ArticlePDF Available

Children's scripts for social emotions: Causes and consequences are more central than are facial expressions

  • Committee for Children


Understanding and recognition of emotions relies on emotion concepts, which are narrative structures (scripts) specifying facial expressions, causes, consequences, label, etc. organized in a temporal and causal order. Scripts and their development are revealed by examining which components better tap which concepts at which ages. This study investigated whether a facial expression or a brief story describing an emotion's cause and consequence was the stronger cue to basic-level and social emotions. Children (N = 120, 4-10 years) freely labelled the emotion implied by faces and, separately, stories for six basic-level emotions (happiness, anger, fear, surprise, disgust, and contempt) and three social emotions (embarrassment, compassion, and shame). Cause-and-consequence stories were the stronger cue overall, especially for fear, disgust, and social emotions. Faces were the stronger cue only for surprise. Younger children assimilated social emotions into basic-level emotion categories (sadness and anger); older children differentiated them. Differentiation occurred earlier for stories than for faces.
Children’sscripts for social emotions:Causes and
consequences aremorecentral than arefacial
Sherri C. Widen* and James A. Russell
Boston College, Chestnut Hill, Massachusetts, USA
Understandingand recognitionofemotionsreliesonemotion concepts,which are
narrativestructures(scripts) specifying facial expressions, causes,consequences, label,
etc. organizedinatemporal andcausalorder.Scripts andtheir developmentare revealed
by examiningwhich components better tapwhich concepts at whichages. This study
investigated whetherafacial expression or abrief story describing an emotion’scause
andconsequence wasthe stronger cuetobasic-level andsocialemotions. Children
( N¼120, 4–10 years) freely labelled theemotion impliedbyfaces and, separately,
storiesfor sixbasic-level emotions (happiness,anger,fear, surprise, disgust, and
contempt)and threesocialemotions(embarrassment, compassion,and shame).Cause-
and-consequencestories were thestrongercue overall, especially forfear, disgust, and
social emotions.Faces were thestrongercue only forsurprise. Yo ungerchildren
assimilatedsocialemotionsintobasic-level emotioncategories(sadnessand anger);older
children differentiated them.Differentiation occurred earlierfor storiesthanfor faces.
Akey developmental task forachild is to cometounderstand the emotions they
experience and witness. Understanding of emotion has been implicated in preschoolers’
cognitive and linguistic development (e.g., Blair,2002), their health (e.g., Rieffe,
Meerum Terwogt,&Jellesma, 2008), and their later school readiness (Brown &Dunn,
1996; Garner &Waajid, 2008; Miller et al.,2006; Raver,Garner, &Smith-Donald, 2007;
Trentacosta&Izard, 2007). Acquisition of emotion understanding is partofthe
development of emotional intelligence (Barrett&Salovey,2002; Zeidner,Matthews,
Roberts, &MacCann, 2003).
One keyaspect of understanding emotion is identifying different emotions:
distinguishing anger from sadnessfrom compassion. As adults, we do so using emotion
scripts. Eachemotion has its own script including its eliciting event, conscious feeling,
facial expression, vocalization, action, physiological manifestation, label, and so on,
aligned in acausal and temporal order.Attributing an emotion to oneself or to another
requires that one’scurrent experience or observation resembles the script forthat
*Correspondence should be addressed to Dr Sherri C. Widen, Department of Psychology,Boston College, McGuinn Hall,
140 Commonwealth Avenue,ChestnutHill, MA 02467, USA (e-mail:
British Journal of Developmental Psychology(2010), 28, 565–581
q2010 The British Psychological Society
emotion(Fehr &Russell,1984). The general question, of which the present study is a
part, is: What is the developmental progression in the formation of emotion scripts? This
question has receivedrelatively little attention, and the field lacks afull treatment of it.
One possibility is that children begin with an innate, or at least prepared, set of
discrete mentalcategories foratleast those emotions with corresponding facial
expressions (e.g., Izard,1971, 1994; To mkins, 1962). On this perspective, certain facial
expressions evolved as an emotion signallingsystem (Ekman, 1994; Izard, 1994;
Kestenbaum &Nelson, 1992; Lenti, Giacobbe, &Pegna, 1997; Pell, 2005; Rinn, 1991;
Susskind, Littlewort, Bartlett, Movellan, &Anderson, 2007; Tomkins, 1962).An
evolutionaryaccount of emotion signalling requires not just the productionofemotion
facesbut their recognition as well (Lenti-Boero, 1994); thereisnoadaptive value in
producing an unrecognizedsignal. Children (like adults)interpret facial expressions in
terms of discrete emotions: the ‘happyface’ is interpreted as happiness, the ‘sad face’as
sadness, and so on (e.g., Denham&Couchoud, 1990; Harrigan, 1984; Hornik,
Risenhoover,&Gunnar, 1987; Izard, 1971; Markham, &Adams, 1992; Repacholi, 1998;
Repacholi &Gopnik, 1997; Wiggers&van Lieshout, 1985).The ability to recognize the
specific emotion conveyed by afacial expression has been theorized to be in place well
within the first half year of life (Izard, 1971)–implying that discrete mentalcategories
are already in place.
With aprepared understanding of the link between emotion and facial expressions, a
child can build ascript of that emotion by addinginformation about the causes,
consequences, label, and so on, as that information is acquired. In other words, the early
recognition of facial expressions in terms of discrete emotions has been assumed to be
the basis –the bedrock –ofyoung children’sunderstanding of emotion and to provide
the foundation on which later learning about emotion is built (e.g., Denham, 1998;
Harris, 1989; Izard,1994; Pons, Harris, &deRosnay,2004; Saarni, 1999; Walker-
There are alternative possibilities. The one we are developing is this: children
initially understandemotions in verybroad mentalcategories and over the course
of development differentiate these categories into narrower,more adult-like ones
(e.g., Bridges, 1930; Fischer,1980; Widen &Russell,2003, 2008). Initially,children
begin with two categories based on asinglepleasure versus displeasure dimension. It is
an empirical question what cues are initiallytied to these categories. The cues could be
faces(smiles and frowns), or theycould be amorescript-based primitive theoryof
mind (you getwhat you want, youfeel good; youdonot getwhat you want, you feel
bad). With time, achild might notice that some negative emotions are caused by a
loss, have facial expressions that involvedowncast eyes, downturnedmouth, and tears,
result in whining and crying,makethe person withdraworseek comfort, and
show signs of lowered arousal. Other negative emotions are caused by one person
blocking another’sgoal, have facial expressions with knittedbrows,staring eyes,
and clenched jaw, result in yelling, make the person approachthreateningly,and have
high levels physiological arousal. As aconsequence, the child would then differentiate
the initially broad negative emotion categoryinto two separate categoriesand
apply different labels to them ( sad,angry)(although at this level bothofthese
categories remain broader than the adult version). Thus,scripts are formed through
differentiation. This account opens up the question of whetherfacial expressions
are indeedthe initial cues forcategorization.
The discrete category account and the differentiation account predict differentroles
forfacial expressions in achild’sidentification of emotion. On the first account, facial
566 SherriC.Widen and James A. Russell
expressions of emotion are asignalling system,and therefore should be strong cues to
emotionfor children. Facial cues are predicted to be the earliestcues to another’s
emotionfor all emotions that have facial expressions and to remain the most definitive
cue at all ages (e.g., Denham, 1998; Harris, 1989; Izard, 1971, 1994).Onthis view,faces
remain the strongest cues because theyare pre-wired to signalaspecific discrete
emotion, whereas the otheraspects of the script are only probabilistically associated
with an emotion and must be learned through association with the facial expression.
On the second account, there is no privileged cue to aspecific emotion.Thus,
at differentagesand fordifferentemotions, one aspectofthe script, such as the
emotion’scause, consequence, or label, may be the strongest cue (Russell &Widen,
2002; Widen &Russell,2004). Aface may or maynot be the strongest cue, forexample,
forthe youngest preschoolersand early emerging broad emotion categories. Acause
(loss vs. frustration) or consequences (slumping vs. hitting) or emotion label (sad vs.
angry) might be stronger cues to emotion than facial expressions forolder children and
later-emerging emotions (e.g., fear,disgust).Thus, as children’sunderstanding of
emotionincreases, the emotion cue that is most powerful in tapping that knowledge
may change.
