrified, apprehen-
sive, worried, to mention a few of thoserelatedto fear), but few to describe
the source of those messages.We do have the terms smile, grin, frown,
squint, but there are relatively few such words that identify particular facial
introduction 9
configurations,distinctive wrinkle patterns,or temporaryshapesof the facial
features. Without terms to refer to the face, we are handicappedin compar-
ing or correctingour interpretationsof facial expression.Commentssuch as
the following, though accurate,are maddeninglycumbersome."I know why
you thoughthe was afraid. It was becausethe inner cornersof his eyebrows
were pulled togetherand raised.But you failed to seethe omegawrinkle in
his forehead.If you had noted that, or the fact that the outer eyebrowcor-
ners were down, not up, you would have known he was sad." At best, it is
not easy to describefacial expression.Picturesare needed,becauseit is a
visual phenomenon. In this book we are able to bring you hundredsof care-
fully selectedphotographswhich show you how the face registersemotion.
We think such a book is neededto accomplishthe following:
-Bring attentionto what you may alreadybe doing without knowing it
--Showwhat you may be missingentirely
-Correctwhat you may be misinterpreting
-Show the subtleties(the families of facial expressions)and the com-
plexities (the blendsof two emotionsin one facial expression)
-Alert you to signs of facial control and teach you how to discover
when a facial qualifier is used,or when an expressionis modulatedor
falsified
-Provide techniquesfor learning whether you show emotion in your
own face in an unusualfashion
The next chapterexplainsa numberof the potentialproblemsencoun-
tered in understandingfacial expression,suggestingwhy mistakesare made
and how to avoid them. You may havebeenbroughtup not to look at faces,
or at certain faces. You may be distractedby other competingbids for your
attention, the words someonespeaks,the sound of his voice, the general
appearanceand movementsof his body. You may not recognize the dif-
ferences between controlled and uncontrolled expressions.You may not
know what to look for in the face, or where to look, to tell if someoneis
angry, or afraid. You may not know exactly what an emotion is, how one
emotion differs from another,or how an emotion differs from a mood, atti-
tude, or charactertrait.
In
why mistakes are
understanding facial
of emotion
made
expressions
The face as a Multisigna/,
Multimessage System
The face provides more than one kind of signal to convey more than
one kind of message.In trying to follow the emotion messages,you may
look at the wrong signal. Or perhapsyou don't clearly differentiatethe emo-
tion messagesfrom the othermessagesconveyedby the face. A very familiar
exampleof a multisignal, multimessagesystemis the systemof road signs. It
employsthree types of signals: shape(triangular, square,circular, rectangu-
lar, octagonal),color (red, yellow, blue, green), and inscription (words,
drawings,numbers).Road signs utilize thesethree types of signals to trans-
mit three types of messages:regulations (stop, no V-turn, yield, etc.),
warnings (school crossing, two-waytraffic, etc.), and information (service
area,bike route, camping,etc.). With road signs, as with facial expressions,
you must focus on a particular type of signal if you wish to learn a particu-
lar type of message.If you want to know if you are approachinga rest area
or a place to camp, you can searchfor the blue or greensigns, becauseit is
thesecolors that give this type of information (yellow signs give warnings,
red signsgive regulations).The parallel in readingfaces is that if you want
to know what emotion someoneis feeling, you must watch the temporary
changesin the face, becauseit is theserapid facial signalswhich give infor-
mation aboutemotions. (If you were trying to determinesomeone'sage,you
would pay attention to more enduring aspectsof the face, such as muscle
tone or permanentwrinkles.)
The face provides three types of signals: static (such as skin color),
slow (such as permanentwrinkles), and rapid (such as raising the eye-
brows). The static signals include many more or less permanentaspectsof
10
why mistakes are made in understanding facial expressions of emotion 11
the face-skinpigmentation, theshapeof the face, bone structure,cartilage,
fatty deposits,and the size, shape,and locationof the facial features (brows,
eyes,nose, mouth). The slow signals include changesin the facial appear-
ance which occur gradually with time. In addition to the developmentof
permanentwrinkles, changesin muscle tone, skin texture, and even skin
coloration occur with age, primarily in the later years of adulthood. The
rapid signalsare producedby the movementsof the facial muscles,resulting
in temporarychangesin facial appearance,shifts in the location and shape
of the facial features, and temporarywrinkles. Thesechangesflash on the
face for a matter of secondsor fractions of a second.
All three types of facial signals can be modified or disguisedby per-
sonal choice, although it is hardestto modify the static and slow signals.
Hair styles are used almost universally to modify thesesignals. Bangs, for
example,can changethe apparentsize of the forehead,or concealperma-
nently etchedwrinkles. Cosmetics,such facial ornamentsas sunglassesand,
at the most extreme,plastic surgery can modify the static and slow facial
signals so that they broadcasta different set of messages.The rapid facial
signalscanbe modified or disguisedby inhibiting the musclemovementsthat
producethem or by maskingone expressionwith another,or by hiding the
face with beardsor sunglasses.Thus, one can be misled, purposefully or
accidentally,by rapid, slow, or static signals.
The face is not just a multisignal system(rapid, slow, static) but also a
multimessagesystem. The face broadcastsmessagesabout emotion, mood,
attitudes,character,intelligence, attractiveness,age, sex, race, and probably
other mattersas well. This book is focusedprimarily on one type of message
and one type of signal-emotionmessagestransmittedby the rapid signals.
When we speakof emotions,we are referring to transitory feelings, such as
fear, anger,surprise,etc. When thesefeelings occur, the facial musclescon-
tract and there are visible changesin the appearanceof theface. Wrinkles
appearand disappear,the location and/orshapeof the eyebrows,eyes,eye-
lids, nostrils, lips, cheeks,and chin temporarily change.Researchhas shown
that accuratejudgmentsof emotioncanbe madefrom the rapid facial signals,
and has recently uncoveredthe particular facial signals-theblueprints-
which distinguish eachof the primary emotions.Photographswere specially
made for this book to reveal and contrastthe facial blueprints that distin-
guish eachof the primary emotionsand the blendsof theseemotions.
