Moral Emotions: An Analysis Guided by Heiderʹs Naive Action Analysis

8/22/2019 Moral Emotions: An Analysis Guided by Heiders Naive Action Analysis

1/24

InternationalJournalofAdvancesinPsychology(IJAP)Volume2Issue2,May2013 www.ijpsychol.org

69

MoralEmotions:AnAnalysisGuidedby

HeidersNaiveActionAnalysisUdoRudolph,KatrinSchulz,NadineTscharaktschiew

InstituteforPsychology,ChemnitzUniversityofTechnology,Chemnitz,Germany

[email protected]

AbstractThe present research proposes a theoretical framework

describingtheemergenceofmoralemotions.Inanalyzingthe

cognitiveantecedentsofmoralemotions,werefertoHeiders

(1958)conceptsofought,goalattainmentandeffort.InStudy

1,participants(N=60)ratedwhethertheyexperiencemoral

observeremotions(i.e.,admiration,anger,contempt,disgust,indignation, pride, respect, schadenfreude, and sympathy)

with respect to situations characterized by different

combinationsofought,goalattainment,andeffort.InStudy2

(N=61),thesameprocedurewasusedtoanalyzemoralactor

emotions (i.e., embarrassment, guilt, pride, regret, and

shame). Studies 3 (N = 75) and 4 (N = 51) replicated these

findings by using techniques characterized by higher

ecological validity. ANOVAS and cluster analyses were

conductedtoanalyzesimilaritiesanddifferenceswithregard

to different subgroups of moral emotions. Taken together,

resultsrevealthatought,goalattainment,andeffortexplain

impressive amounts of variance in the elicitation of moralemotions.Basedontheseresults,anempiricalclassificationof

moralemotionsisproposedwhichisbasedontheevaluative

functionandtargetoftherespectiveemotions.

KeywordsMoralEmotions;Morality;FolkPsychology;AttributionIntroduction

Weknowverywellwhatitisliketofeelsympathyor

anger,admirationorindignation,prideorguilt.Thus,

wenavigate through the landscapeofthesesocalled

moralemotionseasily,andapparentlydonotneedto

engageincomplexconsiderationsto decide howwe

feel.However,thisdoesnotnecessarilymeanthatwe

do not make decisions that guide our emotional

experiences. Nevertheless, to our knowledge, a

comprehensive conceptual framework describing

these decisions and thus explaining the genesis of

moralemotionshasnotbeendeveloped.Recentworks

havedemonstratedthatmoralemotionshaveastrong

impactonmoraldecisions,ascribingtheseemotionsa

strong causal status in the chain from cognition to

action.Many researchershaveelaboratedon specific

moralemotionsorspecificsamplesofmoralemotions,

contributing to a deeper understanding of these

specific emotions. A comprehensive experimental

analysisand synthesisof these findings,however, is

stilllacking.

Inthepresentpaper,wewillfirstprovideaconceptual

analysisofthesocalledmoralemotions,followedbyan experimental analysis of their cognitive

antecedents.

WhatareMoralEmotions?Thus far, a comprehensive and generally accepted

definitionofmoralemotionsdoesnotexist,although

numerous definitions of moral emotions have been

proposed. As a common denominator, these

definitions include aconsideration concerning good

andbad, right and wrong, and ought and should.

Moreover,Haidtdefinesmoralemotionsasthose thatare linked to the interestsorwelfareeitherofsociety

asawholeoratleastofpersonsotherthanthejudge

or agent (p. 276). In addition, Fiske points out that

moral emotions are the immediate motivational

proxies for expected longtermbenefits of important

relationships(p.170).Thesedefinitionscomplement

one another by highlighting different important

elements of moral emotions that is, cognitive,

adaptive,andmotivationalaspects.

Given these definitions, it nevertheless appearsdifficult tocompileadefinite listofmoralemotions.

As a practical approach for identifying moral

emotions, we conducted a PsycInfo and

PhilosophersIndexOnline searchonallwordsofthe

affective lexicon, in combination with the words

moralemotion*.Additionally,wecheckedthelistsof

moral emotions offered by Haidt and Weiner.

According to this literaturesearch,23emotionshave

been labeled as moral emotions: Admiration, anger,

awe, contempt, disgust, elevation, embarrassment,

empathy,envy,gratitude,guilt, indignation,jealousy,

pity, pride, regret, remorse, respect, schadenfreude


2/24

www.ijpsychol.org InternationalJournalofAdvancesinPsychology(IJAP)Volume2Issue2,May2013

70

(joy in the misfortune of others), shame, scorn, and

sympathy.

Ononehand,thislistisprobablyquiteexhaustive,as

wehave includedallpotentialcandidatesmentioned

in thepsychologicalandphilosophical literature.On

the other hand, there is no consensus as to which

emotionscanbereferredtoasmoral,andthereisno

obligatory or selective criterion to distinguish moral

from nonmoral emotions. The abovementioned list

of moral emotions should thus be regarded as

preliminary,andcurrentlycontainsprototypesaswell

asclosesynonyms.AsHaidthasnoted,someofthese

moral emotions are likely more prototypical than

others.We suggest that themoreprototypicalmoral

emotions are exclusively experienced concerning

actions of individuals and occur in a wide range of

differentsituations (e.g.,pride,shame,admirationor

schadenfreude; to mention a few). In contrast,

nonprototypical moralemotions(e.g.,awe,elevation)

arerestrictedtospecificcontexts(e.g.,religion,natural

and moralbeauty, or exceptional human actions or

abilities). In addition, some of these emotions

represent close relatives rather than highly distinct

emotions.

AClassificationofMoralEmotionsWe propose two criteria, target and evaluative

function, for classifyingmoralemotions:First,moral

emotions canbe distinguished with respect to their

target. Some of these emotions are regarded as

selfconscious or selfdirected, as for example guilt,

pride, regret, and shame. In the following, we will

label these emotions as actor emotions, as they are

targeted at ones own actions. In contrast, several

moralemotionsare regardedasbeingotherdirected

(e.g., admiration, anger, gratitude, and sympathy.

Theseemotionshaveincommonthattheyaretargeted

atotherpersonsand theiractions; therefore,wewill

labeltheseemotionsasobserveremotions.

Second, as several authors have pointed out, moral

emotions function to regulate social behaviors. We

therefore propose that moral emotions evaluate a

persons action as either positive or negative. This

criterion applies to both moral actor emotions

(evaluatingonesownactionsaseithergoodorbad),

as well as to moral observer emotions (evaluating

othersactionsaseithergoodorbad).Toputitinother

words,moralemotionsevaluateonesownaswellas

other persons actions as morally praiseworthy or

blameworthy).

We suggest that the evaluative function of a moral

emotion represents the evaluative content of the

respective emotion as a means for communication

(towards others) and information (towards oneself).

For example, with respect to actor emotions, pride

represents a positive selfevaluation, whereas

embarrassment, guilt, regret, remorse, and shame

represent negative selfevaluations. Analogously,

observer emotions represent evaluations of the

observed actions of other persons. For example,

admiration, awe, gratitude, elevation, empathy, pity,

pride, respect and sympathy evaluate the actions of

the observed person as positive, whereas anger,

contempt, disgust, indignation, jealousy, scorn and

schadenfreude are negative evaluations of others

actions1.

Toconclude,theproposedcriteriafortheclassification

of moral emotions, target and evaluative function,

serveasconceptualtoolstodifferentiatethefunctional

aspectsofmoralemotions:Positiveactoremotions(i.e.,

pride) signal the experiencing person has done

something good or praiseworthy and motivate the

actor to continue with this behavior. In contrast,

negative actor emotions (i.e., embarrassment, guilt,

regret,shame)signalthattheexperiencingpersonhasdonesomethingwrongorbad,thatthepersoncanbe

blamed and should change thebehavior. The same

regulative function is applied to observer emotions,

exceptthattheevaluativesignalinthiscaseisdirected

attheobservedperson.

AntecedentsofMoralEmotionsWhichkindsofcognitiveconceptsmightgiverise to

thismultifold landscapeofmoralemotions?Wewill

next describe a theoretical framework based on

Heidersnaveactionanalysisthatprovidesimportantsufficient conditions for eliciting moral emotions 2 .

1Pleasenotethatsomeoftheemotionsdescribedasmoralemotions,

forexampleangeranddisgust,mayalsooccurinnonmoralsettings:

For example, you may feel nonmoral anger because your train

arrived lately; oryou may feel coredisgust watching somebody

eatingapieceofrottenmeat.Rozinetal.refertocoredisgustasthe

guardianofthemouthagainstpotentialcontaminants(p.575).(For

amoredetaileddiscussionofmoralangerandmoraldisgust).

2 Interestingly, Heiders considerations closely resemble the ideas

raised already in 1740 by David Hume in his early and

groundbreaking work ATreatise of Human Nature: Being anattempt to introduce the experimental method of reasoning into

moralsubjects.


