Recovery From Infidelity: Differentiation of Self, Trauma, Forgiveness, and Posttraumatic Growth Among Couples in Continuing Relationships

Abstract

Clinicians and researchers report that individuals can heal from the trauma of infidelity and that forgiveness and even personal growth are possible after infidelity takes place. However, research has not explored specific variables that should relate to these outcomes. We defined personal growth as posttraumatic growth (PTG) and examined the relationships of differentiation of self from family of origin, trauma, forgiveness, and PTG in a sample of individuals remaining in a relationship in which infidelity had occurred. Results showed that differentiation was positively related to forgiveness levels and also moderated the relationship between trauma and forgiveness. The only significant predictor of PTG, however, was forgiveness.

Full text

Recovery From Infidelity: Differentiation of Self, Trauma,
Forgiveness, and Posttraumatic Growth Among Couples in
Continuing Relationships
Ashley Heintzelman, Nancy L. Murdock, Romana C. Krycak, and Larissa Seay
University of Missouri-Kansas City
Clinicians and researchers report that individuals can heal from the trauma of infidelity
and that forgiveness and even personal growth are possible after infidelity takes place.
However, research has not explored specific variables that should relate to these
outcomes. We defined personal growth as posttraumatic growth (PTG) and examined
the relationships of differentiation of self from family of origin, trauma, forgiveness,
and PTG in a sample of individuals remaining in a relationship in which infidelity had
occurred. Results showed that differentiation was positively related to forgiveness
levels and also moderated the relationship between trauma and forgiveness. The only
significant predictor of PTG, however, was forgiveness.
Keywords: infidelity, differentiation of self, forgiveness, posttraumatic growth
Infidelity remains one of the most difficult
issues faced by couples and for the counseling
professionals who work with them. Reviews
consistently document that somewhere between
22 and 25% of men and 11 and 15% of women
are willing to report having sex with someone
other than their spouses while married (Allen et
al., 2005;Mark, Janssen, & Milhausen, 2011).
Additionally, in a study of divorced men and
women, 40% of men and 44% of women re-
ported having more than one extramarital sex-
ual contact during their marriages (Janus &
Janus, 1993). The negative consequences of in-
fidelity are easily and readily identified and
include loss of trust; damaged self-esteem; dis-
ruption to other relationships such as those with
children, friends, or parents; financial conse-
quences; and emotional problems (Charny &
Parnass, 1995;Schneider, Irons & Corley,
1999). Furthermore, infidelity is the most fre-
quently cited cause of divorce reported by cou-
ples (Schneider, Irons & Corley, 1999;Whis-
man, Dixon, & Johnston, 1997;Winek &
Craven, 2003). On the brighter side, Atkins,
Marín, Lo, Klann, and Hahlweg (2010) found
that couples who sought couples therapy after
infidelity experienced the same levels of good
outcomes as did couples who entered counsel-
ing for other reasons.
Despite the well-documented negative effects
of infidelity, researchers and clinicians report
that committed partnerships can survive the
trauma of infidelity, and that further, personal
growth in the wake of infidelity is possible
(Abrahamson, Hussain, Khan, & Schofield,
2012;Gordon & Baucom, 1998). Therefore, the
purpose of this study is to describe a sample of
individuals who remained in a relationship
where infidelity had occurred and to conduct a
preliminary exploration of factors related to re-
covery from infidelity. We defined recovery
from infidelity as forgiveness (Gordon & Bau-
com, 1998) and posttraumatic growth (PTG;
Tedeschi & Calhoun, 2006). Because there are
few empirical studies of couples who stay to-
gether, we were interested in variables that
should theoretically be related to relationship
maintenance and PTG, including levels of
trauma experienced and individuals’ response
tendencies to relationship stress, operational-
ized as differentiation of self from the family of
origin (Kerr & Bowen, 1988). For the purposes
of this study, we focused on individuals who
Ashley Heintzelman, Nancy L. Murdock, Romana C.
Krycak, and Larissa Seay, Division of Counseling and Ed-
ucational Psychology, University of Missouri-Kansas City.
Correspondence concerning this article should be ad-
dressed to Nancy L. Murdock, Counseling and Educational
Psychology, University of Missouri-Kansas City, 5100
Rockhill Road, Kansas City, MO 64110. E-mail:
murdockn@umkc.edu
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Couple and Family Psychology: Research and Practice © 2014 American Psychological Association
2014, Vol. 3, No. 1, 13–29 2160-4096/14/$12.00 DOI: 10.1037/cfp0000016
13

reported that their partners had engaged in sex-
ual intercourse with someone else during a com-
mitted relationship. To better study variables
related to recovery, we required that individuals
still be in the relationship where infidelity oc-
curred and that the infidelity had occurred ⬎6
months before participation in our study.
Healing From Infidelity
Although writers argue that infidelity can of-
fer an “opportunity for growth through insight
and personal and couple understanding” (Bals-
wick & Balswick, 1999, p. 423), it seems that
empirical investigation of this phenomenon is
scarce. A few studies suggest that a small pro-
portion of couples among whom infidelity has
occurred stay together and report improvement
in their relationships and increases in personal
insight and understanding. For example, Charny
and Parnass (1995) asked a sample of therapists
to report on an incident of infidelity with which
they were very familiar. They found that 15% of
the therapists reported that the relationships in
question improved after the infidelity, whereas
the remaining therapists characterized the rela-
tionship as remaining the same or deteriorating.
In a study examining infidelity in the context of
committed heterosexual relationships, Hansen
(1987) asked a sample of 215 participants about
the impact of their own and their partner’s ex-
tradyadic relations on their current committed
relationships. Hansen found that 19.8% of par-
ticipants reported that their own extradyadic
relations improved the quality of their dating
relationship a “great deal,” and 31.3% reported
that the affair improved the dating relationship
“somewhat.” A small percentage of participants
in the study also reported that a partner’s affair
improved the quality of their relationship (6.5%
for a “great deal” and 13.0% for “somewhat”;
Hansen, 1987). Unfortunately, Charny and Par-
nass’ and Hansen’s studies used survey methods
that simply asked participants to rate improve-
ment with no further definition given to partic-
ipants. We located two qualitative studies of the
process individuals experienced after the disclo-
sure of an affair. Olson, Russell, Higgins-
Kessler, and Miller (2002) interviewed 13 indi-
viduals (11 women, 2 men), 11 of whom were
still in the relationship in which the infidelity
had occurred, and reported that some of their
participants described positive outcomes in-
cluding “developing a closer marital relation-
ship, becoming more assertive, realizing the
importance of good marital communication,
placing higher value on the family, and taking
better care of oneself” (p. 430). Abrahamson et
al. (2012) studied the process of couples who
stayed together at least 2 years after an affair
using a narrative qualitative approach. They in-
terviewed seven individuals (four men and three
women), two of whom were the noninvolved
partner. Attending couples counseling, address-
ing intrusive and negative memories, and learn-
ing from others who had weathered infidelity
were important themes that emerged from Abra-
hamson et al.’s data. Both Olson et al. and
Abrahamson et al. found that forgiveness and
understanding the meaning of the affair were
important themes. In sum, it appears that some
evidence exists that relationships can survive
infidelity and that some individuals report ex-
periencing positive outcomes after affairs.
However, in this small amount of research, the
definitions of positive outcomes are fuzzy and
possible predictors of these outcomes were not
explored, although there seems to be agreement
that forgiveness is essential to healing after in-
fidelity (Gordon, Baucom & Snyder, 2005). In
the following sections, we will discuss forgive-
ness as an important process involved in healing
along with variables that might be related to it.
Forgiveness
In the infidelity literature, the most frequently
cited treatment model is the trauma-based
model of forgiveness (Baucom, Gordon, Sny-
der, Atkins, & Christensen, 2006;Gordon &
Baucom, 1998;Gordon et al., 2005). In this
model, forgiveness is an ongoing process that
takes time, rather than a distinct event. Gordon
and Baucom (1998) presented a three-stage for-
giveness model that conceptualized recovery
from an affair as essentially the same as the
process recovery from any interpersonal trau-
ma. The stages in this model are I: dealing with
the impact, II: search for meaning, and III:
recovery or moving forward. Individuals who
are in Stage I report the least amount of forgive-
ness and individuals in Stage III report the high-
est levels of forgiveness. Gordon, Baucom, and
Snyder (2004) examined the model using a rep-
licated case study of six couples who entered
and completed treatment that was designed to
14 HEINTZELMAN, MURDOCK, KRYCAK, AND SEAY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