It is important to note that the emotion messagesare not transmitted
by either the slow or the static facial signals;however, thesemay affect the
implications of an emotion message.If a personhas a thin or fat face, a
wrinkled or smoothface, a thin- or thick-lipped face, an old or young face,
a male or female face, a Black, Oriental, or Caucasianface, that does not
tell you whetherthe personis happyor angry or sad. But it may affect your
12 why mistakes are made in understanding facial expressions of emotion
impression.For example,if the rapid facial signals tell you that a personis
angry,your impressionof why the personis angry andwhat he is likely to do
while angrymaydependin part uponthe informationyou gleanfrom the slow
andstaticfacial signalsabouttheperson'sage,sex,race,personality,tempera-
ment, and character.Thesefurther interpretationsare not discussedin this
bookbecausetoo little is yet known aboutthem, andbecausethere is a prior
questionwith which we are primarily concernedhere. Before consideringthe
matterof what you may expectof different typesof peopleonceyou do know
how theyfeel, thereis thequestionwhich is thefocusof this book: How do you
tell from the face how someonefeels, and how can you tell if his facial
expressionis genuineor phony?
This book also provides some information about the mood messages.
Moods are closely related to emotions, and some are shown in the rapid
facial signals. Moods differ from emotionsin that the feelings involved last
over a longer period. For example,a feeling of angerlasting for just a few
minutes,or evenfor an hour, is called an emotion.But if the personremains
angry all day, or becomesangry a dozentimes during that day, or is angry
for days, then it is a mood. "Irritable" is a word used to describesomeone
in such a mood, although one could just as well say that the personis in
an angry mood. It is possible, although not likely, that a personin such a
mood will show a completeangerfacial expression,with the signalsof anger
registeredover the entire face, throughoutthe period that the mood endures.
More often just a trace of that angry facial expressionmay be held over a
considerableperiod-atightenedjaw or tensedlower eyelid, or lip pressed
againstlip, or brows drawndown and together.Another way a mood may be
registeredin facial expressionis by the frequencywith which the total facial
expressionflasheson andoff within sometime period. It may be obviousthat
a personis in an irritable mood becausehe becameangry so often this after-
noon. Other moodsthat are shownin rapid facial expressionsare depression
(wherethe face showsevidenceof sadness,fear, or a blend of both), anxiety
(wherethe face showsevidenceof fear), and euphoria(where the face shows
evidenceof happinessandexcitement).
The rapid facial signalsalso sendemblematicmessages.In our research
we utilize the term emblemto describesignals the meaningof which is very
specific, the nonverbal equivalentof a common word or phrase.The eye-
wink signal for the messageof agreement("right on," "sure") or flirtation
("will you?") is an example.Facial emblemsare like hand movementsfor
waving "hello" or "goodbye," and like head nods for "yes" and "no." The
movementis always specific and easyto distinguish from other movements.
The meaningis understoodby everyonein a culture or subculture.We will
not discussthe full array of facial emblems,but only thosewhich are closely
related,either in movementor message,to the emotionalexpressions.Rais-
why mistakes are made in understanding facial expressions of emotion 13
ing the brows and holding them while keepingthe rest of the face blank is
an exampleof such an emblem. That brow movementis part of the rapid
facial signal for surprise,but when it is not joined by a movementin the
eyelids and lower face as well, it signalsquestioning.
There are still other emblemsthat might be called emotion emblems,
becausethe messagethey convey is about an emotion. Theseemotion em-
blems look like a facial expressionof emotion, but they are different enough
for the personseeingthem to know that the personmaking them doesn't
feel that way at the moment;he is just mentioningthe emotion.For example,
one of the emotionemblemsfor disgustis to wrinkle the nose,which is part
of the disgust facial expression.When it is used as an emotion emblem it
occursalone,with little raising of the upper lip, and it flasheson and off the
face quickly, and it is thereforenot confusedwith an actual disgust facial
expression.The messageis "disgust (I don't feel that way now)."
The rapid facial signals are used, then, to convey emotion messages
and emblematicmessages.They are also usedas conversationalpunctuators.
Everyoneknows peoplewho use their handsto accentor italicize a word or
phraseas they speak.Peoplecando the samething with the rapid facial sig-
nals, punctuatingwhat is being said in words with facial accents,commas
and periods. The facial punctuatorswill be discussedand shown in later
chapters.
The face sendsmany rapid signalsin addition to the onesthat register
emotion or are used in emotion-relatedemblemsor as punctuators.Facial
grimacing, contorting, and pantomiming; movementsrequired by speech;
suchfacial acts as lip-biting, lip-wiping, etc., will not be considered,because
they are not relatedto or often confusedwith the facial expressionsof emo-
tion. Similarly, the face sends many messagesin addition to those about
emotions and moods. People believe they can read attitudes, personality,
moral character,and intelligence from the face. We will not discussthese,
becauseit is not known whether people even agreeabout such judgments,
or, if so, whether their judgments are correct. Further, it is not known
whether any such judgmentsare basedon the rapid, the slow, or the static
facial signals.Peopleprobablydo agreeand are correct, at leastsomeof the
time, in judging sex, age, and race from the face. Even with these more
obvious messages,not much is known about the precise facial blueprints.
Thoughwe assumethat they are transmittedprimarily by the slow and static
facial signals,we do not know preciselywhat signals transmit thesetypes of
information. For example, in readingwhether someoneis male or female,
is this apparentin the shapeof the upperor lower lip, or in the size of the
lips, or the relativesizeof the two lips, or the pigmentationof the lips? Or is it
not the lips at all, but the amountof hair in the eyebrows,or their shape;or
is it the shapeof the chin; etc.?
