Journal of Obsessive-Compulsive and Related Disorders 1 (2012) 153158

Contents lists available at SciVerse ScienceDirect

Journal of Obsessive-Compulsive and Related Disorders

journal homepage: www.elsevier.com/locate/jocrd

Predictors of mental pollution: The contribution of religion,

parenting strategies, and childhood trauma
Noah C. Berman, Michael G. Wheaton, Laura E. Fabricant, Jonathan S. Abramowitz n
University of North Carolina at Chapel Hill, Campus Box #3270 (Davie Hall), Chapel Hill, NC 27599, United States

a r t i c l e i n f o

abstract

Article history:
Received 12 August 2011
Received in revised form
20 February 2012
Accepted 14 March 2012
Available online 5 May 2012

One presentation of obsessive-compulsive disorder that has begun to receive research attention is
mental pollution, a sense of inward contamination that persists regardless of the presence of external
contaminants. Given the need to better understand this phenomenon, we sought to identify predictor
variables that could inform theory, treatment, and prevention. A large number of undergraduate
students completed a battery of self-report questionnaires assessing personal experiences, religious
variables, parenting strategies, and childhood trauma, that appear theoretically relevant to mental
pollution. Results indicated that mental pollution was not associated with degree of religiosity, yet
positively associated with exposure to childhood trauma and maladaptive guilt induction strategies by
ones parents. Implications for the assessment and cognitive-behavioral treatment of mental pollution,
as well as limitations of the study, are discussed.
& 2012 Elsevier Ltd. All rights reserved.

Keywords:
Mental pollution
Contamination
Guilt
Religion
Trauma

1. Introduction
Obsessive Compulsive Disorder (OCD) consists of persistent
unwanted intrusive thoughts, ideas, or images that provoke
anxiety or distress (i.e., obsessions) and urges to perform behavioral or mental rituals (i.e., compulsions; American Psychiatric
Association, 2000). Although the diagnostic system (DSM-IV; APA,
2000) currently classies OCD as a homogenous condition, obsessions and compulsions are thematically heterogeneous (e.g.,
Abramowitz et al., 2010; Mataix-Cols, Rosario-Camposand
Leckman, 2005; McKay et al, 2004). One person might have
obsessional doubts related to turning off the oven and feel
compelled to compulsively re-check appliances, whereas another
might have aggressive obsessions (e.g., stabbing a family member) leading to avoidance of certain stimuli (e.g., knives). Given
this heterogeneity, it is important to conduct research focusing on
the various symptom presentations.
One manifestation of OCD that has received increased research
attention is mental pollution, dened as a sense of internal uncleanness which can and usually does arise and persist regardless of the
presence or absence of external, observable dirt (Rachman, 1994; p.
311). The sense of uncleanliness is usually triggered by unacceptable
blasphemous or sexual thoughts, aggressive urges, or general feelings
of immorality (Fairbrother, Newth, & Rachman, 2005). Thus, to the

n
Corresponding author at: Department of Psychology, University of North
Carolina at Chapel Hill, Campus Box #3270 (Davie Hall), Chapel Hill, NC 27599,
United States. Tel.: 1 919 843 8170.
E-mail addresses: berman.noah@gmail.com (N.C. Berman),
jabramowitz@unc.edu (J.S. Abramowitz).

2211-3649/$ - see front matter & 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.jocrd.2012.03.005

person, such thoughts are objectionable in a way that creates feelings

of dirtiness that can seemingly be alleviated through hand washing
rituals. Having the intrusive thought God is dead, for example, leads
to feeling inwardly dirty and an urge to wash ones hands. Research
has empirically identied two components of mental pollution:
(a) inward feelings of contamination that are not linked to washing
behaviors (e.g., feeling dirty inside) and (b) washing rituals performed
in response to perceived mental pollution (Cougle, Lee, Horowitz,
Wolitzky-Taylor, & Telch, 2008). Lady Macbeth offers a paragon
presentation of the latter: Following her murder of King Duncan,
Lady Macbeth repeatedly attempts to absolve herself of guilt by
washing her hands (Out damned spot! Out I say); however,
repeated washing is futile as the source of contamination is internal.
Rachman (1994, 2004, 2006) considered mental pollution as a
type of contamination fear, along with fears of contracting an
illness from external sources of contamination (e.g., germs on
doorknobs). Although contamination-related OCD symptoms generally respond well to exposure-based treatment (e.g., Abramowitz,
Franklin, Schwartz, & Furr, 2003), the presence of mental pollution
might explain some instances of non-response (Cougle et al., 2008;
Elliott & Radomsky, 2009). That is, many clinicians who observe
hand washing symptoms (a cardinal feature of OCD) are apt to
presume that these rituals are triggered by external stimuli, rather
than unacceptable thoughts. When exposure to tangible contaminants such as oors or doorknobs does not provoke fear, exposure
might be abandoned and the patient considered treatment refractory despite not using exposures that access the patients actual
obsessional stimuliunacceptable thoughts.
Given the need to better understand mental pollution it is
important to identify variables that serve as predictors of this