help couples to recover from an affair. After
treatment, the couples were found to be less
distressed, reporting fewer posttraumatic stress
symptoms, less depression, less marital distress,
and greater forgiveness regarding the affair.
Gordon and Baucom (2003) found that couples
who report forgiveness after a serious transgres-
sion has taken place show more investment in
their marriages, greater psychological close-
ness, more equal balance of power in their mar-
riages, and higher levels of marital adjustment
compared with couples who have not yet
achieved forgiveness. Other research has dem-
onstrated that forgiveness is related to both re-
lationship and general life satisfaction (Bur-
chard, Yarhouse, Kilian, & Worthington, 2003;
Ripley & Worthington, 2002).
Forgiveness and Trauma
Gordon et al. (2005) likened the forgiveness
process to the recovery from a traumatic event
and discussed various factors that are likely to
influence and complicate the recovery process.
In the current study, we measured trauma in a
generic way, but we were also interested in
several factors thought to influence the level of
trauma one experiences as a result of an event
like an affair. For example, forgiveness is
clearly a process, not an end state (Gordon &
Baucom, 1999;Rusbult, Kumashiro, Coolsen,
& Kirchner, 2004). Hence, time since the infi-
delity appears to be an important component of
the recovery process.
The level of one’s commitment to a relation-
ship may also relate to the level of trauma
experienced as a result of infidelity. Stress that
occurs within relational roles that are particu-
larly significant to an individual’s sense of self
is more likely to have a harmful impact on
psychological health than stress that takes place
in roles that the individual perceives as less vital
(Marcussen, Ritter, & Safron, 2004). Events
that upset salient identities or that disrupt iden-
tities to which individuals are highly committed
will have more destructive effects on psycho-
logical health compared with stressors that dis-
rupt identities that are not as important or to
which individuals are not as committed (Thoits,
1991,1992). Disruption of a relationship by
infidelity may therefore result in higher levels of
trauma for highly committed individuals com-
pared to individuals who are less committed to
their relationships, so we expected that higher
levels of commitment would be related to
higher levels of trauma and less forgiveness.
Differentiation of Self
The challenges of each stage of forgiveness
are often discussed, but variables and processes
that promote forgiveness seem to have received
less attention. One factor that would logically
relate to relationship outcomes such as forgive-
ness is how well individuals generally function
in relationships. Differentiation of self from the
family of origin, a construct originating in Bo-
wen family system theory, is a relationship vari-
able that may provide an understanding of in-
terpersonal functioning (Bowen, 1978;Kerr &
Bowen, 1988). Differentiation of self refers to
the ability to experience both intimacy and au-
tonomy within a relationship. Well-differenti-
ated individuals are able to maintain a clearly
defined sense of self and engage in meaningful
intimacy while allowing others the space for
their own positions, whereas individuals with
lower levels of differentiation tend to fuse in
their interpersonal relationships, reactively dis-
tance themselves, or emotionally cut off, partic-
ularly when the relationship is stressed (Kerr &
Bowen, 1988). Furthermore, well-differentiated
individuals are thought to be more resistant to
the negative effects of stress, to function better
in stressful situations, and to have more satis-
fying relationships compared to less-differenti-
ated individuals (Bowen, 1978;Kerr & Bowen,
1988). Differentiation is also thought to repre-
sent the extent to which individuals can be
objective and think clearly, even under stress.
These inter- and intrapersonal dynamics led
Balswick and Balswick (1999) to suggest that
individuals who display low levels of differen-
tiation of self are (a) more likely to engage in
extramarital affairs and (b) have a more difficult
time recovering from them compared with in-
dividuals to higher levels of differentiation.
Research has documented a positive relation-
ship between differentiation of self and relation-
ship satisfaction (Skowron & Friedlander,
1998). Skowron (2000) found that marital sat-
isfaction was positively associated with various
dimensions of differentiation identified by Bo-
wen; higher levels of satisfaction were corre-
lated with low levels of emotional reactivity
(ER), emotional cutoff (EC), and fusion, along
15RECOVERY FROM INFIDELITY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

with higher levels of ability to take I-positions
(IPs) in relationships. Conversely, couples in
which individuals demonstrated lower levels of
differentiation (high ER, cutoff, and fusion; low
IP) reported greater marital distress. Gubbins,
Perosa, and Bartle-Haring (2010) found similar
relationships between husbands’ and wives’ dif-
ferentiation of self and marital satisfaction, and
also found that differentiation was negatively
related to the tendency to become emotionally
flooded during arguments for both partners.
Gordon and Baucom (2003) postulated that
cognitive processing of the infidelity is critical
to the process of recovering from it. We rea-
soned that the ability to think through situations
objectively along with the ability to stay en-
gaged in relationship despite stress (without re-
acting with extreme emotion or psychologically
cutting off), both aspects of differentiation of
self, should promote the cognitive work needed
for forgiveness. Although limited research has
addressed this aspect of differentiation, Gubbins
et al.’s (2010) finding regarding differentiation
and flooding would seem to support such a
conjecture. Further, a few studies have sup-
ported a link between differentiation and for-
giveness, although not in the context of infidel-
ity. Sandage and Jankowski (2011) tested the
relationships between differentiation of self,
dispositional forgiveness, and psychological
well-being, finding that differentiation of self
mediated the relationship between forgiveness
and psychological functioning. Holeman, Dean,
DeShea, and Duba (2011) demonstrated that
differentiation was related to specific aspects of
forgiveness, namely, inhibition of harmful in-
tention and reduction of negative emotion.
Trauma, Differentiation, and Forgiveness
Although research has documented both me-
diating and moderating roles of differentiation
in the stress–symptom relationship, it is appar-
ent that differentiation is related to how indi-
viduals respond to stressful or traumatic situa-
tions (Krycak, Murdock & Marszalek, 2012;
Murdock & Gore, 2004;Murray, Daniels, &
Murray, 2006;Skowron, Wester, & Azen,
2004). Following Bowen’s assertion that indi-
viduals who are higher in differentiation of self
would react more adaptively in the face of
stress, we expected that level of trauma experi-
enced as a result of infidelity would be related to
forgiveness, and that this relationship would be
moderated by differentiation of self. A moder-
ation rather than a mediation model was tested
because previous research has supported mod-
eration when a global measure of differentiation
is used (Murdock & Gore, 2004;Murray et al.,
2006).
PTG as an Outcome of the
Recovery Process
The possibility of growth arising from the
struggle with suffering and crisis is a theme that
has existed over history (Tedeschi & Calhoun,
2006). Data indicate that for many individuals,
the encounter with very negative events and
trauma can produce positive psychological
change (Tedeschi & Calhoun, 2006), and many
studies have shown that after the experience of
traumatic events, most individuals report posi-
tive life changes (Linley & Joseph, 2004;Te-
deschi & Calhoun, 1995). Given that some re-
search has suggested that positive outcomes can
be observed after infidelity (Charny & Parnass,
1995;Gordon et al., 2004;Hansen, 1987;Olson
et al., 2002), the possibility of PTG is raised as
a way to conceptualize these outcomes.
The phenomenon of PTG is understood to be
a result of intrapersonal struggle to find benefit
and meaning in life after a traumatic experience
(Tedeschi & Calhoun, 2004). Five factors of
PTG have been identified: personal strength,
new possibilities, relating to others, apprecia-
tion for life, and spiritual change (Tedeschi &
Calhoun, 1996). For example, as a result of
experiencing loss and tragedy, many individuals
have reported feeling a greater connection to
other people in general, mainly a greater sense
of compassion for others who suffer (Tedeschi
& Calhoun, 2006). After dealing with traumatic
events, individuals also describe feeling a
greater sense of closeness, intimacy, and free-
dom to be oneself. Furthermore, individuals ex-
perience changed life philosophies and report a
changed sense of what is most important—what
was once trivial is now important and vice ver-
sa. Individuals describe a different sense of life
priorities and an increased appreciation for what
one actually has (Tedeschi & Calhoun, 2006).
The processes and outcomes involved in the
PTG seem to resemble those important in for-
giveness in Gordon and Baucom’s (2003)
model of recovery. Finding meaning in a trau-
16 HEINTZELMAN, MURDOCK, KRYCAK, AND SEAY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