14 why mistakes are made in understanding facial expressions of emotion
In regard to facial expressionsof emotion, however,we know a good
deal. We know that people can make correct judgments. We know the
specific signals in each part of the face that convey the messagesof fear,
surprise, sadness,happiness,anger, disgust, and combinations thereof. If
your understandingof facial expressionsis to be improved so that you can
interpret those shown by others and more adequatelysenseyour own, then
your attention must be focusedupon the rapid facial signals and their dis-
tinctive messages.It is futile to look to the slow and static facial signals for
information about emotion. You will need to learn the subtle and obvious
differencesandsimilaritiesamongtheemotionsin their appearanceon the face
and in how they are experienced.And you will needto learnto distinguish
the facial expressionsof emotion from the facial emotion emblems, other
facial emblems,and facial punctuators.
Not Watching Faces
Problemsin understandingfacialexpressionsarise becausemost of the
time peopledo not watch eAch other'sfaces.Becausemost facial expressions
of emotion are brief, you may often miss an importantmessage.Somefacial
expressionsare extremelyrapid, lasting only a fraction of a second.We call
these micro-expressions.Most people fail to see them or fail to recognize
their importance.Later (Chapter11) we will explain how thesevery quick
micro-expressionscan reveal emotionsthe personis attempting to conceal.
Even the more usual macro-expressionsfrequently last only a few seconds.
It is rare for a facial expressionof emotion to last as long as five or ten
seconds.If it does, the feeling must be intense,so intensethat the feeling is
likely to be simultaneouslyshownin the voice through a cry, laugh, roar, or
in words. Even if you were not looking at a person'sface, you would be
likely not to miss these intense emotions, becauseyou would hear them.
More often, however,the very long facial expressionsare not genuineexpres-
sions of emotion, but mock expressions,in which the personis playing at
showingan emotionin an exaggeratedfashion. It is obvious when you seeit
that the personis playing. Sometimeshe may not be playing, but using the
mock expressionas a way to show the emotionwithout taking responsibility
for doing so. For example,supposeyou have agreedto join someonein an
escapadethat now, as you approach it, looks more risky than you had
anticipated.You might show a mock fear expression.This would let you
show how you feel, get the messageacrossto your companion,so that he
too could expresshesitation;but if he is unsympathetic,he can't ridicule you
becauseyou were showing only mock fear.
It is easy to understandwhy the micro-expressionsmight be missed,
why mistakes are made in understa,,!ding facial expressions of emotion 15
becausethey are so quick. Later we will give someexercisesfor recognizing
them. But even the longer macro-expressions,which last two or three
seconds,are often missed,becausethe face is often not watched.In a sense
this is paradoxical,becausethe face is very commandingof attention.As the
site for the major sensoryinputs-sight,hearing,smell, and taste-andfor
the major communicativeoutput-speech-ithas great importancein social
life. Moreover, most people identify their very selves, in part, with their
facial appearance.Industries exist around enhancingthe appearanceof the
face. A person is most easily identified by his face, more so than by
his body. His privacy is exposedmuch more if a picture takenof his face at
someembarrassingmomentappearsin a newspaper,than if a similar picture
of his body were to be published.And there is curiosity aboutother people's
faces.When told abouta personyou haven'tmet, you will want to seea pic-
ture of his face; you don't usually think aboutwanting to seea picture that
would clearly show his body. When you seea person'sfull face and body,
you tend to look more at the face than at the body, unlessperhapsthere is
somesexualinterest.
When you converse,you seldomlook continuouslyat the otherperson's
face. You probablyspendmore time looking away from the personyou are
talking with than you spendlooking at him. Think abouthow the chairs are
arranged,or how you rearrangethem, when you are about to sit down and
have a conversation.Usually you don't sit facing each other, head-on,but
adjacentto or at anglesto eachother. When your head is aligned with your
body, you and the person you are talking with are each looking off into
space,which is what you probably do most of the time. In order to seethe
face of the other person you must turn your head, which you will do
occasionally.
Researchis presentlybeing carried on to explore just when two people
conversingdo look at eachother. It seemsthat when you speakyou look to
see if the listener agrees,or is amused,angry, interested,bored, etc. You
look to determineif the other personneedsa chanceto get a word in. You
look if the other person has not made any audible listener responsesfor
a while ("mm-hmm," "yes," "good," "uh-huh," 'is that so," etc.) to check
him out. You look to give the other personthe chanceto take the floor. If
you don't intend to give up the floor, you have to be careful not to look at
the other personwhen you are pausing.When you are the listener,you may
look at the speakerwhen he stressesa word, or at the end of his phrase,if
his inflection risesand he seemsto be inviting a verbalor nonverbalresponse
from you.
Therearesomesituationsin which at leastonepersonlooks at the other
person'sface much more of the time. If you are a memberof an audience,
you feel no inhibition about gazing continuouslyat the performer'sface. He
•
16 why mistakes are made in understanding facial expressions of emotion
hasgiven you permissionby virtue of his and your role. If you are an inter-
rogator, you may continuouslylook at the personyou are questioning;that's
your job. Probablythe sameis true in any formal conversationin which one
personhas some clearly defined and acknowledgedauthority over another.
Continuousface-watchingoccurs also in just the oppositetype of situation.
Everyonecan recognizelovers, at least thosewho are still courting, by how
much of the time they unabashedlygazeinto eachother's faces.
Why is it that mostof the time you don't look at the face of the person
you are conversingwith? Why may you actually look away at just the mo-
ment you sensehe may show an emotional facial expression?Part of the
answeris politeness-beingbroughtup not to stare (at least in the U.S. and
someother cultures). You don't want to be rude and intrusive, taking infor-
mation that it is not clear the other personis giving to you. And you don't
want to embarrassthe person you are conversingwith, or yourself. If he
wants you to know how he feels, then he can say it in words; otherwiseyou
would be oversteppingthe boundsof convention.Watching someone'sface
is intimate. You take such a liberty only if the other persongives it to you
by being a public performer,or if your social role bestowsit upon you (as
interrogator,employer, juryman, parent, etc.)-or if you avowedly seek to
shareintimacy, looking and inviting the look of the other person.
