Comparing the effectiveness of brief writing tasks in

reducing feelings of mental contamination.

Phoebe Horrocks

Submitted for the Degree of

Doctor of Psychology(Clinical Psychology)

School of PsychologyFaculty of Health and Medical Sciences

University of SurreyGuildford, SurreyUnited KingdomSeptember 2016

Abstract

1

Background and objectives: Mental contamination (MC) refers to feelings of internal

dirtiness that can arise without direct physical contact with a contaminant. MC is also

associated with a variety of negative emotions including disgust, fear, anger, shame,

guilt and revulsion. Previous research has shown that MC can be evoked by recalling

an autobiographical memory of being the victim of a moral transgression. This study

sought to extend these findings to explore further the MC reaction using

autobiographical memories. In addition, previous research has found that washing

related tasks are ineffective in reducing feelings of MC. This study explored whether

brief writing interventions could help to reduce the MC reaction.

Method: The current study used a non-clinical adult sample and induced MC through

asking participants to recall a time they were the victim of a moral transgression

(N=93). Participants were then randomised into three groups to complete a writing

task (N=74). The writing tasks included: writing about a normal day (control), a self-

compassion writing task and a self-esteem writing task. The study examined first

whether MC would be induced, and second whether writing tasks were effective in

reducing MC.

Results: The autobiographical victim memory recall task induced feelings of MC

(feelings of internal dirtiness, anxiety, shame, guilt, fear, sadness and humiliation).

The largest effect sizes were seen for humiliation, shame and sadness. Feelings of

MC significantly reduced after the writing tasks in the whole sample; no writing task

intervention showed a superior effect.

Conclusions: MC reactions can be induced by the memory of being victim of

‘everyday’ transgressions. The main impact of these is on humiliation, shame and

sadness. The current study supports other literature suggesting that the MC reaction

decays in the absence of active interventions. However, future studies are required

with larger sample sizes and examining other interventions.

Abbreviations: MC=Mental Contamination, MP=Mental Pollution, INE=Internal

negative emotions (such as shame and guilt), ENE=External negative emotions (such

as anger and anxiety.

2

Acknowledgements:

I would like to thank all those who have assisted me throughout the course in various

ways, and without whom my research would not have been possible. In particular, I

would like to say thank you to my dedicated research supervisor Dr Laura Simonds,

who provided essential support, patience and guidance throughout the project.

Without her I would not have survived the last three years and been able to submit

my MRP. Her calm, helpful and encouraging demeanor has been indispensable to me

over the last three years; particularly during difficult times. I would also like to thank

my clinical tutors Dr Sarah Johnstone and Dr Heinz Kobler for supervising my

personal and professional development over the three years of training.

I would like to express my gratitude to all those people who helped me to recruit

participants for my study by distributing it further than I would have been able to

alone. I would also like to thank all the people who volunteered to take part and gave

up their time to complete this study.

Finally, I would like to thank my housemates (Lucy and Sarah), family and friends

for their unwavering support, encouragement, help and patience during the last three

years. In particular, my mother for encouraging me to complete this course even

during times when I know you needed more of my help. To my friends and

colleagues in Cohort 42, you have also helped me throughout the course and I am

glad to have been on this journey with you.

3

Contents

MRP Empirical Paper: Comparing the effectiveness of brief writing tasks in

reducing feelings of mental contamination:

Abstract…………………………………………………pg.5

Introduction……………………………………………. pg. 6

Method………………………………………………….pg. 16

Results…………………………………………………..pg. 23

Discussion………………………………………………pg. 39

References………………………………………………pg. 49

Appendices……………………………………………...pg. 60

MRP Proposal: Investigating factors associated with mental

contamination.................................................................pg. 170

Literature Review: Reviewing the current state of knowledge on the relatively new concept of Mental Contamination…….................pg. 190

Overview of clinical experience………………………..pg. 243

Table of assessments…………………………………….pg. 245

4

Abstract

Background and objectives: Mental contamination (MC) refers to feelings of internal

dirtiness that can arise without direct physical contact with a contaminant. MC is also

associated with a variety of negative emotions including disgust, fear, anger, shame,

guilt and revulsion. Previous research has shown that MC can be evoked by recalling

an autobiographical memory of being the victim of a moral transgression. This study

sought to extend these findings to explore further the MC reaction using

autobiographical memories. In addition, previous research has found that washing

related tasks are ineffective in reducing feelings of MC. This study explored whether

brief writing interventions could help to reduce the MC reaction.

Method: The current study used a non-clinical adult sample and induced MC through

asking participants to recall a time they were the victim of a moral transgression

(N=93). Participants were then randomised into three groups to complete a writing

task (N=74). The writing tasks included: writing about a normal day (control), a self-

compassion writing task and a self-esteem writing task. The study examined first

whether MC would be induced, and second whether writing tasks were effective in

reducing MC.

Results: The autobiographical victim memory recall task induced feelings of MC

(feelings of internal dirtiness, anxiety, shame, guilt, fear, sadness and humiliation).

The largest effect sizes were seen for humiliation, shame and sadness. Feelings of

MC significantly reduced after the writing tasks in the whole sample; no writing task

intervention showed a superior effect.

Conclusions: MC reactions can be induced by the memory of being victim of

‘everyday’ transgressions. The main impact of these is on humiliation, shame and

sadness. The current study supports other literature suggesting that the MC reaction

decays in the absence of active interventions. However, future studies are required

with larger sample sizes and examining other interventions. Abbreviations: MC=Mental Contamination, MP=Mental Pollution, INE=Internal negative emotions

(such as shame and guilt), ENE=External negative emotions (such as anger and anxiety.

5

Introduction

Mental Contamination (MC) was first conceptualised by Rachman in 1994

using the term ‘pollution of the mind’. Mental pollution (MP) is defined as “a sense

of internal un-cleanness which can and usually does arise and persist regardless of

the presence or absence of observable dirt” (Rachman, 1994, p.311). This concept

was first developed from clinical observations of OCD patients who expressed that

they could never feel entirely clean even after repeated washing (Rachman, 2004).

MP was later incorporated into the overarching concept of MC which sits within

Rachman’s theory of fear of contamination (2004, 2006).

The fear of contamination is an intense feeling of dirtiness that is

accompanied by negative emotions such as shame and disgust and is usually

followed by a strong urge to wash. Fear of contamination includes two categories:

contact contamination and MC (Rachman 2006). Traditionally, more focus has been

given to contact contamination which arises after direct physical contact with

something that is perceived to be soiled, impure, infectious or harmful (Rachman

2004). More empirical attention is now being given to Rachman’s theory of MC in

which feelings of fear of contamination arise without direct physical contact with a

perceived contaminant.

Both contact and mental contamination are associated with negative emotions

including disgust, fear and revulsion but, MC also includes feelings of anger, shame

and guilt and is associated with an internal feeling of dirtiness (Rachman 2006).

Rachman (2006) theorised that MC can arise from a physical violation (e.g. sexual

assault), a mental violation (e.g. feeling ashamed, degraded, humiliated) or from self-

contamination (e.g. having unwanted or unacceptable thoughts, images or

memories). A central difference between MC and contact contamination is that MC

6

occurs without physical contact with a contaminant. In addition, Rachman (2006)

hypothesised that the source of contamination can be different for MC and contact

contamination; contact contamination tends to occur after physical contact with a

perceived soiled object or place and is associated with danger, disease and dirt (e.g.

bodily wastes, toxins). In contrast, although MC can also occur from non-human

sources, it is proposed that the source is usually human: a dirty, dangerous, immoral

or harmful person (including the self). In contact contamination the sources are often

seen as a threat to most people (e.g. chemicals) and therefore it is seen as easily

transferable to other people or objects (Tolin, Worhunsky & Maltby, 2004). The

theory of MC suggests that it is unique to the affected person and therefore is not

easily transferred to others (Rachman, 2004). Whilst both contact and mental

contamination are proposed to evoke an urge to wash, washing is only effective, in

the short term, in contact contamination where the feelings of dirtiness are limited to

a specific, identifiable bodily location. In MC, given that the feeling of dirtiness is

internal and not localised, it does not effectively respond to washing but may respond

to mental, neutralising rituals (Rachman, 2006). Although there are differences

between mental and contact contamination they are thought to co-occur and share

some overlapping characteristics including urge to wash and discomfort (Coughtrey,

Shafran, Lee & Rachman 2012b).

MC has been observed in Obsessive Compulsive Disorder (OCD), Post-

Traumatic Stress Disorder (PTSD), and victims of sexual assault giving it important

clinical relevance (Oluntaji, Elwood, Williams & Lohr 2008; Warnock-Parkes,

Salkovskis & Rachman 2008; Fairbrother & Rachman 2004). MC is strongly

conceptually linked to OCD because of the role that fear of contamination plays in

the disorder; studies have found that over half of OCD sufferers have contamination

7

fears (Calamari, Weigartz, Riemann, Cohen, Greer et al 2004). Furthermore,

Coughtrey, Shafran, Knibbs & Rachman (2012) study found that nearly half of

participants with OCD also had clinical levels of MC. OCD is also linked to MC

because feelings of MC can be triggered by intrusive thoughts and images

experienced by people with OCD; in OCD it is common for people to be preoccupied

with obsessive thoughts around moral transgressions which can lead to feelings of

shame and guilt. Thoughts or events that challenge ones moral view of the self may

threaten a person’s self-worth and ones moral self-perceptions are associated with

increased OC contamination concerns (Doron & Kyrios, 2005; Doron, Sar-el &

Mikulincer, 2007). It has also been found that exposure to own and others’ moral

indiscretions can trigger a need for physical cleansing (Zhong & Lilijenquist, 2006).

Despite this, MC is not routinely assessed for or treated separately. Widely used

clinical OCD measures, such as the Yale Brown Obsessive Compulsive Scale

(Goodman et al, 1989), do not assess MC. If MC is a distinct phenomenon in OCD,

and one that might require specific intervention strategies, then it may be that lack of

specific assessment and intervention contribute to the high relapse rate and treatment

resistant nature of conditions like OCD (Ponniah, Magiati & Hollon 2013). For

instance, studies have found that OCD sufferers with cleaning compulsions respond

less well to CBT and fear of contamination is thought to underpin cleaning

compulsions (Coehlo & Whittal 2001). Furthermore, the emotions associated with

MC are also present in a number of other depressive and anxiety related conditions

suggesting that MC has the potential to be a trans-diagnostic factor across several

psychological conditions (Tangney, Wagner & Gramzow, 1992; Fergus, Valentiner,

McGrath & Jencius 2010; Kim, Thibodeau & Jorgensen, 2011; Ille, Schoggl,

Kapfhammer, Arendasy, Sommer & Schienle 2014). If MC is a distinct phenomenon

8

in OCD or other psychological disorders it may need specific treatments tailored

towards it; therefore it is important to understand strategies that might be effective in

reducing or preventing MC.

MC is not just found within clinical populations. Most MC research has been

conducted within the general population, indicating that MC can be induced by moral

transgressions in non-clinical samples (Elliott & Radomsky, 2009, 2012, 2013,

Herba & Rachman, 2007, Rachman et al 2012). Furthermore, studies have found that

MC can be evoked and re-evoked by thoughts and memories of moral transgressions

(Coughtrey, Shafran & Rachman, 2014). As exposure to perceived immoral

situations is inevitable (i.e. individuals will frequently encounter situations where

they perceive they have had wrong done to them/they have transgressed against

others and this will affect their view of themselves) the study of MC is important.