matic event is critical to both models, along
with the evaluation of self and relationships. In
this research, we were interested to see if indi-
viduals who experienced infidelity in their rela-
tionships and stayed in them experienced PTG,
and if they did, what variables might predict it.
Rationale
The goal of our study was to explore vari-
ables that might be related to the recovery from
infidelity among a sample of individuals who
were still involved in the relationship in which
the infidelity occurred. Because little is known
about what contributes to personal growth in the
aftermath of infidelity, we explored the relations
among a number of variables suggested by ex-
isting theory and research on relationship func-
tioning and forgiveness. We hypothesized that
forgiveness, along with specific relationship
variables that might be associated with levels of
trauma (e.g., time since infidelity, commitment,
relationship satisfaction), would be related to
the degree to which individuals experienced
PTG after infidelity. We anticipated that indi-
viduals with higher levels of differentiation of
self are more able to react to stressful situations
without emotionally cutting off or reacting with
high levels of emotionality compared to those
who have lower levels of differentiation, so we
predicted that trauma, differentiation of self,
and the interaction of trauma and differentiation
of self would predict levels of forgiveness. We
also added the interaction of differentiation of
self and trauma to the equation predicting PTG,
although Kerr and Bowen (1988) are largely
silent on predictors of psychological growth.
Method
Participants were 587 individuals recruited
via six main Web sites with online support
forums specifically designed for individuals re-
covering from infidelity. The solicitation script
was posted to these online discussion forums
once a week for 6 months. The questionnaire
was filled out by participants anonymously and
no identifying information was recorded. In-
cluded with the measures was a statement dis-
cussing the volunteer nature of the study and the
potential risks and benefits of participation.
Participants were informed that they were
eligible for the study if they were aged ⬎18
years and currently in a committed relationship
with a partner who had sexual intercourse with
someone other than them during the course of
the relationship. Because we reasoned that for-
giveness and PTG take time to emerge, we
required that the infidelity took place at least 6
months before the time of participation in the
study. Data collection ended once 1018 individ-
uals responded to the survey. In all, 338 respon-
dents did not complete the measures, and an
additional 63 respondents did not meet inclu-
sion criteria of the infidelity occurring 6 months
prior.
The pattern of missing data for the remaining
cases was examined and found to be missing at
random, and very small (⬍5%). Expectation-
maximization imputation was used to estimate
missing values (Tabachnick & Fidell, 2008).
Chi-square tests were used to see if there were
significant differences between individuals who
completed the survey and those who dropped
out (and were thus excluded from the analyses).
Significant differences were found between
the individuals who completed the survey and
those who did not on age, ethnicity, and sexual
orientation, but no differences were found be-
tween the two groups on gender. An indepen-
dent-samples ttest was conducted to test for
differences between the two groups on age.
Results indicated that participants who com-
pleted the survey (M⫽46.15, standard devia-
tion [SD]⫽9.03) were significantly older than
those who did not complete the survey (M⫽
43.33, SD ⫽10.93; t(833) ⫽3.70, p⬍.001).
Additionally, there were significant differences
in the sexual orientation of participants who
completed the survey and those who did not,
with significantly more heterosexual partici-
pants in the group who completed the survey
than in the group that dropped out, ␹
2
(2) ⫽
6.00, p⬍.05.
The final sample of 587 consisted of 86%
women, and 12.6% men (eight individuals did
not respond); average age was 46.15 (SD ⫽
9.03), with a range of 21–74. Race/ethnicity
responses showed the following percentages:
86.9% Caucasian, 3.2% African American,
1.5% Asian, 3.2% Hispanic, .3% Native Amer-
ican, and 3.6% self-classified as “other” (five
individuals did not respond). Of the 587 partic-
ipants, 1.0% identified as bisexual, 96.4% het-
erosexual, and 1.4% homosexual. In all, 93.5%
of the respondents reported being married and
17RECOVERY FROM INFIDELITY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

6.5% were in a committed dating relationship
when the affair took place. Current relationship
status of the sample consisted of 95.1% married
and 4.9% in a committed dating relationship.
These discrepancies appear to result from sev-
eral participants changing their status from
committed relationship to married subsequent
to the affair. On average, the infidelity had taken
place 3.9 years prior, but the range for this
variable was enormous, from 6 months to 38
years (SD ⫽4.98).
1
Measures
The survey package included demographic
items assessing the context of the affair (e.g.,
time since the affair, relationship status when
affair took place) and measures of trauma, re-
lationship commitment, posttraumatic growth,
differentiation of self, current relationship sat-
isfaction, and stage of forgiveness. The demo-
graphic survey asked about the participant’s
age, racial/ethnic background, sexual orienta-
tion, gender, relationship status when the infi-
delity took place, current relationship status,
and the amount of time since the infidelity took
place.
Trauma
Because trauma was conceptualized to be
analogous to stress in the current study, the
Impact of Event Scale (IES; Horowitz, Wilner,
& Alvarez, 1979) was used to assess the impact
of the infidelity. The IES assesses responses to
potentially traumatic events and has been used
in both clinical and nonclinical samples (Briere
& Elliott, 1998). Items for the IES were derived
from statements most frequently used to de-
scribe episodes of distress by persons who had
experienced significant life changes (Horowitz
et al., 1979). Participants were instructed to
respond to the items considering how they felt
as a result of the affair. Two subscales have
been identified in this 15-item instrument, intru-
sion and avoidance. Intrusion items tapped the
experience of unbidden thoughts and images,
troubled dreams, strong pangs or waves of feel-
ings, and repetitive behavior. Avoidance re-
sponses included ideational constriction, denial
of the meanings and consequences of the event,
blunted sensation, behavioral inhibition or
counterphobic activity, and awareness of emo-
tional numbness (Horowitz et al., 1979). On the
measure, participants indicate the frequency of
the experience of each item (1 ⫽not at all,2⫽
rarely,3⫽sometimes, and 4 ⫽often; Horowitz
et al., 1979). In previous research, test–retest
reliability was .87 and internal consistency
ranged from .78 to .82 (unless otherwise noted,
all internal consistency coefficients are Cron-
bach’s alpha; Horowitz et al., 1979). In a study
with nonclinical college students, Cronbach’s
alphas were found to be .89 for the intrusion
scale and .85 for the avoidance scale (Thatcher
& Krikorian, 2005). The instrument was also
found to be internally consistent as a unidimen-
sional scale, with a reliability coefficient of .91
(Thatcher & Krikorian, 2005). In the current
study, the total scale score was used and Cron-
bach’s alpha was found to be adequate at .71.
Differentiation of Self
The Differentiation of Self Inventory–
Revised (DSI-R; Skowron & Schmitt, 2003)
was used to assess differentiation of self from
the family of origin. The DSI-R is a 46-item
self-report measure that focuses on adults, their
significant relationships, and current relation-
ships with their family of origin. The DSI-R
uses a 6-point scale ranging from 1 (not at all
true of me)to6(very true of me). The scale is
made up of four subscales to measure differen-
tiation including EC, ER, fusion with others
(FO), and IP. Skowron and Schmitt (2003)
found that internal consistency of the full scale
was .92 and that subscale internal consistency
ranged from .81 to .89. In the current study,
internal consistency for the full scale was ac-
ceptable; Cronbach’s alpha was .90. Construct
validity is indicated by the inverse relationship
found between DSI-R scores and symptoms,
perceptions of stress, and chronic anxiety (Mur-
dock & Gore, 2004;Skowron & Friedlander,
1998;Skowron, Wester, & Azen, 2004). Given
the moderation hypotheses we proposed in this
study, we used the DSI-R total scores for the
main analyses in order to have sufficient power
(Aiken & West, 1991).
1
Given the range on this variable, we recomputed anal-
yses after removing outliers. However, the results did not
change so we reverted to the original analyses using all data.
18 HEINTZELMAN, MURDOCK, KRYCAK, AND SEAY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Commitment
Participants’ level of commitment was mea-
sured using the Investment Model Scale (IMS;
Rusbult, Martz, & Agnew, 1998). This scale is
made up of 25 items that fall into four sub-
scales: commitment level and three bases of
dependence, namely, satisfaction level, quality
of alternatives, and investment size. Sample
items of the measure include “I want our rela-
tionship to last for a very long time” (i.e., com-
mitment level), “Our relationship is close to
ideal” (i.e., relationship satisfaction), “If I
weren’t dating/married to my partner, I would
find someone else” (i.e., alternative quality),
and “I have put a great deal into our relationship
that I would lose” (i.e., investment size). Higher
scores in the IMS represent higher levels of
commitment to the relationship. Responses
range from 0 (do not agree at all)to8(agree
completely) for each item. In a study sampling
college students, Rusbult et al. (1998) found
that reliability analyses revealed good internal
consistency, with Cronbach’s alphas for the
subscales ranging from .82 to .95. Principal
component analyses (PCAs) performed on the
scale items revealed evidence of four factors
and all items loaded on a single factor with
coefficients exceeding .40, and no items exhib-
ited cross-factor loadings exceeding an absolute
value of .40 (Rusbult et al., 1998). The IMS
variables were moderately associated with other
measures reflecting good couple functioning
(e.g., dyadic adjustment, trust level, inclusion of
other in the self), and were essentially unrelated
to measures assessing personal dispositions
(e.g., need for cognition, self-esteem; Rusbult et
al., 1998). Because we were specifically inter-
ested in assessing commitment level, we only
used the seven items that apply to commitment
level to one’s romantic relationship. Cronbach’s
alpha for the subscale in the current study was
.90.
Relationship Satisfaction
The Dyadic Adjustment Scale (DAS;
Spanier, 1976) is a 32-item measure designed to
assess relational satisfaction. The DAS uses
several different response scales, most of which
are Likert-type scales asking participants to rate
their level of agreement, frequency of types of
interactions with their partners, or level of sat-
isfaction in the relationship. Scores on the items
of the DAS are summed to create a total score
ranging from 0 to 151, with higher scores indi-
cating more positive dyadic adjustment, and
scores ranging from 97 to 102 are used as cut-
offs to indicate distress (Graham, Liu, & Jezi-
orski, 2006). The DAS has good internal con-
sistency (␣⫽.96) and good test–retest
reliability (r⫽.96) after 11 weeks (Spanier,
1976). In the current study, Cronbach’s alpha
for the full scale was .81. Research examining
the validity of the DAS has shown that total
DAS scores have been consistently shown to
discriminate between distressed and nondis-
tressed couples and have been shown to identify
couples with a high likelihood of divorce
(Crane, Busby, & Larson, 1991). PCA indicated
that the DAS is a unidimensional measure.
Forgiveness
The Forgiveness Inventory (FI; Gordon &
Baucom, 2003) is a 23-item questionnaire that
is used to assess injured partner’s progress
through the three-stage forgiveness model. The
FI contains three subscales that assess: (a) Stage
I experiences, (b) Stage II experiences, and (c)
Stage III experiences. Participants rate each
item on a scale ranging from 1 (almost never) to
5 (almost always). To classify individuals into
stages of forgiveness, raw scores for each sub-
scale are converted to z-scores, and based on the
directions from Gordon and Baucom (2003),
participants are assigned to the stages based the
highest of the three subscale z-scores. Low
scores in Stages I and II and higher scores in
Stage III reflect higher levels of forgiveness.
Participants in the scale development study
were couples recruited from a university or mar-
ital clinic and a confirmatory factor analysis ver-
ified the existence of the three subscales, with
each containing cognitive, behavioral, and affec-
tive components (␣⫽.85, .76, and .75, respec-
tively; Gordon & Baucom, 2003). Cronbach’s al-
pha for the three stages were at acceptable levels
of reliability at .85, .76, and .75 for Stages I, II,
and III, respectively (Gordon & Baucom, 2003).
Gordon and Baucom also examined the intercor-
relations between the three factors. As hypothe-
sized, the Stage I and Stage II factors were posi-
tively correlated, r⫽.66; the Stage III factor was
negatively correlated with the Stage I factor, r⫽
⫺.20; and the Stage III factor was positively cor-
19RECOVERY FROM INFIDELITY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