The course of the development of scripts can be revealed by examining the power of
different cues fordifferent emotions at differentages(Balconi&Carrera,2007; Camras
&Allison, 1985; Markham &Adams, 1992; Reichenbach &Masters, 1983;Russell, 1990;
Russell &Widen, 2002; Smith &Walden, 1999; Widen&Russell,2002, 2004, in press).
One study of emotion concepts in 2- and 3-year-olds supported the prediction of the
discrete category view by showing aface superiority effect (Widen &Russell, in press;
Study 1): overall, children were more likely to use the ‘correct’ emotion label when
shown the emotion’sfacial expressionthan when told the emotion’scause and
consequence;this effect wassignificant forsadnessand anger,two early emerging
emotioncategories (Widen&Russell, 2003). In contrast, 10 of 11 studies with slightly
older children found aface inferiorityeffect. Children’s(3years and older) labelling
performance waslower when given afacial expressionthan when given an emotion’s
cause and consequence (Balconi &Carrera, 2007; Reichenbach &Masters, 1983; Smith
&Walden, 1999; Widen&Russell, 2002, 2004, in press, Study 2) or label (Camras&
Allison,1985; Russell, 1990; Russell &Widen, 2002, Widen&Russell,2004).Thisface
inferiority effect was particularly strong forfear (Balconi &Carrera, 2007; Camras &
Allison,1985; Russell, 1990; Russell &Widen, 2002; Widen&Russell,2004) and disgust
(Camras&Allison, 1985; Markham &Adams, 1992; Russell &Widen, 2002;Widen &
Russell,2004), two later-emerging emotions. In only one study didchildren’s(4–8 years)
overall labelling performance not differsignificantlygiven emotions’causes and
consequencesorfacial expressions (Markham &Adams, 1992).
The studies reviewed so farwere restricted to so-called basic-level emotions
(e.g., anger,fear), but otherstudies investigated children’sunderstanding of what have
been called social emotions (e.g., embarrassment, shame, compassion, etc.; Heerey,
Keltner,&Capps,2003; Seidner,Stipek, &Feshbach, 1988; Shamay-Tsoory, Lavidor, &
Aharon-Peretz, 2008). For convenience and in keeping with commonpractice,
we distinguish basic-level from social emotions, and we use that terminology,
although we doubt that the division is clear.Perhaps abetter characterization would
be early- versus later-emerging concepts. Children’sunderstanding of basic-level
emotions does emergeearlier than their understanding of social emotions (e.g., Brody &
Harrison, 1987; Harris, Olthof, Meerum Terwogt, &Hardman, 1987; Wiggers&
van Lieshout, 1985), and forcontinuitywewill also use that terminology.
Scripts for social emotions 567
Prior studies of social emotions have not compared children’sunderstanding of faces
with other cues to social emotions because there was not aset of standardized and
tested facial expressions forsocial emotions.Amajoropportunity to test the assumption
of the power of facesascues to social emotions occurred when facial expressions for
compassion, embarrassment, shame, and contempt were identified (Haidt &Keltner,
1999).Aset of standardizedprototypicalfacial expressions forthese emotions opened
the door to the examination of the power of facial expressions relative to other aspects
of the script to evoke the concept.
Overview of the current study
In the current study,wetested the competing predictions of the discrete emotions
account versus the differentiation account in two ways. First, we compared the strength
of facial expressions relative to cause-and-consequence stories as cues to both basic-
level and social emotions. The discrete emotions account predicts that the faces will be
strongest forall emotions and forall ages (4–10years). The differentiation account
predicts that although faces may be the stronger cue forsome early emerging categories
(e.g., anger), cause-and-consequence stories will be the stronger cue forlater-emerging
categories (e.g., fear,disgust, social emotions).
Second, we investigated whether differentiationdescribes children’sacquisition of
emotionconcepts and, if so, the basis of that differentiation. We propose that children
initially assimilate asocial emotion, such as embarrassment, to aconcept theyalready
possess, such as sadness, but later distinguish the social emotion from the earlier-
emerging emotion concept. Alternatively,onthe discrete emotions account, recognition
of facial expressions is pre-wired and emotion categories are discrete. Thus, once
children have learned the appropriate label foreach face,errorsshould be rare; before
that, errorsshould be random.
Each child participated in two phases. In the first phase,tomake labels forthe target
emotions accessible in the child’sworking vocabulary, the experimenter initiated a
brief conversation in which each label was mentioned. In the second phase, children
were given an emotion cue and then asked to provide alabel of their choice in response
to the question, ‘How does Joan feel?’ fornine emotions.Happiness was presented in
both modes and servedasagatekeeper trial –childrenhad to label this trial happy
(or some close synonym) to be included in the sample. The other eight emotions were
presentedfirstinone of two modes of presentation and then in theother
(counterbalanced) in awithin-subjects design: in one mode, each emotion was
presented as afacial expression and, in the other,asastoryconsisting of astereotypical
cause and behavioural consequence.
Participants were 120 childrenenrolled in preschools and after-school care programmes
in the Greater Boston area. All children were proficient in English. The sample
wasdivided intothree agegroups of 40 childreneach(20 girls,20boys):
preschoolers(48–65 months,mean ¼58:9months), Kindergarten–Grade 1(65–86
months,mean¼72:5months), and Grade 2–3 (85–130 months,mean ¼99:6months).
The sample was representative of the ethnic compositionofthe area: 62.5% were
568 SherriC.Widen and James A. Russell
Caucasian, 16.7%Asian, 5.0% Hispanic, 4.2% of mixedethnicity, and 11.6% other. On the
consent form,parentswereaskedtoindicate the highestlevel of education completed
by each parent (asaproxy of socio-economic status (SES)) on the following six-point
scale: (1) some high school; (2) high school diploma or GED; (3) some college,
vocational degree, associates degree; (4) 4-year collegedegree (BA, BS); (5) master’s
degree (MA, MS); (6) doctorate(PhD,MD, MBA, JD,EdD,ThD). While education alone is
not asufficient indicator of SES, it has been used in the past as amajor component of
indicesofSES (Hollingshead &Redlich,1958; Norton et al.,2005).Parents’mean
education level forthis sample was 4.9 ( SD ¼1:2).
Photographs of facial expressions
The gold standard of facial expressions of emotion are Facial Action Coding System
(FACS)-coded facial expressions published by Ekman and Friesen (1976),but these faces
are now considered dated (Goeleven, de Raedt,Leyman, &Verschuere, 2008).Other
more recent and more modern-looking sets of FACS-coded facial expressions are now
available (e.g., Montreal Set of Facial Displays of Emotion, Beaupre
´,Cheung, &Hess,
2000; Karolinska Directed Emotional Faces, Goelevenet al.,2008; Japanese and
Caucasian Facial Expressions of Emotion, Matsumoto &Ekman, 1988; The NimStim Set
of Facial Expressions, Tottenham et al.,2009). The nine prototypicalfacial expressions
(happy, angry, scared, surprised, disgusted, contemptuous, ashamed, embarrassed, and
compassionate)weused in the current study were developed by Haidt and Keltner
(1999). Thisset has also been FACS coded and validated in that study by adults in two
cultures (USA and India).
Stories of the causes and consequences of emotions
There was one storyfor each emotion (Table 1) which included bothastereotypical
emotion-eliciting event and abehavioural consequence. Stories forhappiness, anger,
fear,surprise, and disgust werebased on priorworkinour laboratoryinwhich children
generated causes and consequences forspecific emotions (Russell, 1990;Russell &
Widen, 2002).Stories forembarrassment, compassion, shame,and contemptwere
elicited in the same wayand then refined through piloting with adults and children. In a
pilot study ( N¼12, 7–8 years), between 58 and 75% of childrenused the target label
(or synonym; see scoring below)for embarrassment, compassion, and shame; these
figures were within the rangeobtained fordisgust and fear (67–83%). Only 17% of
children used the target label forcontempt but an additional 42% used anger forthis
story, which is in keeping with the differentiation proposal that childrenwould
assimilate social emotions to an earlier-emerging emotion category.
On the initial visit to the child carefacility,the experimenter spent time playing with
those children whohad parental consent to participate in the study until each child
seemed comfortable with the experimenter.Onasubsequent visit,the experimenter
invited each child to play agame with her.This‘game’ lastedonaverageless than 10 min
and consisted of two phases. The first phasewas aprimingsession. In the second phase,
the child was asked to label Joan’semotion based on Joan’sfacial expression
Scripts for social emotions 569
or,separately,abrief storydescribing the cause and consequence of Joan’semotion.