The evidence suggests that individuals vary in their sensitivity to MC arising from

perceived immorality and this may be linked with conditions such as anxiety (e.g.

OCD, PTSD) and trauma.

Research supports the idea that MC can be induced experimentally in the

general population through asking participants to imagine scenarios where a moral

transgression takes place. The most common paradigm used to date is a non-

consensual kiss paradigm (the so-called ‘dirty-kiss’). In this paradigm, female

participants are asked to listen to an audiotape which prompts them to imagine being

kissed at a party against their will (Fairbrother, Newth & Rachman 2005; Herba &

Rachman 2007; Elliott & Radomsky 2009, 2012). In these studies women who

imagined a non-consensual kiss reported greater feelings of internal dirtiness, urge to

wash and negative emotions associated with MC than women who imagined a

consensual kiss. Similar findings have been shown in subsequent studies in which

9

men or women were asked to imagine perpetrating a non-consensual kiss (Rachman,

Radomsky, Elliott & Zysk 2012; Waller & Boschen, 2015). This paradigm for

inducing MC, although useful in demonstrating the MC effect, has low ecological

validity given the required level of experimental control. Recent studies have found

MC can also be induced through autobiographical recall of moral transgressions,

both from the victim (i.e. recalling self as the one transgressed against) and

perpetrator (i.e. recalling self as the transgressor) perspective (Coughtrey et al, 2014).

To date, three studies have provided data on potential ways in which feelings

of MC may be ameliorated or amplified. Coughtrey et al (2014) used an

autobiographical memory paradigm to induce MC and found that internal feelings of

dirtiness, urge to wash and anxiety spontaneously decayed after three minutes.

However, re-evoking the memory prevented a return to baseline for as long as twenty

minutes. Similarly, Ishikawa, Kobori, Komuro & Shimuzu, (2014) used the ‘dirty

kiss’ paradigm and explored the persistence of MC and the impact of washing

behaviour. They found that, in both the washing and no washing groups, feelings of

MC decayed after five minutes. However, importantly this study found that ‘internal

negative emotions’ (INEs) that are part of MC (e.g. shame, guilt, humiliation, fear,

sadness) decayed less quickly than ‘external negative emotions’ (ENEs) such as

anxiety, distress, anger, and disgust ’. In past research, INEs have been considered to

be feelings about the self whilst ENEs are feelings about the perpetrator of the

transgression (Elliott & Radomsky, 2009). Taken together, these studies suggest that

some components of the MC response might spontaneously decay more readily than

others. These studies also suggest that washing may have limited effect on feelings

of MC and that re-evoking memories of moral transgressions might prevent the

decay of MC feelings. The findings of Ishikawa et al’s (2014) study, that INEs take

10

longer to decay than ENEs, is important as it suggests interventions might be

targeted usefully at different components of MC.

A third study explored reduction in feelings of MC using the ‘dirty kiss’

perpetrator paradigm with women who imagined kissing an under-age boy (Waller &

Boschen, 2015). The study extended the studies above by exploring the impact of

physical washing (actual mouth-washing) and mental washing (imagined mouth-

washing) as well as atonement on feelings of MC. In contrast to the previous two

studies, Waller and Boschen (2015) included a control group who were asked to

engage in an emotionally ‘neutral’ task (imagining cars on a busy junction).

Importantly, Waller and Boschen found evidence of spontaneous decay of MC

(indexed by significant reduction in the control group) and no difference in reduction

of feelings of mental contamination (using indices of both INEs and ENEs) between

the groups, indicating that active neutralisation strategies were not superior to the

control task.

The current study built on Waller and Boschen’s findings in a number of

ways. Like Waller and Boschen, and in contrast to others studies (Coughtrey et

al.,2014; Ishikawa et al., 2014), the current study assessed INEs and ENEs and did

not aggregate them into a single MC index. This allowed the important question of

differential reduction in specific components of MC to be explored as well as

assessing a broad range of MC indices. With regards to specific MC components,

feelings of internal dirtiness were assessed as this is considered a defining

component of MC. Additionally, Rachman (2006) theorised that the MC response

incorporated negative emotions such as disgust, anxiety, fear, anger, shame, guilt.

Research which has induced mental contamination has supported the idea that these

emotions result after a moral transgression (Fairbrother & Rachman, 2004; Elliott &

11

Radomsky, 2009; Elliott et al 2012; Rachman et al 2012; Lee et al 2013; Coughtrey

et al 2014; Ishiwaka et al; 2014). As INEs have been found in previous research to be

the most problematic reaction for people (i.e. taking longer to decay) these were the

main focus of the current study (Ishiwaka et al, 2014). For the purpose of this study

the INEs were: shame, guilt, humiliation, fear and sadness. Anxiety was included as

an ENE.

To evoke feelings of MC, the current study used the autobiographical ‘victim

memory’ recall paradigm used by Coughtrey et al (2014). This paradigm was used as

it was considered to have higher ecological validity than the ‘dirty kiss’ paradigm

and allowed recruitment of both male and female participants. This paradigm has

been used relatively less frequently than the dirty kiss paradigm. Other studies

exploring the effectiveness of specific strategies on MC have used washing-based

tasks (e.g. duration of washing, imagined washing). The current study aimed to

extend current evidence by focussing on the potential effectiveness of non-washing

based tasks. This was considered important given the accumulating evidence that

washing is likely ineffective in reducing MC, as theorised by Rachman (2004).

Instead, the current study explored whether two writing ‘intervention’ tasks, one

based on increasing self-esteem and the other on self-compassion, might reduce

feelings of MC.

The two tasks were selected because of the intra-psychic nature of MC.

Rachman (2006) theorised that cognitive approaches to MC were more useful than

behavioural ones. This is because, according to Rachman, a sense of pollution (i.e.

feeling dirty through exposure to immorality) is different to feeling dirty as a result

of direct physical exposure to a contaminant: “The direct way to reduce the sense of

pollution is to tackle the nature, source and supports of the cognition of being

12

polluted” (1994; p311). Consistent with Rachman’s ideas, the few intervention

studies which have focussed on MC have used predominantly cognitive interventions

(Jung & Steil, 2009; Warnock-Parkes, Salvoskis & Rachman, 2013; Coughtrey,

Shafran & Lee, 2013).

In the current study, the use of a self-esteem enhancing task was influenced

by cognitive restructuring. This approach helps people to see their negative thoughts,

experiences, misconceptions and assumptions in a different way (Beck, 1970).

Cognitive restructuring is an important component of all cognitive interventions and

has a good evidence base for reducing negative affect related to the MC reaction

(Butler, Chapman, Forman & Beck 2006; Aldao & Noek-Hoeksemo, 2010). A self-

esteem task seemed appropriate as Coughtrey, Shafran & Lee’s (2013) case series

proposed that CBT techniques that improved self-esteem were also effective in

reducing MC. Additionally, Ishikawa, Kobori, Komuro & Shimizu, (2104b) found

that low self-esteem was related to feelings of MC and suggested it may be a

vulnerability factor for MC. The self-esteem task was designed to help individuals

see themselves in a more positive way to ameliorate the MC response.

More recently another strategy found to help people cope with negative

emotions is compassion focussed work. This approach helps individuals to see events

and themselves in a more compassionate, non-judgemental way (Gilbert, 2010).

Compassion theory purports that memories, thoughts and images can stimulate the

threat system which is linked to negative emotions including anger, fear, disgust and

shame (Gilbert, 2010, Macbeth & Gumley, 2012). Theoretically, these emotions can

be reduced through compassion training which involves becoming more accepting of

personal failings. In support of these assertions, studies have found compassion-

focussed therapy is effective in treating a range of psychological difficulties;

13

particularly those linked to high shame and self-criticism (Leaviss & Uttley, 2015).

Having low self-compassion is related to psychological difficulties and high self-

compassion has been found to promote psychological wellbeing and increase

resilience to negative events (Macbeth & Gumley, 2012; Neff & Costigan, 2014;

Leary, Tate, Adams, Allen & Hancock, 2007). A compassion ‘task’ was used in this

study because, theoretically, compassion should reduce negative internal emotions

that are related to the MC reaction, such as shame. The task involved participants

evaluating their experiences differently and without judgement. Studies have shown

that appraisals of events that induce MC impact on feelings of MC, particularly in

terms of perceived responsibility and perceived violation (Elliott & Radomsky 2013;

Radomsky & Elliott 2009). This lends support to the use of a compassion focused

task to reduce MC because this task should help modulate participants’ reaction to,

and appraisal of the victim memory and therefore reduce negative affect. However,

given the findings by Coughtrey et al (2014), that re-evoking a moral violation

prevents a reduction in MC feelings, there was a possibility that the compassion task

would not lead to a reduction in MC feelings as the task involved thinking about the

MC memory, a process which could arguably re-evoke it. The current study therefore

assessed whether self-esteem focussed writing would be superior to compassion-

focussed writing.

Following Waller and Boschen (2015), in the current study the control group

was asked to undertake a neutral task that did not aim to reduce feelings of MC but

which helped to keep participants occupied such that they could not engage in any

internal neutralising behaviours.

In the current study, the ‘interventions’ used were writing tasks. Research by

Pennebaker has shown that writing about difficult life events can have a positive

14

impact on mood and improve wellbeing (Pennebaker & Chung, 2011). The focus of

Pennebaker’s writing work has often been in writing about a difficult negative

experience in depth, exploring all thoughts and feelings related to it every day for a

number of days. Burton and King (2004) found that writing about positive

experiences also led to enhanced positive mood. Writing about both positive and

negative experiences for just two minutes a day for two days has been found to have

an impact on health (Burton & King, 2008). Although these studies focus on writing

over longer periods of time, previous studies have used brief, one off self-

compassion and self-esteem writing tasks to study their impact on affect (Leary et al,

2007; Breines & Chen, 2012). These two studies had differing results with Leary et

al. (2007) finding that a self-compassion writing group reported significantly lower

negative affect than both a self-esteem and a neutral writing control group. In

contrast, Breines and Chen (2012) found no difference between these three groups on

positive affect. An additional rationale for brief writing tasks in the current study was

that brief and remote interventions are being increasingly studied and utilised as

pragmatic and cost-effective means to help individuals with difficult emotions (e.g.

downloadable mindfulness apps, one-session treatments of imagery re-scripting for

intrusive images). Although the current study focused on the everyday occurrence of

MC and did not focus on clinical intervention, the findings may have the potential to

inform the utility of such brief interventions. Exploring strategies that might be

effective in reducing specific aspects of MC may be of particular importance in

addressing the high relapse rate and treatment resistant nature often found in

conditions in which MC is a feature, such as OCD (Ponniah, Magiati & Hollon

2013). Investigating the reduction of feelings of MC was also considered important

in elaborating current knowledge and understanding of MC. The aim of the study

15

was to address whether specific methods might promote the reduction of everyday

MC reactions in the general population and the potential theoretical and clinical

implications of this.

Main hypotheses:

Hypothesis one: Autobiographical recall of being the victim of a moral violation will

induce feelings of MC (operationally defined as an increase in feelings of internal

dirtiness, shame, guilt, sadness, fear, humiliation and anxiety after recall of a

memory of being violated).

Hypothesis two: There will be a greater reduction in feelings of induced mental

contamination following self-esteem or compassion-focussed writing compared to

neutral writing.

Exploratory research question:

Does reduction in mental contamination differ between the self-esteem and self-

compassion-focussed writing groups?