154

N.C. Berman et al. / Journal of Obsessive-Compulsive and Related Disorders 1 (2012) 153158

phenomenon (Elliott & Radomsky, 2009). Herba and Rachman

(2007), for example, found that contact contamination fear,
disgust sensitivity, and prior unwanted sexual contact were
signicant predictors of feelings of inward contamination and
urges to wash. In the present study we extended this work by
examining the relationship between mental pollution and variables from three additional domainsreligious, familial, and
childhood traumathat theoretically might inuence ones sense
of internal uncleanliness and urges to wash away unacceptable
thoughts. These domains are described next.
One such domain is religiosity (i.e., strength of religious faith;
Plante, & Boccaccini, 1997). Certain religious practices, such as
baptism in Christianity or wudu in Islam, instruct followers that
water purication rituals can symbolically cleanse ones body and
conscience (Zhong, & Liljenquist, 2006). The Christian bible, for
example, states: arise and be baptized and wash away your sins
(Acts 22:16). Other instructive Bible verses that reference sin by
thought (e.g., I say to you that everyone who looks on a woman
to lust for her has committed adultery already in his heart;
Matthew 5:2728) may also be considered religious directives
that are in line with the concept of mental pollution (e.g., having
an immoral thought leads to feelings of internal dirtiness or
contamination). Thus, those who internalize the notion that
negative intrusive thoughts are sinful might misinterpret such
thoughts as unacceptable or impure and attempt to absolve
themselves of sin or impurity by washing.
Numerous studies have indeed found associations between
religiosity and OC symptoms (e.g., Abramowitz, Deacon, Woods, &
Tolin, 2004; Sica, Norvara, & Sanavio, 2002); as well as with
thought action fusion (TAF), a cognitive bias that underlies the
misinterpretation of intrusive thoughts as signicant (Berman,
Abramowitz, Pardue, & Wheaton, 2010; Rassin & Koster, 2003;
Siev & Cohen 2007; Siev, Chambless, & Huppert, 2010). Given that
TAF and mental pollution both involve overestimating the signicance of thoughts and are positively correlated (Cougle et al.,
2008), it would not be surprising if they also possessed a similar
relationship with religion. Therefore, we hypothesized that
greater strength of religiosity would be positively related to
mental pollution. It should be noted that due to limited religious
variability at the study site, religious-related hypotheses only
involve members of the Christian faith.
In addition to ones degree of religiosity, his or her motivation
for religion might be related to mental pollution. Allport and Ross
(1967) identied two motivational orientations toward religion:
intrinsic and extrinsic. Individuals with high levels of intrinsic
motivation nd their master motive in religion (Allport, & Ross,
1967, p. 434), perceive it to be a guiding factor in their life, and
believe that it offers a moral foundation. Those with high levels of
extrinsic motivation, on the other hand, can be seen as utilitarian,
assessing the importance of religion based on its relationship to
external rewards (e.g., attending church in order to t in
socially; Ellison, 2008; Meek, Albright, & McMinn, 1995). Given
that intrinsic motivation is associated with the tendency to
internalize religious directives (e.g., sin by thought) to a greater
degree than extrinsic motivation, we hypothesized that mental
pollution would be associated with higher levels of intrinsic, but
not extrinsic, motivation for the Christian religion.
Family factors, such as parenting, might also be associated
with mental pollution. Certain parenting strategies, such as guilt
induction, might mold childrens perception of morality as well as
their understanding of which thoughts and feelings are acceptable. As a result, guilt induction might mirror the hypothesized
relationship between religiosity and mental pollution. Two forms
of parental guilt-induction have been identied: (a) self-serving
elicitation, in which the parent over-emphasizes the sacrices he
or she has made for the child while simultaneously limiting the

childs autonomy; and (b) disparagement, in which the parent

perseverates over the childs (even minor) transgressions and
inappropriately places blames on the child (Donatelli, Bybee, and
Buka, 2007). Research suggests that chronic exposure to high
levels of parental guilt induction interferes with the childs
autonomy, emotional and social growth, and management of
interpersonal conicts (Rakow, Forehand, McKee, Coffelt, and
Fear, 2009). In a previous study, Cougle et al. (2008) found that
feelings of guilt were associated with mental pollution; however,
guilt induction was not examined. It is possible that as a result of
parents instilling persistent feelings of guilt in their child, he or
she would develop feelings of internal dirtiness when confronted
with certain thoughts or feelings. Accordingly, we hypothesized
that maladaptive parental guilt induction would positively predict mental pollution.
Childhood traumas are yet another factor that could be related
to mental pollution (Salkovskis, Shafran, Rachman, and Freeston,
1999). For example, Fairbrother and Rachman (2004) found
elevated mental pollution and post-assault washing behaviors in
a sample of sexual assault victims. Moreover, these participants
reported that the memory of the assault triggered feelings of
dirtiness, urges to wash, and engagement in washing behavior. In
an experimental study, Fairbrother et al. (2005) had female
undergraduates listen to an audiotape describing a consensual
or a nonconsensual kiss. They found that participants who heard
about the non-consensual kiss reported greater internal and
external dirtiness, greater distress, and engaged in washing of
their hands and mouth more frequently than those who heard
about a consensual kiss. Accordingly, we predicted that childhood
trauma would predict mental pollution.
Very little research has been conducted on mental pollution to
date, and elucidating variables that predict this phenomenon can
help shed light on possible development and maintenance factors,
as well as possible considerations for assessment, treatment, and
prevention. Accordingly, in the current study we aimed to evaluate
theoretically relevant religious, familial, and trauma-related factors
as predictors of mental pollution. We hypothesized that mental
pollution would be associated with four factors: (1) Christian
religiosity (2) intrinsic motivation toward the Christian religion,
(3) parental guilt induction, and (4) childhood trauma. As in
previous research on mental pollution (e.g., Elliott, & Radomsky,
2009; Fairbrother et al., 2005; Herba, & Rachman, 2007; Radomsky,
& Elliott, 2009), we examined our hypothesis in a large unscreened
sample of undergraduate students in order to maximize sample
size (the prevalence of mental pollution in clinical OCD samples is
not known, but assumed to be rather small), capitalize on the fact
that OC symptoms, such as mental pollution, occur on a continuum
of severity in the population (e.g., Haslam, Williams, Kyrios, McKay,
& Taylor, 2005; Olatunji, Williams, Haslam, Abramowitz, & Tolin,
2008), and take advantage of the distribution of our predictor
variables in the general population. Moreover, the only psychometrically validated self-report measure of mental pollution, the
Mental Pollution Questionnaire (MPQ; Cougle et al., 2008), was
developed using a nonclinical sample.