3/24


71

Therefore, we take a closer look at his concepts of

ought,goalattainment,andeffort.

1)OughtAs already stated, it is widely agreed upon that

moral emotions require a consideration of goodand bad or right and wrong. In addition, these

judgments involve a general viewpoint beyond

immediatepersonalinterests.Similarly,Heiderhas

stated that ought is an impersonal concept and

referstorelativelystablestandardsofwhatought

to be done or experienced. These standards are

independentoftheindividualswishes: Thisisnot

to say that personal wishes do not influence the

perception of ought forces; it is rather that they

should not; in principle the ought is established

by objective requirements and beyond personal

concerns. The characterization of ought as an

impersonal,normativeconceptalsoimpliesthatit

hasinterpersonalvalidity: Notonlyshouldought

disregard personal desires, not only does ought

appear in principle unchanged in spite of

incidentalsituationalfactors,butitisalsouniversal

andshouldlookaliketoeverybody.

To summarize, the concept of ought refers to

morallygoodversusbadgoalsorsocialstandards,

and isdefinedasacognitiverepresentationofanimpersonal requirement, invariant across many

situations,andtranscendingtheindividualspoint

ofview.

2)GoalAttainmentMoralityisnotonlyamatterofwhatpeopleshould

do, but also a matter of what they actually do.

Furthermore,itisnotonlyimportantwhichgoalsa

person wishes to achieve,but also whether s/he

actuallydoesachievethesegoals.Asfollowsfrom

Heiders nave action analysis, goal attainment isdependent upon a relatively stable relationship

between the person and the environment.

According to this viewpoint, goal attainment

requiresthatapersonsabilitiesandeffortsexceed

bothtaskdifficultyandrandomsituationalfactors.

In the moraldomain, goal attainment informsus

whetherapersonactuallyreachesagivenstandard,

that is, whether a generally positive or negative

goal is either attained or not attained (for a

differentiationbetweenactionsandgoals).

In sum, ought provides the moral standard

concerning what should be done, and goal

attainment defines whether the desirable or

undesirable goal is actually attained or not

attained.

3)EffortGivenamoralstandard(ought)anditsattainmentor nonattainment, motivational forces located

within the acting person may still vary. Within

Heiders nave action analysis, goal attainment is

dependent upon a persons ability and effort,

taskdifficulty, and luck. As Heider states, the

concept of effort is a motivational factor

contributing to theeffectivepersonal force.While

ability, taskdifficulty, and luck are not subject to

volitional change and are thus uncontrollable,

effort has been conceptualized as a controllable

cause.Withinthemoraldomain,freedomofchoiceand personal controllability are necessary

prerequisites for moral evaluations. That is, a

personhasthechoicetoexpendeitherhighorlow

efforttoattainher/hisgoals(e.g.,asacontrollable

causeforsuccessorfailure).Incontrast,apersons

abilityisnotacontrollablecause.

Thus, theconceptofeffort iscloselyconnected to

the volitional forces of the person, expended to

attaina certaingoal.Due to the controllabilityof

effort, judgments concerning this concept (the

investmentofhigheffortversuslackofeffort)exert

adecisiveinfluenceonourmoralemotions.

PredictionsWe argue that different combinations of ought, goal

attainment,andeffortelicitpositiveornegativemoral

evaluations thatconstitute thecognitivebasis for the

experience of moral emotions. As outlined above,

moral emotions communicate positive and negative

evaluationsofapersonsownactions(actoremotions),

ortheactionsofobservedothers(observeremotions).

Ourcentralassumptionisthatthesemoralevaluations

are determined to a large extentby ought (O), goal

attainment (GA), and effort (E), and that different

combinations of ought, goal attainment and effort

provide sufficient conditions for eliciting different

moral emotions. The following predictions can be

derived:

1)PositiveMoralEmotionsAcertainactionshouldberegardedas good if it

conforms to moral standards (O+). A positiveoutcome (GA+) is likely to further increase the

positive evaluation. With respect to effort, it is


4/24


72

widelyagreeduponthatinvestingefforttoattaina

positivestandard(E+)ispositivelyevaluated.Thus,

concerningonesownactions,i.e.,O+,GA+,andE+

should elicit pride, whereas for other persons

actions, the same combination should elicit

admiration,pride,andrespect.Incontrast,whenagoalisnotattained(despitethefactthateffortwas

invested,i.e.,O+,GA,andE+),sympathyshould

betheprototypical emotion.

2)NegativeMoralEmotionsAnactionisevaluatedas bad ifitdoesnotfollow

moral standards (O). Such a negative evaluation

willbestrengthenedwhenapersontrieshard(E+)

toattainanegativegoal,and/orwhenthenegative

goal is actually attained (GA+). In addition, an

action will be also regarded as bad if it doesfollowmoralstandards in thefirstplace(O+),but

when no effort is invested to attain the morally

positivegoal(E).Moreover,lackofeffortbecomes

mostobviouswhenthegoalisnotattained(GA).

Emotional reactions related to ones own actions

include embarrassment, guilt, regret and shame.

With respect to other persons actions, observers

typicallyexperienceanger,contempt,disgust,and

indignation. When other persons either pursue

positivegoalswithout investingeffortor tryhard

toattainanegativelyevaluatedgoal(O+,E orO,E+),schadenfreude iselicitedwhen therespective

goalisnotattained(GA ).

We will now present four studies analyzing the

influenceofought,goalattainment,andefforton

moral emotions. In Study 1 and 2, participants

receive descriptions of events containing

information about ought, goal attainment and

effort; subsequently, moral emotions are assessed

as dependent variables. As actor and observer

emotionsrequireslightlydifferentmethodologies,

Study 1 investigates moral observer emotions,whilemoralactoremotionsareanalyzedinStudy2.

In Study 3 and 4, we will employ a different

method by analyzing autobiographical

recollections of situations that our participants

actuallyexperienced.

STUDY 1

Method1)ParticipantsSixtypsychology students (47 female,13male)ofthe University of Chemnitz (Germany) received

coursecreditsforparticipatinginthisstudy.Their

agerangedfrom19to46(M=23.32)years.Sessions

were run in groups of 20, using a paperpencil

questionnaire, and lasted approximately 30

minutes.

2)ExperimentalDesignAs independent variables, Heiders concepts of

ought, goal attainment, and effort were

manipulated. First, the normative aspect of the

situation (ought)wasvaried; that is,we informed

ourparticipantsaboutanotherperson(referredto

as Max) who wants to attain a highly positive

versus a highly negative goal. Second, we varied

the conceptofgoal attainmentby indicating that

the person either did or did not reach this goal.

Finally, we varied the concept of effort, as this

personeitherdidordidnot investmucheffort to

attainthedesiredgoal.Asforsomeoftheemotions

(e.g.,guiltorpride)closenessofrelationshipmay

be an additional factor contributing to their

elicitation, half of the participants were told that

thetargetpersonwasacloserelativeorbestfriend,

whiletheotherhalfwereinformedthatMaxwasa

stranger to them.This resulted ina2x2x2x2

mixed design, with three withinsubjects factors

(ought, goal attainment, effort) and one

betweensubjectsfactor(closenessofrelationship).

The resulting eight scenarios were presented intwo random sequences toavoidordereffects.All

materialsareavailableuponrequest.

As dependent variables, we presented a set of

moral observer emotions. These emotions were

admiration,anger,contempt,disgust, indignation,

pride(here:beingproudofsomeoneelse),respect,

schadenfreude and sympathy3 . The participants

taskwastoratehowlikelyitwasthattheywould

feel theseemotions.A7pointratingscale labeled

with0 (notatall)and6 (extremely)wasprovided.Thedependentvariableswerepresentedinoneof

tworandomsequences.

3Wedidnotincludeclosesynonyms(remorseasactoremotion,

andempathy,pity,andscornasobserveremotions)aswellas

emotionswhichtypicallyoccurinhighlyspecificcontexts(awe,

elevation, gratitude, jealousy). Moreover, we did not include

envybecausewedonotconsiderthisemotionasamoralemotion:

AsSmith&Kim [11]noted,envy ischaracterizedbysubjective

feelings of inferiority produced by social comparison. This

impliesstandardsforcomparisonarenotuniversal(moral),but

personal. In contrast to moral emotions, envy does notjudge

actionsasmoralornonmoral,butisratherhighlydispositional

andthereforestronglydependentonpersonality.


5/24


73

3)ProcedureWithin each session, participants were randomly

assignedtotherespectiveconditionsandreceived

an instruction sheet, explaining the overall

procedureandintroducingthetargetperson(Max).

Participants received an explanation of a highlypositive versus a highly negative goal. A highly

positive goalwasdefined as an excellent or very

praiseworthy prosocial action, such as helping

someone, whereas a highly negative goal was

defined as a blameworthy or antisocial action,

suchasaggression.Finally,participantswereasked

toratethedegreetowhichtheyexperiencecertain

emotions in different individual scenarios

presentedonthefollowingpages.