Thereis anotherside to it, however.It is not just being polite and not
wanting to embarrassor be embarrassed.Often not looking at the otherper-
son'sface is motivatedby not wanting to be burdenedwith that knowledge
or obligatedto do somethingabout how the personfeels. By not looking at
his face you don't know, or can pretendnot to know. Unlesshe says it in
words, you are not socially obligated to care for or respondto his feelings.
If he is showingannoyanceor angeron his face and you seeit, andhe knows
you seeit, then you may have to find out whether you are the cause,or if
not, why he is angry. If he is showingsadness,you may have to comfort or
aid him; the sameis true if he is showing fear. In many social interactions
the last thing eitherpersonwantsis to acknowledgeor take any responsibility
for having to deal with the feelings of the other person.
In addition to these fairly common reasonsfor not looking at faces,
some people may have had experiencesin their childhood which trained
them not to look at certainpeople'sfacial expressions,or at certain expres-
sions,no matterwho showsthem. A child may learn that it is dangerousto
look at Daddy's face when he is angry, or dangerousto look at anyone's
angry face. Such learningmay occur so early that the grown personavoids
seeingcertainemotions,or certain types of peopleshowingcertainemotions,
without knowing he is engagedin such avoidance.
If your understandingof facial expressionsis to improve, then you must
reexaminewhether or when you want to know how the other personfeels.
You may have to counteracthabits you have learnedand followed without
why mistakes are made in understanding facial expressions of emotion 17
knowing it, which lead you to ignore and miss many facial expressionsof
emotion.
The Communication Barrage
The face is but one sourceof informationavailableduring a conversa-
tion, and the problemis that messagesmay be missedbecauseof the distrac-
tion of the other, more commandingsourcesbidding for attention. In most
conversationsit is sight and sound, what you see and hear, that you rely
upon. In most conversationsamongpeoplefrom Westernindustrializedset-
tings, there is not much tactile stimulation, and precautionsare taken to
camouflageodors.When listening, you gatherinformation from at least three
sourcesin the auditory channel: the actual words used; the sound of the
voice; and such things as how rapidly the words are spoken, how many
pausesthere are, how much the speechis disruptedby words like "aah" or
"ummh." When looking, you gather information from at least four sources
in the visual channel: the face; the tilts of the head; the total body posture;
and the skeletalmusclemovementsof the arms,hands,legs, and feet. Every
one of thesesourcesin both the auditory and visual channel can tell you
somethingabout emotion.
The auditory and visual channelsoffer different advantagesand dis-
advantagesin communicationto the senderand receiver of messages.The
auditory channelcan be completelyturned off by the sender;he can simply
shut up and therebyceaseproviding any information to the other personhe
was conversingwith. This is an advantagefor the sender,a disadvantageto
the receiver. The visual channel is in some sensealways turned on. The
sendermust maintain some body posture,some hand position, some set to
the face; thus, someimpressionis always given to the otherperson.
If the sendercan assumethat the other personis interestedin the con-
versation,the use of the auditory channelprovides greaterlikelihood that a
messagesentwill get through and be received.The listener doesnot needto
point his ears at the sender to receive a message;the listener's auditory
channelis open. Not so with the visual channel.The other personmight be
interested,but fail to glance over at the moment a messageis sent. The
sendermay take advantageof this to "slot" a messagein the visual channel
to commentto an onlookerwithout the knowledgeof the personwith whom
he is conversing.For example, one personmay signal visually to another
personto rescuehim from a boring conversation,without the boring con-
versantnoticing the call for help. On the otherhand, if the senderdoeswant
his messageto get through to the other personwith whom he is conversing,
then he must wait until the otherpersonis visually fixed on him beforesend-
ing a messagein the visual channel.Thus, when a senderand receivershare
18 why mistakes are made in understanding facial expressions of emotion
an interest in communicating,the auditory channel has the advantageof
more certain receipt of messagessent.
On the other hand, the receivercan more easily pretendauditory than
visual interest. Listening can be simulated,with occasionalnods and "um-
hmms" thrown in, when in fact the receiver is listening to another con-
versationor to his own thoughts.Thereare no tell-tale signsot such auditory
disloyalty which parallel the out-of-focuslook to the gaze,if the viewer pre-
tendsto be looking at the senderbut is in fact looking at someonebehind
or to the side of him; nothing to equal the tracking movementswhich show
that the viewer's eyes are following someoneelse's activity. Incidentally,
becauseit is more obviouswho is looking than who is listening, both sender
and receiverhave more protectionfrom eavesdroppingby unwantedlookers
than by unwantedlisteners.
Although there is somerepetitionor overlap in the type of information
transmittedin the two channels,there is a natural division of labor, so that
eachchannelbetterconveyscertainmessages.Words are bestfor most mes-
sages,particularly factual ones. If you are trying to tell someonewhere the
museumis, who played the lead in that movie, whetheryou are hungry, or
how much the meal costs, you use words. You don't resort to the visual
channelas the primary transmitterof such messagesunlessyou can't speak
or be heard,or unless the messageinvolves showing where somethingis in
space(how you get to the post office from here). The voice tone and the
visual channel sourcesadd qualifying messagesto what is transmitted in
words. Nuancesare added,points emphasized,instructionsgiven as to how
seriously to take what is said in words, etc. The visual channelalso can be
used for factual messages,by Indian sign languageor the languageof the
deaf, but the advantageis clearly with words.
Words can also be used to describe feelings or explain them, and
typically are usedin concertwith other sourcesto do so. Here, however, the
advantageis with the visual channel,becausethe rapid facial signalsare the
primary systemfor expressionof emotions.It is the face you searchto know
whethersomeoneis angry, disgusted,afraid, sad, etc. Words cannot always
describethe feelings people have; often words are not adequateto express
what you seein the look on someone'sface at an emotionalmoment. And
words are not trustedas much as faces,when it comesto emotions.If some-
one tells you he is angry and his face showsit as well, nothing is amiss.But
if he sayshe is angry and showsno evidencefacially, you are suspicious.If
the reverseoccurs and he looks angry but doesn'tmention angerfeelings in
his words, you doubt the words but not the anger; you wonder why he
doesn'tadmit how he feels.
