Forgiving the Unforgivable: Couples’ Forgiveness and ExpectedForgiveness of Emotional and Sexual Infidelity From an Error

Management Theory Perspective

Mons Bendixen, Leif Edward Ottesen Kennair, and Trond Viggo GrøntvedtNorwegian University of Science and Technology

From an Error Management Theory (EMT) perspective, we expect a negative forgivenessbias, in which one underestimates a romantic partner’s forgiveness following one’s owntransgressions. The function would be to secure that the transgressions are fully mendedthrough reparative behavior. The extent to which this bias is related to infidelity and typeof infidelity has not yet been examined. Both parties in heterosexual couples (N � 92)responded independently to questionnaires about their reactions to hypothetical emotionaland sexual infidelity scenarios: forgiveness of unfaithful partner’s hypothetical infidelityand likelihood of believing in cheated partner’s forgiveness. Question order was balancedfor self and partner transgressions. The results suggest strong negative forgiveness biaseffects. Participants reported less belief in their infidelity being forgiven compared with 2different criteria of partner forgiveness (self-rated and expressed by partner). This negativeforgiveness bias was present for both emotional and sexual infidelity scenarios, and formale and female participants. However, this bias was clearly less pronounced for emotion-ally unfaithful men. In addition, as predicted, relative to women, men imagining beingvictims of emotional infidelity perceived this to be less threatening to the relationshipcompared with being victim of sexual infidelity: They also expressed more forgiveness andless need of keeping distance. Need for revenge and likelihood of breaking up because ofpartner’s infidelity evinced no sex differentiated patterns. The study provides novel anddeeper insights into how men and women perceive and react to the distress of emotional andsexual infidelity.

Keywords: error management, bias, forgiveness, couples, infidelity, jealousy

Based on signal detection theory (Green &Swets, 1966; Swets, Dawes, & Monahan, 2000),

Error Management Theory (EMT; Haselton &Buss, 2000; Haselton & Nettle, 2006) describeshow natural selection may have designed psy-chological adaptations for judgment under un-certainty. In addition to making correct judg-ments (true positives and true negatives), twotypes of judgmental errors can be committed: Aperson may adopt a belief that is in fact not true(false positive) or fail to adopt a belief that is infact true (false negative). Within domains inwhich the costs of errors have been asymmetri-cal over deep evolutionary time, selection mayhave favored designs that tended to make theless costly error of the two.

Natural selection would produce adaptivelybiased systems that exist in the present becausethey led to survival and reproductive advan-tages for humans in the past. Several protectivebiases have been identified within the percep-tion, attention, and learning domains, and withinsocial- and person-perception domains (Haselton& Nettle, 2006). Within domains where the asym-

This article was published Online First September 28,2017.

Mons Bendixen and Leif Edward Ottesen Kennair, Depart-ment of Psychology, Norwegian University of Science andTechnology; Trond Viggo Grøntvedt, Department of Psychol-ogy and Department of Public Health and General Practice,Faculty of Medicine, Norwegian University of Science andTechnology.

The authors thanks Kyrre Svarva at the IT-section at Nor-wegian University of Science and Technology (NTNU) forformatting machine-readable questionnaires and for his assis-tance in scanning the data, and Joy P. Wyckoff for valuablecomments on the manuscript. We also acknowledge the assis-tance of students at the Bachelor Program in Psychology atNTNU for their recruitment of participants and scanning ofdata. Preliminary findings from this study were presented at theAnnual HBES conference in Vancouver, July 2016.

Correspondence concerning this article should be addressedto Mons Bendixen, Department of Psychology, NorwegianUniversity of Science and Technology, Dragvoll Campus, NO-7491 Trondheim, Norway. E-mail: [email protected]

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anP

sych

olog

ical

Ass

ocia

tion

oron

eof

its

alli

edpu

blis

hers

.T

his

arti

cle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

Evolutionary Behavioral Sciences © 2017 American Psychological Association2018, Vol. 12, No. 4, 322–335 2330-2925/18/$12.00 http://dx.doi.org/10.1037/ebs0000110

322

metry of cost of errors have been similar forwomen and men, no sex difference in bias ispredicted by EMT (Haselton & Buss, 2000;Haselton & Nettle, 2006). However, within do-mains where asymmetry of cost of errors havebeen different for the two sexes, EMT predictssex differences in adaptive biases. For example,in human mating, signals of sexual interest maybe subject to one of two forms of mind-readingbiases: overperception or underperception.Overperception bias would lead a person toinfer there is some sexual interest when there isnone (false positive). An underperception biaswould lead a person to infer there is no sexualinterest albeit some sexual interest being present(false negative). Behavioral outcomes of sexualoverperception would be pursuing disinterestedpotential mates, an activity that is time consum-ing and potentially risky in terms of social em-barrassment, rejection, and violent confronta-tions. Similarly, for sexual underperception itwould be missed sexual (and potential repro-ductive) opportunities. Because of the abun-dance of males willing to mate, the costs ofmissed opportunities for women is assumed tohave been low over deep time relative to thecosts of pursuing disinterested men (Buss,2012; Buss & Schmitt, 1993). Comparably, be-cause of the higher number of potential off-spring a man can leave behind, the costs ofmisses relative to pursuing disinterested womenhave presumably been higher for men. EMTsuggests that selection has favored designs that,though they may produce more errors overall,minimize the costlier error, reducing the overallcosts in the long run. This would produce abiased system that leads men, but not women, tooverperceive signals of sexual interest in poten-tial mates (Haselton & Nettle, 2006). With fewexceptions (Perilloux & Kurzban, 2015; Peril-loux, Muñoz-Reyes, Turiegano, Kurzban, &Pita, 2015), studies using various methods andsamples attest to this male sexual overpercep-tion bias (Bendixen, 2014; Galperin & Hasel-ton, 2012; Haselton, 2003; Koenig, Kirkpatrick,& Ketelaar, 2007; Perilloux, Easton, & Buss,2012). Disagreements have arisen, however, re-garding whether the nature of the sexual over-perception bias is a cognitive or a behavioralone (Galperin & Haselton, 2012; McKay &Efferson, 2010; Perilloux, 2014).

Signals of relationship commitment may alsobe subject overperception and underperception

biases, leading a person to infer that a partner iscommitted when they are not (false positive) orto infer that there is no commitment eventhough the partner is (false negative). Becauseof women’s substantially higher obligatory in-vestment in their offspring compared withmen’s (Trivers, 1972), the cost for ancestralwomen to assume commitment when there wasnone was particularly high in terms of loss ofrecourses allocated to the child and lost futuremating opportunities. Hence, EMT suggeststhat selection has favored designs that leadwomen, but not men, to err on the safe side andunderperceive signals of commitment (Haselton& Buss, 2000; Henningsen & Henningsen,2010).