The happytrial included boththe facial expression and the story, and it served as a
gatekeeping trial: children had to labelthis trial as happiness to be includedinthe
sample. The child was randomly assigned to either the face- or the story-first condition.
There wereeight trials in each modeofpresentation, one foreach emotion, presented in
variousrandom orders.
Simple priming
Although children acquireemotion labels at aearly age(e.g., Bretherton &Beeghly,
1982),accessibility of the associated concepts can be enhanced. The purposeofthe
simple priming procedure was to ensurethat the target emotion labels were as
accessible as possible in each child’svocabulary. In simple priming, the experimenter
initiated aconversation in which each of the target emotion labels was introduced by
Ta ble 1. Stories with causes and consequences for each emotion
Emotion Story
Happiness :::it was Joan’sbirthday. All her friends came to
her birthdaypartyand gave her presents. Joan
jumped up and down and clapped her hands.
Anger :::Joan was waiting in line.Then aboy cut in line in
front of her.Hedidn’tevenask. Joan shoved
him out of line and yelled at him.
Fear :::Joan was walking down the street when abig dog
started growling and chasing her.Joan screamed
and ran away afast as she could.
Surprise :::Joan came home,and her mom’shair was pink. This had
never happened before. Joan just stared and tried to
figure out whyher mom’shair was pink.
Disgust :::Joan took abig bite of an apple. But it was rotten inside.
It tasted awful. Joan spit it out as fast as she could and
threwthe apple on the ground. She did not want to touch it.
Embarrassment :::Joan spilled grape juice all over her white dress. All the kids
laughed at her.Joan’sface turned veryred,and she looked
away from everyone. She wished that she could hide.
Compassion :::Joan was walking on aslipperysidewalk. Joansaw another
kid slip and hurthimself verybadly.Joanwent over to the
boy to see if he was okay.
Shame/Guilt :::Joan took her sister’sfavorite teddy bear and threwitinthe trash.
Her sister cried and cried. Joanwanted to give it back but she
couldn’tbecause her mom had already taken out the trash.
Joan stayedinher room and didn’twant to talk to anyone.
Contempt :::Joan was at school. There was aboy in her class who always did
stuff to get the teachers attention. He was always acting up in
class so she would notice him. Or if the teacher wanted
someone to help her,hewas always wanted to be picked.
Joan didn’ttalk to that boy,and she didn’tsit next to him.
Note.The happystorybegan with, ‘Once upon atime,there was agirl name Joan. One day:::’Each of
the other stories began with, ‘One week later :::
570 SherriC.Widen and James A. Russell
saying, ‘Today we are going to play agame about feelings. Feelings are like whenyou feel
happyorangry. Do you ever feel happy?What about angry?Doyou ever feel angry?’
And so on, until each of the target emotion labels (happy ,angry,scared,surprised,
[for contempt], embarrassed,feel sorryfor someone[for
compassionate], ashamed,and just okay)had been mentioned.The experimenter
didnot discusswhen or whytheseemotionsmight occur. When thechild
spontaneously offered an example of someone feeling aparticular emotion, the
experimenter listened but didnot comment or encouragefurther explanation. When
presenting the emotion words,and throughout the experiment, everyeffortwas made
to use aneutral toneofvoice.
Happy trial
The happytrial was always first. The experimenter began, ‘In this game, I’m going to tell
you some stories about things that happen to Joan. Aftereach one, you gettotell me
how you think Joan feels. How does that sound? Remember:listencarefully,soyou can
tell me how Joan feels’. And then continued, ‘Once uponatime, therewas agirl named
Joan. This is what Joan looked like [showing neutral photo].One day,itwas Joan’s
birthday.All her friendscame to her birthday party and gave her presents. Joan jumped
up and down and clapped her hands. Andshe looked like this [showing happiness
photo].How does Joan feel?’
Up to this point, all the children had been treated identically.Children were now
randomly assigned to either the story- or the face-first condition –which was thus a
between-subjects condition. In each condition, the experimenter continuedseamlessly
on to the next emotion trial, but now the child was presented with either afacial
expression or astory(insteadofboth as in the happytrial). Aftercompleting the first
condition consisting of eight trials, the experimenter introduced the other condition by
saying, ‘Do you know what else Ibrought?Ibrought some picturesof(stories about)
Joan. Wouldyou like to see them?’
The experimenter presented the facial expressions one at atime in arandom order.
While showing each face, the experimenter said, ‘One weeklater,Joan felt like this’, and
asked, ‘How does Joan feel?’
The experimenter presented the stories one at atime in arandom order.Each
storybegan, ‘One weeklater :::’After each story, the experimenter asked, ‘How does
Joan feel?’
We did not expect children in this age rangetoknowthe word ‘contempt’. Even the synonyms forcontempt (e.g., derisible,
derisive,despicable) seemed unlikely to be apartofchildren’svocabularies,since college-educatedadults rarely use any of
these labels to label facial expressions as contempt. Thus,weselected ‘dislike’ as an emotion label that children were likely to
understand and that also came close to the meaning of contempt. The same reasoning wasused in choosing ‘feel sorryfor
someone’ forcompassion.
Scripts for social emotions 571
Responding to children’s responses
In bothmodes of presentation, responses were not corrected and all were mildly
praised(e.g., ‘Good answer’; ‘You are good at this game’). If no responsewas given, the
experimenter used variousprompts (e.g., In the storymode, repeating the story; asking
‘How would you feel if :::’and repeating the storywith the child as the protagonist. In
the face mode, repeating the question, asking the child to ‘Look closely.Ithink youcan
figureitout.’ ‘Why is Joan making this face? How does she feel?’).Ifthe child still did not
respond,the experimenter went on to the next face or story, and, after the other trials in
both modes, returned to any to which the child hadnot responded.Atnotime did the
experimenter use the word emotion,provide any emotion labels, or otherwise direct
the child to trytouse an emotion label beyond asking how Joan wasfeeling.
The participants were allowed to use any label theychose. The scoring keyfor basic-
level emotionsused in this study was drawn from Widenand Russell (2003), who
described the development of ascoring keybased on ratings of two judges blind to the
source of the labels.That same method was used in the current study to develop a
scoring keyfor the social emotions.The labels that children used that were scored as
correct foreach categorywere: forhappinesscontent,excited,glad,good,happy ,
proud;for fear,afraid,freaked out,frightened,nervous,scared,shy,worried;for
surprise, shocked,startled,surprised;for disgust, disgusted,gross,icky,nauseous,sick,
yucky;for anger,angry ,annoyed,cross,frustrated,grumpy,jealous,mad;for sad,
blue,disappointed,discouraged,lonely,sad,upset;for embarrassment, embarrassed;
forcompassionate, concerned,sad forsomeone ,sorryfor;for shame, ashamed,guilty,
sorry ,sorrythat;and forcontemptuous, contempt,does not like.Responses varied
from what was just listed in syntax or by being embedded in aphrase(e.g., very scared ).
These were all the labels children used in the current study that came close to specifying
the specificemotion.
The children had atotal of 2,040 opportunities to provide alabel. Of these, 890 were
the target emotion labels forthe given stimulus, 942 were anon-targetemotion labels for
the stimulus, and 208 were uncodable or non-responses (e.g., ‘I dunno’, silence).
Results and interpretation
Relative power of stories and faces to tap children’semotion concepts
To investigate the relative power of stories vs. faces to tap emotion concepts, we
examinedamixed design ANOVA ( a¼:05) in which agegroup (three levels:
preschoolers, Kindergarten–Grade1,Grade2–3), gender (two levels), and order
of presentation (two levels: storyfirst, face first) were between-subjects factors;
modeofpresentation (two levels: story, face) and emotion (eight levels: anger,fear,
surprise,disgust,contempt, embarrassment, compassion,shame)werewithin-
subjectfactors. The dependent variable waswhether or not the child used the target
label foreachstoryorface, coded 1or0,respectively. There werenosignificant main
or interactive effects forgender or order-of-presentation.
Based on prior research, some effects wereexpected.The main effect forage was
significant, Fð2 ;108 Þ¼59:80, p,:001: Grade2–3s’ (.56)performance was
significantly higher ( p,:001) than the two younger groups’, Kindergarten–Grade 1s’
572 SherriC.Widen and James A. Russell
performance (.38) wassignificantly ( p¼:01) higher than preschoolers’ (.27). Ta ble 2
shows the proportion of children in each agegroup who used the target label foratleast
one of the emotion stimuli (face or story) foragiven emotion. Althoughthe relationship
was not perfectlylinear,performance improved with age. If we take as the criterion
forpossession of the concept 50% of children at agiven agecorrectly using aterm,
we seethatpreschoolerspossess theconcepts of anger,fear, andsurprise;
Kindergarten–Grade 1s add the conceptofcompassion; Grade 2–3s the concepts of
embarrassment, shame, and disgust. Contempt was yet to be acquired in this agerange.