Emotion messagescan also be transmittedby the soundof the voice,
body posture,hand/armmovements,and leg/feet movements.But it is not
why mistakesare made in understandingfacial expressionsof emotion 19
certainwhether they transmit information about emotion with as much pre-
cision as doesthe face. They may moresimply tell you that someoneis upset,
but not whether the upset feeling is anger, fear, disgust, or sadness.The
clarity and precisionof theseother sourcesof emotion messagesremainsto
be studied.We do know that the face is a primary, clear, and precisesignal
systemfor the expressionof the specific emotions.
With two channels-theauditory and visual-transmittinginformation
from sevensources,communicationis a barrage.The speakersendsout a
barrageof signalsand may not carefully attendto or employ all of them in
an optimal fashion, and the listeneris barragedand may not attendto all the
signals available. With the visual channel, people focus more on the face
thanon the othersources;facial expressionsare tunedmore than body move-
ments, and people look at faces more than body movements.But the audi-
tory channel, in particular the words, usually receives the most attention,
both from speakerand listener, becausehere can be transmittedthe richest
and most varied information about everything but emotion.
You can miss important information about emotionsin the face, then,
becauseit is competingwith theseother information sources,and it may be
somewhathandicappedin its bid for attention.Even if you focusedas much
on facial expressionsas on words, the words would receive more attention
becausethey alwaysreachus, while onehas to look to seea facial expression.
The face is, nevertheless,of more importancethan words for transmitting
information about emotion.
Controlling the Face
Facial expressionmay be controlled or uncontrolled. One expression
may be voluntary, another involuntary; one may be truthful and another
false. The problemis to tell which is which.
Supposesomeonehas an emotional experience-somethinghappens,
let us say, to makehim afraid. If an expressionof fear comesover his face,
it will have done so automatically.The persondoesnot think about how to
move his facial musclesto look afraid; the fear expressionis involuntary.
But theremay be interference,dictatedeither by longstandingwell-ingrained
habit or by deliberate,self-consciouschoiceof the moment.The interference
may be minor, only qualifying or modulatingthe expression,or major, inter-
rupting or totally inhibiting the expression.The facial expressionof fear
would be qualified if the personaddedto the fear expressiona bit of a smile,
showingthathecan "grin and bearit." The facial expressionwould be modu-
lated if the personchangedthe apparentintensity of the expression,trying
to ｳｾｯｷ only slight worry when experiencingstrongfear. The facial expres-
20 why mistakes are made in understanding facial expressions of emotion
sion of fear could be interrupted,so that only a trace appeared;or the ex-
pressionmight be totally inhibited so that nothing showedon the face, or
there was only a stiff appearance.Not only can facial expressionsbe inter-
fered with, but peoplecan simulate emotion expressions,attemptingto cre-
ate the impressionthat they feel an emotionwhen they are not experiencing
it at all. A personmay show an expressionthat looks like fear when in fact
he feels nothing, or feels sadnessor someother emotion.
Facial expressionsof emotionare controlledfor various reasons.There
are social conventionsabout what you can show on your face, cultural dis-
play rules that govern how peoplemanagethe appearanceof their faces in
publiC. For example,in the United Statesmany little boys learn the cultural
display rule, "little men do not cry or look afraid." There are also more per-
sonal display rules, not learned by most people within a culture, but the
product of the idiosyncrasiesof a particular family. A child may be taught
never to look angrily at his father, or never to show sadnesswhen dis-
appointed,or whatever. These display rules, whether cultural ones shared
by most peopleor personal,individual ones,are usually so well-learned,and
learnedso early, that the control of the facial .expressionthey dictateis done
automaticallywithout thinking or awareness.
Facial expressionsare also controlled becauseof vocational need or
practice. Some jobs seemto require or selectpeoplewho are expert facial
controllers. Anyone who is successfulin such a job may need to be able
to put on convincingsimulations (the actor, or even the salesman).Or the
requirementmay be never to reveal how you actually feel (the diplomat).
Peoplealso control their facial expressionsof emotion becauseit is to their
advantageat a particular moment. If a pupil cheatson an exam, he may
concealhis apprehensionwhenthe proctor walks by becausehe doesn't want
to get caught.
Someof the confusionabout facial expressionarises,then, becausethe
face conveysboth true and false emotion messages.There are uncontrolled,
involuntary, true expressionsand also qualified, modulated,or false expres-
sions,with lies of omissionthroughinhibition andlies of commissionthrough
simulation. In order to improve your understandingof facial expressionsof
emotion, you will have to distinguish which is which. The first step is to
learn how the actually felt emotionsappearon the face (Chapters4 to 9),
for without that knowledgeyou can't spot the clues to facial control. Later,
in Chapter11 on "Facial Deceit," the display rules and the various manage-
ment techniqueswill be further explained.A number of suggestionswill be
given about how to distinguish felt from modulatedor false facial expres-
sions.Someof the clues are in the shapeof the facial expression,somein its
timing, othersin its location in the conversationalstream,and still othersin
how the facial expressionrelatesto everythingelse the personis doing.
research on
of
facial expressions
#
emotion
There have beenhundredsof experimentson facial expressionsof emotion.
Elsewherewe have analyzed these studies in detail (see Reference5 at
the end of this chapter). Here we will more briefly describe only those
studieswhich are directly relevantto the information presentedin this book.
This chaptershouldhelp resolvethe doubtsof the skepticaboutthe scientific
basis for what is said and shown in subsequentchapters.It is also provided
for those who are curious about how facial expressionsof emotion are
studied.
Which Emotions Does the
Face Show?