2. Methods
2.1. Participants
Four hundred and seventeen undergraduate students in introductory psychology classes at a large university in the southeastern United States completed an
online questionnaire battery. Given that the current study is interested in
variables related to religiosity, only individuals who identied themselves as
afliating with a major religious group were included. This yielded 265 participants (63.5% of total sample) of the following religious groups: Catholics (33.6%)
and Protestants (66.3%). A small number of Jewish, Muslim, and Hindu

N.C. Berman et al. / Journal of Obsessive-Compulsive and Related Disorders 1 (2012) 153158
participants completed the study measures, but were not included in further
analyses because of a sample size below 10. The remaining sample (72 men
[27.2%], 192 women [72.5%], and 1 non-response [.3%]) was 76.4% Caucasian, 4.9%
Hispanic, 9.6% African American, 5.3% Asian, and 3.8% reporting being of other
ethnic backgrounds. The groups mean age was 19.46 years (SD 2.75).

2.2. Procedure
After signing up via an Internet based software program and providing
informed consent, participants were directed to a secure project website where
they completed a demographic questionnaire and study measures (see Measures
section) in the same order. All data were collected using Qualtrics, an online web
survey development tool. Coles, Cook, and Blake (2007) found that the administration of anxiety-related self-report measures using Internet-based and paperand-pencil formats yield highly comparable results. Upon accessing the secure
project website, participants were presented with an instructions page on the
screen. The demographic and study questionnaires then appeared on subsequent
pages. Participants received course credit for completing these measures. At the
end of the last questionnaire, a debrieng statement was presented.

155

3. Data analytic strategy

To test our hypotheses, we rst computed correlation coefcients
to examine zero-order relationships among our predictor variables.
We then performed two hierarchical regression analyses with the
subscales of the MPQ (washing rituals and inward contamination) as
the respective dependent variables. In both of these analyses,
religious afliation (dummy coded with Protestants as the reference
group) was entered in Step 1 to control for possible group differences within Christianity. In Step 2, we added constructs that
assessed the importance of ones religious experience (i.e., strength
of religiosity and intrinsic, extrinsic-personal, and extrinsic-social
motivational orientation). In Step 3 of the regressions, both subscales of parental guilt induction (self-serving elicitation and disparagement) were included. Finally, in Step 4, all 5 subscales of the
CTQ (physical and emotional abuse, emotional neglect, sexual abuse,
and physical neglect) were entered simultaneously.