related with the Stage II factor, r⫽.23. In the
current study, Cronbach’s alphas for the scales
were .81, .62, and .81 for the Stage I, the Stage II,
and the Stage III subscales, respectively. Two of
the three subscale intercorrelations followed ex-
pected patterns (see Table 1 for intercorrelations):
Stage III scores were negatively correlated with
Stage I scores, and Stage I was correlated posi-
tively with Stage II. However, Stage II and III
scores were negatively correlated.
Posttraumatic Growth
The PTGI (Tedeschi & Calhoun, 1996)isa
21-item self-report inventory that measures an in-
dividual’s perception of positive change after a
traumatic life event. Participants are asked to rate
on a scale from 0 (I did not experience this change
as a result of my crisis)to5(I experienced this
change to a very great degree as a result of my
crisis), how much their views have changed as a
result of the trauma they experienced. Participants
were instructed to respond to the items as they
reflected responses to the infidelity they had ex-
perienced. The scale includes five factors: new
possibilities, relating to others, personal strength,
spiritual change, and appreciation for life.
The PTGI was developed and validated in a
sample of college students. Cronbach’s alphas
for the subscales ranged from .67 to .85, and
overall internal consistency for the normative
sample was .90. In the current study, the inter-
nal consistency coefficient was .95. Test–retest
reliability for the PTGI was found to be strong
in the normative sample (r⫽.71). To assess the
concurrent and discriminant validity of the
PTGI, the relationship between the PTGI and
other validated scales measuring individual dif-
ference variables was examined (Tedeschi &
Calhoun, 1996). Results indicated that the PTGI
scores were not significantly correlated with
scores of measures of social desirability or neu-
roticism but were moderately correlated with
variables that would be expected to be related,
such as optimism, positive emotion, and open-
ness (Tedeschi & Calhoun, 1996).
Results
Preliminary Analyses
Before conducting regression analyses, col-
linearity, histograms, and normal probability
plots of the residuals were examined. Standard-
ized residuals, Cook’s distance, leverage, and
Mahalanobis distances were also examined. The
tested assumptions were met, except for the
assumption of normality for relationship satis-
faction and commitment, which were negatively
skewed. Means and SDs for the scales and
intercorrelations among the variables (trauma,
commitment, differentiation of self total and
subscale scores, forgiveness, relationship satis-
faction, and PTG) can be found in Table 1.
Results indicated that DSI-R total scores posi-
tively correlated with Stage III FI scores, as did
three of four of the DSI-R subscales (the excep-
tion was FO). In addition, DSI-R scores corre-
lated positively with DAS and PTGI scores and
negatively with Stage I and II FI scores. PTGI
scores correlated negatively with Stage I and II
FI scores and positively with Stage III scores.
IES scores negatively correlated with DAS and
DSI-R scores. We found significant positive
relations between time since the infidelity and
the DSI-R subscales ER, IP, and FO, and DAS
scores. Stage I and II FI scores were negatively
correlated with time since the infidelity.
We examined the dimensionality of the
DSI-R, DAS, and PTGI using PCA. Results
indicated that each measure was unidimensional
because the first eigenvalue was at least three
times greater than the second eigenvalue in each
case. However, for the DSI-R, the eigenvalue of
the first factor was 2.8 times greater than the
eigenvalue of the second factor, and a visual
inspection of the scree plot indicated a less clear
factor solution than for the other two scales.
Thus, the dimensionality of the DSI-R in this
sample is somewhat subjective. Owing to the
exploratory nature of this study, and given that
there are some logical relationships between
subscales and forgiveness, for example (ER and
EC interfering with forgiveness), we decided to
use full-scale DSI-R scores for the main regres-
sion analyses, and to limit subscale-level anal-
yses to correlations between DSI-R subscales
and forgiveness. Keeping in mind that higher
scores on the DSI-R are indicative of higher
levels of the dimension of differentiation under
consideration, we observed negative correla-
tions between the DSI-R subscale scores (the
strongest for the EC subscale) and Stage I and II
FI scores and positive correlations between
DSI-R subscale scores (the highest again for
EC) and Stage III FI scores. The one exception
20 HEINTZELMAN, MURDOCK, KRYCAK, AND SEAY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Table 1
Distributional Statistics, Cronbach’s Alpha Reliabilities, and Intercorrelations of the Variables in the Model (N⫽587)
Variable 1 2 3 4 5678910111213
1. IMS
a
.90
2. PTGI .166
ⴱⴱ
.95
3. DSI-R .271
ⴱⴱ
.201
ⴱⴱ
.90
4. DSI-R-ER .069 .052 .687
ⴱⴱ
.84
5. DSI-R-IP ⫺.005 .244
ⴱⴱ
.632
ⴱⴱ
.450
ⴱⴱ
.75
6. DSI-R-EC
b
.359
ⴱⴱ
.046 .713
ⴱⴱ
.381
ⴱⴱ
.189
ⴱⴱ
.78
7. DSI-R-FO ⫺.061 ⫺.027 .315
ⴱⴱ
.662
ⴱⴱ
.502
ⴱⴱ
.331
ⴱⴱ
.78
8. DAS
c
.485
ⴱⴱ
.086
ⴱⴱ
.327
ⴱⴱ
.173
ⴱⴱ
.091
ⴱ
.411
ⴱⴱ
.107
ⴱⴱ
.81
9. IES ⫺.060 ⫺.018 ⫺.178
ⴱⴱ
⫺.143
ⴱⴱ
⫺.033 ⫺.229
ⴱⴱ
⫺.092
ⴱ
⫺.114
ⴱⴱ
.71
10. FI Stage I ⫺.476
ⴱⴱ
⫺.221
ⴱⴱ
⫺.449
ⴱⴱ
⫺.363
ⴱⴱ
⫺.204
ⴱⴱ
⫺.422
ⴱⴱ
⫺.205
ⴱⴱ
⫺.351
ⴱⴱ
.187
ⴱⴱ
.81
11. FI Stage II ⫺.046 ⫺.138
ⴱⴱ
⫺.203
ⴱⴱ
⫺.300
ⴱⴱ
⫺.111
ⴱⴱ
⫺.165
ⴱⴱ
⫺.230
ⴱⴱ
⫺.090
ⴱ
.157
ⴱⴱ
.433
ⴱⴱ
.62
12. FI Stage III .495
ⴱⴱ
.365
ⴱⴱ
.431
ⴱⴱ
.246
ⴱⴱ
.271
ⴱⴱ
.325
ⴱⴱ
.067 .342
ⴱⴱ
⫺.089
ⴱ
⫺.657
ⴱⴱ
⫺.136
ⴱⴱ
.81
13. Time ⫺.003 ⫺.052 .080 .117
ⴱⴱ
.083
ⴱ
.065 .100
ⴱ
.084
ⴱ
⫺.061 ⫺.085
ⴱ
⫺.152
ⴱⴱ
.053
Minimum .24 0 107 11 20 10.58 18 50.57 20888.49
Maximum 2 105 213 61 66 17 72 59 45 40 40 35 38
M1.23 60.52 167.71 35.03 43.84 13.53 46.52 54.43 36.72 22.23 30.62 23.70 3.09
SD .55 25.37 17.62 10.65 8.58 1.19 10.34 1.25 4.21 6.28 4.93 5.58 4.98
Note. IMS ⫽Investment Model Scale; PTGI ⫽Posttraumatic Growth Inventory; DSI-R ⫽Differentiation of Self Inventory–Revised; DSI-R-ER ⫽Differentiation of Self
Inventory–Revised–Emotional Reactivity subscale; DSI-R-IP ⫽Differentiation of Self Inventory–Revised–I-Position subscale; DAS ⫽Dyadic Adjustment Scale; DSI-R-EC ⫽
Differentiation of Self Inventory–Revised–Emotional Cutoff subscale; DSI-R-FO ⫽Differentiation of Self Inventory–Revised–Fusion With Others subscale; IES ⫽Impact of Event
Scale; FI ⫽Forgiveness Inventory; Time ⫽Years since infidelity.
a
Transformed with 2 ⫺Log(57 ⫺X).
b
Transformed with 18 ⫺(73 ⫺X)
1/2
.
c
Transformed with 60 ⫺(152 ⫺X)
1/2
.
ⴱ
p⬍.05.
ⴱⴱ
p⬍.01.
21RECOVERY FROM INFIDELITY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