Doesthe face tell us only whethersomeonefeels pleasantor unpleasant,
or does it provide more precise information, conveying which unpleasant
emotion is experienced?If the latter, how many of thesespecific emotions
doesthe face show-six,eight, twelve, or what number?The typical method
used to determinejust which emotionscan be read from the face has been
to show photographsof facial expressionsto observers,who are asked to
say what emotion they seein eachface. The observersmay be given a pre-
determinedlist of emotion words to choosefrom, or left to their own re-
sourcesto reply with whateveremotionword comesto mind. The investigator
analyzesthe answersof the different observersto determinewhat emotions
they agreeabout in describingparticular faces. He might find, for example,
that 80 percentof the observersagreein describinga particularface with the
word "afraid." They might not agreeabout a word to describesomeother
face; for example, a face called "disinterest" by some observersmight be
27
22 research on facial expressions of emotion
labeledwith other emotionsby other observers. On the basisof suchresults,
the investigator reachesa conclusion about which emotions the face can
convey.
The six emotionsthat are the subjectof this book-happiness,sadness,
surprise, fear, anger, and disgust-werefound by every investigator in the
last thirty years who sought to determinethe vocabulary of emotion terms
associatedwith facial expressions.There are probably other emotions con-
veyed by the face-shameand excitement,for example;but thesehave not
yet been as firmly established. Becausewe will be showing not only how
thesesix emotionsappearon the face,but alsohow thirty-threedifferentblends
of thesesix emotionsappear,quite a large portion of the emotional reper-
toire of the face will be represented.
Are Judgments of Emotion Accurate?
It is not enoughto determinewhat emotionsare read from facial ex-
pressions. It is also crucial to discover whether the interpretationsof the
observersare correct or not. When peoplelook at someone'sface and think
that personis afraid, are they right or wrong? Are facial expressionsan ac-
curate reflection of emotional experience?Or are the impressionsgained
from facial expressionmerely stereotypes-allagree about them, but they
are wrong? To study this question the investigator must find some people
whom he knows to be having a particular emotional experience. He must
take some photographs,films, or videotapesof thesepeople and then show
them to observers.If the observers'judgmentsof the facial expressionsfit
with the investigator'sknowledgeof the emotionalexperienceof the persons
being judged, then accuracyis established.
Most of the studies of accuracyin judging facial expressionfailed to
provide conclusiveevidenceone way or another,usually becausethe investi-
gator's knowledge of the emotional experienceof the people being judged
was faulty. In our analysis of the experiments conducted over the last
fifty years,we did find consistentand conclusiveevidencethat accuratejudg-
ments of facial expressioncan be made. Some of these studies were con-
ducted in our own laboratory. In one experiment,photographswere taken
of psychiatric patientswhen they were admitted to a mental hospital, and
againwhen theywere lessupsetand readyfor discharge.Untrainedobservers
were shown thesephotographsand askedwhethereachfacial expressionwas
shown at time of admissionor at time of discharge.The judgmentswere
accurate. Thesesamephotographswere shownto anothergroup of observers
who were not told they were seeingphotographsof psychiatricpatients,but
insteadwere askedto judge whetherthe emotion shownwas pleasantor un-
pleasant.Again accuracywas proven, becausethe facial expressionsshown
research on facial expressions of emotion 23
at admissionwere judged as more unpleasantthan those shown atdischarge
from the hospital. In anotherstudy, other observerswere askedto judge how
pleasantor unpleasantthe facial expressionswere, but the faces shown to
them were of psychiatric trainees undergoing a stress interview. Without
knowing which was which, the observersjudged the facial expressionsduring
stress as moreunpleasantthan the facial expressionsdrawn from a non-
stressfulpart of the interview. In still another experiment,observerswere
showntwo films of collegestudents,one takenwhen they had beenwatching
a very unpleasantfilm of surgery, and one when they had been watching a
pleasant travelogue film. The observersaccurately judged which film the
collegestudentswere watching from their facial expressions.
All these studies were concernedwith spontaneousfacial expressions
which naturally occur when a persondoes not deliberately try to show an
emotion in his face. But what of those situationsin which a persondeliber-
ately tries to show an emotion, to look happy or angry, etc.? Many studies
haveindicatedthat observerscan accuratelyjudge which emotionis intended
when a person deliberately tries to convey an emotion through facial
expression.
Are There Universal facial Expressions
of Emotion?
Are facial expressionsof emotion the samefor peopleeverywhere,no
matterwhat their background?When someoneis angry, will we seethe same
expressionon his face regardlessof his race, culture, or language?Or are
facial expressionsa language,the meaningof which we must learn anew for
eachculture, just as we needto learn the verbal language?A little more than
one hundredyearsago, CharlesDarwin (seeReference1 at the end of this
chapter) wrote that facial expressionsof emotion are universal, not learned
differently in eachculture; that they are biologically determined,the product
of man'sevolution. Since Darwin's time many writers have emphaticallydis-
agreed. Just recently, however, scientific investigations have conclusively
settled this question, showing that the facial appearanceof at least some
emotions,thosecoveredin this book, is indeeduniversal, although there are
cultural differencesin when theseexpressionsare shown.
Researchconductedin our laboratory played a central role in settling
the dispute over whether facial expressionsare universal or specific to each
culture. In one experiment,stress-inducingfilms were shown to college stu-
dentsin the United Statesand to collegestudentsin Japan.Part of the time,
eachpersonwatchedthe film alone and part of the time the personwatched
while talking about the experiencewith a researchassistantfrom the person's
own culture. Measurementsof the actual facial musclemovements,captured
24 researchon facial expressionsof emotion
on videotapes,showedthat when they were alone, the Japaneseand Ameri-
canshad virtually identical facial expressions(see Figure 1). When in the
presenceof anotherperson,however,wherecultural rules aboutthe manage-
ment of facial appearance(display rules) would be applied, there was little
correspondencebetween Japaneseand American facial expressions.The
Japanesemaskedtheir facial expressionsof unpleasantfeelings more than
did the Americans. This study was particularly important in demonstrating
what about facial expressionis universal and what differs for eachculture.