ResultsAnalyses reveal that there is no influence of age,

gender and presentation order on any of the

dependentvariables.Therefore,wewillnot consider

these factors in the followinganalyses.According to

ourexperimentaldesign,weanalyzed the respective

conditionsby repeated measurement ANOVAs with

ought, goal attainment and effort as withinsubjects

factors and closeness of relationship as a

betweensubjects factor. With only one minor

exception, no effects were obtained for closeness of

relationship. (Among11possiblemaineffectsand33

possible first order interactions of closeness of

relationship,justonewassignificant,explainingonly

littlevarianceinthedata(pride:F(1,57)=3.59,p


6/24


74

FIG.1MEANVALUESFORMORALOBSERVEREMOTIONSINSTUDY

When taken together, the three independent

variables on average (that is, across all moral

observer emotions) explain 57 % of variance,

rangingfrom48%(anger)to68%(admiration).As

can be seen from Table 1, the respective

independentvariables(ought,goalattainment,and

effort) as well as their interactions do not

contribute equally to variance explanation:

Differentmaineffectsandinteractionsareobtained

for the respective observer emotions, with someemotions (e.g., admiration, pride and respect)

showing quite similar patterns, while other

emotions (e.g., schadenfreude and sympathy) are

characterized by unique patterns across the

respectiveexperimentalconditions:

Admiration,pride,andrespectareexperiencedwhenapositivegoal isattained,particularlywhena lot

ofefforthasbeen invested (seeFigure 1).Hence,

admiration, pride and respect are determinedby

maineffectsofought(admiration:F=312, 2=0.33,

pride:F=175, 2=0.29,respect:F=250, 2=0.35),goal attainment (admiration: F = 204, 2 = 0.12,

pride:F=128, 2=0.06,respect:F=126, 2=0.06)

andeffort(admiration:F=120, 2=0.08,pride:F=

107, 2=0.09,respect:F=125, 2=0.10).Thesemain

effectsarequalifiedbyinteractioneffectsofought

x goal attainment (admiration: F = 150, 2 = 0.08,

pride:F=122, 2=0.05,respect:F=55, 2=0.03):

Theseemotionsarealsoexperiencedtoamoderate

degreewhenanotherpersonhas investeda lotof

effort to pursue a positive goal,but nevertheless

does not attain this goal, as can be seen frominteractionseffectsofoughtxeffort(admiration:F

=85, 2=0.06,pride:F=89, 2=0.08,respect:F=153,

2=0.10).

Furthermore, observers are most likely to feel

sympathywhentheactorhastriedhardtoattaina

highly positive goal, and nevertheless fails.

Sympathyisalsoexperienced(althoughtoalower

degree)whenapositivegoalisnotattainedandthe

observed person has invested only little effort.

Hence,maineffectsareobtainedforought(F=58,=0.06),goalattainment(F=287, =0.18)andeffort (F = 128, = 0.05). Significant interactions

0

1

2

3

4

5

6

Attainment+ Attainment+ Attainment Attainment

Effort+ Effort Effort+ Effort

Admiration

Ought+

Ought

0

1

2

3

4

5

6

Attainment + Attainment+ Attainment Attainment


Anger

Ought+

Ought

0

1

2

3

4

5

6



Contempt

Ought+

Ought

0

1

2

3

4

5

6



Disgust

Ought+

Ought

0

1

2

3

4

5

6



Indignation

Ought+

Ought

0

1

2

3

4

5

6



Pride

Ought+

Ought

0

1

2

3

4

5

6



Respect

Ought+

Ought

0

1

2

3

4

5

6



Schadenfreude

Ought+

Ought

0

1

2

3

4

5

6



Sympathy

Ought+

Ought


7/24


75

ariseforoughtxgoalattainment(F=250, =0.16),

oughtxeffort(F=92, =0.04),goalattainmentx

effort(F=77, =0.04)aswellasforoughtxgoal

attainmentxeffort(F=100, =0.05).

Anger, contempt, disgust and indignation emerge

predominantly in situations involving a

normativelynegativegoal.Thus, amaineffectof

ought is obtained (anger: F = 185, 2 = 0.39,

contempt:F=209, 2=0.52,disgust:F=150, 2=

0.47, indignation: F = 271, 2 = 0.46). For anger,

additionalvarianceisexplainedbyeffort(F=32, 2

= 0.02). Moreover, anger, contempt, disgust and

indignation are experienced to a higher degree

whenanegativegoalisactuallyattained,ascanbe

seen from the interaction of ought x goal

attainment(anger:F=50, 2

=0.03,contempt:F=32,2=0.01,disgust:F=46, 2=0.01,indignation:F=

72, 2 = 0.04). With regard to positive moral

standards, interactions of ought x effort are

obtained for anger, contempt and indignation:

These emotions are more likely when only little

effort hasbeen invested to attain a positive goal

(anger:F=55, 2=0.04,contempt:F=48, 2=0.01,

indignation:F=64, 2=0.03).Furthermore,slightly

lowerdegreesoftheseemotionsareobtainedwith

regardtonegativemoralstandardswhenefforthas

notbeen invested.Thisoccurs fornegativegoals,regardless of whether they were attained and

nonattained.

Finally,schadenfreudeismostlikelywhenanegative

goal isnotattained. It isalsoexperiencedwhena

positive goal is not attained and the observed

personhas investedonly littleeffort toattain this

goal. Thus, schadenfreude is determinedbyboth

maineffectsofgoalattainment(F=170, =0.29)

andought(F=205, =0.15),aswellasinteractions

ofoughtxgoalattainment(F=127, =0.10)andoughtxeffort(F=18, =0.01).

2)ClusterAnalysesHierarchicalClusterAnalyseswereusedtoanalyze

differences and similarities between moral

emotions. The eight situations presented

(representing the eight cells of our design, with

O+/ * GA +/ * E+/) served as independent

variables. Moral emotions were clustered

according to their intensity in each of the eight

situations.Averagelinkagewaschosenasacluster

algorithm, because (a) it is less influenced by

extreme values than other methods, and (b) it is

resistant to several types of potential errors in

cluster calculation. As we were interested in the

similarities of the profiles of the respective

emotions rather than their absolute differences,

Pearson correlations were used as measure ofsimilarity The elbow criterion was applied to

determinethebestclustersolution.

Figure2showstheoptimalclustersolutionforour

data,consistingofatwoclustersolution(cluster1:

admiration, pride, and respect; cluster 2: anger,

contempt,disgust,and indignation),accompanied

by two distinct emotions (sympathy,

schadenfreude). The vertical line represents the

optimalstoppingpointofthemergingprocessand

wasdeterminedbyusingtheelbowcriterion.

Weassessedthevalidityandglobalfitofthecluster

solution. First, tests of cluster homogeneity were

conductedforeachcluster:AccordingtoBackhaus

Fvalues


8/24

w

76

S

M

w.ijpsychol.

twocluster

AsFigure

representd

the respec

seemstobe

tocluster1emotions

positively

andnegati

0.001). In c

with cluste

with clust

mentioned,

using dista

Eucledian

differences

regard to tgreater di

disgust, in

greater dis

cluster 1 (

valuesisa

FIG.2CLUSTE

O

To conclud

cluster sol

global fit.

sympathy

schadenfrethe presen

analysesof

belongingt

within the

Beforedisc

turn to th

emotions.

UDY 2

thod1)ParticipSixtyonep

org

sand the t

2 shows, sc

istinctemot

ive clusters

closertocl

.Hence,weith the two

orrelatedto

elycorrelat

ontrast, sy

r 1 (r = 0.1

r 2 (r =

we also

ncemeasur

distance,

of the t

he two clustance to cl

dignation),

tance to clu

summary

ailableupo

RANALYSIS(

SERVEREM

e, our resu

tion is ch

Additional

is more si

deismorecluster an

varianced

othesamec

respective

ussing thes

cognitive

ntssychologys

odistinct

adenfreud

onsthatare

. However,

ster2,and

correlatedclusters. S

cluster2(r

edtocluste

pathy corr

,p


9/24


77

TABLE2.ANALYSESOFVARIANCEFORACTOREMOTIONS(STUDY2):FVALUESAND FORMAINEFFECTSAND

INTERACTIONSFOROUGHT,GOALATTAINMENT,ANDEFFORT.

Source

Ought GoalAttainment Effort

Oughtx

GoalAttainment

OughtxEffort

Goal

AttainmentxEffort

Oughtx

GoalAttainment

xEffort

Emotio

nF F F F F F F Treat Between Error

Embarr

assment6.69* 0.00 139.36*** 0.26 3.12 0.00 39.88*** 0.06 15.58*** 0.01 10.53** 0.01


10/24

w

78

w.ijpsychol.