Humans are one of few mammalian specieswhere both parents invest heavily in the off-spring, and where paternal investment isstrongly linked to increased offspring survival.Long-term pair-bonding, which is the predom-inant mating strategy in humans, is consideredan adaptation to recurring problems related tochild rearing (Neuberg, Kenrick, & Schaller,2010). Negative emotional (and physical) reac-tions to romantic breakups are generally strong.Evidently, grief intensity is equally strong inwomen and men, but the expression of griefappears to be sex differentiated (Morris &Reiber, 2011). One particular adaptive problemfor long-term relationships is mate retention.Betrayal represents a serious threat, and anyform of infidelity or unfaithfulness from eitherparty may instigate a breakup. In an ethno-graphic study of 160 cultures, Betzig (1989)found that out of 43 causes of breakups, aspouse’s infidelity was the single most fre-quently cited.

There is evidence that the degree of forgive-ness and likelihood of breakup depend upon thenature of the infidelity: emotional versus sexual.Using hypothetical forced choice scenarios,Shackelford, Buss, and Bennett (2002) foundthat relative to women, more men reported theywould breakup because of their partner’s sexualinfidelity than because of emotional infidelity,and more men than women found it more dif-ficult to forgive sexual infidelity than emotionalinfidelity. This was further sustained by Conferand Cloud (2011) who found that more men(68%) than women (47%) reported that theyactually discontinued their relationship follow-

323COUPLES’ FORGIVENESS AND EXPECTED FORGIVENESS

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anP

sych

olog

ical

Ass

ocia

tion

oron

eof

its

alli

edpu

blis

hers

.T

his

arti

cle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

ing their partner’s sexual infidelity, and byWade (2012) who found men were more likelyto expel a mate because of lack of sexual accesswhile women were more likely to expel a matebecause of lack of emotional access.

The unfaithful party may receive signals offorgiveness of the transgression, and breakupmay be prevented or delayed. However, inter-preting these signals is particularly challengingbecause there are two conceptual levels of for-giveness: (1) the inner, intrapsychic dimensioninvolving emotional state and cognitive and be-havioral accompaniments, and (2) the interper-sonal dimension involving the ongoing relation-ship within which forgiveness takes place(Baumeister, Exline, & Sommer, 1998). It isimportant to consider both dimensions, becausethere may a difference in what people commu-nicate that they have forgiven and how theyreally feel; silent forgiveness is characterized byinternal forgiveness without any behavioral ex-pression, hollow forgiveness is characterized byexpressed forgiveness without any internaltransformation (Baumeister et al., 1998).

EMT maintains that following relationshipmisbehavior, the reproductive costs for thetransgressor of failing to detect genuine nega-tive evaluation signals (false negative) arehigher than assuming negative evaluation whenthere is no signal of such (false positive; Hasel-ton & Nettle, 2006). Transgressions may or maynot be forgiven, but selection has favored de-signs that lead to the tendency not to believeone’s transgressions are forgiven. This tendencyis known as negative forgiveness bias (Friesen,Fletcher, & Overall, 2005; Haselton & Nettle,2006). The evolved function of this biased be-lief is to guide the organism toward reparativebehavior securing that the transgressions arefully mended. Because men and women havefaced similar recurrent problems regardingmate retention, this bias is expected in bothsexes. In a study of 39 university studentcouples Friesen et al. (2005) found supportfor the existence of a general negative for-giveness bias for transgressions that had oc-curred at some time in the relationship re-called by both parties. Both men and womenstrongly underestimated their partner’s for-giveness regardless of this being internal (in-trapsychic) or expressed (communicated).

The Current Study

In the current study, we examine relationshipthreat, likelihood of initiating a breakup, ex-pressed forgiveness, need for keeping distance,and revenge in a sample of couples respondingto imagined infidelity transgressions. In partic-ular, we want to examine negative forgivenessbias following imagined infidelity transgres-sion. Scenarios covered emotional as well assexual infidelity, and one’s own as well as one’spartner’s hypothetical transgressions. In a re-cent study on jealousy using hypothetical sce-narios, (Bendixen, Kennair, & Buss, 2015)found that sex differences in jealousy for con-tinuous measures and forced choice measureswere equally strong (d � .80 –.90) in a largesample of Norwegian students. This findingsupports previous findings in Scandinavia usingthe forced choice method (Bendixen, Kennair,Ringheim, et al., 2015; Kennair, Nordeide, An-dreassen, Strønen, & Pallesen, 2011; Wieder-man & Kendall, 1999). These large sex differ-ences were predicted by Buss, Larsen, Westen,and Semmelroth (1992), who suggested that incultures where fathers invest more, one mightexpect larger sex differences in jealousy. Nor-way has for several years been ranked amongthe top nations of the Global Gender Gap Re-port (World Economic Forum, 2016) with ex-plicit expectations of paternal involvement inchildren (Bendixen, 2014; Bendixen, Kennair,& Buss, 2015; Grøntvedt & Kennair, 2013).

Bendixen, Kennair, and Buss (2015) anal-yses of continuous measures suggest thatwhile men and women did not differ much intheir level of distress/jealousy to sexual infi-delity, relative to men, women found emo-tional infidelity far more distressing. As pre-dicted, within each sex women respondedwith more distress to emotional infidelity thanto sexual infidelity, and men responded withmore distress to sexual infidelity. This sex-differentiated pattern of responses was evi-dent for single and for partnered respondents.Studies of forgiveness largely reflect thesepatterns, where more distress results in lessforgiveness (Confer & Cloud, 2011; Shackel-ford et al., 2002) and increased likelihood ofexpelling a mate or breaking up (Shackelfordet al., 2002; Wade, 2012).

324 BENDIXEN, KENNAIR, AND GRØNTVEDT

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anP

sych

olog

ical

Ass

ocia

tion

oron

eof

its

alli

edpu

blis

hers

.T

his

arti

cle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

Hypotheses and Predictions

Hypothesis 1: In general, regardless of sexand type of infidelity, negative forgivenessbias is expected following one’s owntransgression (Haselton & Nettle, 2006;Friesen et al., 2005).

Hypothesis 1a: Relative to the transgres-sor’s own reports of how likely it is thattheir romantic partner would expresses(communicates) forgiveness (upper left,Figure 1), transgressors are less likely toreport that they believe that their partnerwould forgive their cheating (lower left,Figure 1).

Hypothesis 1b: Relative to their partners’reports of the likelihood of expressing for-giveness (upper right, Figure 1), transgres-sors are less likely to report that their part-ner would forgive their cheating (lowerleft, Figure 1).

Hypothesis 1_Moderation: For the previ-ous hypotheses, we expect participant sexand type of imagined infidelity transgres-sion to moderate the negative forgivenessbias. Relative to men, women are expectedto be more distressed— hence less forgiv-ing of—their partner’s emotional relativeto their partner’s sexual infidelity (Ben-dixen, Kennair, & Buss, 2015; Confer &Cloud, 2011; Shackelford et al., 2002;Wade, 2012). Because unfaithful respon-dents are likely to use their own reaction totheir partner’s unfaithfulness as an “an-chor” when imagining their partner’s reac-tion, we expect diminished negative for-giveness bias in (a) men imagining beingemotionally (compared with being sexu-ally) unfaithful, and in (b) women imagin-

ing being sexually (compared with beingemotionally) unfaithful. Furthermore, andmore explorative, any negative forgivenessbias may in part be because of a generaldisbelief of the other party’s beliefs fol-lowing your expressed forgiveness (lowerright, Figure 1). Still, we would expectnegative forgiveness bias even when ac-counting for this general disbelief.