Method

Design

This study used an experimental between participants design with three writing

groups: a control (neutral writing) group and two ‘intervention’ groups (self-esteem

or compassion-focussed writing). The study had two stages: first, a method was used

to try to induce MC in all participants then, in the second stage, participants were

randomly assigned to one of three writing conditions. The dependent variable was

change in feelings of MC. This was operationalised by changes in self-reported

indices of internal dirtiness, shame, guilt, humiliation, fear, sadness and anxiety. The

study was designed in Qualtrics survey software (Qualtrics, Provo, UT, USA.

http://www.qualtrics.com) permitting participants to take part online at a location and

16

time of their choice. Figure 1 below gives a diagrammatic outline of the study’s

procedure.

Figure 1: Overview of Study Procedure

Participants

Participants did not receive any reimbursement or reward for their participation in

this study and they were aware of this from the beginning. This study recruited a

general population sample of English speaking men and woman over the age of 18.

A general population sample was suitable as previous research has found MC can be

induced in the general population and is not just experienced in clinical populations

(Elliott & Radomsky, 2009, 2012, 2013, Herba & Rachman, 2007, Rachman et al

17

Self Esteem task (control & compassion groups) and debrief

Participants complete MC indices (as above)

Participants randomised to writing task: Self-esteem OR Compassion OR Neutral control

Participants complete MC indices (as above) and vividness ratings

MC induced via ‘'Victim" memory recall task

Participants complete MC indices (shame, guilt, sadness, fear, humiliation, anxiety, internal feelings of dirtiness)

2012, Coughtrey et al, 2014). Participants were recruited from the general population

through snowball sampling from personal contacts, adverts on social media, via

private sector organisations agreeing to distribute an email advertisement about the

study to their employees, advertisements on social media platforms and snowballing

via participants who took part via those sources (See Appendix B). This study

received ethical approval please see Appendix C.

A priori sample size calculation was conducted using G*Power 3.1.7 (Faul,

Erdfelder, Lang & Buchner, 2007). For hypothesis one, previous studies inducing

mental contamination via autobiographical victim memory recall have presented

large effect sizes in the range of d=.79-1.29 (Coughtrey et al 2014). Assuming power

of .8 to detect an effect of at least .62 (2-tailed, alpha=.05) using a dependent samples

t-test (pre vs. post victim memory induction within participants) a priori sample size

calculation indicated a minimum sample size of 14.

In calculating the sample size for hypothesis two it was not possible to draw on

existing literature as no studies have previously used writing tasks. Therefore, sample

size was calculated using Cohen’s (1998) conventions for medium and large effects.

Assuming power of .8 to detect a medium effect (ƞ2=.05) on change in MC indices

from pre to post experimental ‘intervention’ task between groups (2-tailed, alpha

= .05) using a one way ANOVA a priori sample size calculation indicated a sample

size of 159 (i.e. 53 participants per group). To detect a large effect (ƞ2=.13) G*

power calculations indicated a sample size of 60 (20 participants per group).

Therefore, the target sample size was between 60 and 159 participants.

Materials and Measures:

Induction of mental contamination (See Appendix D): To induce feelings of mental

contamination the ‘victim’ memory task presented in Coughtrey et al. (2014) was

18

used. This involved asking participants to think of a time when they had felt

humiliated, ashamed, betrayed, or a time when someone else harmed them in some

way. Participants were first asked to type into a free text box a brief description of

the memory they had chosen. Then they were asked to form a mental image of this

memory in their mind for two minutes imagining it as if it were happening again,

through their own eyes as if they were there and focussing on the negative emotions

that they experienced at the time. This task was chosen because it used real life

memories of events and, arguably, has higher external validity than imagined non-

consensual kiss paradigms used in research to induce MC. It also enabled the study

to use both male and female participants which the ‘kiss’ paradigm study does not.

Coughtrey et al’s (2014) study showed that MC could be induced through

autobiographical memories in the general population making this an appropriate task

for this study. In Coughtrey’s (2014) study, given that remembering this memory for

two minutes was enough to induce MC, the same time was used in the current study.

Participants were asked to write down the memory so the researchers could see what

memories were used and check adherence to the task.

Measures of MC: Participants completed a number of single item Visual Analogue

Scales (VAS) to assess subjective feelings of mental contamination (see Appendix

E). MC was operationalised in this study as internal feelings of dirtiness, shame,

guilt, humiliation, fear, sadness and anxiety. Participants were asked to move a slider

along a scale of 0 (‘not at all’) to 100 (‘extremely’) relating to how they felt in that

moment (e.g ‘How ashamed do you feel?’).

VAS have been found to be a useful measure of mood and have demonstrated high

reliability and validity (Ahearn, 1997). Single item VAS measures of depression and

anxiety have been found to have good validity and reliability and sensitivity to

19

change supporting the contention that VAS scales are useful in measuring mood

(Abend, Dan, Maoz & Bar-Haim, 2014; Van-Rijisbergen, Bockting, Berking, Koeter

& Schene, 2012; Williams, Morlock & Feltner, 2010). VAS are an easy way of

measuring subjective experience and have been used in much of the existing mental

contamination research. A line from 0 to 100 has been found to produce better

reliability (Miller & Ferris, 1993).

Task Engagement Checks: These were included in case the MC induction task was

not successful. These items would permit assessment of whether induction failure

might be due to participants’ inability to clearly imagine the event or how long they

spent thinking about it. Participants were asked on a scale of 0 (‘not at all’) to 100

(‘very’) how easy it was to imagine their memory and how clear the visualisation of

their memory was. They were also asked on a scale of 0 (‘none’) to100 (‘all’) how

much of the allotted time (2 minutes) they spent thinking about their memory. In

addition, to check participants’ adhered to the task correctly, what they wrote was

examined to ensure they had recalled a memory of being a victim of a violation

‘Intervention’ writing tasks (see Appendix F):

Control task: The control group were asked to write about a neutral topic - how

participants spent a ‘normal’ week day where nothing out of the ordinary happened.

Previous research has used this as a neutral control writing task (King, 2001, King &

Meiner, 2000, Pennebaker & Beall, 1986, Burton & King 2004). Previous research

into reducing feelings of MC used a doing ‘nothing’ control group; however,

participants could have been engaging in internal neutralising strategies which

reduced MC feelings. To overcome this, this study included an ‘active’ control task

designed to prevent participants intervening on the MC feelings. A normal day was

chosen as this is unlikely to evoke much of an emotional reaction.

20

Self-compassion writing task: This task involved participants writing about their MC

memory expressing kindness and compassion towards themselves and their feelings

about it. They were told it might help to think about how they would express concern

for a friend who had undergone the same experience. This task was devised based on

compassion-focussed work by Gilbert which focuses on resisting self-blame and

replacing criticism with kindness and understanding for the emotions being

experienced (Gilbert, 2009). The task focused on the self-kindness component of

self-compassion and was similar to a task used successfully by Leary et al. (2007) to

induce self-compassion.

Self-esteem writing task: This task involved participants writing about a time when

they felt worthwhile, valued and empowered. Participants were asked to include any

positive emotions or thoughts they had at the time. This task was to help participants

to see their positive qualities and see that they have also had more positive

experiences which differ to the MC event they described. It was similar to self-

esteem inducing tasks used in previous research which aimed to help participants

acknowledge their positive characteristics, qualities and things they were proud of

(Breiens & Chen, 2012; Leary et al, 2007).

Procedure

The study was delivered by Qualtrics study software which allowed remote

participation online. Once participants clicked on a link to the study they were

directed to the front page of the study which presented the information sheet

(Appendix G). Participants were then asked if they wanted to take part. If they

clicked yes, they were asked to endorse the study consent statements and were

advised they could withdraw from the study at any time (Appendix H).

21

First, some demographic data was collected. Then participants were asked to

fill in the six MC baseline VAS. After this all participants took part in the

autobiographical victim memory MC induction task. Participants were not be able to

proceed to the next part of the study until the two minutes allotted to recalling the

memory had elapsed. This was to try to ensure, as much as possible, all participants

spent the same time on the task. All participants then completed the six MC VAS

measures again and completed the task engagement checks described above.

Following the induction of MC participants were then randomised via Qualtrics into

one of the three writing conditions.

For all three writing tasks participants were unable to move to the next part of

the study until at least two minutes had elapsed but there was no maximum time

limit. The two minute minimum time was set to encourage all participants to engage

in the task. No upper time limit was added because of the abrupt nature of forcing

participants off a task that had the potential to reduce uncomfortable feelings before

they were finished. Forcing the pace of this aspect would arguably also reduce the

ecological validity of the study because ‘therapeutic’ tasks would not have to be

completed under a time pressure. Instead, Qualtrics recorded how long all

participants spent completing the tasks so this could be compared across the three

writing groups. Engagement in the tasks was also measured by the number of words

participants wrote.

After participants had completed their assigned writing task they were asked to

complete the six MC VAS measures again. As the victim memory MC induction task

had the potential to elicit negative feelings, once measures had been completed

participants in the control group and compassion group were invited to complete the

self-esteem task (i.e. thinking of a time they felt valued and empowered) as a way to

22

induce more positive feelings. Participants wrote this down and there was no time

limit but it was not analysed. The self-esteem task was chosen because it was

considered to be the task that had the potential to promote a positive self-view

without the risk of re-evoking the original victim memory given that it asked

participants to focus on a different experience. At the end of the study participants

were de-briefed on the purpose of the study (Appendix I).

Ethical Issues

During this research participants were asked to recall a negative memory

designed to induce feelings of MC which is associated with unpleasant emotions.

Another issue was that participants were asked to recall a moral transgression they

have experienced and the wording was quite open including a time they have felt

‘harmed in some way’. This had the potential to bring up difficult memories for

people. However, the study also looked at how to reduce these negative feelings. All

participants at some point in the study undertook a self-esteem task which was

designed to induce a memory of a time associated with positive feelings.

Before taking part in the experiment participants were told that the study may

induce some unpleasant but short lived feelings so they could decide whether to take

part or not. Participants were also made aware that they could withdraw from the

study at any time. Participants were given guidance around the type of memories that

they could recall. This was to help demonstrate to the participants that they could

choose how significant the moral transgression was that they remembered.

Participants were not asked to remember very distressing events. They were given a

full de-brief at the end of the experiment and information on where they could get

further support if the study brought up anything difficult for them.

23

Results

Induction of MC

Participants:

In total, 197 people accessed the study online. After reading the information sheet,

146 of these consented to take part in the study. Between consenting to take part and

completing the MC induction task 37 participants dropped out. This left a total of

109 participants who completed the MC induction task. Eleven of these participants

had to be removed subsequently due to timing: the Qualtrics software imposed a 120

second time limit for the MC induction task however, due to technical difficulties

this varied for some participants. The decision was made to remove any participants

whose timings were 120 seconds +/- 3 seconds. This was to ensure fidelity across

participants in this task. After this process, 98 participants were then left. However,

four of these did not complete the post-induction MC measures resulting in an

eligible sample of 94. Please see appendix J for participation and drop out/ exclusion

over the course of MC induction task.

The 94 participants (35 male and 59 female) had a mean age of 31.72 (SD=9.77;

range=23 to 67). The majority of participants identified themselves as either: ‘White

British’ or ‘White other’ (N=89), with the remaining five identifying as either: Black

African, Black British, Indian or Chinese. The sample was highly educated with the

majority having an undergraduate level (N=53) or post-graduate level (N=34)

education. The remainder was educated to GCSE (N=1) or A-level (N=6).