(embarrass

0.13,sham

Thedatap

of embarr

thereare s

notonlyd

30, =0.0

butalsoby

Thus,regre

forallnega

the nonat

moresoif

observatio

oughtxgo

effort(F=

xgoalattai

Pride is e

negativeg

forpositiv

invested.T

(F=544,

effort (F =

further qu

attainment

=0.01),a

0.02).2)ClusterAsinStud

x 2 x 2 r

factors ou

addition,

Analyses t

between t

statistical

Analyses

analysesre

presentatio

variables.

Ascanbes

into the cl

guilt,ands

as also su

regretstill

0.71, p


11/24


79

the observer perspective; therefore, moral standards

areappliedmorestrictlytootherpeople.

As expected, admiration, pride (as an observer

emotion),andrespect typicallyoccurwhenobserved

others attain morally positive goals (O+, GA+),

especially when effort has been invested (E+).

Furthermore,sympathy istypicallyelicitedwhen the

observed person has tried hard to attain a morally

positive goal (O+, E+),but nevertheless fails (GA).

Anger,contempt,disgust,and indignationemerge in

situations either characterized by a negative moral

standard(O),orapositivemoralstandard(O+)given

thatonlylittleefforthasbeeninvested(E).Similarly,

schadenfreude is predominantly elicited when a

negative goal is not attained (O, GA), or when a

positive goals is not attained (O+, GA) due to low

effort(E).

Moreover,pride (as an actor emotion) ismost likely

whenamorallypositivegoalwasattained(O+,GA+)

andhighefforthasbeen invested (E+).Thenegative

moralactoremotionsofembarrassment,guilt,shame,

and regret are typically elicited in case of negative

goals (O), but also with regard to nonattained

positive goals (O+; GA), especially when only little

efforthasbeeninvested.

Thus far, we have used abstract scenarios, and

participantsreceivedonlyminimalinformationabouttherespectivesituations.Althoughpreviousresearch

indicates that this method does not necessarily

influencetheobtainedresults[e.g.,10,57,seealso58],

our goal now is to investigate whether the present

findings areobtained formore reallife situations as

well. Therefore, in Study 3 and 4, we will use

situationsderivedfromautobiographicalrecollections

involvingmoralemotions.

STUDY 3

Method

1)ParticipantsSeventyfive psychology students (60 female, 15

male) of the University of Chemnitz received

coursecreditsforparticipatinginthisstudy.Their

agerangedfrom18to46(M=22.00).

2)ExperimentalDesignAs inStudy 1 and 2,Heiders conceptsof ought,

goal attainment, and effort were varied as

independentvariables.IncontrasttoStudies1and

2,weused16scenariosdescribingreallifeevents

which manipulated the three independent

variables.Furthermore,wevariedtheinvolvement

oftheobserver.Thatis,weselectedeightsituations

characterized by a receiver perspective (i.e.,

participantsbeing the target of the action of the

observedperson).

Incontrast,theremainingeightsituationsinvolved

an observer perspective (i.e., participantsbeing a

mere observer). The resulting selection of

situationsisavailableuponrequest.Asdependent

variables,participants rated towhichdegree they

felt admiration, anger, contempt, indignation,

pride, respect, schadenfreude, and sympathy

elicitedby thescenario4.Situationswerepresented

intworandomsequences,witheachscenarioona

separatesheet.

3)ProcedureTo generate a sample of situations manipulating

ought,goalattainment,andeffort,weproceededas

follows: (1) We generated five scenarios for each

combinationofought,goalattainment,andeffort

byaskingrandomlyselectedrespondents(N=24)

torecollectautobiographicalevents inwhich they

had experienced any moral emotion. (2)

Subsequently, we asked 10 independent raters to

assess these situationswith regard toought,goal

attainment,andeffort,ratingthenormativeaspectof the situation (ought), the degree of goal

attainment,andtheinvestedeffort(effort)foreach

scenariobyusing6pointratingscalesanchoredat

bothends(ought:1=verynegative,6=verypositive;goalattainment:1=goalnotattainedatall,6=goalcompletelyattained;effort:1=noeffort,6=higheffort).(3) We chose the 16 scenarios (two for each

combinationofought,effortandgoalattainment,

respectively)providingthebestfittotheintended

manipulations.

Participants received an instruction sheet

explainingtheoverallprocedureandratingscales.

They were asked to imagine the situations as

vividly as possible and to evaluate the degree to

which they would experience the different moral

emotions in the respective situation. A 6point

rating scale anchored atbothends (1 =notatall)and (6 = very strongly) was provided. As a4We excluded disgust from Study 3 because this emotion is

connected to highly specific contexts [17] and thus difficult toassesswithscenarioswhichshouldelicitawiderangeofpossible

emotions.


12/24


80

manipulationcheck (using thesame rating scale),

participants were asked to rate the normative

aspectofthesituation(ought),thegoalattainment,

andtheinvestedeffortforeachspecificscenario.

ResultsAsforStudies1and2,weconducted2x2x2repeated

measuresANOVASandhierarchicalclusteranalyses.

Furthermore, to allow for comparisons between

Studies1and3,correlationanalyseswereperformed.

Additionally, mixed ANOVAS and ttests were

conducted to investigate potential influences of

perspective (receiver vs. observer), control variables

and to assess the quality of our experimental

manipulation.

1)ControlVariablesandManipulationChecksAnalysesrevealthatthereisnoinfluenceofgender

and presentation order on any of the dependent

variables. Therefore, we will not consider these

factors in the following analyses. In addition, we

conducted manipulation checks to ensure that

ought, goal attainment, and effort were

manipulatedas intended.For thispurpose, ttests

fordependentsampleswereconducted:

(1) The positive ought conditions received

significantlyhigherratings(M=5.78,SD=0.27)as

compared to the negative ought conditions (M =1.29,SD=0.40),t(74)=64.90,p


13/24


81

TABLE3.ANALYSESOFVARIANCEFOROBSERVEREMOTIONSINREALISTICSCENARIOS:FVALUESAND(MAINEFFECTSAND

INTERACTIONS) FOROUGHT,GOALATTAINMENT,ANDEFFORT.

Source

Ought GoalAttainment

Effort

Oughtx

GoalAttainment

OughtxEffort

Goal

AttainmentxEffort

Oughtx

Goal

AttainmentxEffort

Emotion F F F F F F F Treat

Bet

ween

Erro

r

Admiratio

n1138.36*** 0.41 479.31*** 0.18 89.12*** 0.02 423.66*** 0.16 72.45** 0.02 39.34*** 0.01 53.44*** 0.01 0.81 0.03 0.15

Anger652.95*** 0.43 140.81*** 0.08 146.93*** 0.04 233.62*** 0.10

114.37*

*0.03 53.02*** 0.02 124.79** 0.03 0.73 0.07 0.20

Contempt 899.98*** 0.61 6.89* 0.00 5.23* 0.00 64.51*** 0.03 66.13** 0.03 34.74*** 0.01 16.91*** 0.00 0.68 0.11 0.21

Indignatio

n1020.89*** 0.56 56.76*** 0.03 82.37*** 0.03 134.08*** 0.06 77.61** 0.02 70.00*** 0.02 56.08*** 0.02 0.74 0.08 0.18

Pride 177.02*** 0.21 76.82*** 0.09 96.41*** 0.04 114.50*** 0.12 119.49**

0.04 0.39*** 0.00 0.55 0.00 0.50 0.13 0.37

Respect892.11*** 0.43 279.59*** 0.14 150.30*** 0.04 260.16*** 0.13

100.64*

*0.03 22.90*** 0.00 36.24*** 0.01 0.78 0.04 0.18

Schadenfr

eude105.48*** 0.13 158.00*** 0.20 10.00** 0.00 60.80*** 0.09 3.16 0.00 18.68*** 0.01 9.78** 0.00 0.43 0.08 0.47

Sympathy 20.83*** 0.03 183.05*** 0.11 64.24*** 0.04 146.47*** 0.11 50.50** 0.04 15.37*** 0.01 49.l6*** 0.03 0.37 0.21 0.41

M() 0.36 0.10 0.02 0.10 0.02 0.02 0.02 0.64 0.90 0.26a. =explainedvarianceforeachindividualfactorintheexperimentaldesign; Treat=explainedvariancebytreatmentfactor; Between=

percentageintersubjectvariance; Error=percentageerrorvariance;*p


14/24


82

andoughtxgoalattainment (anger:F=234, 2=

0.10,contempt:F=65, 2=0.03,indignation:F=134,

2=0.06).Angerandindignationarealsoreported

(again, to a much lower degree) when a person

doesnotattainapositivegoalbuthasinvestedalot

of effort. Thus, main effects for goal attainment(anger:F=140, 2=0.08, indignation:F=57, 2=

0.03) and effort (anger: F = 146, 2 = 0.04,

indignation:F=82, 2=0.03)areobtainedforthese

twoemotions.