There werethus largedifferences in the ageatwhich different emotion concepts
emerged, with anger already in place forpreschoolersand with embarrassment, shame,
and disgustnot emerging until 5years later.
The main effect foremotion wasalso significant, Fð7;756Þ¼142:71, p,:001. The
social emotions werelabelled ‘correctly’ more frequently than disgustand contempt,
which some theorists have called basic-level emotions. The rank order foremotion
(from highesttolowest) was: anger (.93), fear (.62), surprise (.60), compassion (.31),
embarrassment (.25), shame (.23),disgust(.22), and contempt(08). Both ageand
emotionmain effects were qualified by interaction effects, which are described below.
The main effect formode of presentation was significant, Fð1 ;102 Þ¼115:14,
p,:001. Children’sperformance was significantly higher in the storymode than in the
face mode: story, .49; face, .32. The mode-of-presentation £emotion interaction
(Figure 1) wassignificant, Fð7;756Þ¼30:13, p,:001. The storymodewas more
powerful forfive of the eight emotions (fear,compassion, embarrassment, disgust, and
shame). For anger and contempt, therewas no significant difference between modes;
these two emotions were near the ceiling and floor,respectively. For surprise, the
overall patternwas reversed (Figure 1), showing aface superiority effect.
The age£mode-of-presentation interaction wasalso significant, Fð2;108Þ¼59:80,
p,:001. Children’sperformance increased with ageinboth modes, but more rapidly in
the storymodethan in the face mode (Figure 2). In the storymode, the increase in
performance was significant between each agegroup ( p,:001). In the face mode, the
increase in performance was more gradual than in the storymode, though performance
Ta ble 2. Percentage of children who correctlylabelled each emotion category
Age HappyAnger Fear Surprise Compassion Embarrass Disgust Shame Contempt
Preschool 100.0 100.0 92.5 65.0 32.5 25.0 25.0 7.5 5.0
–Grade 1
100.0 100.0 95.0 77.5 52.5 37.5 25.0 35.0 5.0
Grade 2–3 100.0 100.0 97.5 97.5 82.5 77.5 67.5 67.5 30.0
Mean 100.0 100.0 95.0 80.0 55.8 46.7 39.2 36.7 13.3
Note.Bold indicates an age at which at least 50% of the childrenlabelled that emotion correctly. To be
‘correct’, children had to correctlylabel at least one storyorface for agiven emotion. Each child used
at least one label from the set of labels deemed correct for that emotion concept; labels that children
used that werescored as correct for each categorywere: for happiness content,excited,glad,good,
happy,proud;for fear,afraid,freaked out,frightened,nervous,scared,shy ,worried;for surprise,shocked ,
startled,surprised;for disgust, disgusted,gross,icky,nauseous,sick,yucky;for anger,angry ,annoyed,cross,
frustrated,grumpy,jealous,mad;for sad, blue,disappointed,discouraged,lonely,sad,upset;for
embarrassment, embarrassed;for compassionate,concerned,sad for someone,sorryfor;for shame,
ashamed,guilty,sorry ,sorrythat;and for contemptuous, contempt,does not like.
Scripts for social emotions 573
did increase significantly between each agegroup ( p,:05). At each age, the advantage
forthe storymodewas significant ( p,: 01).
Number of labels used increases with age more for stories than forfaces
Aricher view of children’scategorization of emotion can be had by looking more
broadly at all the responses (both correct and incorrect) the children made. Thus, for
each emotion stimulus, we tallied the differentemotion categorylabels used. Figure 3
shows the modal labels children used at each agewhen the stimuli werefaces
(Figure 3A) and stories (Figure 3B). (For any stimulus forwhich there were two modal
responses, we used agenerous decision rule and treated the ‘correct’label as the mode.)
Proportion correct
Figure1.Proportion of children who ‘correctly’ labelled each emotion.
Proportion 'correct'
Preschool K-Gr1 Gr2-3 Preschool K-Gr1 Gr2-3
Age group
Face Story
Figure2.Proportion of children in each age group who ‘correctly’ labelled emotions in each mode.
574 SherriC.Widen and James A. Russell
The number of different labels that weremodal increased with agemore rapidly with
stories than with faces.
Consider Figure 3A in which children werepresented with faces. Preschoolers
divided all faces into only four categories, which theylabelled anger,surprise,fear,and
sadness.(This column illustrates our contention that children use each emotion label to
Labeling Faces
Angry face (37)
Contempt face (8)
Disgusted face (32)
Surprised face (23)
Scared face (13)
Surprised face (29)
Scared face (21)
Surprised face (35)
Scared face (24)
Angry face (39)
Contempt face (16)
Disgusted face (36)
Angry face (39)
Contempt face (8)
Disgusted face (37)
Embarrassed face (19)
Compassionate face (12)
Ashamed face (26)
Embarrassed face (15)
Compassionate face (12)
Ashamed face (16)
Embarrassed face (22)
Compassionate face (5)
Ashamed face (13)
Kindergarten &Grade 1s Grade 2-3s
Embarrassed story(14) Embarrass
Embarrassed story(31)
Labeling Stories
Angry story(35)
Contempt story(25)
Disgusted story(14)
Ashamed story(13)
Angry story(38)
Contempt story(24)
Disgusted story(27)
Preschoolers Kindergarten &Grade 1s Grade 2-3s
Angry story(36)
Contempt story(23)
Disgusted story(15)
Ashamed story(14)
Ashamed story(24)
Surprised story(12)
Scared story(28)
Compassionate story(12)
Embarrassed story(21)
Surprised story(15)
Scared story(24)
Surprised story(29)
Scared face (39)
Compassionate story(33)
Compassionate story(20)
Figure3.The modal label that was used for each (A) face or (B) storybyeach age group (the number
of children that used the label is in parentheses). By 6years, children did not use sadness modally for
any of the stories in this study,but, based on prior research (e.g. Russell &Widen, 2002), we assume
that at all ages theywould have labelled sad stories as sadness.
Scripts for social emotions 575
cover abroader range of phenomena than do adults.) No changeisseen when we move
to the Kindergarten–Grade1s. Some progress is seen in Grade 2–3s, who now
differentiate the sadnesscategoryinto sadnessand shame.That is, Grade 2–3s showed
five modal responsesrather than four.
Now turntoFigure 3B. Here, preschoolersdivided all emotion stories into five
categories, which theylabelled anger,sadness,surprise,fear,and compassion.
Kindergarten–Grade 1s differentiated the anger categoryinto anger and sadnessand the
sadnesscategoryinto sadnessand embarrassment. Thus,the Kindergarten–Grade 1s
have six categories. The Grade 2–3s differentiated the anger categoryinto anger and
disgust, and the sadness categoryinto sadnessand shame. Thatis, the Grade 2–3s had
eight categories. Thisisthe quantitative patternanticipated by adifferentiation account.
The current study was designed to investigate the development of children’semotion
scripts and contrasted two perspectives: one in which facial expressions are assumed to
be primaryinall emotioncategories that have facial expressions and one in which all
aspects of the script are assumed to be acquired in tandem and later-emerging emotion
categories differentiate from earlier-emerging categories. The results of this study
support aspects of bothperspectives.
Based on the results of the current and prior research, we propose that it is probable
that the basis of very young children’searliestemotion categories –those broademotion
categories theylabel happy ,sad,and mad –and also surprise is afacial expression. This
face superiority effect has been shown forpreschoolersfor sadness (Widen &Russell,
in press), anger (Russell &Widen, 2002; Widen &Russell,2002, in press), and surprise
(Russell &Widen, 2002; Wiggers&van Lieshout, 1985).Inthe currentstudy,wefound
the face superiority effect forsurprise, but notfor anger.