Hypothesis 2: In general, relative towomen, men will find sexual infidelitymore distressing than emotional infidelity,and vice versa relative to men, women willfind emotional infidelity more distressingthan sexual infidelity aspect (Bendixen,Kennair, & Buss, 2015; Buss, 2013; Sa-garin et al., 2012). For specific indicatorsof distress, we predict the following:

Hypothesis 2a: Relative to women, menwill report more threat to the relationshipimagining their partner’s sexual infidelitythan imagining their partner’s emotionalinfidelity, and vice versa, relative to men,women will report more threat to the rela-tionship imagining their partner’s emo-tional infidelity than imagining their part-ner’s sexual infidelity (Bendixen, Kennair,& Buss, 2015; Buss, 2013; Confer &Cloud, 2011; Shackelford et al., 2002; Sy-mons, 1979).

Hypothesis 2b: Relative to women, menwill less likely express forgiveness be-cause of their partner’s sexual infidelitythan because of their partner’s emotionalinfidelity, and vice versa, relative to men,women will be less likely to express for-giveness because of their partner’s emo-tional infidelity than because of their part-ner’s sexual infidelity (Bendixen, Kennair,& Buss, 2015; Buss, 2013; Confer &Cloud, 2011; Symons, 1979).

Hypothesis 2c: Relative to women, menwill more likely keep their distance be-cause of their partner’s sexual infidelitythan because of their partner’s emotionalinfidelity, and vice versa, relative to men,women more likely keep their distance be-cause of their partner’s emotional infidelitythan because of their partner’s sexual infi-delity (Bendixen, Kennair, & Buss, 2015;

Figure 1. Variables included in testing negative forgive-ness bias.

325COUPLES’ FORGIVENESS AND EXPECTED FORGIVENESS

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anP

sych

olog

ical

Ass

ocia

tion

oron

eof

its

alli

edpu

blis

hers

.T

his

arti

cle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

Buss, 2013; Confer & Cloud, 2011; Sy-mons, 1979).

Hypothesis 2d: Relative to women, menwill more likely seek revenge because oftheir partner’s sexual infidelity than be-cause of their partner’s emotional infidel-ity, and vice versa, relative to men, womenwill be more likely seek revenge becauseof their partner’s emotional infidelity thanbecause of their partner’s sexual infidelity(Bendixen, Kennair, & Buss, 2015; Buss,2013; Confer & Cloud, 2011; Symons,1979).

Hypothesis 2e: Relative to women, menwill more likely break up because of theirpartner’s sexual infidelity than to theirpartner’s emotional infidelity, and viceversa, relative to men, women will morelikely break up because of their partner’semotional infidelity than because of theirpartner’s sexual infidelity (Bendixen, Ken-nair, & Buss, 2015; Confer & Cloud, 2011;Shackelford et al., 2002; Wade, 2012).

Method

Participants

A total of 92 heterosexual couples studying atthe Norwegian University of Science and Tech-nology partook in this study. Their age rangedfrom 19 –30 years (women: M � 22.0; SD �1.8; men: M � 22.9; SD � 2.2). Mean length ofcurrent relationships was reported to be 21months (ranging from 1 month to 9 years, SD �19 months) by both parties (r � .988).

Procedure

Participants were recruited by research assis-tants on the university campus. Flyers includinga short description of the study were handed outduring lecture breaks, and oral information wasgiven at lectures. To partake, participants had tobe in a relationship and bring their partner to adesignated lab. At arrival, they received writteninformation about the task, and were informedthat the questionnaires included several hypo-thetical scenarios concerning infidelity. Theyfilled out questionnaires in separate rooms. Af-ter the questionnaire was completed, put in asealed envelope and handed in to the research

assistants, the couples were rejoined and re-ceived two movie tickets in exchange for theirparticipation. The research assistants wereavailable for questions during and after com-pleting the questionnaire.

Measurements

Outcome variables. Based on our prior ap-plication of jealousy scenarios (Bendixen, Ken-nair, & Buss, 2015), we developed a total offour scenarios for one’s own and one’s partner’sinfidelity respectively. Two of the scenarios re-flected exclusively sexually unfaithful behavior,the other two reflected exclusively emotionallyunfaithful behavior (for a full description of thefour scenarios, see Appendix).

Following each of the scenarios describingthe respondent being unfaithful, participantsfirst rated the level of threat/severity of theincidents. Response alternatives ranged fromnot at all (1) to extremely (7). Each participantthen rated the likelihood that their partner wouldexpress (communicate) that they were forgiven;very unlikely (1) to very likely (7) and immedi-ately after, they rated the likelihood that theywould believe that they were forgiven; veryunlikely (1) to very likely (7). Finally, partici-pants rated the likelihood they themselves (ortheir partner) would be breaking up; very un-likely (1) to very likely (7). All questions weretranslations of items used by Friesen et al.(2005).

Following each of the scenarios describingtheir partner being unfaithful, each participantfirst rated the level of threat/severity of theincidents as described previously followed by arating of (a) the likelihood that they wouldexpress (communicate) forgiveness to theirpartner, and (b) the likelihood their partnerwould believe he or she was forgiven. Partici-pants then rated the likelihood that they them-selves or their partner would end the relation-ship. All items were translated items used byFriesen et al. (2005). For measuring internalmotivational aspects of forgiveness, we appliedfour items reflecting keeping distance and threeitems reflecting revenge from the modified andtranslated version of the Transgression-RelatedInterpersonal Motivations (TRIM) inventory(McCullough et al., 1998) as described by Fri-esen et al. (2005). Internal consistency (Cron-bach’s alpha) for keeping distance was excellent

326 BENDIXEN, KENNAIR, AND GRØNTVEDT

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anP

sych

olog

ical

Ass

ocia

tion

oron

eof

its

alli

edpu

blis

hers

.T

his

arti

cle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

for both sexual infidelity (women: � � .85;men: � � .90) and emotional infidelity (women:� � .89; men: � � .93). Alphas for the threerevenge items were acceptable for both sexualinfidelity (women: � � .66; men: � � .66) andemotional infidelity (women: � � .75; men:� � .64). Item scores were summed and aver-aged. High scores reflected more distance andhigher levels of revenge.1

Question order manipulation. Respondingto questions regarding one’s own or one’s part-ner’s (hypothetical) infidelity may provide a men-tal context or frame for subsequent infidelity judg-ments. This alludes to studies on the effect ofpeople reporting about themselves first, versusreporting about others first (Bless & Schwarz,2010; Kruger, 1999), but the specificity of thispotential contextual effect is not known. To con-trol for possible effects of responding to owninfidelity on judgments of partner’s infidelity, webalanced the question order. Half of the coupleswere randomly assigned to respond to partner’sinfidelity scenarios first, and the second half toone’s own infidelity scenarios first.

Ethics Statement

The study was carried out in line with theAmerican Psychological Association’s ethicalprinciples of psychologists and code of conduct.Prior to the data collection, the Norwegian DataProtection Official for Research (Personver-nombudet, NSD) was consulted. Any formalnotification of the research was deemed unnec-essary. All project assistants signed a confiden-tiality form.