In the MC induction task participants were asked to write a short description

of the ‘victim’ memory they were going to think about. This was done in order to

check adherence to the paradigm. After reviewing these descriptions, one participant

24

was removed from the data analysis as they had not adhered to the paradigm

instructions. This participant recalled a time they had cheated on a partner which was

considered to be a perpetrator memory. Participants reported a range of memories in

response to the instructions (see Table 1). The most frequently recalled memory was

being cheated on by a partner.

Table 1: Victim Memories (N=93)

Victim Memory NBeing cheated on: 24Physically assaulted (strangers or family) and being mugged: 13Bullied by friends or in work place: 10Being lied to: 8Being dumped or rejected in romantic relationships or friendships: 8Being teased: 7Being betrayed by others: 7Being embarrassed by others publicly: 3Being criticised or told off: 3Being blamed for something you didn’t do: 3Being burgled: 2Losing job: 1Demotion at work: 2Sexually abused: 2

Histograms were examined and Kolmogorov-Smirnov statistics calculated to assess

the distribution of MC scores pre and post induction. As would be expected, the pre-

induction MC indices demonstrated marked positive skew. This was also evident

when examining the post-induction indices (See Appendix K and L). As the data was

not normally distributed a non-parametric Wilcoxon Signed Rank Test was used to

explore any significant differences between MC indices pre and post the induction

task. Descriptive statistics and comparison tests for pre-post induction MC indices

are shown in Table 2.

25

Table 2: Descriptive Statistics and comparison tests (N=93)MC indices Median Minimum Maximum Wilcoxon

signed rank test statistic

Significance Effect size

Pre Dirtiness 3 0 74Z=-4.32 p=<0.001 r= 0.36

Post Dirtiness 9 0 95Pre Anxiety 20 0 92

Z=3.32 p=0.001 r=0.27Post Anxiety 30 0 91Pre Shame 2 0 77

Z=-6.04 p=<0.001 r=0.5Post Shame 24 0 97

Pre Guilt 8 0 100Z=-1.98 p=0.048 r=0.16Post Guilt 10 0 83

Pre Fear 3 0 68Z=-4.10 p=<0.001 r=0.34Post Fear 11 0 100

Pre Sadness 5 0 83Z=-7.46 p=<0.001 r=0.62Post Sadness 40 0 100

Pre Humiliation 1.5 0 84Z=-7.45 p=<0.001 r=0.62Post Humiliation 49 0 100

Table 2 shows that before MC induction, anxiety was the highest of all the indices

(Mdn=20) and humiliation was the lowest (Mdn=1.5). It would be expected that

anxiety would be the highest given the potential anticipation of the unknown task

ahead and that state anxiety is common generally. Table 2 also displays the results of

comparisons between MC indices pre and post the MC induction task. Results

indicated that there was a significant increase on all MC indices suggesting that

feelings of MC had been evoked by the autobiographical recall of ‘victim’ memories.

26

For feelings of guilt a small effect size was found; it is understandable that guilt had

the smallest effect size as participants recalled an event where they were a victim

rather than a perpetrator. A medium effect size was found for feelings of internal

dirtiness, anxiety, shame and fear. The largest effect sizes were found for humiliation

and sadness.

Writing tasks:

Participants:

After the MC induction task, participants were randomised into one of three writing

task groups (see Appendix M for drop out and exclusions form this point). At this

point 14 participants dropped out of the study leaving 79 participants who completed

the writing tasks component of the study. Of these 79 participants, three were

removed because the time taken to complete the task compared to how much had

been written suggested they had not engaged in the writing tasks. These participants

had taken almost 17 minutes to write less than six words. The text that remaining

participants had written during the task was examined to ensure it was consistent

with the task instructions. Following this, three further participants were removed

from the ‘compassionate writing’ group because they had written about what they

would say or do to someone else in this situation but had not related this to

themselves or their memory. Examples from each writing condition can be seen in

Appendix N.

Once these six participants had been removed, the total sample for analysis was 73.

Although the Qualtrics system allows equal randomisation to groups, subsequent

drop-out and exclusion due to task non-engagement/inappropriate engagement

resulted in unequal group sizes: 30 control, 25 self-compassion, and 18 self-esteem.

G-power calculations indicated that, for a large effect, 20 participants were needed in

27

each group. Between-group comparisons in age, gender, ethnicity and education

showed no differences in participant composition between the three writing groups

(see Tables 3 and 4).

Table 3: Participant characteristics by condition

N=Control (N=30), Self-esteem (N=18), Compassion (N=25)

28

Participant characteristics

Writing Condition

Frequencies

Gender:Control Male= 11

Female=19Self-esteem Male=7

Female=11Compassion Male=8

Female=17

Ethnicity: Control White British=25

White Other=3 Indian=1 Not Stated=1

Self-esteem White British=16White Other=2

Compassion White British=24Black British or Black other=1

Education:Control G.C.S.E=0

A-level=1Undergraduate=17Post graduate=12

Self-esteem G.C.S.E=1A-level=1Undergraduate=10Post graduate=6

Compassion G.C.S.E=0A-level=3Undergraduate=13Post graduate=9

A chi-squared test showed there were no significant difference in gender

composition between the three groups (X2(2)=2.41, p=0.89). A Kruskal-Wallis test

showed there were also no significant difference in the participant age between

groups (H(2)= .261, p=0.88).

Age was notnormally distributed as it was a predominantly younger sample (see

Appendix O); therefore, Table 4 shows the median ages across the groups.

Table 4: Participant characteristics- Age

There were also no significant differences between the three groups for any of the

MC indices taken after the MC induction task (i.e. before the writing task

commenced) - see Table 5.

Table 5: Pre writing task MC scores per group

29

Writing Condition Median Minimum MaximumControl (N=30)

28 24 58

Self-esteem (N=18)

27.50 23 67

Compassion (N=25)

30 24 65

30

31

MC INDEX Writing condition Median Mean Ranks

Kurskal-wallis

StatisticDegrees of freedom

Significance level

Internal dirtiness

Control 9.5 35.82 H=4.36 2 p=.11Self-esteem 3.5 30.08Compassion 15 43.40

AnxietyControl 29 37.45 H=.80 2 p=.67Self-esteem 23 33.33Compassion 31 39.10

ShameControl 21.5 34.88 H=1.51 2 p=0.47Self-esteem 24 34.67Compassion 24 41.22

GuiltControl 8 33.53 H=1.77 2 p=0.41Self-esteem 13.5 37.00Compassion 17 41.16

FearControl 8 33.30 H=1.62 2 p=0.44Self-esteem 22 40.53Compassion 11 38.90

SadnessControl 34 34 H=3.39 2 p=0.18Self-esteem 27.5 33.22Compassion 52 43.32

HumiliationControl 39.5 35.52 H=2.12 2 p=0.34Self-esteem 33 32.81Compassion 50 41.78

Engagement Analysis:Given that extent of engagement in the writing task might conceivably impact on

extent of reduction in MC, the number of words written and the time spent writing

was compared between the groups. This was to ascertain whether these factors

needed to be controlled when investigating the impact of the tasks on reducing

feelings of MC following the writing task.

Table 6: Descriptive statistics- time spent in secondsWriting Condition Median Minimum MaximumControl 177.34 121.75 619.34Self-Compassion 175.39 123.20 334Self-Esteem 187.51 121.77 486.59

Histograms and normality plots were examined and Kolmogorov-Smirnov

statistics calculated to assess the distribution of time spent in the three groups (see

Appendix P). The distributions did not indicate marked skew but the homogeneity of

variance assumption was breached so non-parametric analysis was chosen.

Consequently, Table 6 shows the median, minimum and maximum as a comparison

for each group. The minimum time given to complete the task was 120 seconds for

all three groups; after this participants could move on whenever they had finished.

Table 6 shows that, on average, the self-esteem group took the longest to complete

the writing task. It also suggests that in all three groups participants were engaged in

the tasks because, on average, they spent more than the minimum of 120 seconds. As

Levene’s test indicated that the assumption of homogeneity of variance was violated

(F (2, 20)=4.893, p=.010) and given the group sizes were small and unequal, a non-

parametric test was conducted. A Kruskal-Wallis test showed that there was no

significant difference between the three groups in time spent on the writing tasks

(H(2)=0.290, p=.86).

32

Word Count:

Table 7 indicates that, on average, the control group wrote the most and the self-

esteem group wrote the least. Histograms and normality plots were examined and

Kolmogorov-Smirnov statistics calculated to assess the distribution of word count in

the three groups (see Appendix Q). Tests showed that word count was not normally

distributed in any of the groups. As Levene’s test indicated the assumption of

homogeneity of variance was violated (F (2, 70)=4.039, p=.022) and the group sizes

were small and unequal, a non-parametric test was conducted. A Kruskal-Wallis test

showed there was a significant difference in word count between the three groups

(H(2)=9.09, p=0.01). Mann-Whitney tests were used to follow up this finding. A

Bonferroni correction was applied and so all effects were tested at a .0167 level of

significance. Pairwise comparisons showed that the control group (mean rank=33.02)

wrote significantly more than the self-compassion group (mean rank=21.98)

(U=224.500, p=0.01). The control group (mean rank=28.27) also wrote significantly

more than the self-esteem group (mean rank=18.22) (U=157, p=0.02). The self-

esteem (mean rank=19.86) and self-compassion groups (mean rank=23.54) did not

differ significantly (U=186.50, p=.34).

Table 7: Descriptive Statistics- Word Count

Writing Condition Median Minimum MaximumControl 75 21 300Cognitive 35 3 202Compassion 51 23 162

This analysis showed that although the groups did not differ in time spent writing,

the control group produced significantly more words than the other two groups. This

could be explained by the different nature of the tasks. Given the control group wrote

about a typical day, it would be expected that they would be drawing on readily

33

accessible routine and repeated events. In contrast, people do not routinely think or

write about memoires from a ‘self-esteem’ or ‘compassionate’ perspective so these

may have been less readily accessible. Furthermore, these tasks require more thought

and consideration than describing an average day. These findings also lend support to

the integrity of the self-esteem and self-compassion tasks; it would be expected that

fewer words would be produced in the same time frame compared to the control

condition because it would take participants longer to generate the appropriate

material. In view of this, time taken was considered to be a more valid measure of

task engagement. Given this did not differ significantly between the groups, it was

not controlled when comparing reduction of MC indices between the three groups.

Reduction in feelings of MC:Table 8 displays descriptive statistics for change in MC indices for the three groups

after participants took part in the writing tasks. The change score calculation was:

score on MC index taken after the writing task minus score on the MC index after the

induction task; the higher the negative score, the greater the magnitude of decrease in

the MC index following writing. All MC indices showed a reduction, but of differing

magnitude. Shame, sadness and humiliation evidenced the greatest reductions in all

three groups. These were the indices that showed the greatest increase during the

induction phase of the study.