Schadenfreude is most likely to occur when a

normatively negative goal is not attained,

especially if the other person has tried hard to

attain thisgoal.Thus,schadenfreude isdetermined

bybothmaineffectsofgoalattainment(F=158, 2

= 0.20) and ought (F = 105, 2 = 0.13), as well as

interactionsofoughtxgoalattainment(F=61, 2=

0.09)andgoalattainmentxeffort(F=19, 2=0.01).

3)ClusterAnalysesAsforStudies1and2,weusedhierarchicalCluster

Analyses to explore differences and similarities

between different moral emotions. To allow for

comparisonsbetween Studies 1 and 2, the same

statistical procedures and methods were used.

Apart from the fact that disgust is no longer

investigated,thestructureoftheclustersolutionis

identical to the cluster solution in Study 1:

Admiration,pride,andrespectconstitutecluster1;

cluster2containsanger,contempt,andindignation.

Schadenfreudeandsympathyaredistinctemotions.

(Due to space limitations, we do not present a

Figureforthisclustersolution;thedendrogramis

availableuponrequest.)

Tests of cluster homogeneity, again conducted

separately for the eight cells of our experimental

design,indicatethatbothclustersarehomogenous:

Only two of the 16Fvalues were slightly higherthan1 (bothcasesoccurred forO+,GA,E+;F=

1.01forcluster1,F=1.04forcluster2).Forcluster1

(admiration, pride, respect) a mean value of F =

0.35 is obtained. For cluster 2 (anger, contempt,

indignation) the mean value is F = 0.48. Thecophenetic correlation of rcoph = 0.95 (p


15/24


83

the items from the observer into an actor

perspective. For example, a situation with a

positive ought, high effort, and attained goal is:

Youwanttobeagoodmother/father. Youhaveto

workveryhard inyourjob,butyou spendevery

spareminutewithyourchildrenandcarealot.Thechildrenaredoingverywellandareveryhappy.

As dependent variables, we presented the moral

actoremotionsembarrassment,guilt,pride,regret,

andshame.

3)ProcedureWeused thesame instructionsandproceduresas

in Study 3, the only difference being that

participants received two situations for each

combination of ought, effort and goal attainment

(as there is only one perspective for actorsituations).

ResultsAs for theprevious studies,we conducted2 x2 x2

repeatedmeasuresANOVASandhierarchicalcluster

analyses. Again, we also conducted correlation

analyses to assess the similaritiesbetween Studies 2

and 4. Again, mixed ANOVAS and ttests were

conducted to assess the quality of our experimental

manipulations and the potential influence of control

variables.

1)ControlVariablesandManipulationChecksAgain,analysesrevealthatthereisnoinfluenceof

gender and presentation order on any of the

dependent variables. Therefore, we will not

considerthesefactorsinthefollowinganalyses.As

isStudy3,we conductedmanipulation checks to

ensurethatought,effort,andgoalattainmentwere

manipulatedas intended.For thispurpose, ttests

fordependentsampleswereconducted,comparing

means of the two variations of each independent

variable. The positive ought conditions received

significantlyhigherratings(M=5.77,SD=0.35)as

compared to the negative ought conditions (M =

1.28,SD=0.34,t(50)=50.82,p


16/24


84

TABLE4.ANALYSESOFVARIANCEFOROBSERVEREMOTIONSINREALISTICSCENARIOS:FVALUESAND(MAINEFFECTSAND

INTERACTIONS) FOROUGHT,GOALATTAINMENT,ANDEFFORT.

Source

Ought GoalAttainment Effort

Oughtx

GoalAttainment

OughtxEffort

Goal

AttainmentxEffort

Oughtx

GoalAttainment

xEffort

Emotion F F F F F F F Treat Between

Error

Embarras

sment34.67** 0.05 492.55*** 0.47 2.91 0.00 62.61*** 0.06 25.41** 0.01 2.83 0.00 39.37** 0.02 0.61 0.15 0.23

Guilt 85.61*** 0.13 355.52*** 0.27 31.14** 0.01 574.50*** 0.25 2.21 0.00 33.35** 0.01 10.34* 0.00 0.67 0.12 0.20

Pride 493.81*** 0.19 821.83*** 0.48 49.94*** 0.01 475.62*** 0.19 0.03 0.00 49.33*** 0.01 1.35 0.00 0.88 0.03 0.10

Regret 115.54*** 0.21 132.70*** 0.16 44.92*** 0.02 178.31*** 0.13 22.33** 0.01 41.02*** 0.01 33.58** 0.02 0.56 0.17 0.27

Shame 113.28*** 0.17 346.12*** 0.27 3.29 0.00 189.82*** 0.14 48.96*** 0.01 3.89 0.00 28.48** 0.01 0.60 0.17 0.23

M() 0.15 0.33 0.01 0.15 0.00 0.00 0.01 0.66 0.13 0.21a. =explainedvarianceforeachindividualfactorintheexperimentaldesign; Treat=explainedvariancebytreatmentfactor; Between=

percentageintersubjectvariance;Error=percentageerrorvariance;*p


17/24


85

moral actoremotionsobtained for abstract scenarios

(Study2)anddescriptionsofreallifeevents(Study4).

Large effects were obtained for all emotions

(embarrassment:r=0.96,p


18/24


86

antecedentcognitions.Inwhatfollows,wewillbriefly

summarizetheseresults.

1)PositiveObserverEmotionsAdmiration, pride, and respect are strongly

connected to the attainment of highly positivegoals, especially when high effort has been

invested.Toamoderatedegree,theseemotionsare

alsoexperiencedwhenpositivegoalsareattained

intheabsenceofeffort.Likely,effortandabilityare

regardedasbeingcompensatory.Hence,effortless

goal attainmentwillbeattributed tohigh ability:

We do not only admire, respect or feel proud of

otherpeopleforinvestingeffort(acauseregarded

as being controllable, but also for their ability

(regarded as being uncontrollable). In addition,

these emotions also emerge (although to a lesserdegree) when a normatively positive goal is not

attained while high effort was present, thus

underlining the positive evaluation of the action

(investingeffort).

Sympathyisconnectedtoahighlyspecificpattern

of ought, goal attainment, and effort: We

predominantlyfeelsympathywhensomeonetries

hardtoattainapositivegoalandneverthelessfails,

representing a positive evaluation of the actors

behavior. As effort and ability are regarded as

compensatory,thisresultalsoconfirmsthatpeople

are sympathetic toward others who are unable

andwhoseplight isuncontrollable, suchas the

physically or mentally handicapped. However,

sympathy is also experienced (although to a

substantially lower degree) when effort had not

beeninvested.Fromanevolutionarypointofview,

sympathy (or compassion) has evolved as a

response to the suffering of others (both kin and

nonkin individuals). Applied to our data, the

failing person could be regarded as sufferingbecause of the nonattainment of her/his goal.

Consequently,sympathyiselicitedwhen(positive)

goals were not attained. However, sympathy is

furtherincreased ifthepersontriedhardtoattain

the goal, which also implies an evolutionary

adaptive signal indicating that the persons

behavior was good, encouraging the person to

continuewiththisbehaviorandtryagain.

Interestingly, moderate degrees of positive moral

emotions even occur for nonattained positive

goals (given the presence of high effort), andnegativemoralemotionsevenoccurwhenpositive

goalsareattained(giventheabsenceofeffort).Itis

inthesecasesthattheevaluativefunctionofmoral

emotionsbecomesmost obvious: Despite the fact

that a person was not successful, the morally

positivebehavior (investing effort) is appreciated

bypositivemoralemotions.2)NegativeObserverEmotionsAnger, contempt, disgust, and indignation are

moststronglyexperiencedvisvisnegativegoals

that are actually attained. Moderate degrees of

these emotions are also obtained when a person

pursues a negative goal without attaining it

(especially for indignation and anger). Finally, a

person elicits a certain degree of negative moral

emotionsifthepersondidnotinvestefforttoattain

a positive goal, communicating that thebehavior(notinvestingeffort)wasmorallynegative.Again,

these findings underline that anger is a value

judgmentfollowingfromthebeliefthatonecould

and should have done otherwise. In contrast,

schadenfreudeismoststronglyexperiencedwhen

another person does not attain a negative goal,

regardlessofwhethertheactortriedhardornot.In

addition, some degree of schadenfreude is also

experienced when a positive goal is not attained

because the actor invested only little effort.

Apparently, people do not deserve to attain anynegativegoal(regardlessofhowmuchtheytried),

and (to a lesser degree) they do not deserve to

attainapositivegoalwheneffortisabsent.

3)PositiveActorEmotionsSimilar to its observer counterpart, pride is most

strongly experienced when highly positive goals

are attained given high effort, underlining the

positive evaluative function of pride. Moderate

pride is also experienced for the attainment of

positivegoals in theabsenceofeffortand for theattainment of negative goals. Tangney as well as

Tracyetal.distinguishbetweenalphapride(pride

relatedtoonespersonalself;andbetapride(being

proudofonesownbehavior).Whereasbetapride

results from attributions to internal, unstable,

controllable causes (i.e., effort, according to our

analysis), alphapride results from attributions to

internal,stable,uncontrollablecauses(e.g.,ability,

asderivedfromeffortlessgoalattainment).Almost

noprideisexperiencedforsituationscharacterized

by nonattainment of a goal (regardless of oughtandeffort).