2.3. Measures

4. Results
Demographics. At the onset of the online questionnaire, participants were
asked to report their gender, age, and their identied racial or ethnic group and
religious afliation.
Childhood Trauma Questionnaire Short Form (CTQ-SF; Bernstein et al. 2003).
The CTQ-SF is a 28-item self-report questionnaire that uses retrospective report to
assess for child maltreatment. Participants rate responses on a 5-point Likert-type
scale (0 Never True to 4 Very Often True). Five subscales are included in this
measure (reported reliabilities are based on community samples): physical abuse
(e.g., People in my family hit me so hard that it left me with bruises or marks,
a .83) and emotional abuse (e.g., People in my family said hurtful or insulting
things to me, a .87), emotional neglect (e.g., There was someone in my family
who helped me feel that I was important or special, reverse coded, a .91),
sexual abuse (e.g., Someone threatened to hurt me or tell lies about me unless I
did something sexual with them, a .92), and physical neglect (e.g., I had to
wear dirty clothes growing up, a .61). As reported, the subscales possess
moderate to high internal consistency. Although, the physical neglect subscale
demonstrates weak internal consistency and should thus be interpreted with
caution.
Intrinsic Extrinsic Revised (I/E - R; Gorsuch & McPherson, 1989). The I/E-R is a
14-item self-report scale that measures intrinsic (e.g., I try hard to live all my life
according to my religious beliefs; a .83) and extrinsic motivation for religion.
Two types of extrinsic motivation are assessed: social extrinsic (Es; I go to
religious services mostly to spend time with my friends; a .58) and personally
extrinsic (Ep; What religion offers me most is comfort in times of trouble and
sorrow; a .57). Each item is rated on a 5-point Likert scale (1 strongly disagree,
5 strongly agree). It is recognized that the reliabilities of the extrinsic scales are
not strong, but this measure of religious motivation best assesses the constructs of
interest.
Maladaptive Guilt-Induction Measure (MGI; Donatelli et al., 2007). MGI is a 12item self-report questionnaire assessing youths perceived experiences with
parental guilt induction. Respondents are presented with several statements
regarding guilt induction and are asked to rate the truthfulness of each statement
on a Likert scale from 1 (not at all true) to 7 (very true). The MGI has two
empirically demonstrated subscales. The disparagement subscale assesses child
directed criticism and blame (e.g., [My primary caregiver] makes me feel guilty
even when its not my fault; a .84). The self-serving elicitation subscale assesses
the frequency that parents exaggerate sacrices that they have made for the child
(e.g., [My primary caregiver] always reminds me of favors and sacrices he/she
has made; a .77). As demonstrated, both subscales have good internal
consistency.
Mental Pollution Questionnaire (MPQ; Cougle et al., 2008). MPQ is an eight-item
self-report questionnaire. Respondents are presented with several statements and
are asked to rate agreement on a Likert scale from 1 (strongly disagree) to 7
(strongly agree). The MPQ has two empirically supported subscales. The washing
rituals subscale assesses washing rituals performed in response to perceived
mental pollution (I wash my hands when I feel guilty; a .86). The inward
contamination subscale pertains to a sense of inward contamination not linked to
washing behaviors (Sometimes I feel dirty inside; a .85).
Santa Clara Religious Faith Scale (SCRFS; Plante & Boccaccini, 1997). The SCRFS
is a 10-item self-report scale, which provides a reliable and valid measure of ones
strength of religiosity (e.g., I pray daily; My relationship with God is extremely
important to me). Total scores range from 1040, with higher scores indicating
greater religiosity. Psychometric research demonstrates that scoring at or above
33 indicates high religiosity (Plante & Boccaccini, 1997). The SCRFS has good
reliability (a .92.95) and converges with other valid measures of religiosity
(Plante & Boccaccini, 1997).

4.1. Sample Characteristics

Table 1 presents the mean, standard deviation, and range for
each study measure. For these measures, the descriptive statistics
approximate those reported in other nonclinical samples
(Bernstein et al., 2003; Cougle et al., 2008; Donatelli et al.,
2007; Gorsuch & McPherson, 1989; Plante & Boccaccini, 1997).
4.2. Correlations among study Measures
Zero-order correlations among the study variables are presented in Table 2. As can be seen, the MPQ subscales were
signicantly correlated with only one measure of religiosity:
IER-Es. Both MGI subscales demonstrated moderately strong
positive correlations with the MPQ-WR, whereas relationships
with the MPQ-IC were weaker in magnitude, but still signicant.
Similarly, all CTQ subscales demonstrated strong positive correlations with the MPQ-WR and weaker, positive, and signicant
correlations with the MPQ-IC subscale.
4.3. Regression Analyses
Table 3 presents the results of the regression analysis predicting scores on the MPQWR subscale. As can be seen, in step one,
Table 1
Descriptive statistics for study measures.
Measure
Religiosity
Santa Clara Religious Faith Scale (SCRFS)
Motivational Orientation
Intrinsic
Extrinsic Social
Extrinsic Personal
Guilt Induction
Disparagement
Self-Serving Elicitation
Childhood Trauma
Emotional Abuse
Physical Abuse
Sexual Abuse
Emotional Neglect
Physical Neglect
Mental Pollution
Inward Contamination
Washing Rituals

SD

Range

29.03

8.19

10-40

25.98
7.29
9.98

6.53
2.57
2.73

12-40
3-15
3-15

10.83
12.20

7.35
8.13

6-42
6-42

2.27
1.51
.55
7.07
1.34

3.39
2.60
2.19
6.08
2.62

019
016
017
032
014

11.23
5.85

6.45
3.46

4 - 27
4-25

156

N.C. Berman et al. / Journal of Obsessive-Compulsive and Related Disorders 1 (2012) 153158

Table 2
Zero-Order Correlations Among Study Measures.
Variable

Int M

Ext - S

Ext - P

MGI-D

MGI-SS

CTQ-EA

CTQ-PA

CTQ-SA

CTQ-EN

CTQ-PN

MPQ-WR

MPQ-IC

SCRFS
Int M
Ext -S
Ext - P
MGI-D
MGI-SS
CTQ-EA
CTQ-PA
CTQ-SA
CTQ-EN
CTQ-PN
MPQ-WR

.81nn

-

.17n
.09
-

.36nn
.14n
.30nn
-

 .12
 .07
.07
.03
-

 .14n
 .10
.12nn
.09
.87nn
-

 .10
 .07
.02
.02
.59nn
.53nn
-

.06
.03
.12
.14n
.42nn
.39nn
.72nn
-

.02
.02
.16n
.05
.30nn
.28nn
.48nn
.61nn
-

 .12
 .09
 .09
 .04
.49nn
.43nn
.62nn
.41nn
.21nn
-

 .04
 .004
.22nn
.04
.43nn
.37nn
.58nn
.58nn
.50nn
.61nn
-

.07
.04
.31nn
.12
.43nn
.36nn
.37nn
.46nn
.49nn
.29nn
.48nn
-

.08
.08
.13n
.05
.16n
.19nn
.25nn
.19nn
.22nn
.13n
.10
.39nn

Note. SCRFS Religiosity, Int M Intrinsic Motivation, Ext-S Extrinsic Social, Ext-P Extrinsic Personal, MGI-D Guilt Induction-Disparagement, MGI-SS Self Serving,
CTQ-EA Emotional Abuse, CTQ-PA Physical Abuse, CTQ SA Sexual Abuse, CTQ-EN Emotional Neglect, CTQ-PNPhysical Neglect, MPQ-WR Washing Rituals, MPQIC Inward Contamination.
p o.001.
p o .05.