In contrast to this face superiority effect forearly developing categories and surprise,
we found aface inferiorityeffect forfive emotions –fear,compassion, embarrassment,
disgust, and shame. Children’shigherperformance forthese emotions on stories than
on the corresponding face suggests that childrenare learning aboutthese later-emerging
emotioncategories from their observations of the events that cause them and the
behavioursthat result, more than from the facial expression. Thus,causes and
consequencesofemotions help children learnthe script forthese emotions.This
advantage forstories supportsprior findings that stories are often stronger cues to
emotionthan are faces (e.g,Balconi &Carrera,2007; Camras &Allison, 1985;
Reichenbach &Masters, 1983; Widen &Russell, 2002), especially forfear and disgust
(Camras&Allison,1985;Russell &Widen, 2002; Widen&Russell,2002, 2004). Here,
we extend the storysuperiority effect to social emotions.
The discrete categories account (e.g., Izard,1971, 1994; Kestenbaum &Nelson,
1992; Lenti et al.,1997;Susskind et al.,2007) predicts that facial expressions are
primaryfor all emotions that have corresponding facial expressions –including all the
emotions tested in the current study.Thatis, children should recognize that afacial
expression represents acertain discrete emotion regardless of the level of development
of other aspects of the emotion script. On this view,the face should be the strongest cue
to each emotion we studied. In addition, this account does not predict apatternof
differentiation in children’s responses. Instead, on alabelling task, children’s‘erroneous’
responses should be random because theydonot yetassociate the correct labelwith
576 SherriC.Widen and James A. Russell
that emotion, and then upon learning its label, responses should hereafter be correct.
This all-or-nothing patternwas not observed.
The patternofdifferentiation observedinthe currentstudy better supports the
secondaccount in which children assimilated events that adults label disgust,
embarrassment, or shametotheir earlier-emerging emotion categories (e.g., anger,
sadness). Then, children graduallydifferentiated the later-emerging categories from
these through observation of emotions’ causes,consequences, arousal levels, etc. but
less so from observations of the facial expressions. For example, Kindergarten and
Grade 1s differentiated sadness from shame (something preschoolers didnot do), but
only when the emotion was presented as astory, not as aface. In the face mode, this
differentiation occurred in Grades 2and 3s. Other aspects of differentiationthat
occurred in the storymode had not yet occurred in Grade2and 3inthe face mode.
The finding that differentiation occurred earlier in the storymodethan in the face
moderaises questions about the specific version of the differentiationmodel of
children’sunderstanding of emotion, especially the proposed specific ages forthe
splitting of specific categories (Widen &Russell,2003). Thatversion was based
on facialexpressions of basic-level emotions.The difficulty is that children’s
differentiation of facial expressions may be limited to asmall subset of basic-level
facial expressions –those that were originally tested –and differentiationmay occur
more rapidly whennon-facial cues are examined. In the currentstudy,the facial
expressions tested in the original studies (anger,fear,surprise,disgust) were
differentiated, with the exception of disgustwhich was stillassimilated to anger even
forthe oldestage group, but only one facial expression foralater-emerging social
emotiondifferentiated from an earlier-emerging category(shamefrom sadness). The
initialversion mightmoreaccurately describethe development of children’s
understanding of emotion if it were based on their understanding of other aspects of
the emotion script, such as the causes and consequences of emotions. In the story
mode, children differentiated moreofthe later-emerging emotion categories from the
early emerging ones at earlier ages and the oldestage group had differentiated all but one
category(contempt from anger).
It is possible to argue that the faces that were used in the current study wereweaker
than the stories because theywere not clear,prototypicalfacial expressions of the
emotions. This possibility is aparticular hazard forthe social emotions,the proposed
facial expression forwhich have been less tested. It is also possible that clearer examples
of facial expressions of social emotionswould provide stronger cues to these emotions
than stories. However,the facial expressions forthe social emotions used in the current
study were carefully developed and have been tested: the embarrassed and ashamed
faceswere based on astudy in which participants experienced the target emotions,and
then other participants identified the emotions from the facial expressions (Keltner,
1995).The compassion face was based on criteria set in an observational study by
Eisenberg et al. (1989).The fullset of faces wasthen tested cross-culturally in the USA
and India (Haidt&Keltner, 1999): happiness,anger,fear,surprise,and disgust,and
embarrassment were each the modal responsefor the corresponding faces in both
cultures; shamewasthe modal response forashamed in the USA ( sadness was modal,
guilt was second, and shame was third in India); and contempt was the modal response
forcontemptuous in India (disgust was modal, contempt was second in the USA). Only
forthe compassion face was compassion not the modal responseineither culture, but it
was the second most frequent responsefor both(sadness was modal in the USA, awe in
India).Ofcourse, the possibility remains that more recognizable facial expressions for
Scripts for social emotions 577
compassion, shame,and embarrassment will one day be found and published. Still, the
theorythat these emotions are innately and universally expressed in the face does
suggest that the most recognizable expressions should have been easily found long ago.
One might arguethat inferring emotionfrom its facial expressionisamore difficult
task than doingsofrom astory. Faces were presented visually,for example, whereas the
storywas verbal as was the response mode. In asense, this argument is simply a
restatement of our results: formost emotions, storieswere the better cue. The argument
is inconsistent with the emotion signallingaccount in which recognition of specific
emotions from facial signals should be easy.
Our results are consistent with the idea that achild’sunderstanding of emotion is
embedded in anarrative structure (what we called scripts and operationalizedas
stories). Thus, the stories have an advantage herebecause theyevoke in the child that
narrative structure. Viewing afacial expression provides no such structure. Our results
point to aperspective different from that commonly taken in emotion research. From
our perspective, the study of emotion understanding must examinethe role of language
and cognition. For example, languagedevelopment and emotion understanding are
related.Children’slevel of language development was not measured in the current
study,but othershave demonstratedthis relationship (e.g., Pons, Lawson, Harris, de
Rosnay,2003;Ruffman, Slade, Rowlandson,Rumsey, &Garnham, 2003). Pons et al.
showed that children’s(4–11years)receptive grammar abilities was correlated with
their level of emotion understanding. Ruffman et al. showed that children’s(3–5 years)
receptive syntax and semantic understanding was correlated with their recognition of
facial expressions.
This study investigated which aspect of children’semotion scripts better tapped
their understanding of basic-level and social emotions: facial expression or cause-and-
consequence stories. Overall, cause-and-consequence stories were the stronger cue,
especially forfear,disgust, and social emotions. Faces were the stronger cue only for
surprise. When all of children’sresponses (both ‘correct’ and ‘incorrect’) were
considered, younger childrenassimilated social emotions into basic-level emotion
categories; older children differentiated them. Thispatternofdifferentiation, which
occurred earlier forthe stories than forthe faces, is contrarytothe predictions of the
discrete categories account.
We thank the staff, parents, and children at the daycares and after-school care programmes that
participated in this study: without their help, this researchcould not havebeen done. We thank
Gillian Cohen, Ryan Winton, and Maria-Paz Rodriguez fortheir help with developing the stories for
social emotions, data collection, and coding, and Nicole Nelson, Jennifer Gallucci, and Hannah
Tappen fortheir help with manuscript preparation. This study was funded by agrant from the
National Science Foundation.
Balconi, M., &Carrera, A. (2007). Emotional representation in facial expression and script:
Acomparison between normal and autistic children. Research in Developmental Disabilities,
Barrett, L. F. ,&Salovey,P.(2002). Thewisdom in feeling: Psychological processes in emotional
intelligence.New York:Guilford Press.
578 SherriC.Widen and James A. Russell
´,M.G., Cheung, N.,&Hess, U. (2000). TheMontreal Set of Facial Displays of
Emotion [CD].Montreal, QC: Department of Psychology,University of Quebec.
Blair,C.(2002). School readiness: Integrating cognition and emotion in aneurobiological
conceptualization of children’sfunctioning at school entry. American Psychologist,57,
Bretherton, I., &Beeghly,M.(1982). Talking about internal states: The acquisition of an explicit
theoryofmind. Developmental Psychology ,18,906–921.
Bridges, K. M. B. (1930). Agenetic theoryofthe emotions. Journal of Genetic Psychology ,37,
Brody,L.R., &Harrison, R. H. (1987). Developmental changes in children’sabilities to match and
label emotionally laden situations. Motivation and Emotion,11,347–365.
Brown, J. R., &Dunn, J. (1996). Continuities in emotion understanding from three to six years.
Child Development,67,789–802.
Camras, L. A., &Allison, K. (1985). Children’sunderstanding of emotional facial expressions and
verbal labels. Journal of Nonverbal Behavior,9,84–94.
Denham, S. A. (1998). Emotional development in young children.New Yo rk: Guilford Press.
Denham, S. A., &Couchoud, E. A. (1990). Yo ung preschoolers’ ability to identify emotions in
equivocal situations. Child Study Journal,20,153–169.