19/24


87

4)NegativeActorEmotionsEmbarrassment,guilt,regret,andshamearemost

strongly experienced for the nonattainment of

positive goals, when only little effort has been

invested5. As for negative observer emotions, the

negative evaluation of lack of effort becomesobvious. In addition, moderate degrees of these

emotions are obtained for negative goals,

regardless of goal attainment and effort. In

previous research, it is agreed upon that shame,

guilt, and embarrassment as selfconscious

emotions are experienced in social contexts and

functionasameansofbehavioral regulationandto

avoidnegativesanctions fromothers.Thepresent

datasuggestthatagreatamountofthesimilarities

in these negative moral actor emotions can be

tracedback to thecognitiveantecedentsofought,goalattainment,andeffort.

Inshort,ourresultscanbesummarizedasfollows:

Positive moral observer emotions (admiration,

pride, respect, sympathy) constitute positive

evaluation of other persons actions and

predominantlyoccurforpursuingmorallypositive

goals(ought)aswellasforinvestingefforttoattain

thesegoals.

In contrast, negative moral observer emotions

(anger, contempt, disgust, indignation,schadenfreude)communicatenegativeevaluations

ofotherpersonsactionsandare typicallyelicited

by pursuing morally negative goals or not

investingefforttoattainmorallypositivegoals).

Positivemoralactoremotions(pride)functionasa

positive evaluation of a persons own action and

are most likely to occur for the attainment of

positive goals, especially in the presence of high

effort.

Finally, negative moral actor emotions (shame,guilt, regret, embarrassment) are negative

evaluations of a persons own action and are

typically causedbymorallynegativegoalsor the

nonattainmentofpositivegoals,especiallywhen

onlylittleeffortwasinvested.

In sum, we suggest that ought, goal attainment,

and effort are powerful sufficient conditions

determining the emergence of moral emotions.

5

Thereisonlyoneexception;inStudy2,incaseofnonattainedpositivegoals,shamewas felttoa fairlysimilarextentforboth

conditions(higheffortaswellaslackofeffort).

However, this is not to say that these factors are

their only determinants. For example, much

research has been devoted to the differences

betweenguiltandshame orangerandcontempt,

offeringimportantadditionalinsightswithregard

to specific emotions, for example envy andschadenfreude,admirationorpride.

Furthermore,variousemotionalappraisaltheories

haveanalyzeddifferentcognitiveappraisals (e.g.,

pleasantness, fairness, certainty of the outcome,

suddenness, personal importance, controllability,

effort,modifiability, familiarityetc.)asantecedent

conditions for the elicitation of a large variety of

different emotions (such as sadness, fear,

disappointment,shame,anger,joy,surprise,relief,

disgust or hope. However, note that within the

moraldomain,theemergenceofemotionsstrongly

depends on appraisals that are referring to

volitionalcontrol.Withregardtotheevaluationof

goal attainment, only ought (the positive or

negativemoralstandard)andeffort(theamountof

invested effort to attain the respective goal) are

controllable to the person and thus elicit moral

emotions which function as evaluations of our

intensions(ought)andactions(effort).

Moreover,animportantcontributionofthepresent

research is to offer a comprehensive frameworkthat applies to all moralemotions, andonwhich

more specific analyses, including additional

determinantsofspecificemotions,canbebased.As

a starting point, we suggest that moral emotions

constitute positive (positive moral emotions) or

negative (negativemoralemotions)evaluationsof

our own (actor emotions) or others (observer

emotions)actionsintheprocessofgoalattainment.

Theseevaluations ofactionsarebasedonuniversal

moral standards (ought), the attainment of a

persons goal (goal attainment), and the investedefforttoattainthegoal(effort).

FutureResearchIn the present studies, we have analyzed two new

conceptsinthedomainofmoralemotions,thatis,goal

attainment and effort. Previous studies mainly

focused (1) on the moral standards underlying the

actions under consideration, and (2) the reasoning

processesgiving rise tomoraljudgments [see1,84].

Accordingtoourresults,moralstandardsindeedlieat

the

very

heart

of

moral

judgments.

However,

theonesided focus on moral standards disregards

essentialinformationthatisaccessibletonaveactors


20/24


88

and observers. In light of our findings, it becomes

apparentthatgoalattainmentandeffortarepowerful

(additional)predictorsofmoralemotionsaswell.

Severalresearchquestionsariseinthiscontext:Firstof

all, future research should analyze the influence of

ought,goalattainment,andeffortonmoralbehavior.It seems plausible to assume that approach and

avoidanceaswellasrewardandpunishmentvaryasa

function of the moral standard of a goal, goal

attainment, and effort. Closely connected to this

question, research is needed to elucidate the

motivational(i.e.,functional)statusofmoralemotions

with respect to moral actions. From an evolutionary

pointofview,weassumethatmoralemotionsevolved

tomotivatemoralbehavior [see32]. In linewith this

view, studies on helping behavior and aggression

from an attributional viewpoint revealed that thesebehavioral reactions arepredominantly mediatedby

the experience of sympathy and anger (prototypical

moralemotions;Furthermore,itwasfoundthatguilt,

shame, regret and schadenfreude also influence

helping behavior or similar prosocial actions.

Equivalent empirical research on the motivational

function of other moral emotions is still missing,

however.

Second, it becomes apparent that the Heiderian

conceptsofought,goalattainment,andeffortdividethe landscape of moral emotions into meaningful

clusters.Furtherresearchisneededtounderstandthe

(probablymoresubtle)differencesof thoseemotions

thatmapintooneandthesamecluster.Forexample,

weneed to fullyunderstand thedifferencesbetween

admiration,respect,andpride,buildingajointcluster

butpresumablydiffering in someotherwaysand/or

withrespecttotheirbehavioral consequences.

Third, it becomes evident that one and the same

emotion can be elicited by a variety of different

antecedent conditions.For example,anger is elicitedvisvispersonspursuingnegativegoals,butalsofor

nonattainment of a positive goal in the absence of

effort.Thus,apromisingdirectionforfutureresearch

is to analyze prototypical situations eliciting the

respective emotion especially clearly.

Autobiographicalrecollectionsofmoralemotionsand

situations are a promising methodology to answer

thesequestions.

Fourth, the selectionofmoralemotionsdiscussed in

our study wasbased on a literature search. On the

basisoftheproposedtheoreticalframeworkitisnow

possible to design studies analyzing the differences

betweenmoralandnonmoralemotions.Forexample,

it seems reasonable that for the elicitation of

nonmoral emotions (e.g., joy or sadness) goal

attainment should be more important than for the

elicitationofmoralemotions.

Insum,webelievethatthecomprehensivetheoretical

frameworkpresentedhereprovidesauseful tool for

futureresearch.Fromthisbasis,wewillbeable(1)to

betterunderstandthecommonunderlyingconceptsof

moral emotions, and (2) to further explore the

idiosyncraticdeterminantsofspecificmoralemotions

beyond their common characteristics outlined in the

presentresearch.

ACKNOWLEDGMENT

We are extremely grateful to Frank Schssow andBernard Weiner for their comments and insights

related to former versions of the present paper.

Moreover, special thanks to Dieter Reichelt for his

invaluable help with data collection, and Andr

KrnerandStephanieLauxforeditorialassistance.

REFERENCES

A.BenZeev,Pleasure inothersmisfortune.TheJerusalem

PhilosophicalQuarterly.41(1992).

A.H.FischerandI.J.Roseman,Beatthemorbanthem:The

characteristicsandsocialfunctionsofangerandcontempt.

JournalofPersonalityandSocialPsychology,93,1 (2007)

doi:10.1037/00223514.93.1.103.

A.Ortony,G.L.CloreandA.Collins,Thecognitivestructure

of emotions, NY US: Cambridge University Press, New

York,(1988).

A.P.Fiske,Sociomoralemotionsmotivateaction tosustain

relationships. Self and Identity, 1, 2 (2002)

doi:10.1080/152988602317319357.

B.F.Malle,Theactorobserverasymmetry inattribution:A

(surprising)metaanalysis.PsychologicalBulletin,132,6

(2006)doi:10.1037/00332909.132.6.895.

B.Weiner,Anattributionaltheoryofachievementmotivation

and emotion. Psychological Review, 92, 4 (1985)

doi:10.1037/0033295X.92.4.548.

B. Weiner and A. Kukla, An attributional analysis of

achievementmotivation,JournalofPersonalityandSocial

Psychology,15,1(1970)doi:10.1037/h0029211.