Results

In testing Hypothesis 1a, we performed afour-way (2 � 2 � 2 � 2) Mixed Model (Pro-file) Analysis (Tabachnick & Fidell, 2014) withBias (transgressor’s likelihood of believing vs.transgressor’s reports of likelihood that partnerexpress) � Infidelity type (sexual vs. emo-tional) � Couple Sex (she vs. he) as withinsubject factors, and Question order (transgres-sor first vs. victim first) as between subjectfactor. See Table 1 for means and SDs forrelevant variables.

Relative to the transgressor’s reports of like-lihood that their cheated partner would expressforgiveness (i.e., what they say; M � 2.9),

transgressors had a strong negative bias towardnot believing in the expressed forgiveness (M �2.2), F(1, 90) � 79.49, p � .001, �p2 � 0.469.This bias was not moderated by Sex, F(1, 90) �2.17, by Infidelity type, F(1, 90) � 0.26, or bythe combined effect of Sex by Infidelity type,F(1, 90) � 1.24. Also, the overall bias was notaffected by question order, F(1, 90) � 0.13.

In testing Hypothesis 1b, we reran the anal-yses for Hypothesis 1a substituting only theBias factor (transgressor’s likelihood of believ-ing vs. cheated partner’s likelihood of express-ing forgiveness; Figure 1). Relative to the like-lihood that their cheated partner would expressforgiveness (M � 3.0), unfaithful respondentshad a strong bias toward not believing the ex-pressed forgiveness (M � 2.2), F(1, 90) �49.26, p � .001, �p2 � 0.354. This bias waspartly moderated by Sex, F(1, 90) � 3.21, p �.077 (somewhat stronger bias for women), andby Sex of transgressor � Infidelity type, F(1,90) � 5.28, p � .05. A closer inspection of thescores for each group suggest that this three-way interaction was because of cheated men’selevated likelihood of forgiving their partner’semotional relative to sexual infidelity in combi-nation with unfaithful men’s relatively strongerbelief in being forgiven for their emotionaltransgression. Post hoc analyses evinced weakand nonsignificant bias for emotionally (d �.31, p � .136) relative to sexually (d � .77)unfaithful men. The bias for sexually unfaithfulwomen was as strong as for men (d � .83). Thestrongest bias was found for emotionally un-faithful women (d � 1.14). In this analysis,question order moderated the overall bias, F(1,90) � 5.57, p � .05.2

1 The single item measuring cheated partner’s expressedforgiveness correlated substantially with the scales keepingdistance and revenge across scenarios and respondent sex(mean rs � �.47, ranging from r � �.31 to r � �.62). Theaverage correlation for keeping distance and revenge wasr � .43.

2 The bias was relatively stronger for participants whoresponded to victim scenarios following transgressor sce-narios (�p2 � 0.470) than for respondents responding tovictim scenarios before transgressor scenarios (�p2 � 0.211).A closer inspection of the mean scores suggest that thetransgressors’ beliefs were not affected by question order(M � 2.2 vs. M � 2.3), but the likelihood that cheatedpartners expressed forgiveness was lower if the coupleresponded as transgressors first (M � 2.7) rather than asvictims first (M � 3.3).

327COUPLES’ FORGIVENESS AND EXPECTED FORGIVENESS

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anP

sych

olog

ical

Ass

ocia

tion

oron

eof

its

alli

edpu

blis

hers

.T

his

arti

cle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

To examine whether or not the negative for-giveness bias reported above may in part bebecause of a general disbelief of the other par-ty’s beliefs following your expressed forgive-ness, we performed a five-way (2 � 2 � 2 �2 � 2) Mixed Model (Profile) Analysis(Tabachnick & Fidell, 2014) with Actor (trans-gressor vs. cheated partner) � Bias (believe vs.express) � Infidelity type (sexual vs. emo-tional) � Respondent Sex (she vs. he) as withinsubject factors, and Question order (transgres-sor first vs. victim first) as between subjectfactor. See Figure 1 for an illustration. Theessential finding from this complex model (Fig-ure 2) was that the transgressor believed theircheated partner’s expressed forgiveness less,compared with their estimates of how likelytheir partner would express forgiveness (i.e.,Transgressor’s bias). In comparison, thecheated party’s estimates of forgiving the trans-gressor did not differ much from their disbelief;their estimates of how likely it was that theirpartner believed in their forgiveness (Actor �Bias interaction effect), F(1, 90) � 18.74, p �.001, �p2 � 0.172. Hence, the negative forgive-ness bias for the transgressor was evident evenwhen accounting for general disbelief of thecheated partner. This pattern was not moderatedby type of infidelity, nor by question ordermanipulation, but the pattern tended to differsomewhat for women and men, F(1, 90) � 3.03,p � .085. Relative to men, the bias appeared to

be stronger for women when they responded astransgressors, and less when they responded ascheated partners (men reported more disbeliefthan women as cheated partner).

Table 1Response Means (and SDs) Across the Four Infidelity Scenarios

Variable

As transgressor As cheated partner

Sexual (SC 1) Emotional (SC 2) Sexual (SC 3) Emotional (SC 4)

Women Men Women Men Women Men Women Men

Threat/severity 6.4 (.8) 6.4 (.9) 6.1 (1.2) 5.8 (1.2) 6.3 (.9) 6.3 (1.1) 6.3 (1.0) 5.8 (1.2)Likelihood

You break up 3.7 (1.8) 3.6 (1.9) 4.4 (1.7) 3.8 (1.8) 5.5 (1.6) 5.3 (1.7) 5.2 (1.6) 4.7 (1.7)Partner express forgiveness 2.6 (1.4) 2.8 (1.5) 3.2 (1.7) 3.2 (1.7)You express forgiveness 2.9 (1.6) 2.7 (1.7) 2.9 (1.7) 3.5 (1.8)Believe forgiveness 1.9 (1.3) 2.1 (1.3) 2.3 (1.4) 2.6 (1.5)Partner believe forgiveness 2.8 (1.4) 2.4 (1.3) 2.8 (1.6) 3.0 (1.7)

TRIMDistance 5.1 (1.2) 4.9 (1.5) 5.2 (1.3) 4.5 (1.6)Revenge 3.1 (1.1) 3.2 (1.3) 2.9 (1.4) 2.8 (1.3)

Note. SC � scenario; TRIM � transgression-related interpersonal motivation. All means and SDs are based on a 7-pointscale. The forgiveness scores were reversed (high scores � more forgiveness).

Figure 2. Mean likelihood (95% confidence interval) ofexpressed forgiveness and beliefs in forgiveness (1 � veryunlikely, 7 � very likely). T � transgressor’s estimates;CP � Cheated partner’s estimates.