34

Table 8: Descriptive statistics: MC Indices change scoresCondition MC Indices Median Minimum MaximumControl Feelings of

dirtiness-2.5 -85 6

Cognitive Feelings of dirtiness

-3.44 -21 1

Compassion

Feelings of dirtiness

-3 -51 6

Control Anxiety -18 -69 2Cognitive Anxiety -11.5 -21 1Compassion

Anxiety -11 -58 32

Control Shame -17.5 -83 5Cognitive Shame -14 -79 54Compassion

Shame -19 -90 32

Control Guilt -20 -76 10Cognitive Guilt -5 -53 27Compassion

Guilt -7 -90 32

Control Fear -4.5 -90 1Cognitive Fear -3.5 -46 5Compassion

Fear -10 -93 0

Control Sadness -25.5 -100 4Cognitive Sadness -16.5 -88 6Compassion

Sadness -22 -92 6

Control Humiliation -28.5 -89 3Cognitive Humiliation -23.5 -100 7Compassion

Humiliation -19 -84 5

Examination of histograms, normality plots and Kolmogorov-Smirnov statistics (see

Appendix R & S) indicated there was not a constant pattern of normal distribution

across the three groups for any of the MC indices. Levene’s tests indicated that the

MC indices shame, guilt, fear, sadness and humiliation demonstrated homogeneity of

variance. However, this assumption was violated for anxiety and feelings of internal

dirtiness (see Appendix T). As the three groups had an unequal number of

participants, there was not a clear pattern of normally distributed data, and not all

35

indices displayed homogeneity of variance it was decided that a Kruskal-Wallis test

was the most appropriate way of analysing the data for all indices.

Table 9: MC change score comparison testsMC INDEX Writing

conditionMean Ranks Kurskal-

wallisStatistic

Degrees of freedom

Significance level

Internal dirtiness

Control 35.30 H=3.12 2 p=0.21Cognitive 44.39Compassion 33.72

AnxietyControl 31.78 H=3.69 2 p=0.16Cognitive 43.61Compassion 38.50

ShameControl 38.30 H=0.44 2 p=0.80Cognitive 38.00Compassion 34.72

GuiltControl 38.25 H=2.68 2 p=0.26Cognitive 42.11Compassion 31.82

FearControl 38.25 H=1.08 2 p=0.58Cognitive 39.27Compassion 33.54

SadnessControl 33.25 H=1.25 2 p=0.54Cognitive 41.83Compassion 35.62

HumiliationControl 34.38 H=0.88 2 p=0.64Cognitive 40.06Compassion 37.94

Table 9 illustrates that there were no significant differences between the three groups

in any MC change scores. Given that no group differences in MC change scores were

found between the three groups, the pre and post writing MC scores for the whole

sample were combined to investigate whether there was a significant change in MC

indices. The results of this analysis can be seen in Table 10.

36

Table 10: MC index scores pre to post writing taskMC indices Media

nMinimum

Maximum

Wilcoxon signed rank test statistic

Significance

Effect size

Pre writing task: Internal dirtiness

9 0 95Z=-5.50

p=0.000r=-

0.46Post writing task: Internal dirtiness

1 0 71

Pre writing task: Anxiety

30 0 91Z=-6.20 p=0.000 r=-

0.51Post writing task: Anxiety

10 0 75

Pre writing task: Shame

24 0 97Z=-5.71 p=0.000 r=-

0.47Post writing task: Shame

5 0 54

Pre writing task: Guilt

10 0 83Z=-5.04 p=0.000 r=-

0.42Post writing task: Guilt

5 0 52

Pre writing task: Fear 11 0 100Z=-6.34 p=0.000 r=-

0.52Post writing task: Fear

1 0 81

Pre writing task: Sadness

40 0 100Z=-6.98 p=0.000 r=-

0.58Post writing task: Sadness

4 0 83

Pre writing task: Humiliation

49 0 100Z=-6.94 p=0.000 r=-

0.57Post writing task: Humiliation

3 0 90

Table 10 shows that all indices of MC significantly reduced following the writing

task. Sadness and humiliation had large effect sizes and all other indices had medium

effect sizes.

Post Hoc Analyses:

37

Participants in the control condition were asked to write about an average day as this

was considered to be a neutral task. However, understandably, some participants had

written about showering or washing as part of their daily routine. As feelings of

mental contamination are associated with feeling ‘dirty’, there was a possibility that

writing about cleansing might have had an impact on feelings of MC measured.

Ishikawa et al (2014) found that ENEs and feelings of internal dirtiness were reduced

by washing; although this was no more effective than waiting. However, INEs did

not decrease with washing behaviours but persisted for longer. Waller and Boschen

(2015) found that imaginal cleansing was not superior to control in reducing mental

contamination.

To explore this, the control group was split into ‘control-washing’ and ‘control-non-

washing’ groups and the analysis of change scores was re-run. There were 16 people

in the control condition who wrote about washing and 14 who did not. Results were

similar to the three group analysis and indicated no significant differences between

the four groups in change on any of the MC indices (Table 11). Therefore, writing

about washing did not explain why the control group was as effective at reducing

feelings of MC as the two intervention groups.

38

39

Table 11: MC change score comparison tests, including control wash group.MC INDEX Writing

conditionMean Ranks Kurskal-

wallisStatistic

Degrees of freedom

Significance level

Internal dirtinessControl-N-W 37.64 H=3.45 4 p=0.33Control-W 33.25Cognitive 44.39Compassion 33.72

AnxietyControl-N-W 33.68 H=3.90 3 p=0.27Control-W 30.13Cognitive 43.61Compassion 38.50

ShameControl-N-W 40.86 H=0.83 3 p=0.84Control-W 36.06Cognitive 38.00Compassion 34.72

GuiltControl-N-W 42.71 H=3.86 3 p=0.28Control-W 34.34Cognitive 42.11Compassion 31.82

FearControl-N-W 37.68 H=1.10 3 p=0.78Control-W 38.75Cognitive 39.27Compassion 33.54

SadnessControl-N-W 37.07 H=1.44 3 p=0.70Control-W 33.66Cognitive 41.83Compassion 35.62

HumiliationControl-N-W 37.39 H=1.41 3 p=0.70Control-W 31.75Cognitive 40.06Compassion 37.94N-W=Control did not write about washing, W=Control wrote about washing

40

Summary:

In summary, the autobiographical victim memory recall task induced feelings of MC,

as operationalised by: internal dirtiness, anxiety, shame, guilt, fear, sadness and

humiliation, in all participants. The largest effect of the MC induction task was for:

humiliation, shame and sadness.

Randomisation checks showed that the three experiment groups were similar in terms

of age, gender, education and race. There were also no significant differences

between the three groups in how long participants spent on the task or in scores on

MC indices after MC induction. Feelings of MC significantly reduced after the

writing tasks for all participants and there was no significant difference between the

three writing task intervention groups.

Discussion:

Inducing feelings of MC:

Overall, the findings of this study provide experimental evidence to support

the assertion that feelings of mental contamination can result after recalling an

unpleasant or immoral event. These findings support Hypothesis one and add to

evidence supporting the theory of MC (Rachman, 2006).

The study demonstrated that feelings of MC can be evoked in a non-clinical

population via the recall of autobiographical memories of being a victim of a moral

violation. Following recalling these memories, participants reported a significant

increase in a number of feelings associated with MC, particularly shame, sadness and

humiliation. Consistent with Rachman’s (1994) theory, feelings of internal dirtiness

can manifest in the absence of observable dirt and MC can be evoked through self-

contamination from recalling unpleasant memories. This study also provides more

evidence of the varied emotional reactions associated with MC as theorised by

41

Rachman (2006) and consistent with previous research (Fairbrother & Rachman,

2004; Elliott & Radomsky, 2009; Elliott et al 2012; Rachman et al 2012; Lee et al

2013; Ishiwaka et al; 2014). The replicated finding that MC can be induced by

memories may have important clinical implications when addressing events from

people’s past and understanding the impact of these memories. This also has

implications for future MC research as it provides more evidence that this is a valid

and ecologically improved way of inducing MC.

The study showed that recalling a victim memory had the strongest impact on

feelings of humiliation, shame and sadness and the weakest impact on guilt. Guilt is

generally an emotion experienced when a person feels a sense of responsibility for

causing harm to others and judges their behaviour negatively (Lee, Scragg & Turner,

2001). As participants were recalling a memory of being a victim it is understandable

that this had the weakest impact on feelings of guilt; this may be a more prominent

feature of MC when people recall memories of having committed a moral violation

themselves. In contrast, both shame and humiliation can be felt even when a person

feels they have done nothing wrong and does not blame themselves (Gilbert, 1998).

Humiliation occurs from experiences of being treated unfairly by others in some

way; being in a powerless position, ridiculed or abused (Gilbert, 1997, 98). Shame is

theorised to be a social emotion related to how others perceive us and how we judge

ourselves. It often accompanies times when social status is impacted such as being

degraded or devalued (Gilbert, 1997). In the ‘victim’ memories recalled, the events

all relate to social and relational breaches that are likely to have led to the victim

feeling devalued. As such, these experiences are likely to have the strongest impact

on feelings of humiliation and shame. Gilbert (1997, 98) described two types of

shame: internal (negative evaluation of the self) and external (negative evaluation

42

from others). Shame was not separated in this study and it might be interesting to

explore whether victim memories differ in how they impact these two types of

shame. It is possible that participants reported more shame because they were asked

to write down their memories rather than privately recalling them. However,

participants were aware the study was conducted anonymously.

The finding that humiliation, shame and sadness were the most salient aspects

of the MC reaction is important and has not been highlighted in previous research.

This should be considered when addressing MC in clinical populations and guiding

the focus of interventions. These may be particularly salient when working with

people who have been the victims of assault and trauma. Research has linked MC to

the development and maintenance of PTSD symptoms following a sexual assault

(Fairbrother & Rachman, 2004, Olantunji et al 2008). There is emerging research

that the relationship between MC and PTSD, in female victims of sexual assault,

might be moderated by the tolerance of negative emotions (Fergus & Bardeen,

2016). By highlighting the key emotional reactions experienced in MC this study can

contribute to this area, although research with a clinical population is needed.

This study has shown the potential for people to experience the negative

consequences of MC after being violated in relatively commonly experienced

situations (betrayal, cheating, theft, bullying etc). The emotional reactions found to

have the strongest impact- shame, humiliation, sadness- are common to many forms

of emotional distress and are associated with psychopathology. Shame has been

associated with depression (Kim et al, 2011), anxiety (Fergus et al, 2010; Gilbert,

2000; Valentiner & Smith, 2008) and PTSD (Taylor, 2015; Lee et al, 2001).

Humiliation has been linked to the onset of depressive and anxiety episodes

(Kendler, Hetteman, Butera, Gardener & Prescott, 2003). Therefore, the question of

43

whether some people experience more MC from these events than others and

whether MC may pre-dispose people to anxiety, depression and general difficulties

with wellbeing continue to be vital empirical questions given the strong negative

emotional reactions found following MC induction procedures. This study has

highlighted that it is important to consider how people in the non-treatment seeking

population manage these unpleasant effects of everyday violations and avoid

potential difficulties with mental wellbeing.

Although it could be argued that induction task has increased ecological

validity through the focus on actual memories rather than imagined events, results of

the induction may have been subject to experimenter demand effects. The wording of

the VAS scales before and after induction could have cued participants to the

purpose of the study. In addition, the wording of the MC induction task included the

words ‘humiliated’ and ‘ashamed’ which may have contributed to these two

emotional reactions having the strongest effects. The study also relied on participants

reporting subjective levels of distress. Some people may find it difficult to

distinguish and report accurately these accurately. Tangney (1996) reported that

people were sometimes inaccurate at distinguishing shame from other emotions.

However, these criticisms are tempered by the finding that the MC indices were not

uniformly affected by the induction procedure. For example, guilt was least

responsive, as would be predicted based on theory. If these effects had been present,

it would likely have influenced all indices.