Eisenberg, N.,Fabes, R. A., Miller,P.A., Fultz, J.,Shell, R., Mathy,R.M., et al. (1989). Relation of
sympathy and distress to prosocial behavior: Amultimethod study.Journal of Personality
and Social Psychology ,57,55–66.
Ekman, P. (1994). Strong evidence foruniversals in facial expressions: Areply to Russell’smistaken
critique. Psychological Bulletin,115,268–287.
Ekman, P. ,&Friesen, W. V. (1976). Pictures of facial affect.Palo Alto, CA: Consulting
Psychologists Press.
Fehr,B., &Russell, J. A. (1984). Concept of emotion viewed from aprototype perspective.
Journal of Experimental Psychology: General,113,464–486.
Fischer,K.W.(1980). Atheoryofcognitivedevelopment: The control and construction of
hierarchies of skills. Psychological Review,87,477–531.
Garner,P.W., &Waajid, B. (2008). The associations of emotion knowledgeand teacher–child
relationships to preschool children’sschool-related developmental competence. Journal of
Applied Developmental Psychology ,29,89–100.
Goeleven, E., de Raedt, R., Leyman, L., &Verschuere, B. (2008). The Karolinska directed emotional
faces: Avalidation study.Cognition and Emotion,22,1094–1118.
Haidt, J.,&Keltner,D.(1999). Culture and facial expression: Open-ended methods find more
expressions and agradient of recognition. Cognition and Emotion,13,225–266.
Harrigan, J. A. (1984). The effect of task order on children’sidentification of facial expressions.
Motivation and Emotion,8,157–169.
Harris, P. L. (1989). Children and emotion: Thedevelopment of psychological understanding.
Oxford: Basil Blackwell.
Harris, P. L., Olthof, T. ,Meerum Terwogt, M., &Hardman, C. E. (1987). Children’sknowledge
of the situations that provoke emotion. International Journal of Behavioral Development,
Heerey, E. A., Keltner,D., &Capps, L. M. (2003). Making sense of self-conscious emotion: Linking
theoryofmind and emotion in children with autism. Emotion,3,394–400.
Hollingshead, A. B. B., &Redlich, F. C. (1958). Social class and mental illness: Community
study.Hoboken, NJ: John Wiley&Sons.
Hornik, R., Risenhoover,N., &Gunnar,M.(1987). The effects of maternal positive, neutral, and
negative affectivecommunications on infant responses to new toys. Child Development,58,
Izard, C. E. (1971). Theface of emotion.New Yo rk: Appleton CenturyCrofts.
Izard, C. E. (1994). Innate and universal facial expressions: Evidence from developmental and
cross-cultural research. Psychological Bulletin,2,288–299.
Scripts for social emotions 579
Keltner,D.(1995). Signs of appeasement: Evidence forthe distinct displays of embarrassment,
amusement, and shame. Journal of Personality and Social Psychology ,68,441–454.
Kestenbaum, R., &Nelson, C. A. (1992). Neural and behavioral correlates of emotion recognition
in children and adults. Journal of Experimental Child Psychology ,54,1–18.
Lenti, C., Giacobbe,A., &Pegna, C. (1997). Aneuropsychological approach to depression in
children and adolescents: The discrimination of emotional facial expressions. Italian Journal
of Psychiatryand Behavioral Sciences,7,121–127.
Lenti-Boero, D. (1994).Oltre le parole:Lacommunicazioneumana nonverbal come
comunicazione universal. Teoria Sociologica,2/3,94–122.
Markham, R., &Adams, K. (1992). The effect of type of task on children’sidentification of facial
expressions. Journal of Nonverbal Behavior,16,21–39.
Matsumoto, D.,&Ekman, P. (1988). Japanese and Caucasian facial expressions of emotion and
neutral faces (JACFEE): 56 jpeg photos of facial expressions [CD]. San Francisco: Intercultural
and Emotional ResearchLaboratory.
Miller,A.L., Fine, S. E., Gouley, K. K, Seifer,R., Dickstein, S., &Shields, A. (2006). Showing and
telling about emotions: Interrelations between facets of emotional competence and
associations with classroom adjustment in Head Startpreschoolers. Cognition and Emotion,
Norton, A., Winner,E., Cronin, K., Overy, K., Lee, D. J.,&Schlaug, G. (2005). Are there pre-
existing neural, cognitive, or motoric markersfor music ability? Brain and Cognition,59,
Pell, M. D. (2005). Nonverbal emotion priming: Evidence from the ’facial affect decision task’.
Journal of Nonverbal Behavior,29,45–73.
Pons, F. ,Harris, P. L., &deRosnay,M.(2004). Emotion comprehension between 3and 11 years:
Developmental periods and hierarchical organization. European Journal of Developmental
Psychology ,1,127–152.
Pons, F. ,Lawson, J.,Harris, P. L., &deRosnay,M.(2003). Individual differences in children’s
emotion understanding: Effects of ageand language. Scandinavian Journal of Psychology ,44,
Raver,C.C., Garner,P., &Smith-Donald, R. (2007). The roles of emotion regulation and emotion
knowledgefor children’sacademic readiness: Are the links causal? In R. C. Pianta, M. J. Cox,
&K.L.Snow (Eds.), School readiness and the transition to kindergarten in the eraof
accountability (pp. 121–147). Baltimore, MD: Paul H. Brookes Publishing.
Reichenbach, L., &Masters, J. C. (1983). Children’suse of expressive and contextual cues in
judgements of emotion. Child Development,54,993–1004.
Repacholi, B. M. (1998). Infants’ use of attentional cues to identify the referent of another person’s
emotional expression. Developmental Psychology ,34,1017–1025.
Repacholi, B. M., &Gopnik, A. (1997). Early reasoning about desires: Evidence from 14- and
18-month-olds. Developmental Psychology ,33,12–21.
Rieffe,C., Meerum Terwogt, M., &Jellesma, F. C. (2008). Emotional competence and health in
children. In A. Vingerhoets, I. Nyklı
´ cek, &J.Denollet (Eds.), Emotion regulation: Conceptual
and clinical issues (pp. 184–201). New York: Springer Science þBusiness Media.
Rinn, W. E. (1991). Neuropsychology of facial expression. In R. S. Feldman &B.Rime
Fundamentals of nonverbal behavior.Studies in emotion and social interaction (pp. 3–30).
New York:CambridgeUniversity Press.
Ruffman, T. ,Slade, L., Rowlandson, K., Rumsey, C., &Garnham, A. (2003). How languagerelates to
belief, desire, and emotion understanding. CognitiveDevelopment,18,139–158.
Russell, J. A. (1990). The preschooler’sunderstanding of the causes and consequences of
emotion. Child Development,61,1872–1881.
Russell, J. A., &Widen, S. C. (2002). Words versus faces in evoking children’sknowledgeofthe
causes of emotions. International Journal of Behavioral Development,26,97–103.
Saarni, C. (1999). Thedevelopment of emotional competence.New Yo rk: Guilford Press.
580 SherriC.Widen and James A. Russell
Seidner,L.B., Stipek, D. J. ,&Feshbach, N. D. (1988). Adevelopmental analysis of elementary
school-aged children’sconcepts of pride and embarrassment. Child Development,59,
Shamay-Tsoory, S. G.,Lavidor,M., &Aharon-Peretz, J. (2008). Social learning modulates the
lateralization of emotional valence. Brain and Cognition,67,280–291.
Smith, M., &Walden, T. (1999). Understanding feelings and coping with emotional situations:
Acomparison of maltreated and nonmaltreated preschoolers. Social Development,8,93–116.
Susskind, J. M., Littlewort, G.,Bartlett, M. S., Movellan, J.,&Anderson, A. K. (2007). Human and
computer recognition of facial expressions of emotion. Neuropsychologia,45,152–162.
Tomkins, S. S. (1962). Affect, imagery, consciousness: Vo l. I. Thepositive affects.New York:
Springer Publishing.
Tottenham, N.,Tanaka, J.,Leon, A. C., McCarry,T., Nurse, M., Hare, T. A., et al. (2009). The
NimStim set of facial expressions: Judgments from untrained researchparticipants. Psychiatry
Research,168(3), 242–249.
Trentacosta, C. J.,&Izard, C. E. (2007). Kindergarten children’semotion competence as a
predictor of their academic competence in first grade. Emotion,7,77–88.