328 BENDIXEN, KENNAIR, AND GRØNTVEDT

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anP

sych

olog

ical

Ass

ocia

tion

oron

eof

its

alli

edpu

blis

hers

.T

his

arti

cle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

In testing Hypothesis 2a through 2e (Figure3) that, relative to women, men will find thesexual infidelity aspect more distressing thanthe emotional infidelity aspect, and vice versarelative to men, women will find the emotionalinfidelity aspect more distressing than the sex-ual infidelity aspect, we performed a five sepa-rate three-way (2 � 2 � 2; Profile) Analysiswith Infidelity type (sexual vs. emotional) �Couple sex (she vs. he) as within subject fac-tors, and Question order (transgressor first vs.victim first) as between subject factor. For Hy-pothesis 2a (Level of threat) we found that apartner’s sexual infidelity was more threateningthan a partner’s emotional infidelity, F(1, 90) �8.89, p � .01, �p2 � 0.090 and that infidelity wasgenerally more threatening for women than formen, F(1, 90) � 4.49, p � .05, �p2 � 0.048. Asevident from Figure 3, Panel 2a, the level ofthreat for sexual infidelity did not differ forwomen and men (M � 6.3 for both sexes), butmen (M � 5.8) found their partner’s emotionalinfidelity more threatening than did women(M � 6.3), producing a significant Sex � Infi-delity type interaction, F(1, 90) � 4.48, p � .05,�p

2 � 0.047.For Hypothesis 2b (expressed forgiveness)

we found that respondents imagining their part-ner’s unfaithfulness reported higher levels ofexpressed forgiveness for emotional comparedwith sexual infidelity, F(1, 90) � 11.31, p �.001, �p2 � 0.112. This effect was qualified by aSex � Infidelity type interaction, F(1, 90) �7.63, p � .01, �p2 � 0.078, suggesting thatrelative to women, men were more likely toforgive emotional infidelity than sexual infidel-ity.

For Hypothesis 2c (keeping distance), wefound no effect for infidelity type, F(1, 90) �2.21, p � 14, but men kept less distance thanwomen (M � 4.7), F(1, 90) � 8.63, p � .01,�p

2 � 0.088. This effect was qualified by aSex � Infidelity type interaction, F(1, 90) �6.58, p � .05, �p2 � 0.068, suggesting thatrelative to women, men kept their distance lessfor emotional infidelity than for sexual infidel-ity.

For Hypothesis 2d (revenge), we found nosex difference, F(1, 89) � 0.18, but higherlevels of revenge were reported for sexual com-pared with emotional infidelity, F(1, 89) � 13.91, p � .001, �p2 � 0.135. However, respondent

sex did not moderate the above type of infidelityeffect.

Finally, in testing Hypothesis 2e (likelihoodof breakup) we found that partner’s sexual in-fidelity was more likely to result in breakup thanpartner’s emotional infidelity, F(1, 90) � 6.80,p � .05, �p2 � 0.070, and that women weremarginally more likely to break up than men,F(1, 90) � 3.45, p � .067. However, the like-lihood of breakup for sexual over emotionalinfidelity was not significantly different forwomen and men (p � .147).

Discussion

Hypothesis 1 was strongly supported: Wefound a robust negative forgiveness bias follow-ing one’s own imagined infidelity for both maleand female transgressors. Relative to the likeli-hood of being forgiven, transgressors reportedthat they believed less that their partner wouldforgive their cheating This bias was evident foranalyses using two different criteria: the trans-gressor’s reports only (Hypothesis 1a, and foranalyses comparing the transgressor’s reportswith their partner’s external forgiveness, Hy-pothesis 1b). The moderation part of Hypothesis1 was supported for emotional, but not for sex-ual unfaithfulness. We found diminished nega-tive forgiveness bias for emotionally unfaithfulmen, but not for sexually unfaithful women.Emotionally unfaithful men evinced less bias inthe analyses of their partner’s expressed for-giveness. Relative to women, men not onlyseem to be more willing to forgive emotionalinfidelity by their partner, they also tend tobelieve more that their emotional infidelity willbe forgiven. It is not that they are naïve aboutemotional infidelity; men do understand this is atransgression and that it might negatively affecttheir relationship, but the sex difference in re-sponse to both one’s own and one’s partner’semotional infidelity is striking. This supportsthe evolutionary perspective on infidelity fromthe jealousy literature that women are moredistressed than men by emotional infidelity(Bendixen, Kennair, & Buss, 2015). The simi-larity of findings across the two tests of negativeforgiveness bias suggest that this bias can berelatively accurately estimated even withoutany response from the respondent’s partner. Fri-esen et al. (2005) considered their findings of aforgiveness bias consistent with Error Manage-

329COUPLES’ FORGIVENESS AND EXPECTED FORGIVENESS

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anP

sych

olog

ical

Ass

ocia

tion

oron

eof

its

alli

edpu

blis

hers

.T

his

arti

cle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

Figure 3. Mean scores (95% confidence interval) for women and men imagining theircurrent partner’s sexual or emotional infidelity. Scores ranged between 1 and 7, and scalemidpoint was 4 for all outcomes.

330 BENDIXEN, KENNAIR, AND GRØNTVEDT

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anP

sych

olog

ical

Ass

ocia

tion

oron

eof

its

alli

edpu

blis

hers

.T

his

arti

cle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

ment Theory preliminary until replicated. Thecurrent findings lend support to negativeforgiveness bias being a functional response torelationship uncertainty. Further evidence of atrue forgiveness bias is found when controllingfor cheated partner’s belief in transgressor be-lieving in being forgiven. This general disbeliefdid not account for the negative forgivenessbias, which remains strong. The overall func-tion of the negative forgiveness bias seems to beat least partly dependent upon subjective insightinto the seriousness of the transgression. Thereare sex-differentiated perceptions of transgres-sions. In the current study, men understand thatemotional infidelity is a problem, they just donot have insight into how great a problem theirpartner finds it to be. Perhaps in a more ecolog-ically valid setting, their partner’s distresswould be more clearly communicated, resultingin better grounds for assessing the need for aforgiveness bias also for emotional infidelity formen.

In support of Hypothesis 2a we found thatrelative to women, men reported more threatimagining a partner’s sexual infidelity com-pared with emotional infidelity. While level ofthreat to the relationship for sexual infidelity didnot differ for women and men, women reportedemotional infidelity to be more threatening thanmen did. In support of Hypothesis 2b and 2c,expressed and internal (keeping distance) for-giveness of partner’s infidelity mirrors findingsof sex differences in jealousy responses for con-tinuous measures (Bendixen, Kennair, & Buss,2015) and forced choice (Bendixen, Kennair, &Buss, 2015; Confer & Cloud, 2011; Shackelfordet al., 2002). As predicted, relative to women,men found it harder to express forgiveness andinternally forgive sexual infidelity comparedwith emotional infidelity. Bendixen, Kennair,and Buss (2015) suggested, in line with previ-ous predictions by Buss et al. (1992), that thismay be because of greater expectancy and vari-ability in father investment in more gender egal-itarian nations and cultures.