Reduction in feelings of MC:

All three groups showed a reduction in feelings of MC after completing the

writing tasks. The analysis of the full sample indicated a significant reduction on all

indices of MC following writing. However, contrary to Hypothesis 2, analysis by

44

writing group indicated that neither of the intervention tasks demonstrated a greater

reduction in MC indices compared to the control task. In answer to the exploratory

research question, there were also no differences in reduction of MC found between

the self-esteem and self-compassion writing groups.

The finding that MC reduced in the control group as well as the two

‘intervention’ writing groups is consistent with previous research in non-clinical

populations (Coughtrey et al, 2014; Ishakawa et al, 2014; Waller & Boschen, 2015).

Similar to Waller and Boschen (2015) the current study included a control group

designed to try to counteract any internal neutralising strategies participants may

have been engaging in to regulate emotional arousal. As with Waller and Boschen

(2015), the findings of the current study indicate that reduction in feelings of MC

was equivalent across the ‘intervention’ and control groups. The current study

therefore demonstrates that feelings of MC reduce in the absence of a direct

challenge. The current study replicated Waller and Boschen’s (2015) findings using

real life memories. This might be argued to have greater ecological validity than the

‘dirty-kiss’ paradigm utilised by these authors. The current study therefore extends

the existing evidence by showing that MC resulting from personally relevant

memories of violation reduce without intervention.

However, although the study aimed to include a ‘no intervention’ control by

giving participants a neutral task to complete, it is difficult to be sure that there were

not other internal neutralising strategies being used. It would be helpful in future

research to ask participants after the experiment about any emotional regulation

strategies they used. Furthermore, although writing about a normal day was thought

to be ‘neutral’ it could be that this generated feelings of normalcy and security which

helped to reduce MC.

45

The current study aimed to advance the literature by exploring the potential

impact of cognitive interventions given Rachman’s view that pollution of the mind

should target the cognitive underpinnings of a sense of pollution. Rachman’s ideas

are supported by the relative ineffectiveness of washing-based interventions on MC

(Coughtrey et al, 2014; Ishakawa et al, 2014; Waller & Boschen, 2015). Therefore,

the study used writing tasks that would promote a positive self-view as a way to

reduce MC. As MC promotes shame and other negative emotions then thinking of

the self in a more positive and compassionate ways should reduce these negative

effects and restore a more positive view of the self. Although these tasks did reduce

MC, they did not reduce it more than writing about a neutral event.

In a different context, other studies have looked at the impact of compassion

and self-esteem writing tasks on emotions. The findings of this study are consistent

with Brienes and Chen (2012) who found no significant difference in positive affect

(sadness, contentment and upset) between a control, compassion and self-esteem

writing task. In contrast, Leary et al (2007) found participants in a self-compassion

writing group reported significantly less negative affect ( sadness, anger and anxiety)

than a self-esteem or control writing group. The difference may be because the

compassion and self-esteem tasks in this study differed somewhat from Leary et al’s

(2007) tasks: the self-compassion task in the current study asked participants to write

to themselves with compassion, understanding, kindness and concern in the same

way they might speak to a friend who had been through the same experience. Leary

et al’s (2007) study included this but also asked participants to list the ways other

people experience similar events. They also included a mindfulness element which

asked participants to describe their feelings in an objective and unemotional way.

46

Coughtrey et al (2014) found that re-evoking memories prevented a reduction

in feelings of MC. In this study the self-compassion condition involved writing about

the memory; this could have acted to re-evoke the memory and therefore maintain

feelings of MC. However, as feelings of MC reduced in the self-compassion group

and there was no significant difference between the self-compassion group and the

other two groups it could be proposed that, although the task asked people to focus

again on the victim memory, this did not prevent a reduction in MC.

Another explanation for finding no group differences could be that the study

was underpowered to detect more subtle effects. The study had just enough power to

detect a large effect size; it might be that self-esteem/compassionate writing tasks

produce a modest benefit that could not be detected without a larger sample size.

However, if such tasks only produce a modest benefit in reducing MC relative to

other forms of writing then their clinical utility would not be persuasive.

Furthermore, it may be that one ‘dose’ of a writing task was not sufficient to

have an impact on reducing MC. Previous research to have found an effect of writing

tasks has asked participants to engage in the tasks once a day for a number of days

(Burton & King, 2004, 08) Therefore, it is too early to make a conclusion about the

effectiveness of the writing tasks without testing longer durations and exploring

writing tasks with a bigger sample size.

Further Limitations of study:

There are other limitations that have not thus far been raised that should be

addressed in future research. The first is how well participants were able to

understand what was required of them in the intervention writing tasks. There were

no measures of how easy the tasks were to complete. To assist development of tasks

in the future, such feedback would be useful.

47

The study had a high drop-out rate which meant that the final sample size was

relatively small (N=74) and the group sizes unbalanced. As noted, the study was

underpowered to detect smaller effects. It is possible that more guidance on the

writing tasks might have prevented loss of participants although this needs to be

balanced with not making participation seem too onerous.

Another limitation of this study is how far findings can be transferred to

clinical samples. Due to ethical considerations, instructions asked participants to

think of low-level violations rather than very traumatic memories. The effect sizes

for MC reported in this study are from these low level violations and the effects of

recalling more traumatic memories might be larger. The extent to which the MC

response might be stronger or different in clinical samples needs to be investigated

further. It could be that writing intervention tasks would be more effective than the

control group in a clinical sample where feelings of MC are stronger. Future research

into the reduction of feelings of MC need to include clinical populations.

Further implications of findings and future research:

This research contributes to the theoretical understanding of MC. The study

has shown that some indices of MC show a greater response to recalling victim

memories; this has not been shown in previous research. The findings also suggest

that neither of the writing tasks reduced MC more than the control group. This adds

to the accumulating evidence base which seems to suggest that, in non-clinical

samples, MC does not need an active intervention to reduce. However, given the

limitations of this study a clear conclusion on this cannot be made and further

research with larger sample sizes, longer writing interventions and clinical samples is

needed to elucidate this further. If MC spontaneously reduces in a non-clinical

population, this may lend support for the use of exposure with response prevention

48

when working with people susceptible to MC; if feelings habituate in the same way

as anxiety, this may be an effective way to address MC as has been shown with

contact contamination. Further research is needed to support this assertion.

The current study was not designed to elucidate whether the writing tasks

might have a longer-term impact on feelings of MC. Previous research has shown

that re-evoking MC memories prevents the reduction of MC (Coughtrey et al 2014).

Therefore, it would be interesting to extend the current study to see whether any of

the writing tasks prevent a return of MC following second recall of the victim

memory. Although group differences were not evident in impact of writing on

feelings of MC, it is feasible that the writing tasks could serve a protective effect on

subsequent feelings of MC when a memory is revisited (e.g. by allowing the person

to experience the original memory in a more compassionate way). If re-evoking

triggers MC, future research should continue to explore how to reduce this.

Most MC research has focussed on ‘victims’ of a moral violation. The current

study might be extended in future research to investigate interventions that might be

effective for reducing MC in ‘perpetrators’. This work might also extend to

individuals high in obsessive-compulsiveness who imagine perpetrating moral

violations and who subsequently feel contaminated and distressed by such

cognitions.

This study tentatively suggests that brief writing tasks focussing on

improving self-esteem and self-kindness may not reduce MC more than neutral

writing tasks. However, this is the first study of writing interventions and replication

is needed. Other techniques may also be more effective. Imagery rescripting was

developed to reduce the distress associated with recall of past events (Wheatley &

Hackman, 2011). It has been found to be helpful in treating PTSD (Grunet, Weis,

49

Smucker & Christianson, 2007), social anxiety (Wild, Hackman & Clark, 2007;

Norton & Abbot, 2016), depression (Wheatley & Hackman, 2011) and there is some

emerging evidence for its usefulness with OCD (Veale, Page, Woodward &

Salkovskis, 2015). Given the overlap with the emotional components of these

difficulties and MC, and the finding that recalling distressing memories can induce

MC, future research could explore the impact of imagery re-scripting on MC.

.

50

References

Abend, R., Dan, O., Maoz, K., Raz, S., & Barhaim, Y. (2014). Reliability, validity and

sensitivity of a computerized visual analog scale measuring state anxiety. Journal of

Behavior Therapy and Experimental Psychiatry,45, 447-453.

Ahern, E. (1997). The use of visual analogue scales in mood disorders: a critical review.

Journal of Psychiatric Research, 31:569–579.

Aldao, A., & Noek-Hoeksemo, S. (2010). Specificity of cognitive emotion regulation

strategies: A transdiagnostic examination. Behaviour Research and Therapy, 48,

974-983.

Beck, A. T. (1970). Cognitive therapy: Nature and relation to behavior therapy. Behavior

Therapy, 1, 184-200.

Breines, J., & Chen, S. (2012). Self-Compassion Increases Self-Improvement Motivation.

Personality and Social Psychology Bulletin, 38: 1133-1143.

Burton, C., & King, A. (2004). The health benefits of writing about intensely positive

experiences. Journal of Research in Personality, 38,150-163.

Burton, C., & King, A. (2008). Effects of (very) brief writing on health: The two-minute

miracle. British Journal of Health Psychology, 13, 2044-8287.

51

Butler, A., Chapman, J., Forman, E., & Beck, A. (2006). The empirical status of cognitive-

behavioral therapy: A review of meta-analyses. Clinical Psychology Review, 26, 17-

31.

Calamari, J. E., Weigartz, P.S., Riemann, B.C., Cohen, R. J., Greer, A., Jacobi, D. M., Jahn,

S.C., & Carmin, C. (2004). Obsessive-compulsive disorder subtypes: An attempted

replication and extension of a symptom based taxonomy. Behaviour Research and

Therapy, 42, 647-670.

Cohelo, J. S., & Whittal, M. L. (2001). Are subtypes of OCD differentially responsive to

treatment? Vancouver, Canada: World Congress of Behavioural and Cognitive

Therapies

Coughtrey, A. E., Shafran, R., Knibbs, D., & Rachman, J. (2012). Mental contamination in

obsessive-compulsive disorder. Journal of Obsessive-Compulsive and Related

Disorders, 1(4), 244-250.

Coughtrey, A. E., Shafran, R., Lee, M., & Rachman, S. J. (2012b). It's the feeling inside my

head: A qualitative analysis of mental contamination in obsessive-compulsive

disorder. Behavioural and Cognitive Psychotherapy,40(2), 163-173.

Coughtrey, A. E., Shafran, R., & Lee, M. (2013). The treatment of mental contamination: A

case series. Cognitive and Behavioral Practice 20(2), 221-231.

52

Coughtrey, A. E., Shafran, R., & Rachman, S. J. (2014). The spontaneous decay and

persistence of mental contamination: An experimental analysis. Journal of Behavior

Therapy and Experimental Psychiatry, 45(1), 90-96.

Doron, G., & Kyrios, M. (2005). Obsessive compulsive disorder: A review of possible

specific internal representations within a broader cognitive theory. Clinical

Psychology Review, 25, 415-432.

Doron, G., Sar-El, D., & Mikulincer, M. (2012). Threats to moral self-perceptions trigger

obsessive compulsive contamination-related behavioral tendencies. Journal of

Behavior Therapy and Experimental Psychiatry, 43, 884-890.

Elliott, C. M., & Radomsky, A. S. (2009). Analyses of mental contamination: Part I,

experimental manipulations of morality. Behaviour Research and Therapy, 47(12),

995-1003.

Elliott, C. M., & Radomsky, A. S. (2012). Mental contamination: The effects of imagined

physical dirt and immoral behaviour. Behaviour Research and Therapy, 50(6), 422-

427.