nn
n

Table 3
Regressions Predicting Washing Rituals subscale of the MPQ.
Variable
MPQ Washing Rituals
Step 1
Catholic
Step 2
Strength of Religiosity
Intrinsic Motivation
Extrinsic-Social
Extrinsic-Personal
Step 3
Guilt Induction - Disparagement
Guilt Induction Self-Serving
Step 4
Emotional Abuse
Physical Abuse
Sexual Abuse
Emotional Neglect
Physical Neglect

R2

Table 4
Regressions Predicting Inward Contamination subscale of the MPQ.
t

.001
.03

.37

.02
.009
.33
 .001

.15
.08
4.88
 .02

.47
 .09

4.42
 .80

 .11
.16
.25
 .01
.15

 1.11
1.71
3.31
 .12

.11

.27

.40

Variable

n.s.
n.s.
o .001
n.s.
n.s.
o .001
n.s.
o .001
o .001
n.s.
o .001
n.s.
n.s.
o .001
n.s.
n.s.

MPQ Inward Contamination

Step 1
Catholic
Step 2
Strength of Religiosity
Intrinsic Motivation
Extrinsic-Social
Extrinsic-Personal
Step 3
Guilt Induction Disparagement
Guilt Induction Self-Serving
Step 4
Emotional Abuse
Physical Abuse
Sexual Abuse
Emotional Neglect
Physical Neglect

religious afliation (with Protestants as the reference group)

explained a negligible and nonsignicant portion of the variance
(R2 .001, p 4.05). Religious-related variables entered in Step 2,
however, explained a signicant amount of variance in MPQ-WR
scores (DR2 .11, p o.001), with the IER-ES subscale emerging as
a signicant individual predictor. In Step 3, adding the MGI
subscales accounted for a signicant amount of additional variance (DR2 .16, p o.001), with the disparagement subscale
emerging as a signicant individual predictor. Finally, in Step 4,
adding the subscales of the CTQ accounted for a signicant
amount of additional variance (DR2 .16, po.001). Of all the
childhood trauma subscales, only the sexual abuse subscale
emerged as a signicant individual predictor. The nal model
with all variables entered, accounted for 40% of the variance in
washing rituals scores.
In the prediction of MPQ-IC scores (Table 4), results of the
hierarchical regression analyses were as follows: In step one,
religious afliation (with Protestants as the reference group)
explained a negligible and nonsignicant portion of the variance
(R2 .001, p4.05). Religious-related variables entered in Step
2 did not explain a signicant amount of additional variance
(DR2 .019, p4 .05). In Step 3, however, the addition of the MGI
subscales accounted for a signicant amount of additional variance (DR2 .04, p o.05). Finally, in Step 4, adding the subscales of

R2

.001

 .006

.04
.13
.11
.02

.55
1.07
1.56
.29

.02
.18

.15
1.48

.28
 .03
.15
.02
 .16

2.36
 .27
1.59
.24
 1.57

.001
.02

.06

.12

n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
o.05
n.s.
n.s.
o.05
o.05
n.s.
n.s.
n.s.
n.s.

the CTQ accounted for further variance (DR2 .06, p o.05). Of all
the childhood trauma subscales in Step 4, only the emotional
abuse subscale emerged as a signicant individual predictor of
MPQ-IC scores. The nal model with all variables entered,
accounted for 12% of the variance in inward contamination scores.

5. Discussion
Mental pollution is an understudied presentation of OC symptoms in which the person experiences a feeling of internal
uncleanness, often accompanied by ritualistic washing to
expunge the thought and mental contamination. As a less
understood symptom presentation, mental pollution also represents a potential stumbling block for clinicians treating individuals with OCD. Accordingly, it is important to better understand
experiences that predict this phenomenon. The present investigation examined the relationship between mental pollution and
several variables that, at a conceptual level, appear to be associated with this phenomenon: degree of religiosity, motivational
orientation toward religion, parental guilt induction, and childhood trauma. Given the lack of research on such relationships, our
aim was to gain a clearer understanding of factors that explain
mental pollution in hopes of providing avenues for future