to this pattern of significant relationships is that
FO scores were not significantly correlated with
Stage III FI scores.
Based on Gordon and Baucom’s (2003) in-
structions, each scale of the FI (Gordon & Bau-
com, 2003) was considered separately. Raw
scores for each subscale were converted to z-
scores and then compared. Each participant was
assigned to a forgiveness stage based on his or
her highest z-score (Gordon & Baucom, 2003).
Participants in this study could be classified into
all three forgiveness stages (Stage I: 33.7%,
Stage II: 27.4%, and Stage III: 38.8%). Because
we were most interested in individuals’ progress
toward the higher levels of forgiveness, Stage
III scores were used to run the regression anal-
yses described next.
Predictors of Forgiveness
To test the hypothesis that differentiation of
self and trauma predict forgiveness, a hierarchi-
cal linear regression was conducted with DSI-R
and IES scores as predictors of Stage III FI
scores (Table 2). To help prevent multicol-
linearity that is introduced by the creation of the
interaction term, the predictor variables were
centered by subtracting the mean for the vari-
able from all scores. The predictors (centered
IES scores and centered DSI-R scores) were
entered first, followed by the interaction term
(centered IES ⫻DSI-R) in the second step.
Results indicated that the regression equation
with IES and DSI-R scores significantly pre-
dicted forgiveness scores (F(2, 584) ⫽66.62,
p⬍.001, R
2
⫽.186). Examination of the beta
weights showed that differentiation of self was
a significant predictor (␤⫽.429, p⬍.001) but
trauma (IES) was not (␤⫽⫺.013, p⬎.05).
The second step of the regression equation in
which the interaction term was entered was also
statistically significant, (F(3, 583) ⫽48.83, p⬍
.001, R
2
⫽.201) and the addition of the inter-
action term produced a significant increase in R
2
(⌬R
2
⫽.02; F(1, 583) ⫽3.14, p⬍.001; ad-
justed R
2
⫽.20, p⬍.05; Table 2). Examination
of the beta weights in Step 2 indicated that
differentiation of self (␤⫽.429, p⬍.001) and
the interaction of differentiation of self and
trauma (␤⫽⫺.124, p⫽.001) were significant
predictors, but trauma was not (␤⫽.007, p⬎
.05). Figure 1 shows the graph of the modera-
tion of the effect of trauma on forgiveness by
differentiation of self with differentiation set at
minus 1 SD, plus 1 SD, and at the mean. An
analysis of the simple slopes of regression plotted
for forgiveness at the three levels of differentiation
demonstrated that the high-differentiation and
low-differentiation groups’ slopes were signifi-
cantly different from zero (t⫽⫺2.3478, p⫽.05
for high differentiation; t⫽2.2646, p⬍.05 for
low differentiation) and from one another (95%
confidence intervals around their slope values did
not overlap; CI
95
⫽⫺.025 to ⫺.28 and .023 to
.322 for high and low differentiation, respec-
tively).
Predictors of PTG
To examine what predicts PTG, we con-
ducted a hierarchical linear regression with PTG
(PTGI) as the criterion variable and time since
infidelity, differentiation of self (DSI-R),
trauma (IES), relationship satisfaction, commit-
Table 2
Summary of Hierarchical Regression With Predictors of Forgiveness
Variable B
Standard
error B ␤R
2
Adjusted
R
2
⌬R
2
Step 1 .186 .183 .186
DSI-R .136 .012 .429
ⴱ
IES ⫺.017 .050 ⫺.013
Step 2 .201 .197 .015
ⴱ
DSI-R .136 .012 .429
ⴱ
IES .010 .050 .007
IES ⫻DSI-R ⫺.009 .003 ⫺.124
ⴱ
Note. DSI-R ⫽Differentiation of Self Inventory–Revised; IES ⫽Impact of Event Scale;
IES ⫻DSI-R ⫽Interaction term of Impact of Event Scale and Differentiation of Self
Inventory–Revised.
ⴱ
p⬍.001.
22 HEINTZELMAN, MURDOCK, KRYCAK, AND SEAY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

ment (IMS), and Stage III FI scores as predic-
tors in Step 1. In Step 2 of the regression, we
added the interaction of IES and DSI-R (Table
3). To protect against multicollinearity that is
introduced by the creation of interaction terms,
we centered each predictor variable by subtract-
ing the mean score from individual scores on
each variable. Results indicated that the regres-
sion equation containing time since infidelity,
DSI-R, IES, IMS, relationship satisfaction, and
Stage III FI scores significantly predicted PTGI
scores (F(6, 58) ⫽16.23, p⬍.001, R
2
⫽.144).
However, examination of the beta weights
showed that only the beta for Stage III FI scores
was significantly different from zero (␤⫽.359,
p⬍.001); regression coefficients for time since
infidelity, DSI-R, IMS, IES, and DAS scores
were not. The second step of the regression
equation in which the interaction term was en-
tered was also statistically significant (F(7,
579) ⫽14.13, p⬍.001, R
2
⫽.146), but the
addition of the interaction term did not produce
a significant increase in R
2
(⌬R
2
⫽.002; F(1,
579) ⫽1.47, p⬎.05; Table 3). Examination of
the beta weights in Step 2 indicated that Stage
III FI score remained the only significant pre-
dictor (␤⫽.352, p⬍.001).
Discussion
Because little is known about what variables
contribute to personal growth and relational im-
provement in the aftermath of infidelity, the aim
of this study was to explore specific variables
that are thought to influence the trajectory of the
recovery process from infidelity. The results
help to shed light on variables that may be
critical to the healing process and the ability to
move forward after the experience of infidelity.
The hypothesis that trauma, differentiation of
self, and the interaction between trauma and
differentiation of self would predict forgiveness
was partially supported. Level of trauma was
not found to be a predictor of forgiveness, ex-
Figure 1. Moderation of the effect of trauma on forgiveness by differentiation of self.
DSI-R ⫽Differentiation of Self Inventory–Revised.
23RECOVERY FROM INFIDELITY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