Parts d and e of Hypothesis 2 (revenge andbreaking up) were not supported. The responsesto a partner’s hypothetical sexual or emotionalinfidelity were not significantly different forwomen and men. It may be that any transgres-sion of this kind is largely unforgiveable; atleast as a deterrent. Our participating coupleswere aware of their partner simultaneously an-

swering the same questionnaire as they were.Maybe some of their responses reflect a hypo-thetical a priori deterrent effect to questions like“what did you answer to . . .” when meeting upafterward. Despite this, we found no evidenceof a ceiling effect for breakup or revenge. It ispossible that actual post hoc behavior would beless vengeful. Further, deterring threats ofbreakup a priori might work as retention tactics(Buss & Shackelford, 1997), but for emotionalinfidelity actually breaking up would be an in-effective tactic. Our results for breakup thusdovetail with the findings of Shackelford et al.(2002) who did not find a robust effect of sex.As such, breakup differs somewhat from threat,distress, jealousy, and forgiveness.

We did not anticipate that men should under-estimate the severity of emotional infidelity tosuch a degree. In Bendixen, Kennair, and Buss(2015), it seemed that the large sex differencesin jealousy responses were because of women’sincreased distress after emotional infidelity. Thecurrent finding may reflect men’s lack of aware-ness of how upset their female partner would befollowing men’s emotional infidelity. Alterna-tively, men were more self-centered in theirbeliefs when estimating their partner’s reactionto emotional infidelity. However, because for-giveness bias is the difference between trans-gressor’s and victim’s evaluation, any one orboth of these evaluations will influence thestrength of the bias. As such, it is worth notingthat for sexual infidelity the bias was similar formen and women, while men evaluated emo-tional infidelity as less severe.

Limitations, Implications, andFuture Research

There were generally low levels of expressedforgiveness and internal forgiveness in terms ofkeeping distance in our hypothetical infidelityscenarios. Transgressions involving sexual oremotional infidelity were generally not forgivenby our participants (mean scores between 2.6and 3.5, all well below the scale midpoints).That is perhaps the nature of infidelity; forgive-ness is rare. However, it is important to note thatthese are hypothetical cases of infidelity. It ispossible that one in real life would be moreforgiving, as shown by Friesen et al. (2005) forthe internal aspect of forgiveness. The re-sponses predicted by our participants may be

331COUPLES’ FORGIVENESS AND EXPECTED FORGIVENESS

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anP

sych

olog

ical

Ass

ocia

tion

oron

eof

its

alli

edpu

blis

hers

.T

his

arti

cle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

their moral and emotional ideal responses. Pos-sibly, respondents fail to consider various rela-tionship factors when imagining transgressions(McCullough & Hoyt, 2002). Also, the effect ofreligiosity, attributions, and negative emotionsare shown to be stronger in hypothetical for-giveness compared with actual forgiveness(Riek & Mania, 2012). Hence, findings fromstudies of forgiveness using hypothetical sce-narios may not always inform real life for-giveness. Future research needs to considerthe extent the current findings reflect actualforgiveness of sexual and emotional infidel-ity. Furthermore, future research might bene-fit from investigating differences in negativeforgiveness bias and sex differences based ontransgression descriptions (as in this study)versus emotional displays of distress.

Although we believe that our scenarios havecaptured the essence of sexual and emotional in-fidelity—the former describing a one-shot en-counter, future research would benefit from usingdescriptions of sexual infidelity scenarios that in-volve numerous sexual interactions to increaseequivalence to the emotional infidelity scenarios.Our study might also suffer from an a priori mateguarding, deterrent effect: Possible future infidel-ity might be met with clearer communication ofultimatums, where the severity of reaction anddrama is intensified. Despite answering the ques-tionnaires in separate rooms, the participantsmight have considered the hypothetical transgres-sions from a communicative perspective wherethey attempted to deter any future infidelity.

We note that for only one of the hypotheses(H1b) question order showed a significant effect;victims who answered as transgressors first ex-press less forgiveness resulting in a stronger for-giveness bias. This is maybe counterintuitive; onemight have thought that if anything, one was moreforgiving. Responding as victims first may reflectthoughts about what one ideally ought to do, whileresponding after one’s own transgression reducedthis benign effect. Some indirect support for this isfound in Kruger (1999) and Bless and Schwarz(2010). They suggest that when people reportabout themselves first (self-other) their responsesbecome egocentrically biased (increasing self-other differences), while reporting about othersfirst may attenuate self-other differences. Possibly,responding as transgressor is more similar to re-sponding as “self,” and responding as victim ismore similar to responding about “others,” and

that the reduced bias observed in our study may beattributed to more egocentric biased responses forthose who responded as transgressors first (theywere less likely to express forgiveness when re-sponding as victims). This might prove not to be arobust finding, but future research needs to con-sider the effects of whether answering questionsabout own transgressions before considering part-ner’s infidelity may affect results.

We did not ask women whether they wereusing hormonal contraception, which could influ-ence the response to infidelity. As Geary, DeSoto,Hoard, Sheldon, and Cooper (2001) suggested,women using hormonal contraceptives tend to bemore distressed by partners sexual infidelity thanwomen who did not use such contraceptives. Arecent student sample collected by the authorsshows that, in a comparative sample of femalestudents (n � 439) in relationships, 82% of matedwomen used hormonal contraceptives (Grøntvedt& Kennair, unpublished data). Given that the ma-jority of women the current sample are likelycontraceptive users, the effects reported are likelyto be slightly underestimated. Sex differencesfrom samples with naturally cyclic women areexpected to be larger. Also, a recent study byGrøntvedt, Grebe, Kennair, and Gangestad (2017)suggests that there are important differences withregards to the specific type of hormonal contra-ception and relative relationship investment thatmight led to other outcomes than suggested byGeary et al. (2001).

There is a practical and clinical implication ofthe current findings that might be relevant forcouple counselors to be aware of: The significantdifference in how the two sexes perceive thisspecific class of acts. Even though both men andwomen perceive both emotional and sexual infi-delity as relationship threats, they have very dif-ferent appreciations of the severity of especiallyemotional infidelity. This is true for both own andpartner’s transgressions. This may potentially be asource of misunderstanding, conflict and miscom-munication in couples, and maybe a topic thatcouple counselors need to address.

Conclusions

The current study, combines forgiveness bias(EMT) with theory on evolved sex differences injealousy. By testing the current predictions withcouples’ reports of both own and partners’ trans-gressions and forgiveness, the current study pro-

332 BENDIXEN, KENNAIR, AND GRØNTVEDT

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anP

sych

olog

ical

Ass

ocia

tion

oron

eof

its

alli

edpu

blis

hers

.T

his

arti

cle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

vides novel and deeper insights into how men andwomen perceive and react to the distress of emo-tional and sexual infidelity.

The general finding suggests that negative for-giveness bias following infidelity transgressions isrobust across multiple comparisons. Regardless ofwhether one compares the transgressor’s beliefswith his or her own estimates of likelihood ofbeing forgiven, or their partner’s expressed orinternal likelihood of forgiving them, the bias isstrong. However, when making comparisons withpartner’s scores, the bias appears to be less pro-nounced for emotionally unfaithful men. Forgive-ness bias appears to be an adaptive response pre-dicted by EMT with the function to maintainreparative behavior to secure the relationship afterrelationship threatening transgressions. It is hardto consider forgiveness bias as anything but acognitive bias. It is the biased belief that mayinfluence adaptive restorative behavior rather thanthe accurate belief (Haselton & Nettle, 2006).