Elliott, C. M., & Radomsky, A. S. (2013). Meaning and mental contamination: Focus on

appraisals. Clinical Psychologist, 17(1), 17-25.

Fairbrother, N., & Rachman, S. (2004). Feelings of mental pollution subsequent to sexual

assault. Behaviour Research and Therapy, 42(2), 173-189.

53

Fairbrother, N., Newth, S. J., & Rachman, S. (2005). Mental pollution: Feelings of dirtiness

without physical contact. Behaviour Research and Therapy, 43(1), 121-130.

Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G*Power 3: A flexible statistical

power analysis for the social, behavioral, and biomedical sciences. Behavior

Research Methods, 39, 175-191.

Fergus, T. A., Valentiner, D. P., McGrath, P. B., & Jencius, S. (2010). Shame and guilt

proneness: relationships with anxiety disorder symptoms in a clinical sample.

Journal of Anxiety Disorders, 24, 811-815.

Fergus, T. A., & Bardeen, J. R. (2016). Main and Interactive Effects of Mental

Contamination and Tolerance of Negative Emotions in Relation to Posttraumatic

Stress Symptoms Following Sexual Trauma. Journal of Psychopathology and

Behavioural Assessment, 38, 274-283.

Gilbert, P. (1997). The evolution of social attractiveness and its role in shame,

humiliation, guilt and therapy. British Journal of Medical Psychology, 70, 113-

147.

54

Gilbert, P. (1998). What is shame? Some core issues and controversies . In P. Gilbert &

B. Andrews (Eds). Shame. Interpersonal Behaviour, Psychopathology and

Culture. (pp. 3-39). Oxford, England: Oxford University Press.

Gilbert, P. (2000). The relationship of shame, social anxiety and depression: the role of

the evaluation of social rank. Clinical Psychology & Psychotherapy, 7, 174-189.

Gilbert, P. (2009). Introducing compassion-focused therapy. Advances in Psychiatric

Treatment, 15, 199-208.

Gilbert, P. (2010). The Compassionate Mind. London, England: Constable & Robinson.

Goodman, W., Price, L., Rasmussen, A., Mazure, C., Fleischmann, R., Hill, C., Heninger,

G., & Charney, D. (1989). The Yale-Brown Obsessive-Compulsive Scale.

Development, use and reliability. Archive of General Psychiatry, 46, 1006-1011.

Grunet, B., Weis, J., Smucker, M., & Christianson, H. Imagery rescripting and

reprocessing therapy after failed prolonged exposure for post-traumatic stress

disorder following industrial injury. Journal of Behaviour Therapy and Experimental

Psychiarty, 38, 317-328.

Herba, J. K., & Rachman, S. (2007). Vulnerability to mental contamination. Behaviour

Research and Therapy, 45(11), 2804-2812.

55

Ille, R., Schoggl, H., Kapfhammer, H., Arendasy, M., Sommer, M., & Schienle, A. (2014).

Self-disgust in mental disordes- symptom related or disorder-specific?

Comprehensive Psychiatry, 55, 938-943,

Ishikawa, R., Kobori, O., Komuro, K. & Shimizu, E. (2014). Comparing the roles of

washing and non-washing behaviour in the reduction of mental contamination.

Journal of Obsessive-Compulsive and Related Disorders, 3(1), 60-64.

Ishikawa, R., Kobori, O., Komuro, K. & Shimizu, E. (2014b). Examining the Beliefs that

Contribute to Mental Contamination. Journal of Depression and Anxiety, 3, 1-7.

Jung, K., & Steil, R. (2012). The feeling of Being Contaminated in Adult Survivors of

Childhood Sexual Abuse and its treatment via a two-session program of cognitive

restructuring and imagery modification. A case study. Behaviour Modification,

36(1): 67-86.

Kendler, K., S., Hettema, J. M., Butera, F., Gardner, C. O., Prescott, C., A. (2003). Life

event dimensions of loss, humiliation, entrapment, and danger in the prediction of

onsets of major depression and generalized anxiety, Archives General Psychiatry,

60, 8, 789-796.

Kim, S. Thibodeau, R., & Jorgensen, R. (2011). Shame, guilt and depressive symptoms; a

meta-analytic review. Psychological Bulletin, 137 (1), 68-96.2011;

56

Leary, M., Tate, E., Adams, C., Batts, A., & Hancock, J. (2007). Self-compassion and

reactions to unpleasant self-relevant events: The implications of treating oneself

kindly. Journal of Personality and Social Psychology, 92, 887-904.

Leaviss, J., & Uttley, L. (2015). Psychotheraputic benefits of compassion focussed therapy:

An early systemic review. Psychological Medicine, 45, 927-945.

Lee, M., Shafran, R., Burgess, C., Carpenter, J., Millard, E., & Thorpe, S. (2013). The

induction of mental and contact contamination. Clinical Psychologist, 17(1), 9-16.

Lee, D., Scragg, P., & Turner, S. (2001). The role of shame and guilt in traumatic events: a

clinical model of shame-based and guilt-based PTSD. British Journal of Medical

Psychology, 74, 451-466.

Macbeth, A., & Gumley, A. (2012). Exploring compassion: a meta-analysis of the

association between self-compassion and psychopathology. Clinical Psychology

Review, 32, 545-552.

Miller, M., & Ferris, D. (1993). Measurement of subjective phenomena in primary care

research: the Visual Analogue Scale. Journal of Family Practice and Research, 13,

15-24.

Neff, K., & Costigan, A. (2014). Self-compassion, well-being and happiness. Psychologie in

Österreich, 2, 114-119.

57

Norton, A., & Abbott, M. (2016). The efficacy of imagery rescripting compared to cognitive

restructuring for social anxiety disorder. Journal of Anxiety Disorders, 40, 18-28.

Olatunji, B. O., Elwood, L. S., Williams, N. L., & Lohr, J. M. (2008). Mental pollution and

PTSD symptoms in victims of sexual assault: A preliminary examination of the

mediating role of trauma-related cognitions. Journal of Cognitive Psychotherapy,

22(1), 37-47.

Pennebaker, J. W., & Beall, S. (1986). Confronting a traumatic event: towards an

understanding of inhibition and disease. Journal of Abnormal Psychology, 95, 274-

281.

Pennebaker, J. W., & Chung, C. K. (2011). Expressive writing: connections to physical and

mental health. In H. S. Friedman (Ed.), Oxford handbook of health psychology. New

York, NY: Oxford University Press

Ponniah, K., Magiati, I., Hollon, S. (2013). An update on the efficacy of psychological

treatments for obsessive-compulsive disorder in adults. Journal of Obsessive-

Compulsive and Related Disorders, 2 (2), 207-218.

Rachman, S. (1994). Pollution of the mind. Behaviour Research and Therapy, 32(3), 311-

314.

58

Rachman, S. (2004). Fear of Contamination. Behaviour Research and Therapy, 42, 1227-

1255.

Rachman, S. (2006). Fear of contamination: Assessment and treatment. Oxford University

Press.

Rachman, S., Elliott, C. M., & Zysk, E. (2012). Mental contamination: The perpetrator

effect. Journal of Behavior Therapy and Experimental Psychiatry, 43(1), 587-593.

Radomsky, A. S., & Elliott, C. M. (2009). Analyses of mental contamination: Part II,

individual differences. Behaviour Research and Therapy, 47(12), 1004-1011.

Radomsky, A. S., Rachman, S., Shafran, R., Coughtrey, A. E., & Barber, K. C. (2014). The

nature and assessment of mental contamination: A psychometric analysis. Journal of

Obsessive-Compulsive and Related Disorders, 13 (2), 181-187.

Tangney, J., Wagner, P., & Gramzow, R. (1992). Proneness to shame, proneness to guilt and

psychopathology. Journal of Abnormal Psychology, 101, 469-478.

Tangney, J. (1996). Conceptual and methodological issues in the assessment of shame and

guilt. Behaviour and Research Therapy, 34, 741-54.

59

Taylor, A. (2015). The influence of shame on post trauma disorders: have we failed to see

the obvious? European Journal of Psychotraumatology, 6,

doi: 10.3402/ejpt.v6.28847

Valentiner, D., & Smith, S. (2008). Believing that Intrusive Thoughts Can be

Immoral Moderates the Relationship Between Obsessions and Compulsions

for Shame-prone Individuals. Cognitive Therapy and Research, 32, 714-720.

Van-Rijsbergen, G., Bockting, C., Berking, M., Koeter, M., & Schene, A. (2012).

Can a One-Item Mood Scale Do the Trick? Predicting Relapse over 5.5-Years in

Recurrent Depression. PLoS ONE 7(10): e46796.doi:10.1371/journal.pone.0046796.

Veale, D., Page, N., Woodward, E., & Salkovskis, P. (2015). Imagery Rescripting for

Obsessive Compulsive Disorder: A single case experimental design in 12 cases.

Behavior Therapy and experimental, 49, 230-236.

Waller, K., & Boschen, M. (2015). Evoking and reducing mental contamination in female

perpetrators of an imagined non-consensual kiss. Journal of Behavior and

Experimental Psychiatry, 49, 195-202.

Warnock-Parkes, E., Salkovskis, P., & Rachman, J. (2012). When the Problem is Beneath

the Surface in OCD: The Cognitive Treatment of a Case of Pure Mental

Contamination. Behavioural and Cognitive Psychotherapy, 40(4), 383-399.

60

Wheatley, J., & Hackmann, A. (2011). Using Imagery Rescripting to Treat Major

Depression: Theory and Practice. Cognitive and Behavioral Practice, 18, 444-453.

Wild, J., Hackman, A., & Clark, D. (2007). When the present visits the past: Updating

traumatic memories in social phobia. Journal of Behavior Therapy and Experimental

Psychiatry, 38, 386-401.

Williams, V., Morlock, R., & Feltner, D. (2010). Psychometric evaluation of a visual analog

scale for the assessment of anxiety. Health and Quality of Life Outcomes, 8, 57:

http://www.hqlo.com/content/8/1/57

Zhong, C., & Lilijenquist, K. (2006). Washing away your sins: threatened morality and

physical cleansing. Science, 313, 1451-1452.

61

List of Appendices

Appendix A: Behaviour Research and Therapy Journal- Guidelines for authors

Appendix B: Recruitment to study materials.

Appendix C: Ethics Approval

Appendix D: MC Induction task instructions

Appendix E: MC Visual Analogue Scales

Appendix F: Writing task instructions.

Appendix G: Information sheet

Appendix H: Consent form

Appendix I: De-brief

Appendix J: Participation exclusion/drop out flow chart for MC induction task

Appendix K: Distribution of data- Pre MC and Post MC induction all indices (histograms and

normality plots).

Appendix L: Normality tests for Pre MC and Post MC induction all indices.

Appendix M: Participation drop out/exclusion for writing tasks.

Appendix N: Examples of writing task answers

Appendix O: Distribution of age per group (histograms and normality plots).

Appendix P: Distribution of time spent per group (histograms and normality plots)

Appendix Q: Distribution of word count per group (histograms and normality plots).

Appendix R: Distribution of MC change scores (histograms and normality plots).

Appendix S: Normality tests for MC change scores.