N.C. Berman et al. / Journal of Obsessive-Compulsive and Related Disorders 1 (2012) 153158

research on potential risk and maintenance factors, as well as

effective prevention and treatment programs.
In summary, our ndings were as follows: Contrary to our
predictions, mental pollution was not associated with degree of
religiosity in our sample, and only to the social subtype of
extrinsic motivation for religiona surprising nding. As
expected, both subscales of mental pollution were positively
associated with participants parents maladaptive guilt induction
strategies, as well as with all types of childhood trauma.
Our regression analyses indicated that religious afliation
(Step 1) within Christianity was not a meaningful predictor of
either component of mental pollution. However, strength of
religiosity and motivation toward religion (entered together in
Step 2) predicted the washing rituals dimension of mental
pollution, but not inward contamination. Of these variables, being
extrinsically motivated by social factors, such as power, politics,
or friends, was found to possess unique predictive value for
washing rituals. Maladaptive parental guilt induction (Step 3)
signicantly predicted both types of mental pollution, with
disparagement acting as an independent predictor of washing
rituals. Finally, childhood traumas (Step 4) signicantly predicted
both types of mental pollution, with sexual abuse emerging as a
signicant predictor of washing rituals and emotional abuse as a
predictor of inward contamination.
Whereas our hypotheses relating to religiosity were not
supported, the results of the present study generally provided
support for our other predictions regarding correlates of mental
pollution. Unexpectedly, an extrinsic motivation for religion,
wherein individuals utilize their faith to enhance social support
or power, predicted greater engagement in hand washing following thoughts of inward contamination. This relationship may be
due to each constructs observable nature that functions to elicit a
favorable judgment from oneself and others. Washing rituals are
overt behaviors that, by denition, are observable. Thus, one
might be motivated to allow others to see them washing their
hands in order to be perceived as a good person. Similarly,
observable behaviors are the crux of an extrinsically motivated
individual (e.g., wanting to be seen at church), thus, being seen
by others while going to church might also lead to being
perceived as a good and moral person.
Contrary to the hypotheses related to religion, our prediction that
parental maladaptive guilt induction would be positively associated
with mental pollution (both inward contamination and washing
rituals) was supported. In fact, guilt induction was the strongest
independent predictor of inward contamination. This is consistent
with Cougle et al.s (2008) ndings that general feelings of guilt
were strongly correlated with inward contamination, as well as with
Salkovskis et al.s (1999) assertion that rigid rules within the home
contribute to the development of dysfunctional beliefs and misinterpretations of intrusive thoughts, feelings, and behaviors. Moreover, Zhong and Liljenquist (2006) demonstrated that threats to
ones moral image (e.g., blaming the child inappropriately for moral
transgressions) could lead to the psychological need to engage in
cleansing behaviors. In this vein, persistent or intensied feelings of
guilt, perhaps induced by parents, would lead to stronger urges or
attempts to morally cleanse themselves (Zhong & Liljenquist, 2006).
This hypothesis, however, requires further study.
Our hypotheses related to childhood trauma were also generally supported, indicating that experiences with childhood
trauma positively predict mental pollution and explain the greatest amount of variance in mental pollution-related washing
rituals. Consistent with ndings from Fairbrother and Rachman
(2004), Fairbrother et al. (2005), and Olatunji, Elwood, Williams,
and Lohr (2008), thoughts related to sexual trauma in particular
predicted mental pollution, specically engagement in washing
rituals. Perhaps this type of trauma elicits feelings of moral

157

uncleanliness, which as Zhong and Liljenquist (2006) argued,

activate a need for physical cleansing. Thus, the cleansing behaviors in mental pollution might be conceptualized as attempts to
repair ones sense of moral integrity.
Childhood trauma was also positively associated with feelings of inward contamination. For this dimension of mental
pollution, however, emotional abuse had the greatest predictive
value. This relationship may be explained by the repeated
internalization of emotionally hurtful remarks. Given that emotional abuse tends to involve repeated statements or behaviors
that target the victims worth or sense of self (e.g., insults, verbal
assaults, name calling, feeling hated by family), the child victim
might internalize the abuse and come to view the comments or
behaviors as an accurate reection of him or herself as intrinsically disgusting or worthless, leading to feelings of inward
contamination.
Inasmuch as the results of the present study can be generalized to treatment-seeking individuals with clinical levels of
mental pollution and contamination-related OC symptoms, the
present study has implications for the assessment and cognitivebehavioral treatment. First, it seems critical to thoroughly assess
for mental pollution in patients with contamination-related OC
symptoms. This is likely overlooked by clinicians trained to assess
for largely external contaminants that might contain germs or
poisons, such as bathrooms, door knobs, chemicals, and the like.
Second, in patients with mental pollution, clinicians should assess
for possible childhood trauma (specically emotional and sexual
abuse), and inquire about exposure to parenting strategies that
involve maladaptive guilt induction (e.g., blaming the child for
negative outcomes when he/she was not at fault). Interventions
addressing guilt and shame associated with these events might
help the patient engage in, and benet from, imaginal exposures
to intrusions that provoke mental pollution (e.g., asking the
patient to write statements that evoke feelings of mental pollution; Rachman, 2004). If the patient is still living with his or her
family and high levels of parental guilt induction are present, it
might be useful to utilize family-based therapy approaches to
increase effective communication strategies (Dattilio, & Epstein,
2004).
For a number of reasons, caution is warranted regarding the
conclusions drawn from the present ndings. First, although
mental pollution appears to exist on a continuum of severity in
the general population, the use of a nonclinical sample potentially
limits the generalizability of our ndings to clinical populations.
Additional studies with those seeking treatment for obsessional
problems are warranted. Second, the current study used a
correlational design, which limits conclusions regarding causality.
Longitudinal designs or experimental manipulations are necessary next steps to better understand how the experiences that we
examined relate to mental pollution.
Third, only self-report measures were used. As it is optimal to
have multiple measures or assessment modalities (Kazdin, 2002),
in vivo measures of mental pollution (Elliott, & Radomsky, 2009) or
guilt induction (e.g., Donatelli et al., 2007) would improve the study
of these psychological constructs. In a similar vein, the use of
retrospective self-report questionnaires for assessing childhood
traumas and childrearing practices is problematic. Bernstein et al.
(1994) note that retrospective report of trauma can be inuenced by
repression, denial, or embarrassment. Thus, additional assessments
could be used to corroborate or objectively verify reports, such as
the Childhood Trauma Interview (Fink, Bernstein, Foote, Lovejoy,
Ruggiero, & Handelsman, 1993) or examining Child Protective
Service records. Fourth, the limited types of religious afliations
restricted a thorough investigation of the relationship between
religion and mental pollution. The current study only examined
religious group differences between Protestants and Catholics.