The hypothetical acts of infidelity committed inthe current scenarios were in general not forgiven.Infidelity, whether emotional or sexual, is not eas-ily forgiven; it is probably one of the most severetransgressions one may commit toward one’s part-ner. What is most striking with our results is howmen do not quite understand how serious womenperceive and deem emotional infidelity to be;while men cannot be described as naïve about thisaspect of their relationship, they certainly are notas concerned with emotional infidelity as womenare.

References

Baumeister, R. F., Exline, J. J., & Sommer, K. L.(1998). The victim role, grudge theory, and twodimensions of forgiveness. In E. L. Worthington,Jr. (Ed.), Dimensions of forgiveness (pp. 79–104).Philadelphia, PA: Templeton.

Bendixen, M. (2014). Evidence of systematic bias insexual over- and underperception of naturally oc-curring events: A direct replication of Haselton(2003) in a more gender-equal culture. Evolution-ary Psychology, 12, 1004–1021. http://dx.doi.org/10.1177/147470491401200510

Bendixen, M., Kennair, L. E. O., & Buss, D. M.(2015). Jealousy: Evidence of strong sex differ-ences using both forced choice and continuousmeasure paradigms. Personality and IndividualDifferences, 86, 212–216. http://dx.doi.org/10.1016/j.paid.2015.05.035

Bendixen, M., Kennair, L. E. O., Ringheim, H. K.,Isaksen, L., Pedersen, L., Svangtun, S., & Hagen,K. (2015). In search of moderators of sex differ-ences in forced-choice jealousy responses: Effectsof 2d:4d digit ratio and relationship infidelity ex-periences. Nordic Psychology, 67, 272–284. http://dx.doi.org/10.1080/19012276.2015.1013975

Betzig, L. (1989). Causes of conjugal dissolution: Across-cultural study. Current Anthropology, 30,654– 676. http://dx.doi.org/10.1086/203798

Bless, H., & Schwarz, N. (2010). Mental construaland the emergence of assimilation and contrasteffects. Advances in Experimental Social Psychol-ogy, 42, 319–373. http://dx.doi.org/10.1016/S0065-2601(10)42006-7

Buss, D. M. (2012). Evolutionary psychology: Thenew science of the mind (4th ed.). Boston, MA:Allyn & Bacon.

Buss, D. M. (2013). Sexual jealousy. PsihologijskeTeme, 22, 155–182.

Buss, D. M., Larsen, R. J., Westen, D., & Semmel-roth, J. (1992). Sex differences in jealousy: Evo-lution, physiology, and psychology. PsychologicalScience, 3, 251–255. http://dx.doi.org/10.1111/j.1467-9280.1992.tb00038.x

Buss, D. M., & Schmitt, D. P. (1993). Sexual strat-egies theory: An evolutionary perspective on hu-man mating. Psychological Review, 100, 204–232.http://dx.doi.org/10.1037/0033-295X.100.2.204

Buss, D. M., & Shackelford, T. K. (1997). Fromvigilance to violence: Mate retention tactics inmarried couples. Journal of Personality and SocialPsychology, 72, 346–361. http://dx.doi.org/10.1037/0022-3514.72.2.346

Confer, J. C., & Cloud, M. D. (2011). Sex differencesin response to imagining a partner’s heterosexualor homosexual affair. Personality and IndividualDifferences, 50, 129–134. http://dx.doi.org/10.1016/j.paid.2010.09.007

Friesen, M. D., Fletcher, G. J. O., & Overall, N. C.(2005). A dyadic assessment of forgiveness inintimate relationships. Personal Relationships, 12,61–77. http://dx.doi.org/10.1111/j.1350-4126.2005.00102.x

Galperin, A., & Haselton, M. G. (2012). Error man-agement and the evolution of cognitive bias. InJ. P. Forgas, K. Fiedler, & C. Sedikides (Eds.),Social thinking and interpersonal behavior (pp.45– 64). New York, NY: Psychology Press.

Geary, D. C., DeSoto, M. C., Hoard, M. K., Sheldon,M. S., & Cooper, M. L. (2001). Estrogens andrelationship jealousy. Human Nature, 12, 299–320. http://dx.doi.org/10.1007/s12110-001-1001-2

Green, D. M., & Swets, J. A. (1966). Signal detectionand psychophysics. New York, NY: Wiley.

Grøntvedt, T. V., Grebe, N. M., Kennair, L. E. O., &Gangestad, S. W. (2017). Estrogenic and proges-togenic effects of hormonal contraceptives in rela-

333COUPLES’ FORGIVENESS AND EXPECTED FORGIVENESS

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anP

sych

olog

ical

Ass

ocia

tion

oron

eof

its

alli

edpu

blis

hers

.T

his

arti

cle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

tion to sexual behavior: Insights into extendedsexuality. Evolution and Human Behavior, 38,283–292. http://dx.doi.org/10.1016/j.evolhumbe-hav.2016.10.006

Grøntvedt, T. V., & Kennair, L. E. O. (2013). Agepreferences in a gender egalitarian society. Journalof Social, Evolutionary, & Cultural Psychology, 7,239–249. http://dx.doi.org/10.1037/h0099199

Haselton, M. G. (2003). The sexual overperceptionbias: Evidence of a systematic bias in men from asurvey of naturally occurring events. Journal ofResearch in Personality, 37, 34– 47. http://dx.doi.org/10.1016/S0092-6566(02)00529-9

Haselton, M. G., & Buss, D. M. (2000). Error man-agement theory: A new perspective on biases incross-sex mind reading. Journal of Personalityand Social Psychology, 78, 81–91. http://dx.doi.org/10.1037/0022-3514.78.1.81

Haselton, M. G., & Nettle, D. (2006). The paranoidoptimist: An integrative evolutionary model ofcognitive biases. Personality and Social Psychol-ogy Review, 10, 47– 66. http://dx.doi.org/10.1207/s15327957pspr1001_3

Henningsen, D. D., & Henningsen, M. L. M. (2010).Testing Error management theory: Exploring thecommitment skepticism bias and the sexual over-perception bias. Human Communication Research,36, 618– 634. http://dx.doi.org/10.1111/j.1468-2958.2010.01391.x

Kennair, L. E. O., Nordeide, J., Andreassen, S., Strønen,J., & Pallesen, S. (2011). Sex differences in jealousy:A study from Norway. Nordic Psychology, 61, 20–34. http://dx.doi.org/10.1027/1901-2276/a000025

Koenig, B. L., Kirkpatrick, L. A., & Ketelaar, T.(2007). Misperception of sexual and romantic in-terests in opposite-sex friendships: Four hypothe-ses. Personal Relationships, 14, 411– 429. http://dx.doi.org/10.1111/j.1475-6811.2007.00163.x

Kruger, J. (1999). Lake wobegon be gone! The “be-low-average effect” and the egocentric nature ofcomparative ability judgments. Journal of Person-ality and Social Psychology, 77, 221–232. http://dx.doi.org/10.1037/%2f0022-3514.77.2.221

McCullough, M. E., & Hoyt, W. T. (2002). Trans-gression-related motivational dispositions: Person-ality substrates of forgiveness and their links to thebig five. Personality and Social Psychology Bulle-tin, 28, 1556–1573. http://dx.doi.org/10.1177/014616702237583

McCullough, M. E., Rachal, K. C., Sandage, S. J.,Worthington, E. L., Jr., Brown, S. W., & Hight,T. L. (1998). Interpersonal forgiving in close rela-tionships: II. Theoretical elaboration and measure-ment. Journal of Personality and Social Psychol-ogy, 75, 1586–1603. http://dx.doi.org/10.1037/0022-3514.75.6.1586

McKay, R. T., & Efferson, C. (2010). The subtletiesof error management. Evolution and Human Be-

havior, 31, 309–319. http://dx.doi.org/10.1016/j.evolhumbehav.2010.04.005

Morris, C., & Reiber, C. (2011). Frequency, intensityand expression of post- relationship grief. TheJournal of the Evolutionary Studies Consortium, 3,1–11.