The universalfeatureis the distinctive appearanceof the face for eachof the
primary emotions.But people in various cultures differ in what they have
beentaughtaboutmanagingor controlling their facial expressionsof emotion.
In anotherexperimentwe showedphotographsof the different emotion
expressionsto observersin the United States,Japan,Chile, Argentina, and
Brazil. The observersin these different cultures had to chooseone of the
six primary emotion words for eachphotographthey saw. If facial expres-
sions were a languagethat differs from culture to culture, then a facial ex-
pressionsaid to be angry by Americansmight be called disgust or fear by
peoplein Brazil, or might not meananything to them. Just the oppositewas
found. The samefacial expressionswere judged as showing the sameemo-
tions in all thesecountries,regardlessof languageor culture (seeFigure 2).
Essentiallythe sameexperimentwas carried out independentlyat the same
Fig. 1 Example of spontaneousfacial
expressionof a Japanese(on the left)
and an American (on the right) shown
when watching a stressfilm.
Fig. 2 Examples of the photographsutilized in a study of how
emotions are judged across literate cultures
PercentageAgreementin How Photograph Was Judged
Across Cultures
Fear
Disgust
Happiness
Anger
UNITED
STATES
(J=99)
85%
92%
97%
67%
BRAZIL
(J=40)
67%
97%
95%
90%
CHILE
(J=119)
68%
92%
95%
94%
ARGEN-
TINA
(J=168)
54%
92%
98%
90%
JAPAN
(J=29)
66%
90%
100%
90%
26 research on facial expressions of emotion
time by Carroll Izard (Reference8) with observersin eight different cultures,
and the same evidenceof universality was found.
Although we wished to interpret our findings as evidencethat some
facial expressionsareuniversal,oneloopholeremained.All the peoplestudied
had some sharedvisual contact, usually not directly but through the mass
media. It was still possiblethat facial expressionsmight really differ in all
the culturesstudied,but the peoplemight havelearned,throughmovies, tele-
vision, andpicturemagazines,what eachother'sfacial expressionsof emotion
looked like. Or facial expressionsof emotionmight be similar in all the cul-
tures we studied precisely becausethe people had all learnedhow to show
emotionon their face by watchingthe sameactorsin the movies or television
and imitating their facial expressions.We had not eliminated the possibility
that, amongpeople who did not have the opportunity to view mass-media
portrayalsof facial expressionsof emotion,emotionswould be shownby en-
tirely different facial muscularmovements.The only way to settle this ques-
tion was to study visually isolatedpeoplewho had no contactwith the mass-
media,andlittle if any contactwith the outsideworld.
We conducteda series of experimentsin the SoutheastHighlands of
New Guinea,wherewe were able to find peoplewho met thesecriteria. Be-
causethesepeoplewere in no way accustomedto taking psychologicaltests
or participatingin experiments,and becausewe did not know their language
but had to work throughtranslators,we had to modify our experimentalpro-
cedure.In othercountrieswe hadshowna singlephotographof oneor another
of the emotion expressionsand given the observera choice amonga list of
emotionwords. In New Guinea,we showedthe personthreephotographsat
once, had a translatorread an emotion story, such as "A person'smother
died," and askedthe observerto point to the photographthat fit the story.
We found that thesepeopleselectedthe sameface for the sameemotion as
did peoplein all the other cultures we had studied.There was but one ex-
ception: the New Guineansfailed to distinguishbetweenthe fear and surprise
facial expressions.
In a related experiment, other New Guineanswere told an emotion
story and eachwas askedto show the emotion on his own face. Videotapes
were taken of theseintendedemotion expressions,someexamplesof which
are shown in Figure 3. Analysis of theseNew Guineans'facial expressions
showedagain that the samefacial expressionswere producedfor the same
emotionsas had beenfound in other cultures,with the exceptionof fear and
surprise,which were confusedwith each other. Further confirmation of the
universalityof facial expressionswas obtainedby a study of anotherculture
in West Irian, t.lte western portion of the island of New Guinea. Karl and
EleanorHeider,who wereskepticalof our evidenceof universality,conducted
the sameexperimentswith peopleeven more visually isolatedthan thosewe
had studied,and they also obtainedevidenceof universality.
research on facial expressions of emotion 27
Taken together, our studies, those of Izard, the Heiders' study, and
evidencefrom Eibl-Eibesfeldt (an ethologist using very different methods),
showedquite conclusivelythat Darwin was correct in claiming that there are
universalfacial expressionsof emotion.
Although the appearanceof the face for eachof theprimary emotionsis
commonto all peoples,facial expressionsdo vary acrossculturesin at least
two respects.What elicits or calls forth an emotionwill usually differ; people
Fig. 3 Video frames of attemptsto poseemotionby subjects
from the Fore of New Guinea.The instruction for the
top left photographwas "your friend has come and
you are happy"; for the top right "your child has
died"; for the bottom left "you are angry and about
to fight"; for the bottom right "you seea dead pig
that has been lying there for a long time."
28 research on facial expressions of emotion
may becomedisgustedor afraid in responseto different things in different cul-
tures.Also, culturesdiffer in the conventionspeoplefollow in attemptingto
control or managethe appearanceof their faces in given social situations.
Peoplein two different culturesmay feel sadnessat the deathof a loved one,
but one culture may prescribethat the chief mournersmask their facial ex-
pressionwith a mildly happy countenance.
How Does Each Emotion Appear
on the face?
As we beganto find evidencethat there are somefacial expressionsof
emotionwhich are universal, and before all the studieswere completed,we
beganto investigatejust what these universal facial expressionsof emotion
look like. We soughtto constructan Atlas of the face, which would depict
photographicallyeachof the universalfacial expressionsof emotion. It is this
Atlas (Reference4) which forms the basisfor the photographsshownin the
subsequentchaptersof this book. Our first stepin developingthe FacialAtlas
was to study what othershad said about the appearanceof the face for each
of the primary emotions.Some writers had describedwhich muscleswere
contractedin particular emotions, while others concernedthemselvesonly
with the appearanceof the surfaceof the face. None had systematicallycon-
sideredall the musclesor all the consequentchangesin the surfaceappear-
anceof the face for the six primary emotions.