cept in interaction with differentiation of self.
This result would seem to have implications for
the trauma-based model of forgiveness pro-
posed by Baucom et al. (2006). However, this
was a retrospective study, and it is possible that
over time, levels of trauma experienced at the
time of the infidelity become less salient as
individuals move through the forgiveness pro-
cess. To participate in our study, we required
that at least 6 months have passed since the
affair ended and the average time since the
affair was about 3 years. Given that our cou-
ples were still together, it perhaps makes
some sense that experiences of intrusion and
avoidance are not a strong predictor of for-
giveness that far out. The observation that the
highest score in the sample on the IES was 45
and the mean was 22, when the scale has a
potential high score of 60, might support the
hypothesis that the retrospective nature of our
measure resulted in lower levels of recalled
trauma.
As hypothesized, differentiation of self was a
significant predictor of forgiveness. Inspection
of Figure 1 and the correlation between differ-
entiation and trauma indicate that individuals
with higher levels of differentiation reported
more forgiveness and less trauma overall than
did respondents who were lower in differentia-
tion. In addition, three of four subscales of the
DSI-R were significant predictors of forgive-
ness: less EC, less ER, and more IP were asso-
ciated with higher levels of forgiveness. An
important facet of Stage III in the Gordon et al.
(2004) model of forgiveness of infidelity in-
cludes partners’ increased understanding of
each other, themselves, and their relationship to
“free themselves from being dominated by neg-
ative thoughts, feelings, and behaviors”
(p. 216). Accordingly, individuals who are less
emotionally reactive and more able to stay con-
nected (i.e., lower EC) and who have the ability
to calmly state their positions (i.e., more IP)
may deal better with the immediate issues and
emotions that arise as a result of the infidelity
because they are more able to stay engaged in
the needed processing over time.
However, we also thought that the relation-
ship between trauma and forgiveness would be
moderated by differentiation of self, with par-
ticipants higher in differentiation being less af-
fected by the level of trauma than those lower in
differentiation of self. Previous research has
indicated that under high stress, individuals
with lower differentiation show more psycho-
logical symptoms than do individuals higher in
differentiation of self—that is, they react more
to the stress. Inverting this reasoning, we ex-
Table 3
Summary of Hierarchical Regression With Predictors of Posttraumatic Growth
Variable B
Standard
error B ␤R
2
Adjusted
R
2
⌬R
2
Step 1 .144 .135 .144
Time since infidelity ⫺.362 .197 ⫺.071
DSI-R .11 .063 .073
IMS ⫺.283 2.205 ⫺.006
IES .097 .236 .016
DAS ⫺1.00 .919 ⫺.050
FI Stage III 1.632
ⴱ
.216 .359
Step 2 .146 .136 .002
Time since infidelity ⫺.361 .197 ⫺.071
DSI-R .110 .063 .076
IES ⫺.408 2.207 ⫺.009
IES .143 .239 .024
DAS ⫺.959 .920 ⫺.047 ⫺1.043
FI Stage III 1.603
ⴱ
.217 .352 7.393
IES ⫻DSI-R ⫺.016 .013 ⫺.048 ⫺1.214
Note. DSI-R ⫽Differentiation of Self Inventory–Revised; IMS ⫽Investment Model Scale;
IES ⫽Impact of Event Scale; DAS ⫽Dyadic Adjustment Scale; FI Stage III ⫽Forgiveness
Inventory, Stage III score; IES ⫻DSI-R ⫽Interaction term of Impact of Event Scale and
Differentiation of Self Inventory–Revised.
ⴱ
p⬍.001.
24 HEINTZELMAN, MURDOCK, KRYCAK, AND SEAY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

pected that individuals lower in differentiation
would show lower levels of forgiveness under
highly traumatic conditions compared with less
traumatic ones. In our data, the differentiation
of self moderated the relationship between
trauma and forgiveness, and inspection of Fig-
ure 1 indicated the significant interaction re-
sulted from the difference in slopes between the
high- and low-differentiated groups; partici-
pants lower in differentiation of self showed
more forgiveness under high-trauma conditions
than in lower-trauma conditions. In contrast,
those with high differentiation showed less for-
giveness under high-trauma conditions than un-
der lower-trauma levels.
The relationships between trauma, differ-
entiation, and forgiveness are, at first glance,
puzzling. However, several aspects of these
findings invite consideration. First, and per-
haps most prominently, we conceptualized
trauma as analogous to stress in this study,
and it is possible that the trauma related to an
affair as measured in this study is not com-
parable with the distress caused by stressful
life events measured in previous research.
The IES assesses the classic symptoms result-
ing from trauma (i.e., intrusion and avoidance
responses) rather than perceived stress as as-
sessed in other studies (Krycak et al., 2012;
Murdock & Gore, 2004). Also, trauma was
measured retrospectively, so it is possible that
over time, distortions in recall occurred, ob-
scuring the relationships between trauma, dif-
ferentiation, and forgiveness that existed at
the time of the affair. In addition, levels of
trauma were relatively low in this research.
Finally, it could be that for individuals lower
on the differentiation continuum, instances of
infidelity that were recalled as less traumatic
simply called for less forgiveness, hence the
increased levels of forgiveness when higher
trauma levels were reported. Higher-differen-
tiated individuals seemed to be indicating that
the more they recalled experiencing intrusion
and avoidance related to the trauma of the
infidelity, the less they were willing to for-
give, which we suppose could be understand-
able. Also, it is important to reiterate that
high-differentiated individuals still showed
more forgiveness than did those lower in dif-
ferentiation over all levels of trauma. Future
research could investigate these questions
more closely, perhaps by using qualitative or
mixed-methods approaches.
Apparently, forgiveness trumps all in terms
of PTG. None of the other variables we tested
emerged as significant predictors of PTG. The
lack of relationship between commitment and
relationship satisfaction and growth may be ac-
counted for by the fact that only current level of
commitment and relationship satisfaction were
assessed, which may not be as informative as
commitment and relationship satisfaction as-
sessed around the time of the infidelity. In ad-
dition, time since infidelity varied widely
among participants, and although it may seem
surprising that time since infidelity does not
correspond to PTG, this lack of relationship
may point the critical role of other variables in
whether one grows as a result of traumatic
events. Time alone does not produce growth.
The only variable that predicted PTG was
Stage III FI score; those who were more able to
forgive their partners for the infidelity also ex-
perienced more growth after the event. This
finding is consistent with Gordon and Baucom’s
(2003) model, in that individuals who evidence
high levels of forgiveness are expected to expe-
rience higher levels of relational and life satis-
faction compared with individuals reporting
lower levels of forgiveness. Indeed, those who
report higher levels of Stage III forgiveness
might be said to be “in recovery” from the
affair, and experience greater understanding, a
nondistorted view of the relationship, fewer
negative emotions, and decreased urges to pun-
ish their partners (Gordon & Baucom, 2003).
These changes resulting from the forgiveness
process seem to parallel the process of PTG,
which involves both affective changes and cog-
nitive restructuring of beliefs and assumptions
after a traumatic event (Tedeschi & Calhoun,
2004). Thus, it is fitting that those who are more
able to forgive their partners’ infidelity would
develop the emotional relief and cognitive clar-
ity that is characteristic of PTG (Tedeschi &
Calhoun, 2004).
Implications
The results of this study indicate that differen-
tiation of self from the family of origin, forgive-
ness, and PTG are involved in the process of
recovery from infidelity. Therefore, this study has
implications for counselors working with clients
25RECOVERY FROM INFIDELITY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