Neuberg, S. L., Kenrick, D. T., & Schaller, M.(2010). Evolutionary social psychology. In S. T.Fiske, D. T. Gilbert, & G. Lindzey (Eds.), Hand-book of social psychology (5th ed., Vol. 2, pp.761–796). Hoboken, NJ: Wiley. http://dx.doi.org/10.1002/9780470561119.socpsy002021

Perilloux, C. (2014). (mis)reading the signs: Men’sperception of women’s sexual interest. In V. A.Weekes-Shackelford & T. K. Shackelford (Eds.),Evolutionary perspectives on human sexual psy-chology and behavior (pp. 119–133). New York,NY: Springer New York. http://dx.doi.org/10.1007/978-1-4939-0314-6_6

Perilloux, C., Easton, J. A., & Buss, D. M. (2012). Themisperception of sexual interest. Psychological Sci-ence, 23, 146–151. http://dx.doi.org/10.1177/0956797611424162

Perilloux, C., & Kurzban, R. (2015). Do men overper-ceive women’s sexual interest? Psychological Sci-ence, 26, 70–77. http://dx.doi.org/10.1177/0956797614555727

Perilloux, C., Muñoz-Reyes, J. A., Turiegano, E., Kurz-ban, R., & Pita, M. (2015). Do (non-American) menoverestimate women’s sexual intentions? Evolution-ary Psychological Science, 1, 150–154. http://dx.doi.org/10.1007/s40806-015-0017-5

Riek, B. M., & Mania, E. W. (2012). The antecedentsand consequences of interpersonal forgiveness: Ameta-analytic review. Personal Relationships, 19,304–325. http://dx.doi.org/10.1111/j.1475-6811.2011.01363.x

Sagarin, B. J., Martin, A. L., Coutinho, S. A., Edlund,J. E., Patel, L., Skowronski, J. J., & Zengel, B.(2012). Sex differences in jealousy: A meta-analytic examination. Evolution and Human Be-havior, 33, 595– 614. http://dx.doi.org/10.1016/j.evolhumbehav.2012.02.006

Shackelford, T. K., Buss, D. M., & Bennett, K.(2002). Forgiveness or breakup: Sex differences inresponses to a partner’s infidelity. Cognition andEmotion, 16, 299–307. http://dx.doi.org/10.1080/02699930143000202

Swets, J. A., Dawes, R. M., & Monahan, J. (2000).Psychological science can improve diagnostic deci-sions. Psychological Science in the Public Interest, 1,1–26. http://dx.doi.org/10.1111/1529-1006.001

Symons, D. (1979). The evolution of human sexual-ity. New York, NY: Oxford University Press.

Tabachnick, B. G., & Fidell, L. S. (2014). Usingmultivariate statistics (6th ed.). Essex, UnitedKingdom: Pearson.

Trivers, R. (1972). Parental investment and sexual

334 BENDIXEN, KENNAIR, AND GRØNTVEDT

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anP

sych

olog

ical

Ass

ocia

tion

oron

eof

its

alli

edpu

blis

hers

.T

his

arti

cle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

selection. In B. Campbell (Ed.), Sexual selectionand the descent of man, 1871–1971 (pp. 136–179).Chicago, IL: Aldine.

Wade, T. J. (2012). Mate expulsion and sexual con-flict. In T. K. Shackelford & A. Goetz (Eds.), TheOxford handbook of sexual conflict in humans (pp.315–327). New York, NY: Oxford UniversityPress. http://dx.doi.org/10.1093/oxfordhb/9780195396706.013.0020

Wiederman, M. W., & Kendall, E. (1999). Evolution,sex, and jealousy: Investigation with a samplefrom Sweden. Evolution and Human Behavior, 20,121–128. http://dx.doi.org/10.1016/S1090-5138(98)00046-4

World Economic Forum. (2016). The global gendergap report 2016. Retrieved from http://www3.weforum.org/docs/GGGR16/WEF_Global_Gender_Gap_Report_2016.pdf

Appendix

The Four Infidelity Scenarios

Scenario 1. Your Sexual Infidelity

Imagine that you met someone at a party.You felt a strong attraction to this person andyou danced intimately and flirted throughout theevening. Your partner was not present, butlearnt through mutual friends a few weeks laterthat you slept with this person that night. Yourpartner gets very upset and confronts you withthis. After being pressured and cornered youadmit that you had sex that night, but that youwere not in love. You show remorse, apologize,and promise that it will never happen again.

Scenario 2. Your Emotional Infidelity

Imagine that you met someone at a party.You felt a strong attraction to this person andyou danced intimately and flirted throughout theevening. Your partner was not present, butlearnt through mutual friends a few weeks laterthat you have met this person several timessince the party, and that it looks like you havefallen in love. Your partner gets very upset andconfronts you with this. After being pressuredand cornered you admit that you have met thisperson secretly, but that you have not had sex.You show remorse, apologize, and promise tobreak all contact.

Scenario 3. Your Partner’s Sexual Infidelity

Imagine that your partner met someone at aparty. S/he felt a strong attraction to this person

and they danced intimately and flirted through-out the evening. You were not present, butlearnt through mutual friends a few weeks laterthat your partner slept with this person thatnight. This makes you very upset and you con-front your partner with this. After being pres-sured and cornered your partner admits that s/hehad sex that night, but that s/he were not in love.Your partner shows remorse, apologizes, andpromises that it will never happen again.

Scenario 4. Your Partner’sEmotional Infidelity

Imagine that your partner met someone at aparty. S/he felt a strong attraction to this personand they danced intimately and flirted through-out the evening. You were not present, butlearnt through mutual friends a few weeks laterthat your partner has met this person severaltimes since the party, and that it looks like s/hehas fallen in love. This makes you very upsetand you confront your partner with this. Afterbeing pressured and cornered your partner ad-mits that s/he has met this person secretly, butthat they have not had sex. Your partner showsremorse, apologizes, and promises to break allcontact.

Received June 12, 2017Revision received September 1, 2017

Accepted September 2, 2017 �

335COUPLES’ FORGIVENESS AND EXPECTED FORGIVENESS

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anP

sych

olog

ical

Ass

ocia

tion

oron

eof

its

alli

edpu

blis

hers

.T

his

arti

cle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.