Walker-Andrews, A. S. (1997). Infants’ perception of expressivebehaviors: Differentiation of
multimodal information. Psychological Bulletin,121,437–456.
Widen, S. C., &Russell, J. A. (2002). Gender and preschoolers’ perception of emotion. Merrill-
Palmer Quarterly,48,248–262.
Widen, S. C., &Russell, J. A. (2003). Acloser look at preschoolers’ freely produced labels forfacial
expressions. Developmental Psychology ,39,114–128.
Widen, S. C., &Russell, J. A. (2004). The relative powerofanemotion’sfacial expression, label,
and behavioral consequences to evoke preschoolers’ knowledgeofits cause. Cognitive
Widen, S. C., &Russell, J. A. (2008). Yo ung children’sunderstanding of other’semotions.
In M. Lewis, J. M. Haviland-Jones, &L.F.Barrett (Eds.), Handbook of emotions (3rd ed.,
pp. 348–363). New York: Guilford Press.
Widen, S. C., &Russell, J. A. (in press). Differentiation in preschooler’scategories foremotion.
Wiggers, M., &van Lieshout, C. F. M. (1985). Development of recognition of emotions:
Children’sreliance on situational and facial expressive cues. Developmental Psychology ,
Zeidner,M., Matthews, G.,Roberts, R. D.,&MacCann, C. (2003). Development of emotional
intelligence: Towards amulti-level investment model. Human Development,46,69–96.
Received 23 July 2008; revised version received 20 May 2009
Scripts for social emotions 581
... initially including "happy" more than other labels and then gradually adding anger and sadness, followed by fear and surprise, and lastly disgust (e.g., Widen, 2013). This sequential unfolding of emotion labels has been observed across a variety of face labeling and sorting tasks (e.g., Widen & Russell, 2003, 2010a, 2010b. ...
... Studying emotion labeling within relational contexts provides further insight of early emotion understanding. Indeed, preschoolers' emotion labeling is influenced by the context, such as hearing a corresponding emotion story (e.g., Camras & Allision, 1985;Denham et al., 1994;Dunn & Hughes, 1998), and their accuracy improves when particular emotions, specifically fear, compassion, embarrassment, disgust, and shame, are described in stories rather than presented solely as facial expressions (Wang et al., 2014;Widen & Russell, 2010b). For example, Leitzke and Pollak (2016) found that 4-year-olds more accurately identified images of disgust facial expressions presented in context (i.e., body holding a dirty object) than the face alone, but this pattern did not hold for anger. ...
... The emotion expressions were of moderate intensity, and all images were previously validated as conveying the intended emotion (> 80% agreement for the target emotion; see Knothe & Walle, 2018). All emotional contexts were selected to be familiar events for children (e.g., receiving a gift; disliking a food) and thematically similar to previous vignette studies (e.g., Widen & Russell, 2010b). Example images are presented in Fig. 1, and further detailed descriptions of the images can be found in Knothe and Walle (2018, Fig. 1 and Appendix Tables A1 and A2). ...
Emotion understanding involves appreciating the significance of the relational context; the “aboutness” of the emotion. This study examined how children labeled emotions and described relational elements of discrete emotion contexts. Preschool children (3.5-year-olds, n = 22; 4.5-year-olds, n = 23) described images of 5 emotion contexts (anger, sadness, disgust, fear, and joy). Researchers assessed children’s (1) correct labeling of discrete emotions, and (2) differential mentioning of the emoter (person displaying the emotion) and the referent (the elicitor of the emotion) across discrete emotions. Children’s pattern of accurately labeling discrete emotions was similar to prior research, with both age groups correctly labeled anger, sadness, and joy more often than disgust or fear. Novel to the present study, we found that older children differentially highlighted emotional elements (i.e., the emoter, the referent) when describing discrete emotion contexts. Specifically, 4.5-year-olds emphasized the emoter more when describing anger, sadness, and joy than fear and disgust contexts, and mentioned the referent more in disgust, fear, and joy than anger and sadness contexts. Differential emphasis of relational elements was not observed for 3.5-year-olds. These findings highlight the importance of examining children’s appreciation of relational contexts and indicate important differences in how children differentially emphasize relational elements when viewing discrete emotion contexts. Potential developmental mechanisms, opportunities for further empirical research, and implications for emotion theory are discussed.
... We measured children's emotion recognition via emotion labeling in fully visible faces in two tasks. In one task, children freely labeled emotional facial expressions of adults (Adult Faces Task; task adapted from Widen and Russell, 2010). In a second task, the children did the same with emotional facial expressions of children (Child Faces Task; task adapted from Streubel et al., 2020). ...
... The data collected before the COVID-19 pandemic (no-COVID-19-experience sample) were taken from studies by Streubel et al. (2020) (Child Faces Task) and Widen and Russell (2010) (Adult Faces Task) with the kind permission of the authors . The sample by Streubel et al. (2020) was collected in Germany in 2019 and consisted of 30 children (11 girls, 19 boys) at the age of 4.54-5.59 ...
... In this sample, 71% children had at least one parent with a college degree. The sample by Widen and Russell (2010) was collected in the United States before 2010 and consisted of 40 children (20 girls, 20 boys) at the age of 4.00-5.75 years (M = 4.91 years, SD = 0.46 years). ...
Full-text available
During the COVID-19 pandemic people were increasingly obliged to wear facial masks and to reduce the number of people they met in person. In this study, we asked how these changes in social interactions are associated with young children's emotional development, specifically their emotion recognition via the labeling of emotions. Preschoolers labeled emotional facial expressions of adults (Adult Faces Task) and children (Child Faces Task) in fully visible faces. In addition, we assessed children's COVID-19-related experiences (i.e., time spent with people wearing masks, number of contacts without masks) and recorded children's gaze behavior during emotion labeling. We compared different samples of preschoolers (4.00–5.75 years): The data for the no-COVID-19-experience sample were taken from studies conducted before the pandemic (Adult Faces Task: N = 40; Child Faces Task: N = 30). The data for the with-COVID-19-experience sample ( N = 99) were collected during the COVID-19 pandemic in Switzerland between June and November 2021. The results did not indicate differences in children's labeling behavior between the two samples except for fearful adult faces. Children with COVID-19-experience more often labeled fearful faces correctly compared to children with no COVID-19 experience. Furthermore, we found no relations between children's labeling behavior, their individual COVID-19-related experiences, and their gaze behavior. These results suggest that, even though the children had experienced differences in the amount and variability of facial input due to the pandemic, they still received enough input from visible faces to be able to recognize and label different emotions.
... al, 2005), koji podrazumeva da se ispitanicima prvo čitaju priče koje opisuju dejstvo nekog pobuđivača, a da se zatim od njih traži da predvide ili objasne osećanje nekog aktera. Rezultati ovako osmišljenih istraživanja (Harris;Mirić, 2019;Rieffe et al., 2005;Rosnay & Harris, 2002;Russell, 1990;Wellman & Banerjee, 1991;Widen & Russell, 2010), pokazali su sledeće: 1. Deca na ranijem uzrastu počinju da povezuju emocije sa željama nego sa uverenjima; 2. Sa uverenjima se lakše povezuje strah, a sa željama (njihovim ispunjenjem ili osujećenjem) radost, tuga i bes; 3. I želje i uverenja se pre povezuju sa osnovnim nego sa složenim emocijama; 4. Brojni kontekstualni činioci utiču na povezivanje mentalnih stanja i kauzacije emocija. ...
... Umesto toga, korišćene su priče koje opisuju kako emocije nastaju usled dejstva drugačijih (nesimboličkih) pobuđivača, kao što su: početak kiše, fino jelo, nalaženje zatvorene kutije u ormaru, poseta baki (Wellman & Banarjee, 1991), neobična buka, bolest kućnog ljubimca, poklon (lopta) od strane roditelja, nepoznata osoba u kući, igranje u dvorištu (Rieffe et. al, 2005), poseta prijatelja za rođendan, otimanje mesta u redu za čekanje, susret sa opasnim psom (Widen & Russell, 2010), pas koji je ranije napao kućnog ljubimca, igračka koju je drugo dete uništilo, igračka sa kojim se igrala mačka koja je kasnije uginula (Lagattuta et al., 1997) itd. ...