who are dealing with infidelity. Formally or infor-
mally assessing differentiation of self may help to
provide the counselors with a rough idea of how
clients are emotionally responding to the infidel-
ity. Furthermore, the results of these assessments
could help counselors to gauge how far along the
clients are in the recovery process (e.g., signs of
EC or ER may indicate less progress) and provide
the counselor with information about specific
emotional aspects to focus on during counseling
sessions, such as helping partners stay connected
and to clearly state their positions (i.e., addressing
EC and increasing IP) or focusing on managing
strong emotional responses to the infidelity (i.e.,
examining ER). In addition, providing clients with
psychoeducation regarding differentiation of self
and the importance of maintaining a healthy bal-
ance between emotional connection and separate-
ness may help clients stay engaged with their
partners during the difficult process of repairing
the relationship and ultimately facilitate growth
and self-awareness in clients who have experi-
enced infidelity from their partners. In essence, by
increasing levels of differentiation of self, partners
may be able to weather the processing needed to
reach forgiveness. Although Bowen (1978) was
somewhat vague when discussing specific inter-
ventions aimed at increasing differentiation of
self, he noted that psychotherapy could help to
increase differentiation by expanding awareness
of the difference between emotional functioning
and intellectual functioning and also allowing in-
dividuals to examine the dynamics of important
interpersonal relationships. Because this assertion
has not been tested empirically, future counseling
process and outcome research should examine
whether psychotherapy with these components
leads to an increase in differentiation of self.
Our results would encourage clinicians to
educate clients on the process of recovery from
infidelity that is associated with forgiveness.
Educating counselors in training about forgive-
ness models would therefore seem helpful (for
two frequently cited and empirically supported
models used to help promote forgiveness, see
Worthington et al., 2000, and McCullough &
Worthington, 1995). Our results would also en-
courage clinicians to facilitate the forgiveness
process during couples therapy. Greenberg,
Warwar, and Malcolm (2010) have found that
emotion-focused couples therapy is an effective
treatment orientation for promoting forgiveness
and alleviating marital distress. To provide
counselors with direction on how to use emo-
tion-focused couples therapy to lead couples to
forgiveness, Meneses and Greenberg (2011) de-
veloped a preliminary model of forgiveness.
Our study supports the importance of future
research focusing on testing the validity of the
model. Counselors may also want to encourage
clients to join group-therapy sessions with other
clients dealing with infidelity because psychoe-
ducational group interventions have been found
to be effective in promoting forgiveness (Mc-
Cullough & Worthington, 1995). The online
support forums specifically designed for indi-
viduals dealing with infidelity may also be a
useful suggestion for clients. Online support
groups have been found to help people effec-
tively cope with a variety of problems and foster
well-being by promoting personal empower-
ment, improving understanding and knowledge,
and developing social relationships (Barak,
Boniel-Nissim, & Suler, 2008).
Limitations
There are limitations to this study that should
be noted. The characteristics of the sample af-
fect the ability to generalize the results of this
study. Our sample consisted of 85.3% women
and 86.7% Caucasian participants. Most partic-
ipants were married. Participants who com-
pleted the study were older than those who did
not and were largely heterosexual. Therefore,
the recovery process for unmarried, male, those
of sexual orientations other than heterosexual,
or individuals of other races and ethnicities may
be different from that observed in this study.
In addition, for the purposes of this study, we
focused on individuals who reported that their
partners had engaged in sexual intercourse with
someone else during a committed relationship. To
better study variables related to forgiveness and
PTG, we required that individuals still be in the
relationship where infidelity occurred and that the
infidelity had occurred ⬎6 months before partic-
ipation in our study. This restriction resulted in
retrospective data collection, meaning that our re-
sults may have been affected by other, unknown,
processes occurring over time. The study was also
cross-sectional, so even though from a theoretical
perspective the directionality of the relationships
we found is sensible (differentiation predicts for-
giveness; forgiveness predicts PTG), it cannot be
assumed to be the only possibility.
26 HEINTZELMAN, MURDOCK, KRYCAK, AND SEAY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Only nonparticipating partners were included
in this study. Characterized as the “injured
party” by Gordon et al. (2004), they are proba-
bly more willing to participate in research be-
cause they are not the guilty ones. By restricting
our definition to situations involving sexual in-
tercourse only, we used a narrow definition of
infidelity, which has the benefit of precision but
the drawback of leaving other types of infidelity
unexplored.
The data collection method may have had an
effect on results of this study. All of the partic-
ipants in this study were obtained from online
support forums for infidelity. McCullough and
Worthington’s (1995) findings suggest that
group support may promote the experience of
forgiveness. Further, Teo (2001) and Weiser
(2000) found that women are more likely to use
the Internet for support, interpersonal commu-
nication, and educational assistance, whereas
men typically use the Internet for entertainment,
leisure, and purchasing activities. The large ma-
jority of female participants in this study would
fit with their data. Therefore, to tease out the
relationship among these factors and how they
relate to forgiveness, replication of this study
with participants who are not part of online
support groups would be helpful.
The effect sizes found for each hypothesis
were small (Cohen, 1988). Therefore, it is a
possibility that the significant effects found in
this study may be an artifact of the large sample.
However, effect sizes for moderator terms are
known for being small (Aiken & West, 1991;
McClelland & Judd, 1993), as were those ob-
served in the analysis of differentiation and
trauma as predictors of forgiveness.
Conclusion
This study has helped to identify factors that
may be important in the recovery process of
infidelity and therefore provides counselors
with some empirically grounded suggestions to
use in their work with individuals recovering
from infidelity. Important areas for future re-
search include further exploration of variables
that have been empirically linked to forgiveness
and the recovery process from infidelity that
were outside of the scope of this study, such as
empathy, acceptance, religiosity, external sup-
port, presence of children, and how long the
infidelity lasted (Blow & Harnett, 2005;Rusbult
et al., 2004;Worthington, 1998,2005). Further-
more, the interaction between differentiation of
self and trauma level should be clarified in
future research, and the links between cognitive
and interpersonal process and forgiveness ex-
plored. Finally, to improve the ability to gener-
alize the results of this study, replication is
needed with participants who are more diverse
in gender, ethnicity, sexual orientation, and re-
lationship status. Longitudinal research designs
that use forgiveness interventions in a group
format may also be a particularly fruitful area of
research on the recovery process from infidelity.
References
Abrahamson, I., Hussain, R., Khan, A., & Schofield,
M. J. (2012). What helps couples rebuild their rela-
tionship after infidelity? Journal of Family Issues, 33,
1494 –1519. doi:10.1177/0192513X11424257
Aiken, L. S., & West, S. G. (1991). Multiple regres-
sion: Testing and interpreting interactions. Thou-
sand Oaks, CA: Sage.
Allen, E. S., Atkins, D. C., Baucom, D. H., Synder,
D. K., Gordon, K., & Glass, S. P. (2005). Intrap-
ersonal, interpersonal, and contextual factors in
engaging in and responding to extramarital in-
volvement. Clinical Psychology: Science and
Practice, 12, 101–130. doi:10.1093/clipsy.bpi014
Atkins, D. C., Marín, R. A., Lo, T. T. Y., Klann, N.,
& Hahlweg, K. (2010). Outcomes of couples with
infidelity in a community-based sample of couple
therapy. Journal of Family Psychology, 24, 212–
216. doi:10.1037/a0018789
Balswick, J., & Balswick, J. (1999). Extramarital
affairs: Causes, consequences, and recovery. Mar-
riage and Family: A Christian Journal, 2, 419 –
427. Retrieved from http://www.aacc.net/
Barak, A., Boniel-Nissim, M., & Suler, J. (2008).
Fostering empowerment in on-line support groups.
Computers in Human Behavior, 24, 1867–1883.
doi:10.1016/j.chb.2008.02.004
Baucom, D., Gordon, K., Snyder, D., Atkins, D., &
Christensen, A. (2006). Treating affair couples:
Clinical considerations and initial findings. Jour-
nal of Cognitive Psychotherapy, 20, 375–392. doi:
10.1891/jcpiq-v20i4a004
Blow, A. J., & Harnett, K. (2005). Infidelity in com-
mitted relationships I: A methodological review.
Journal of Marital and Family Therapy, 31, 183–
216. doi:10.1111/j.1752-0606.2005.tb01556.x
Bowen, M. (1978). Family therapy in clinical prac-
tice. New York, NY: Jason Aronson.
Briere, J., & Elliott, D. M. (1998). Clinical utility of
the Impact of Event Scale: Psychometrics in the
27RECOVERY FROM INFIDELITY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