158

N.C. Berman et al. / Journal of Obsessive-Compulsive and Related Disorders 1 (2012) 153158

Consequently, the relationship between strength of religiosity and

motivational orientation towards religion is only relevant for individuals who afliate with these branches of Christianity. The current
study only examined individuals afliating with Christianity, therefore, we cannot draw broad conclusions about the predictive value
of religiosity in other faiths. However, given that (a) Christian
religiosity did not currently emerge as a signicant predictor of
mental pollution and (b) Christian religiosity tends to more strongly
predict OCD-related cognitive processes when compared to other
religious faiths (Siev et al., 2010; Berman et al., 2010), we hypothesize that religious afliation and degree of religiosity, more broadly,
would not be signicant predictors of mental pollution. In order to
empirically evaluate this question, future research is needed.
An additional weakness is the limited number of constructs
evaluated. The amount of variance explained in mental pollution by
the predicted variables in our regression models, leaves open the
possibility that additional factors contribute to mental pollution.
Moreover, it is clear from our regression models that the included
constructs better predict washing rituals (40%), when compared to
inward contamination (12%). This prediction pattern indicates that
inward contamination and washing rituals do not possess the same
developmental correlates and potential risk factors, which aligns
with Cougle et al.s (2008) ndings that these types of mental
pollution are distinct factors, at least in unscreened, nonclinical
samples. In order to better understand inward contamination
specically, future research should examine how theoretically
relevant developmental constructs, like parental psychological control (i.e., attempts to control a childs psychological development by
limiting the childs thinking processes, emotional expression, and
autonomy; Barber, 1996) or other pathways to inated responsibility (Coles, & Schoeld, 2008) predict mental pollution.
References
Abramowitz, J., Deacon, B., Olatunji, B., Wheaton, M., Berman, N. C., Losardo, D.,
et al. (2010). Assessment of obsessive-compulsive symptom dimensions:
Development and validation of the Dimensional Obsessive-Compulsive Scale.
Psychological Assessment, 22(1), 180198.
Abramowitz, J. S., Deacon, B., Woods, C., & Tolin, D. (2004). Association between
protestant religiosity and obsessive compulsive symptoms and cognitions.
Depression and Anxiety, 20(2), 7076.
Abramowitz, J. S., Franklin, M., Schwartz, S., & Furr, J. (2003). Symptom presentation and outcome of cognitive-behavioral therapy for obsessive-compulsive
disorder. Journal of Consulting and Clinical Psychology, 71(6), 10491057.
Allport, G., & Ross, J. (1967). Personal religious orientation and prejudice. Journal of
Personality and Social Psychology, 5, 424443.
American Psychiatric Association (2000). Diagnostic and statistical manual of
mental disorders ((4th ed.). Washington, DC: American Psychiatric Association.
Barber, B. K. (1996). Parental psychological control: Revisiting a neglected
construct. Child Development, 67, 32963319.
Berman, N. C., Abramowitz, J., Pardue, C., & Wheaton, M. (2010). The relationship
between religion and thought action fusion: Use of an in vivo paradigm.
Behavior Research and Therapy, 48, 670674.
Bernstein, D. P., Fink, L., Handelsman, L., Foote, J., Lovejoy, M., Wenzel, K., et al.
(1994). Initial reliability and validity of a new retrospective measure of child
abuse and neglect. American Journal of Psychiatry, 151, 11321136.
Bernstein, D., Stein, J., Newcomb, M., Walker, E., Pogge, D., Ahluvalia, T., et al.
(2003). Development and validation of a brief screening version of the
Childhood Trauma Questionnaire. Child Abuse and Neglect, 27, 169190.
Coles, M. E., & Schoeld, C. A. (2008). Assessing the development of inated
responsibility beliefs: The Pathways to Inated Responsibility Beliefs Scale.
Behavior Therapy, 39(4), 322335.
Cougle, J. R., Lee, H., Horowitz, J. D., Wolitzky-Taylor, K. B., & Telch, M. (2008). An
exploration of the relationship between mental pollution and OCD symptoms.
Journal of Behavior Therapy and Experimental Psychiatry, 39, 340353.
Dattilio, F., & Epstein, N. (2004). Cognitive behavioral couple and family therapy.
In: T. L. Sexton, G. R. Weeks, & M. S. Robbins (Eds.), The Family Therapy
Handbook. New York: Routledge.