Putting together what was written by Darwin, Duchenne,a French
anatomistwhom Darwin had quoted extensively,Huber (Reference7), an
American anatomistwriting thirty yearsago, and Plutchik (Reference9), an
American psychologistconcernedwith emotion, we saw part of the picture
emerge.We constructeda table which listed all the facial musclesand the
six emotions,enteringinto the table what thesemen had written aboutwhich
muscleswere involved in what way for each emotion.Therewere many gaps,
however, where no one had said anything about the·involvement of a par-
ticular muscle in a particular emotion. Working with Silvan Tomkins (Ref-
erence10), we filled in thosegapswith information from our cross-cultural
studiesand our sharedimpressions.
The next stepwas to photographmodels,who were instructedto move
particular facial muscleslisted in the table. We separatelyphotographedthe
three areas of the face which are capableof independentmovement-the
brow/forehead;the eyes/lidsand root of the nose; and the lower face, in-
cluding the cheeks,mouth, most of the nose,and chin. The completedAtlas
consistsof a seriesof photographsof thesethree different areasof the face,
eachphotographkeyedto oneof the six emotions.As might well be expected,
for eachof the emotionsthereis more than oneAtlas photographfor at least
research on facial expressions of emotion 29
one facial area.For example,for surprisethere is one brow/forehead,one
eyes/lids/rootof nose,but four different Atlas photographsof the lower face.
The next obviousquestionwas whetherthe Atlas is correct.Are the six
emotions-happiness,sadness,anger,fear, disgust,and surprise-inactuality
composedof the facial appearanceslisted in the Atlas? Or does the Atlas
appearanceof disgustactually occur with anger,and so forth? We havecon-
ductedfour experimentson the validity of the Atlas. Two of the experiments
attemptedto prove its validity by showingthat measurementsof the face with
the Atlas correspondedwith other evidenceof the subjectiveemotionalex-
perienceof the personswhose faces were measured.Theseexperimentsin-
vestigatedthe experientialvalidity of the Atlas.
The other two experimentsinvestigatedthe social validity of the Atlas.
Ratherthan attemptingto prove that the Atlas measurementscorrespondto
the person'sexperience,thesestudiesinvestigatedwhetherthe Atlas measure-
mentscan predict what observersthink a personis feeling when they look at
his face. Although experiential and social validity should be related, they
neednot necessarilybe so. We may not look to othershow we actually feel,
at least all the time. Thus, it was necessaryto study both experientialand
social validity.
The studiesof experientialvalidity drew from materialsgatheredin one
of the cross-culturalstudies of facial expressionsdescribedearlier. College
studentsin Japanand in the United Stateshad individually watchedpleasant
and unpleasantmovies while we videotapedtheir facial expressions.From
their answersto questionnairesafter the experiment,it was clear that they
experiencedvery different emotionswhile watchingthe two films. In describing
their reactionsto the travelogue,the subjectshad said it was interestingand
pleasant,and causedthem to feel moderatehappiness.In describingtheir re-
actionsto the surgical film, the subjectssaid they had unpleasant,disgusted,
pained,fearful, sad, and surprisedfeelings. If the Facial Atlas is valid, then
measurementsbasedon it should be able to distinguish betweenthe facial
expressionsshownwhenthesetwo differentsetsof emotionswereexperienced.
All the facial muscularmovementsvisible on the videotapeswere iso-
lated, their duration was measured,.and they were classifiedin terms of the
Atlas. This measurementprocedurewas done in slow motion, with the mea-
surementsmadeseparatelyfor the three areasof the face, by three separate
technicians.Such precise measurementrequired about five hours for each
minute of videotapedfacial behavior.The resultswere very clear-cut.Mea-
surementswith the Facial Atlas clearly distinguishedthe two emotionalcon-
ditions-whethersubjectshadbeenwatchinga stressfulfilm or a travelogue.
And the Atlas was equally successfulwith the facial expressionsof Japanese
subjectsand with Americans,as it should be, becauseit was built to show
the universalfacial expressionsof emotion.One limitation of this experiment,
however, is that it didn't determinewhether the Atlas correctly depicts the
30 research on facial expressions of emotion
facial appearancesfor eachof the six emotions.It only showsthat the Atlas
is correct in distinguishingbetweenunpleasantand pleasantexperiences.
The secondexperientialvalidity study provided a partial remedyto this
limitation. Recentresearchon the physiology of emotionssuggeststhat there
are markedly different patternsof heart rate accelerationand deceleration
with the emotionsof surpriseand disgust. Measuresof heart rate and skin
conductancehad beengatheredon the Japaneseand Americansubjectswhen
they were watching the pleasantand stressfulfilms. If the Atlas is correct in
what it says a surprise face and a disgust face look like, then when the
Atlas says such facial expressionsoccur, there should be a different pattern
of heart rate for each.When we examinedthe changesin heart rate which
coincidedwith facial expressionsthe Atlas had designatedas either surprise
or disgust, the results showed the predicteddifference.
Although this secondstudy doesprovide evidenceof the validity of the
Atlas for surpriseand disgust, it doesn'tshow that the Atlas is necessarily
valid in what it says about the other emotions-anger,happiness,sadness,
and fear. Logically, if it is shown to be valid for surpriseand disgust, the
Atlas shouldbe equally valid for the other emotions,becauseit was derived
by the samemethodfor all six emotions.But evidenceis still required, and
for that we turn to the third study, which examinedthe Atlas in terms of
social validity. Could the Atlas predict how observerswill interpret facial
expressions?
Photographsthat had been taken by many different investigatorsof
facial expressionwere obtained. These pictures were shown to observers,
who wereaskedto judgewhich of the six emotionswasshownin eachpicture.
Only thoseon which the observershad agreedabout the emotion expressed
in the face were further