Full-text available
Iako je odavno poznato da simbolički stimulusi mogu da budu pobuđivači emocija, dosadašnja istraživanja u oblasti socijalne kognicije (tj. teorije uma) bila su usmerena gotovo isključivo na prirodne pobuđivače. Zato smo ispitanicima različite starosti, koristeći metodu strukturisanog intervjua, postavljali pitanja o tome šta izaziva 4 osnovne (radost, strah, tuga, bes) emocije, kako bismo utvrdili kolika će biti relativna učestalost simboličkih pobuđivača u njihovim spontanim odgovorima. Relativna učestalost takvih pobuđivača uzeta je kao indikator za njihov značaj za razumevanje kauzacije emocija. Učestvovalo je ukupno 120 ispitanika, po 20 na različitim uzrastima od predškolskog do studentskog. Dobijene odgovore kategorisala su dva nezavisna procenjivača; ukupno slaganje među njima bilo je 96%. Rezultati su pokazali da su simboličke pobuđivače u visokom procentu navodili ispitanici na svim ispitivanim uzrastima. Ovaj nalaz smo protumačili kao pokazatelj toga da su takvi pobuđivači od najranijih uzrasta deo implicitne teorije o kauzaciji emocija i da je u dosadašnjim istraživanjima napravljen „propust“ time što oni nisu u većoj meri uzeti u obzir. Jedina značajna uzrasna razlika u učestalosti simboličkih pobuđivača dobijena je između dece od 5 godina (46.4%) i starijih ispitanika (65.9% u proseku), što se poklapa sa prelaskom sa predškolskog na školski uzrast. Relativna učestalost takvih pobuđivača ostaje približno ista na starijim uzrastima. Analize uzrasnih razlika u učestalosti simboličkih pobuđivača iz različitih potkategorija pokazale su da je na uzrastu od 5 godina veća učestalost ikoničkih pobuđivača nego na starijim uzrastima, dok je obrnut slučaj sa društveno-normativnim pobuđivačima; u oba slučaja, kasnije uzrasne razlike nisu statistički značajne.
... During early childhood, children can recognize and understand basic emotions such as happiness, anger, fear, surprise, and disgust (Denham et al., 2011). For instance, we expect preschoolers to accurately label their own emotions and to understand how basic emotions are linked to behavior within their cultural norm group (e.g., identify that they would feel happy if given a present; Widen & Russell, 2010;. By middle childhood, we expect children to describe complex self-conscious emotions (e.g., shame, guilt, and pride; Carr, 2017) and to develop awareness that they may feel conflicting emotions about the same person (e.g., being mad at someone they like). ...
... To illustrate this type of emotionshaping in action, I turn to theories and practices in early childhood development. Many preschools and elementary schools in the U.S. incorporate "emotion dolls" into the classroom (see Widen and Russell 2010a). There may be multiple dolls, each with a different face. ...
Full-text available
Can we read emotions in faces? Many studies suggest that we can, yet skeptics contend that these studies employ methods that unwittingly help subjects in matching faces with emotions. Some studies present subjects with posed faces, which may be more exaggerated than spontaneous ones. And some studies provide subjects with a list of emotion words to choose from, which forces them to interpret faces in specific emotion terms. I argue that the skeptics’ challenge rests on a false assumption: that once subjects leave the lab, they no longer receive help in matching faces with emotions. I contend that people receive as much help in the wild as they do in the lab. People unconsciously amplify their spontaneous expressions in the presence of others, thereby making them easier to read. And people teach children to interpret faces in the same specific emotion terms found in the experimenters’ word lists. I argue that we are good at readings emotions in faces because we can normally count on a little help from our friends.
El reconocimiento de expresiones facialesemocionales (REFE) es unahabilidad básica para la interacciónsocial adecuada, aunque es poco investigadoen población infantil utilizandoestímulos de niños expresandolas 6 emociones básicas. En el estudioparticiparon 165 niños entre 4 y 16años, realizaron la Tarea de REFE (70fotografías de niños mexicanos expresandoseis emociones básicas –alegría,miedo, sorpresa, tristeza, asco, enojo– y neutral). Los resultados indican queel reconocimiento de las expresionesde Sorpresa, Asco y Neutral parecemejorar de acuerdo a la edad. Alegríaes la expresión más reconocida, mientrasMiedo es la menos reconocida.Encuanto al género, las niñas reconocenmejor Asco. En conclusión, el REFE seve influido tanto por la maduraciónbiológica como por el ambiente en quese desarrollen los individuos.
A tanulmány a negatív érzelmek szerepét tárgyalja Franz Kafka Az átváltozás című novellájában a kognitív narratológia eszközeivel. Alapvetően a főszereplőre adott érzelmekkel foglalkozik. Elsőként a családtagok fiktív érzelmi reakcióit, azaz az ábrázolt érzelmeket, majd a szövegstratégiák által támogatott befogadói érzelmeket vizsgálja. A szövegstratégiákat azonosítva bemutatja, hogy a szöveg az olvasóban Gregorral szemben egy a fikcionális szereplők által bejárt érzelmi úttal – a Gregortól való távolodás állomásait leíró félelem-undor-düh érzelmi ívvel – ellentétes, közelítő érzelemív kiváltódását támogatja. Továbbá a szöveg értelmezésén keresztül amellett érvel, hogy Az átváltozás alapvetően érzelmeket ábrázol és érzelmek kiváltására törekszik. Miközben a mű az értelmi megértés szintjén rejtélyes és szemantikailag széttartó, az érzelmi megértés szintjén egy irányba mutat, és egy univerzálisan érthető és elérhető megértési élményt nyújt.
This chapter provides the current knowledge of the scientific literature on the development of emotional competencies, from infancy to adolescence. It aims to trace the child's journey through this acquisition process, which is as complex as any other form of academic learning. The chapter discusses the role and influence of language skills in the development of emotional competence. The scientific approach to the study of emotions was born thanks to the work of Charles Darwin, the famous naturalist who, in his book entitled The Expression of the Emotions in Man and Animals , defended an evolutionary perspective on the origin of emotions. Studies show that, as early as 16 months, children stare longer at a happy face when hearing the corresponding language label and improve their performance for the emotion of anger by 28 months.
Research with adults has increasingly moved beyond the focus on a small set of allegedly basic emotions, each associated with a signature facial expression. That expansion has been accompanied by a greater emphasis on the potential variability of emotion concepts across different cultural settings. In this conceptual review of children’s understanding of emotion, we argue that it is also important in developmental research to look beyond the small set of emotions associated with distinctive facial expressions. At the same time, we caution against any premature rejection of a universalist approach to children’s understanding of emotion. We review three different lines of evidence in support of this stance: (1) children’s ability to appropriately cite situational elicitors for emotions beyond the basic set; (2) their developing understanding of the relations between emotions and other mental processes; and (3) their realization that a person’s facially expressed emotion may not indicate their felt emotion. In each of these three domains, we target studies that have included children from a variety of cultures to assess how far they respond similarly or differently. We conclude that there is robust evidence for similar conceptual progress in children’s understanding of emotion across a range of cultural settings.
Full-text available
Children (N = 160), aged 3 to 4 years, generated stories describing the causes of six different emotions: happiness, surprise, fear, anger, disgust, and sadness. The emotion was specified to the child either by a word (such as scared or disgusted) or by a photograph of a facial expression said to be a universal, biologically based signal for that emotion. For no emotion did the face produce significantly better performance than did the word. For fear and disgust, the word produced significantly better performance than did the face.
Children's performance on free labeling of prototypical facial expressions of basic emotions is modest and improves only gradually. In 3 data sets (N = 80, ages 4 or 5 years; N = 160, ages 2 to 5 years; W = 80, ages 3 to 4 years), errors remained even when method factors (poor stimuli, unavailability of an appropriate label, or the difficulty of a production task) were controlled. Children's use of emotion labels increased with age in a systematic order: Happy, angry, and sad emerged early and in that order, were more accessible, and were applied broadly (overgeneralized) but systematically. Scared, surprised, and disgusted emerged later and often in that order, were less accessible, and were applied narrowly.
This study explored 14- and 18-month-old infants' ability to identify the target of an emotional display. In the visual task, infants were presented with 2 boxes. Each box contained an object that could be identified by opening the box lid and looking inside. In the tactile task, the objects had to be pulled out of the boxes before they could be seen. An experimenter expressed happiness as she looked or put her hand inside one box, and disgust as she repeated this action with the other box. Infants were then allowed to explore the boxes. Infants touched both boxes but preferred to search for the happy object. Thus, regardless of age or task, infants identified the target of each emotional display as something inside a box and not the box itself. Infants appeared to use the experimenter's attentional cues (gaze and action) to interpret her emotional signals and behaved as if they understood that she was communicating about the objects.