B. Weiner,Judgments of responsibility: A foundation for a


21/24


89

theoryofsocialconduct,GuilfordPress,NewYork(1995)

B.Weiner,Socialmotivation,justice,andthemoralemotions:

An attributional approach, NJ US: Lawrence Erlbaum

AssociatesPublishers,Mahwah,(2006).

B.Weiner,Theclassroomasacourtroom.SocialPsychologyofEducation,6,1(2003)doi:10.1023/A:1021736217048.

C. A. Hutcherson andJ.J. Gross, The moral emotions: A

socialfunctionalist account of anger, disgust, and

contempt.JournalofPersonality andSocialPsychology,

100,4(2011)doi:10.1037/a0022408.

C.A.SmithandR.S.Lazarus,Appraisalcomponents, core

relational themes, and the emotions. Cognition And

Emotion,7,34(1993).doi:10.1080/02699939308409189.

C.Reyna

and

B.Weiner,

Justice

and

utility

inthe

classroom:

An attributional analysis of the goals of teachers`

punishment and intervention strategies. Journal of

Educational Psychology, 93, 2 (2001)

doi:10.1037/00220663.93.2.309.

E.Turiel,Thecultureofmorality:Socialdevelopment,context,

and conflict, Cambridge University Press, New York

(2002).

E. Turiel, The Development of Social Knowledge: Morality

andConvention,CambridgeUniversityPress,Cambridge

(1983).

F. Heider, The psychology of interpersonal relations,

LawrenceErlbaumAssociatesPublishers,Mahwah (US)

(1958).

G. L. Clore, A. Ortony and M. A. Foss, The psychological

foundationsoftheaffectivelexicon.JournalofPersonality

and Social Psychology, 53, 4 (1987).

doi:10.1037/00223514.53.4.751.

G.W.Milligan,Anexaminationof theeffectofsix typesof

error perturbation on fifteen clustering algorithms.

Psychometrika,45,3(1980).doi:10.1007/BF02293907.

H. B. Lewis, Shame and guilt in neurosis. Psychoanalytic

Review,58,3(1971).

H. H. Kelley, Commons sense psychology and scientific

psychology. Annual Review of Psychology, 43, (1992)

doi:10.1146/annurev.ps.43.020192.000245.

I.E.deHooge,M.ZeelenbergandS.M.Breugelmans,Moral

sentiments and cooperation: Differential influences of

shame and guilt. Cognition and Emotion, 21, 5, (2007)

doi:10.1080/02699930600980874.

I.E.deHooge,M.ZeelenbergandS.M.Breugelmans.Restore

andprotectmotivationsfollowingshame.CognitionAnd

Emotion,24,1(2010)doi:10.1080/02699930802584466.

I.J.RosemanandA.Evdokas,Appraisalscauseexperienced

emotions: Experimental evidence. Cognition AndEmotion,18,1(2004)doi:10.1080/02699930244000390.

I.J. Roseman, C. Wiest and T. S. Swartz Phenomenology,

behaviors, and goals differentiate discrete emotions.

JournalofPersonalityandSocialPsychology,67,2(1994)

doi:10.1037/00223514.67.2.206.

I. Kant, Grundlegung zur Metaphysik der Sitten. Reclam,

Ditzingen(1998)(Originalworkpublished1785).

J. Bacher, Clusteranalyse, R. Oldenbourg Verlag, Munich

(2002).J.Bentham,An introduction to theprinciplesofmoralsand

legislation. Oxford University Press, New York (2005)

(Originalworkpublished1789).

J. Clatworthy, D. Buick, M. Hankins, J. Weinman and R.

Horne.Theuseandreportingofclusteranalysisinhealth

psychology: A review, British Journal of Health

Psychology.10,3(2005)doi:10.1348/135910705X25697.

J.Cohen,Etasquaredandpartialetasquaredin fixedfactor

ANOVA designs, Educational and Psychological

Measurement, 33, 1, (1973)

doi:10.1177/001316447303300111.

J.Cohen,Statisticalpoweranalysisforthebehavioralsciences

(2nded.),AcademicPress,NewYork(1988).

J.Chipperfield,R.Perry,B.WeinerandN.Newall,Reported

CausalAntecedentsofDiscreteEmotionsinLateLife,The

InternationalJournalOfAgingAndHumanDevelopment.

68,3(2009).

J.D.Greene,L.E.Nystrom,A.D.Engell,J,M.DarleyandJ.D.

Cohen,Theneuralbasesofcognitiveconflictandcontrol

inmoraljudgment.Neuron (ScienceDirect),44,2 (2004)

doi:10.1016/j.neuron.2004.09.027.

J.D.Greene,Opinion:Fromneuralistomoralought:What

are the moral implications of neuroscientific moral

psychology? Nature Reviews Neuroscience, 4, 10 (2003)

doi:10.1038/nrn1224.

J.D.Greene,R.B.Sommerville,L.E.Nystrom,J.M.Darley

and J .D. Cohen, An fMRI investigation of emotional

engagementinmoraljudgment.Science,293,5537(2001)

doi:10.1126/science.1062872.


22/24


90

J. Haidt, The emotional dog and its rational tail: A social

intuitionist approach to moraljudgment. Psychological

Review,108,4(2001)doi:10.1037/0033295X.108.4.814.

J. Haidt,Thenew synthesis inmoralpsychology.Science,

316,5827,(2007)doi:10.1126/science.1137651.J. Haidt, The moral emotions. In Handbook of affective

sciences, Edited R.J. Davidson, K. R. Scherer, & H. H.

Goldsmith,NYUS:OxfordUniversityPress:New York

(2003),pp.852870.

J.L. Goetz, D. Keltner andE. SimonThomas, Compassion:

An evolutionary analysis and empirical review.

PsychologicalBulletin,136,3(2010)doi:10.1037/a0018807.

J. L. Tracy and R. W. Robins, The prototypical pride

expression:Developmentofanonverbalbehaviorcodingsystem.Emotion,7,4(2007)doi:10.1037/15283542.7.4.789.

J.L.Tracy,Robins,R.W.,&Lagattuta,K.H.,Canchildren

recognize pride? Emotion, 5, 3, (2005) 251257.

doi:10.1037/15283542.5.3.251.

J.P.Tangney,Moralaffect:Thegood,thebad,andtheugly.


doi:10.1037/00223514.61.4.598.

J. P. Tangney and K. W. Fischer (Eds.), Selfconscious

emotions:Thepsychologyofshame,guilt,embarrassment

andpride.,NYUS:GuilfordPress,NewYork,(1995).

J.P.Tangney,J.StuewigandD.J.Mashek,Moralemotions

and moralbehavior. Annual Review of Psychology, 58,

(2007)doi:10.1146/annurev.psych.56.091103.070145.

J.P.Tangney andR. L Dearing. Shame andguilt., NYUS:

GuilfordPress,NewYork,(2002).

Jr., D. J. Ketchen and T. M. Hult, Validating cluster

assignments. Psychological Reports, 87, 3 (2000)

doi:10.2466/PR0.87.7.10571058.

J. Shao, S. W. Tanner, N. Thompson and T. E. Cheatham,

Clustering molecular dynamics trajectories: 1.

characterizing the performance of different clustering

algorithms.JournalofChemicalTheoryandComputation,

3,6(2007)doi:10.1021/ct700119m.

J. Stuewig, J.P. Tangney, C. Heigel, L. Harty and, L.

McCloskeyShaming,blaming,andmaiming:Functional

linksamongthemoralemotions,externalizationofblame,

andaggression.JournalofResearch inPersonality,44,1

(2010)doi:10.1016/j.jrp.2009.12.005.

J. S. Mill, Utilitarianism, Books LLC, Memphis, (2009)

(Originalworkpublished1863).

K. Backhaus, Multivariate Analysemethoden. Springer

Verlag,Berlin(2000).

Keltner andJ. Haidt, Approaching awe, a moral, spiritual,

andaestheticemotion.Cognition&Emotion,17,2(2003)doi:10.1080/02699930302297.

K. Scherer (Ed.), A. Schorr (Ed.) and, T. Johnstone (Ed.),

Appraisal processes in emotion: Theory, methods,

research,OxfordUniversityPress,NewYork,2001.

L. Kohlberg, Stage and sequence: The

cognitivedevelopmental approach to socialization. In

HandbookofsocializationtheoryandresearchEditedD.

A.Goslin,RandMcNally,Chicago,(1969).pp.347480.

M. C. Matteucci, Teachers facing school failure: The socialvalorization of effort in the school context. Social

Psychology of Education, 10, 1 (2007)

doi:10.1007/s112180069011x.

M. D. Robinson and G. L. Clore, Episodic and semantic

knowledge in emotional selfreport: Evidence for two

judgment processes. Journal of Personality and Social

Psychology,83,1(2002)doi:10.1037/00223514.83.1.198.

M.E.McCullough,S.D.Kilpatrick,R.A.EmmonsandD.B.