general population. Assessment, 5, 171–180. doi:
10.1177/107319119800500207
Burchard, G. A., Yarhouse, M. A., Kilian, M. K., &
Worthington, E. L. (2003). A study of two marital
enrichment programs and couples’ quality of life.
Journal of Psychology and Theology, 31, 240.
Charny, I., & Parnass, S. (1995). The impact of
extramarital relationships on the continuation of
marriages. Journal of Sex and Marital Therapy,
21, 100 –115. doi:10.1080/00926239508404389
Cohen, J. (1988). Statistical power analysis for the
behavioral sciences. Mahwah, NJ: Lawrence Erl-
baum Associates.
Crane, D. R., Busby, D. M., & Larson, J. H. (1991).
A factor analysis of the Dyadic Adjustment Scale
with distressed and nondistressed couples. Ameri-
can Journal of Family Therapy, 19, 60 – 66. doi:
10.1080/01926189108250835
Gordon, K., & Baucom, D. (1998). Understanding
betrayals in marriage: A synthesized model of for-
giveness. Family Process, 37, 425– 449. doi:
10.1111/j.1545-5300.1998.00425.x
Gordon, K., & Baucom, D. (1999). A multitheoreti-
cal intervention for promoting recovery from ex-
tramarital affairs. Clinical Psychology: Science
and Practice, 6, 382–399. doi:10.1093/clipsy.6.4
.382
Gordon, K., & Baucom, D. (2003). Forgiveness and
marriage: Preliminary support for a measure based
on a model of recovery from a marital betrayal.
American Journal of Family Therapy, 31, 179 –
199. doi:10.1080/01926180301115
Gordon, K., Baucom, D., & Snyder, D. (2004). An
integrative intervention for promoting recovery
from extramarital affairs. Journal of Marital and
Family Therapy, 30, 213–231. doi:10.1111/j.1752-
0606.2004.tb01235.x
Gordon, K., Baucom, D., & Snyder, D. (2005). Treat-
ing couples recovering from infidelity: An integra-
tive approach. Journal of Clinical Psychology, 61,
1393–1405. doi:10.1002/jclp.20189
Graham, J. M., Liu, Y. J., & Jeziorski, J. L. (2006).
The Dyadic Adjustment Scale: A reliability gen-
eralization meta-analysis. Journal of Family and
Marriage, 68, 701–717. doi:10.1111/j.1741-3737
.2006.00284.x
Greenberg, L., Warwar, S., & Malcolm, W. (2010).
Emotion-focused couples therapy and the facilita-
tion of forgiveness. Journal of Marital and Family
Therapy, 36, 28 – 42. doi:10.1111/j.1752-0606
.2009.00185.x
Gubbins, C. A., Perosa, L. M., & Bartle-Haring, S.
(2010). Relationships between married couples’
self-differentiation/individuation and Gottman’s
model of marital interactions. Contemporary Fam-
ily Therapy, 32, 383–395. doi:10.1007/s10591-
010-9132-4
Hansen, G. L. (1987). Extradyadic relations during
courtship. Journal of Sex Research, 23, 382–390.
doi:10.1080/00224498709551376
Holeman, V. T., Dean, J. B., DeShea, L., & Duba,
J. D. (2011). The multidimensional nature of the
quest construct forgiveness, spiritual perception, &
differentiation of self. Journal of Psychology and
Theology, 39, 31– 43.
Horowitz, M., Wilner, N., & Alvarez, W. (1979).
Impact of Event Scale: A measure of subjective stress.
Psychosomatic Medicine, 41, 209 –218. Retrieved
from http://www.psychosomaticmedicine.org/
Janus, S., & Janus, C. (1993). The Janus report on
sexual behavior. Oxford, England: Wiley.
Kerr, M., & Bowen, M. (1988). Family evaluation:
An approach based on Bowen theory. New York,
NY: Norton.
Krycak, R. C., Murdock, N. L., & Marszalek, J. M.
(2012). Differentiation of self, stress, and emo-
tional support as predictors of psychological dis-
tress. Contemporary Family Therapy, 34, 495–
515. doi:10.1007/s10591-012-9207-5
Linley, P. A., & Joseph, S. (2004). Positive changes
following trauma and adversity: A review. Journal
of Traumatic Stress, 17, 11–21. doi:10.1023/B:
JOTS.0000014671.27856.7e
Marcussen, K., Ritter, A., & Safron, D. J. (2004). The
role of identity salience and commitment in the
stress process. Sociological Perspectives, 47, 289 –
312. doi:10.1525/sop.2004.47.3.289
Mark, K. P., Janssen, E., & Milhausen, R. R. (2011).
Infidelity in heterosexual couples: Demographic,
interpersonal, and personality-related predictors of
extradyadic sex. Archives of Sexual Behavior, 40,
971–982. doi:10.1007/s10508-011-9771-z
McClelland, G. H., & Judd, C. M. (1993). Statistical
difficulties of detecting interactions and moderator
effects. Psychological Bulletin, 114, 376 –390. doi:
10.1037/0033-2909.114.2.376
McCullough, M., & Worthington, E. (1995). Promot-
ing forgiveness: A comparison of two brief psy-
choeducational group intervention with a waiting-
list control. Counseling and Values, 40, 55– 68.
doi:10.1002/j.2161-007X.1995.tb00387.x
Meneses, C. W., & Greenberg, L. S. (2011). The
construction of a model of the process of couples’
forgiveness in emotion-focused therapy for cou-
ples. Journal of Marital and Family Therapy, 37,
491–502. doi:10.1111/j.1752-0606.2011.00234.x
Murdock, N. L., & Gore, P. A. (2004). Differentia-
tion, stress, and coping: A test of Bowen theory.
Contemporary Family Therapy, 26, 319 –335. doi:
10.1023/B:COFT.0000037918.53929.18
Murray, T. L., Daniels, M. H., & Murray, C. E.
(2006). Differentiation of self, perceived stress,
and symptom severity among patients with fibro-
myalgia syndrome. Families, Systems, & Health,
24, 147–159. doi:10.1037/1091-7527.24.2.147
28 HEINTZELMAN, MURDOCK, KRYCAK, AND SEAY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Olson, M. M., Russell, C. S., Higgins-Kessler, M., &
Miller, R. B. (2002). Emotional processes follow-
ing disclosure of an extramarital infidelity. Journal
of Marital and Family Therapy, 28, 423– 434. doi:
10.1111/j.1752-0606.2002.tb00367.x
Ripley, J. S., & Worthington, E. L. (2002). Hope-
focused and forgiveness-based group interventions
to promote marital enrichment. Journal of Coun-
seling & Development, 80, 452– 463. doi:10.1002/
j.1556
Rusbult, C., Kumashiro, M., Coolsen, M., & Kirch-
ner, J. (2004). Interdependence, closeness, and rela-
tionships. In D. J . Mashek & A. P. Aron (Eds),
Handbook of closeness and intimacy (pp. 137–
161). Mahwah, NJ: Erlbaum.
Rusbult, C. E., Martz, J. M., & Agnew, C. R. (1998).
The Investment Model Scale: Measuring commitment
level, satisfaction level, quality of alternatives, and in-
vestment size. Personal Relationships, 5, 357–387. doi:
10.1111/j.1475-6811.1998.tb00177.x
Sandage, S. J., & Jankowski, P. J. (2011). Forgive-
ness, differentiation of self, and mental health. In M. M .
Maamri, N. Nevin, & E. L. Worthington, Jr. (Eds),
A journey through forgiveness (pp. 87–98). Oxford,
England: Inter-Disciplinary Press.
Schneider, J., Irons, R., & Corley, M. (1999). Disclo-
sure of extramarital sexual activities by sexually exploit-
ative professionals and other persons with addictive or
compulsive sexual disorders. Journal of Sex Education
and Therapy, 24, 277–287. Retrieved from https://
catalyst.library.jhu.edu/catalog/bib_3562196
Skowron, E. A. (2000). Therole of differentiation of self
in marital adjustment. Journal of Counseling Psychol-
ogy, 47, 229 –237. doi:10.1037/0022-0167.47.2.229
Skowron, E., & Friedlander, M. (1998). The Differ-
entiation of Self Inventory: Development and ini-
tial validation. Journal of Counseling Psychology,
45, 235–246. doi:10.1037/0022-0167.45.3.235
Skowron, E. A., & Schmitt, T. A. (2003). Assess-
ing interpersonal fusion: Reliability and validity
of a new DSI Fusion with Others subscale. Jour-
nal of Marital and Family Therapy, 29, 209 –222. doi:
10.1111/j.1752-0606.2003.tb01201.x
Skowron, E. E., Wester, S. R., & Azen, R. (2004).
Differentiation of self mediates college stress and ad-
justment. Journal of Counseling and Development, 82,
69 –78. doi:10.1002/j.1556-6678.2004.tb00287.x
Spanier, G. B. (1976). Measuring dyadic adjustment:
New scales for assessing the quality of marriage
and similar dyads. Journal of Marriage and the
Family, 38, 15–28. doi:10.2307/350547
Tabachnick, B. G., & Fidell, L. S. (2008). Using
multivariate statistics (5th ed.). New York, NY:
Pearson Incorporated.
Tedeschi, R., & Calhoun, L. (1995). Trauma and
transformation: Growing in the aftermath of suffering.
Thousand Oaks, CA: Sage.
Tedeschi, R., & Calhoun, L. (1996). The posttrau-
matic growth inventory: Measuring the positive
legacy of trauma. Journal of Traumatic Stress, 9,
455– 471. doi:10.1002/jts.2490090305
Tedeschi, R., & Calhoun, L. (2004). Posttraumatic
growth: Conceptual foundations and empirical evi-
dence. Psychological Inquiry, 15, 1–18. doi:
10.1207/s15327965pli1501_01
Tedeschi, R., & Calhoun, L. (2006). Time of change?
The spiritual challenges of bereavement and loss. Ome-
ga: Journal of Death and Dying, 53, 105–116.
doi:10.2190/7MBU-UFV9-6TJ6-DP83
Teo, T. (2001). Demographic and motivation vari-
ables associate with internet usage activities. In-
ternet Research: Electronic Networking Applica-
tions and Policy, 11, 125–137. doi:10.1108/
10662240110695089
Thatcher, D. L., & Krikorian, R. (2005). Exploratory
factor analysis of two measures of posttraumatic stress
disorder (PTSD) symptoms in a non-clinical sample of
college students. Journal of Anxiety Disorders, 19,
904 –914. doi:10.1016/j.janxdis.2004.11.004
Thoits, P. A. (1991). On merging identity theory and
stress research. Social Psychology Quarterly, 54,
101–112. doi:10.2307/2786929
Thoits, P. A. (1992). Identity structures and psycholog-
ical well-being: Gender and marital status comparisons.
Social Psychology Quarterly, 55, 236 –256. doi:
10.2307/2786794
Weiser, E. B. (2000). Gender differences in Internet use
patterns and Internet application preferences: A two-
sample comparison. CyberPsychology and Behavior,
3, 167–178. doi:10.1089/109493100316012
Whisman, M., Dixon, A., & Johnson, B. (1997). Ther-
apists’ perspectives of couple problems and treatment
issues in couple therapy. Journal of Family Psychol-
ogy, 11, 361–366. doi:10.1037/0893-3200.11.3.361
Winek, J., & Craven, P. (2003). Healing rituals for
couples recovering from adultery. Contemporary
Family Therapy: An International Journal, 25,
249 –266. doi:10.1023/A:1024518719817
Worthington, E. (1998). An empathy-humility-
commitment model of forgiveness applied within fam-
ily dyads. Journal of Family Therapy, 20, 59 –76.
doi:10.1111/1467-6427.00068
Worthington, E. (2005). Handbook of forgiveness.
New York, NY: Taylor & Francis Group.
Worthington, E., Kurusu, T., Collins, W., Berry, J.,
Ripley, J., & Baier, S. (2000). Forgiving usually
takes time: A lesson learned by studying interven-
tions to promote forgiveness. Journal of Psychol-
ogy and Theology, 28, 3–20. Retrieved from http://
journals.biola.edu/jpt
Received April 18, 2013
Revision received December 9, 2013
Accepted December 30, 2013 䡲
29RECOVERY FROM INFIDELITY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.