Donatelli, J., Bybee, & Buka, S. (2007). What do mothers make adolescents feel
guilty about? Incidents, reactions, and relation to depression. Journal of Child
and Family Studies, 16, 859875.
Elliott, C. M., & Radomsky, A. S. (2009). Analyses of mental contamination: Part I,
experimental manipulations of morality. Behavior Research and Therapy, 47,
9951003.
Ellison, L. (2008). Identifying overguilt: An investigation of the relationship
between guilt, religious motivation, and adaptation to college in religious
college students. Dissertation Abstracts International Section A: Humanities and
Social Sciences, 68, 7 A.
Fairbrother, N., Newth, S. J., & Rachman, S. (2005). Mental pollution: Feelings of
dirtiness without physical contact. Behavior Research and Therapy, 43,
121130.
Fairbrother, N., & Rachman, S. (2004). Feelings of mental pollution subsequent to
sexual assault. Behavior Research and Therapy, 40, 173189.
Fink, U., Bernstein, D., Foote, J., Lovejoy, M., Ruggiero, J., & Handelsman, U. (1993).
The Childhood Trauma Interview: reliability, validity, and relationship to personality disorders. Proceedings of the Third International Congress on the Disorders of
Personality. Roskilde, Denmark: International Society for the Study of Personality Disorders.
Gorsuch, R., & McPherson, S. (1989). Intrinsic/extrinsic measurement: I/E-revised
and single-item scales. Journal for the Scientic Study of Religion, 28, 348354.
Haslam, N., Williams, B., Kyrios, M., McKay, D., & Taylor, S. (2005). Subtyping
obsessive-compulsive disorder: A taxometric analysis. Behavior Therapy, 36,
381391.
Herba, J. K., & Rachman, S. (2007). Vulnerability to mental contamination. Behavior
Research and Therapy, 45, 28042812.
Kazdin, A. (2002). Research Design in Clinical Psychology ((4th ed.). Boston: Allyn
and Bacon.
Mataix-Cols, D., Rosario-Campos, M. C., & Leckman, J. F. (2005). A multidimensional model of obsessive-compulsive disorder. American Journal of Psychiatry,
162, 228238.
McKay, D., Abramowitz, J. S., Calamari, J. E., Kyrios, M., Radomsky, A. S., Sookman, D.,
Taylor, S., & Wilhelm, S. (2004). A critical evaluation of obsessive-compulsive
disorder subtypes: Symptoms versus mechanisms. Clinical Psychology Review, 24,
283313.
Meek, K., Albright, J., & McMinn, M. (1995). Religious orientation, guilt, confession,
and forgiveness. Journal of Psychology and Theology, 23(3), 190197.
Olatunji, B., Elwood, L., Williams, N., & Lohr, J. (2008). Mental pollution and PTSD
symptoms in victims of sexual assault: A preliminary evaluation of the
mediating role of trauma-related cognitions. Journal of Cognitive Psychotherapy,
22(1), 3747.
Olatunji, B., Williams, B., Haslam, N., Abramowitz, J., & Tolin, D. (2008). The latent
structure of obsessive-compulsive symptoms: A taxometric study. Depression
and Anxiety, 25, 956968.
Plante, T. G., & Boccaccini, M. (1997). The Santa Clara Strength of Religious Faith
Questionnaire. Pastoral Psychology, 45(5), 375387.
Rachman, S. (1994). Pollution of the mind. Behavior Research and Therapy, 32,
311314.
Rachman, S. (2004). Fear of contamination. Behavior Research and Therapy, 42,
12271255.
Rachman, S. (2006). The Fear of Contamination: Assessment and Treatment (Cognitive
Behavior Therapy: Science & Practice). Oxford: Oxford University Press.
Radomsky, A. S., & Elliott, C. M. (2009). Analyses of mental contamination: Part II,
individual differences. Behavior Research and Therapy, 47, 10041011.
Rakow, A., Forehand, R., McKee, L., Coffelt, N., Champion, J., Fear, J., et al. (2009).
The relation of parental guilt induction to child internalizing problems when a
caregiver has a history of depression. Journal of Child and Family Studies, 18,
367377.
Rassin, E., & Koster, E. (2003). The correlation between thoughtaction fusion and
religiosity in a normal sample. Behavior Research and Therapy, 41, 361368.
Salkovskis, P., Shafran, R., Rachman, S., & Freeston, M. H. (1999). Multiple pathways to inated responsibility beliefs in obsessional problems: Possible
origins and implications for therapy and research. Behavior Research and
Therapy, 37, 10551072.
Sica, C., Norvara, C., & Sanavio, E. (2002). Religiousness and obsessive-compulsive
cognitions and symptoms in an italian population. Behavior Research and
Therapy, 40, 813823.
Siev, J., & Cohen, A. B. (2007). Is thoughtaction fusion related to religiosity?
Differences between christians and jews. Behavior Research and Therapy, 45,
829837.
Siev, J., Chambless, D., & Huppert, J. (2010). Moral thought-action fusion and OCD
symptoms: The moderating role of religious afliation. Journal of Anxiety
Disorders, 24, 309312.
Zhong, C., & Liljenquist, K. (2006). Washing away your sins: Threatened morality
and physical cleansing. Science, 313, 14511452.