Larson,Isgratitudeamoralaffect?PsychologicalBulletin,

127,2(2001).doi:10.1037/00332909.127.2.249.

M.Hauser,Moralminds:Hownaturedesignedouruniversal

sense of right and wrong. NY US: Ecco/HarperCollins

Publishers,NewYork(2006).

M. Higuchi and H. Fukada, A comparison of four causal

factorsofembarrassmentinpublicandprivatesituations.

Journal of Psychology, 136, 4 (2002)

doi:10.1080/00223980209604166.

M. Lewis, Selfconscious emotions: Embarrassment, pride,

shame, and guilt, in Handbook of emotions (3rd ed.),

Edited M. Lewis, J. M. HavilandJones, L. Barrett, M.

Lewis,J. M. HavilandJones, L. Barrett, Guilford Press,

NewYork,(2008)pp.742756.

M. Silfver Coping with guilt and shame: A narrative

approach. Journal of Moral Education, 36, 2 (2007)

doi:10.1080/03057240701325274.

N.H.Frijda,P.KuipersandE. terSchure,Relationsamong

emotion, appraisal, and emotional action readiness.

JournalOfPersonalityAndSocialPsychology,57,2(1989)

doi:10.1037/00223514.57.2.212.


23/24


91

N. T. Feather and I. R. McKee, Differentiating emotions in

relation to deserved or undeserved outcomes: A

retrospective study of reallife events. Cognition and

Emotion,23,5(2009)doi:10.1080/02699930802243378.

N. T. Feather and R. Sherman, Envy, resentment,schadenfreude,andsympathy:Reactionstodeservedand

underserved achievement and subsequent failure.

Personality and Social Psychology Bulletin, 28, 7 (2002)

doi:10.1177/01467202028007008.

N.T.Feather,Effectsofobserversownstatusonreactionsto

a high achievers failure: Deservingness, resentment,

schadenfreude, and sympathy. Australian Journal of

Psychology,60,1(2008)doi:10.1080/00049530701458068.

N.T.Feather,Values,achievement,andjustice:Studiesinthe

psychology of deservingness, Kluwer Academic

Publishers.DordrechtNetherlands(1999).

P.M.Fromson,Evoking shame andguilt:A comparisonof

two theories. Psychological Reports, 98, 1 (2006)

doi:10.2466/pr0.98.1.99105.

P. C. Ellsworth and K. R. Scherer, Appraisal processes in

emotion, inHandbookof affective sciences,EditedR.J.

Davidson,K.R.Scherer,H.Goldsmith,R.J.Davidson,K.

R. Scherer &H. Goldsmith., NY US: Oxford University

Press.NewYork(2003),pp.572595.

P. Ekman, Expression and the nature of emotion, in

ApproachestoEmotion,EditedK.S.Scherer&P.Ekman

NJ:Erlbaum,Hillsdale(1984),pp.319344.

P.M.Niedenthal,J.PTangneyand, I.Gavanski, Ifonly I

werentversusIfonly Ihadnt:Distinguishingshame

and guilt in counterfactual thinking. Journal of

Personality and Social Psychology, 67, 4 (1994)

doi:10.1037/00223514.67.4.585.

P.Rozin,L.Lowery,S. ImadaandJ.Haidt,TheCAD triad

hypothesis: A mapping between three moral emotions

(contempt, anger, disgust) and three moral codes

(community,autonomy,divinity).JournalofPersonality

and Social Psychology, 76, 4 (1999)

doi:10.1037/00223514.76.4.574.

P.RussellandR.GinerSorolla,Moral,anger,butnotmoral

disgust,responds to intentionality.Emotion,11,2 (2011)

doi:10.1037/a0022598.

R.F.Baumeister,A.M.StillwellandT.F.Heatherton,Guilt:

An interpersonalapproach.PsychologicalBulletin.115,2

(1994).doi:10.1037/00332909.115.2.243.

R. Gutierrez and R. GinerSorolla, Anger, disgust, and

presumption of harm as reactions to taboobreaking

behaviors. Emotion, 7, 4 (2007)

doi:10.1037/15283542.7.4.853.R. H. Smith and S. H. Kim Comprehending envy.

Psychological Bulletin, 133, 1 (2007).

doi:10.1037/00332909.133.1.46.

R.S.Lazarusand,C.A.Smith,Knowledgeandappraisalin

the cognition motion relationship. Cognition and

Emotion,2,4(1988).doi:10.1080/02699938808412701.

S.B.AlgoeandJ.Haidt,Witnessingexcellenceinaction:The

otherpraising emotions of elevation, gratitude, and

admiration.TheJournalOfPositivePsychology,4,2(2009)doi:10.1080/17439760802650519.

S. Graham, A review of attribution theory in achievement

contexts. Educational Psychology Review, 3, 1 (1991)

doi:10.1007/BF01323661.

S. Hareli and B. Weiner, Social emotions and personality

inferences:Ascaffoldforanewdirectionin thestudyof

achievementmotivation.EducationalPsychologist, 37,3

(2002)doi:10.1207/S15326985EP3703_4.

S. Hareli and B. Parkinson, Whats social about social

emotions?.JournalfortheTheoryofSocialBehaviour,38,

2(2008)doi:10.1111/j.14685914.2008.00363.x.

S.SheikhandR.JanoffBulman,Theshouldsandshould

notsofmoralemotions:Aselfregulatoryperspectiveon

shame and guilt. Personality and Social Psychology

Bulletin,36,2(2010)doi:10.1177/0146167209356788.

Tangney,J.P.,Assessingindividualdifferencesinproneness

to shame and guilt: Development of the selfconscious

affect and attribution inventory. Journal of Personality

and Social Psychology, 59, 1 (1990) 102111.

doi:10.1037/00223514.59.1.102.

Tangney,J. P., Miller, R. S., Flicker, L., & D. Barlow, Are

shame, guilt, and embarrassment distinct emotions?


doi:10.1037/00223514.70.6.1256.

T. D. Wilson and P. W. Linville, Improving the academic

performance of college freshmen: Attribution therapy

revisited.JournalOfPersonalityAndSocialPsychology,

42,2(1982)doi:10.1037/00223514.42.2.


24/24


T.KetelaarandW.Au,Theeffectsoffeelingsofguiltonthe

behaviorofuncooperative individuals inrepeatedsocial

bargaininggames:Anaffectasinformationinterpretation

oftheroleofemotioninsocialinteraction.Cognitionand

Emotion,17,3(2003)doi:10.10 80/02699930143000662.T. R. Levine and C. R. Hullett, Eta squared, partial eta

squared,andmisreportingofeffectsizeincommunication

research.HumanCommunication Research, 28, 4 (2002)

doi:10.1111/j.14682958.2002.tb00828.x.

U.Rudolph, S.C. Roesch, T.Greitemeyer andB.Weiner,A

metaanalyticreviewofhelpgivingandaggressionfrom

an attributional perspective: Contributions to a general

theoryofmotivation.Cognition&Emotion,18,6 (2004)

doi:10.1080/02699930341000248.

W.W.vanDijk,J.W.Ouwerkerk,Y.M.WesselingandG.M.

vanKoningsbruggen,Towardsunderstandingpleasureat

the misfortunes ofothers: The impactof selfevaluation

threatonschadenfreude.Cognition&Emotion,25,2(2011)

doi:10.1080/02699931.2010.487365.

W.W.vanDijk,J.W.Ouwerkerk,S.Goslinga,M.Niewegand

M.Gallucci,Whenpeoplefallfromgrace:Reconsidering

the roleof envy in schadenfreude.Emotion, 6,1 (2006).

doi:10.1037/15283542.6.1.156.

Z. Aiqing, Z. Fanglian and L. Huashan, The Relationships

Among ResponsibilityJudgment, Schadenfreud and the

WillofHelpGiving.PsychologicalScience(China).27,1

(2004).

L.F.Martinez,M.ZeelenbergandJ.B.RijsmanBehavioural

consequences of regret and disappointment in social

bargaining games. Cognition & Emotion, 25(2), (2011).doi:10.1080/02699931.2010.485889.

UdoRudolph is Professor in General and

Biopsychology at the Department of

Psychology at the Chemnitz University of

Technology (Faculty of Human and Social

Sciences). Research interests include both

basic and applied research on causality,

attribution,andmoralemotions.

Katrin Schulz is lecturer in General and

Biopsychology at the Department ofPsychology at the Chemnitz University of

Technology (Faculty ofHuman and Social

Sciences). Research interests include

applied research on the development of

emotional and social skills during

childhood.

Nadine Tscharaktschiew is lecturer in

General and Biopsychology at the

Department of Psychology at the Chemnitz

UniversityofTechnology(FacultyofHuman

and Social Sciences). Research interestsincludebasicresearchonmoralemotions.

Documents

Moral Emotions: An Analysis Guided by Heiderʹs Naive Action Analysis