Appendix T: Levenes test for MC change scores

62

Appendix A: Behaviour Research and Therapy Journal: Guidelines for authors

BEHAVIOUR RESEARCH AND THERAPY

AUTHOR INFORMATION PACK

ISSN: 0005-7967

DESCRIPTION

An International Multi-Disciplinary Journal

The major focus of Behaviour Research and Therapy is an experimental psychopathology

approach to understanding emotional and behavioral disorders and their

prevention and treatment, using cognitive, behavioral, and psychophysiological

(including neural) methods and models. This includes laboratory-based

experimental studies with healthy, at risk and subclinical individuals that inform

clinical application as well as studies with clinically severe samples. The following

types of submissions are encouraged: theoretical reviews of mechanisms that

contribute to psychopathology and that offer new treatment targets; tests of novel,

mechanistically focused psychological interventions, especially ones that include

theory-driven or experimentally-derived predictors, moderators and mediators; and

innovations in dissemination and implementation of evidence-based practices into

clinical practice in psychology and associated fields, especially those that target

underlying mechanisms or focus on novel approaches to treatment delivery. In

addition to traditional psychological disorders, the scope of the journal includes

behavioural medicine (e.g., chronic pain). The journal will not consider manuscripts

dealing primarily with measurement, psychometric analyses, and personality

63

assessment.The Editor and Associate Editors will make an initial determination of

whether or not submissions fall within the scope of the journal and/or are of

sufficient merit and importance to warrant full review.

AUDIENCE

. For clinical psychologists, psychiatrists, psychotherapists, psychoanalysts, social workers,

counsellors, medical psychologists, and other mental health workers.

IMPACT FACTOR

. 2015: 3.798 © Thomson Reuters Journal Citation Reports 2016

AUTHOR INFORMATION PACK 17 Jul 2016 www.elsevier.com/locate/brat 2

ABSTRACTING AND INDEXING

. BIOSIS Elsevier BIOBASE Current Contents/Social & Behavioral Sciences MEDLINE® LLBA

EMBASE PASCAL/CNRS PsycINFO Psychological Abstracts PsycLIT PsycSCAN

Psychology Abstracts Social Sciences Citation Index Social Work Research &

Abstracts Scopus

EDITORIAL BOARD

. Editor Michelle. G. Craske, Dept. of Psychology, University of California at Los Angeles

(UCLA), 405 Hilgard Avenue, Los Angeles, CA 90095-1563, California, USA Associate

Editors Emily A. Holmes, Emotion Research Group, Cambridge, UK Anita Jansen,

University of Maastricht, Maastricht, Netherlands Allison M. Waters, Griffith

University, Mt Gravatt, Queensland, Australia Ed R. Watkins, University of Exeter,

Exeter, UK Michael J. Zvolensky, University of Houston, Houston, Texas, USA

Founding Editor H.J. Eysenck † Statistical Review Board T. A. Brown, Boston

University, Boston, Massachusetts, USA A. P. Field, University of Sussex, East Sussex,

64

UK T. B. Hildebrandt, Mount Sinai School of Medicine, New York, New York, USA J. L.

Krull, UCLA, Los Angeles, California, USA J. M. Prenoveau, Loyola University

Maryland, Baltimore, Maryland, USA D. Rosenfield, Southern Methodist University,

Dallas, Texas, USA E. Stice, Oregon Research Institute, Eugene, Oregon, USA

Editorial Board J. S. Abramowitz, University of North Carolina at Chapel Hill, Chapel

Hill, North Carolina, USA G. Andersson, Linköping University, Linköping, Sweden M.

M. Antony, Ryerson University, Toronto, Ontario, Canada J. J. Arch, University of

Colorado Boulder, Boulder, Colorado, USA A. R. Arntz, Universiteit van Amsterdam,

Amsterdam, Netherlands A. Bandura, Stanford University, Stanford, California, USA

D. H. Barlow, Boston University, Boston, Massachusetts, USA J. G. Beck, University

of Memphis, Memphis, Tennessee, USA C. B. Becker, Trinity University, San Antonio,

Texas, USA B. C. Chu, Rutgers University, Piscataway, New Jersey, USA D. M. Clark,

University of Oxford, Oxford, UK M. E. Coles, State University of New York (SUNY),

Syracuse, New York, USA P. Cuijpers, VU University, Amsterdam, Netherlands B. J.

Deacon, University of Wollongong, Wollongong, New South Wales, Australia S.

Dimidjian, University of Colorado, Boulder, Colorado, USA A. Ehlers, University of

Oxford, Oxford, UK C. A. Espie, University of Oxford, Oxford, UK J. C. Fournier,

University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, USA R. O.

Frost, Smith College, Northampton, Massachusetts, USA P. A. Garety, King''s College

London, London, UK A. Ghaderi, Karolinska Institutet, Stockholm, Sweden

AUTHOR INFORMATION PACK 17 Jul 2016 www.elsevier.com/locate/brat 3

C. M. Grilo, Yale University, New Haven, Connecticut, USA A. G. Harvey, University of

California at Berkeley, Berkeley, California, USA S. C. Hayes, University of Nevada,

65

Reno, Nevada, USA R. G. Heimberg, Temple University, Philadelphia, Pennsylvania,

USA S. G. Hofmann, Boston University, Boston, Massachusetts, USA D. R. Hopko,

University of Tennessee-Knoxville, Knoxville, Tennessee, USA J. L. Hudson,

Macquarie University, North Ryde, New South Wales, Australia E. H. W. Koster,

Universiteit Gent, Gent, Belgium W. Kuyken, University of Oxford, Oxford, UK A. J.

Lang, University of California at San Diego (UCSD), San Diego, USA J. D. Latner,

University of Hawaii at Mãnoa, Manoa, Hawaii, USA K. L. Loeb, Fairleigh Dickinson

University, Teaneck, New Jersey, USA D. E. McCarthy, Rutgers University, New

Brunswick, New Jersey, USA D. McKay, Fordham University, Bronx, New York, USA

R. J. McMahon, Simon Fraser University, Vancouver, British Columbia, Canada R. J.

McNally, Harvard University, Cambridge, Massachusetts, USA J. Miranda, UCLA, Los

Angeles, California, USA M. L. Moulds, UNSW Australia, Sydney, New South Wales,

Australia R. O. Nelson-Gray, University of North Carolina at Greensboro,

Greensboro, North Carolina, USA L.-G. Ost, Stockholms Universitet, Stockholm,

Sweden M. L. Perlis, University of Pennsylvania, Philadelphia, Pennsylvania, USA J.

B. Persons, Center for Cognitive Therapy, Oakland, California, USA M. B. Powers,

University of Texas at Austin, Austin, Texas, USA A. S. Radomsky, Concordia

University, Montréal, Quebec, Canada R. M. Rapee, Macquarie University, North

Ryde, New South Wales, Australia E. G. C. Rassin, Erasmus Universiteit, Rotterdam,

Netherlands B. Renneberg, Freie Universität Berlin, Berlin, Germany K. A. Rimes,

King's College London, London, UK L. Roemer, University of Massachusetts Boston,

Boston, Massachusetts, USA J. I. Ruzek, VA Palo Alto Health Care System, Menlo

Park, California, USA Y. Sakano, Waseda University, Tokorozawa-Shi, Japan D. M.

66

Sloan, Boston University School of Medicine, Boston, Massachusetts, USA J. A.

Smits, University of Texas at Austin, Austin, Texas, USA L. C. Sobell, Nova

Southeastern University (NSU), Ft. Lauderdale, Florida, USA S. H. Spence, Griffith

University, Nathan, Queensland, Australia P. Spinhoven, Universiteit Leiden, Leiden,

Netherlands G. Steketee, Boston University, Boston, Massachusetts, USA R. Sysko,

Columbia University, New York, USA D. J. Taylor, University of North Texas, Denton,

Texas, USA M. A. van den Hout, Utrecht University, Utrecht, Netherlands M. L.

Whittal, University of British Columbia, Vancouver, British Columbia, Canada S. A

Wonderlich, University of North Dakota, Fargo, North Dakota, USA W. Yule, King's

College London, London, UK Associate Editorial Board A. R. Ashbaugh, University of

Ottawa, Ottawa, Ontario, Canada J. Bomyea, University of California, San Diego, La

Jolla, California, USA J. F. Boswell, University at Albany, SUNY, Albany, New York,

USA J. D. Clapp, University of Wyoming, Laramie, Wyoming, USA A. E Coughtrey,

University College London (UCL), London, UK H. F. Dodd, University of Reading,

Reading, UK M. W. Gallagher, University of Houston, Houston, Texas, USA E.

Hedman, Karolinska Institutet, Stockholm, Sweden A. Hereen, Université Caholique

de Louvain, Louvain-la-Neuve, Belgium H. Hesser, Linköping University, Linköping,

Sweden K. J. Hsu, University of California at Los Angeles (UCLA), Los Angeles,

California, USA K. Kircanski, National Institute of Mental Health (NIMH), Bethesda,

Maryland, USA N. Koerner, Ryerson University, Toronto, Ontario, Canada J. R. Kuo,

Ryerson University, Toronto, Ontario, Canada R. T. LeBeau, University of California,

Los Angeles, California, USA B. Ljotssons, Karolinska Institutet, Stockholm, Sweden

K. Naragon-Gainey, University at Buffalo, State University of New York, Buffalo, New

67

York, USA J. M. Newby, University of New South Wales, Darlinghurst, New South

Wales at St Vincent''s Hospital, Sydney, Australia A. J. Ouimet, University of Ottawa,

Ottawa, Ontario, Canada J. R. Peters, Alpert Medical School at Brown University,

Providence, Rhode Island, USA D. J. Robinhaugh, Massachusetts General Hospital,

Boston, Massachusetts, USA

AUTHOR INFORMATION PACK 17 Jul 2016 www.elsevier.com/locate/brat 4

A. J. Rosellini, Harvard Medical School, Boston, Massachusetts, USA S. Sauer-Zavala, Boston

University, Boston, Massachusetts, USA C. T. Taylor, University of California, San

Diego, La Jolla, California, USA K. R. Timpano, University of Miami, Coral Gables,

Florida, USA M. Treanor, University of California at Los Angeles, Los Angeles,

California, USA A. A. Vujanovic, University of Texas Health Science Center at

Houston, Houston, Texas, USA B. E Wisco, University of North Carolina at

Greensboro, Greensboro, North Carolina, USA K. B. Taylor, University of California

at Los Angeles, Los Angeles, California, USA Q. J. J. Wong, Macquarie University,

New South Wales, Australia K. S. Young, University of California at Los Angeles

(UCLA), Los Angeles, California, USA

AUTHOR INFORMATION PACK 17 Jul 2016 www.elsevier.com/locate/brat 5

GUIDE FOR AUTHORS

. INTRODUCTION The major focus of Behaviour Research and Therapy is an experimental

psychopathology approach to understanding emotional and behavioral disorders

and their prevention and treatment, using cognitive, behavioral, and

psychophysiological (including neural) methods and models. This includes

laboratory-based experimental studies with healthy, at risk and subclinical

68

individuals that inform clinical application as well as studies with clinically severe

samples. The following types of submissions are encouraged: theoretical reviews of

mechanisms that contribute to psychopathology and that offer new treatment

targets; tests of novel, mechanistically focused psychological interventions,

especially ones that include theory-driven or experimentally-derived predictors,

moderators and mediators; and innovations in dissemination and implementation

of evidence-based practices into clinical practice in psychology and associated

fields, especially those that target underlying mechanisms or focus on novel

approaches to treatment delivery. In addition to traditional psychological disorders,

the scope of the journal includes behavioural medicine (e.g., chronic pain). The

journal will not consider manuscripts dealing primarily with measurement,

psychometric analyses, and personality assessment.

The Editor and Associate Editors will make an initial determination of whether or not

submissions fall within the scope of the journal and/or are of sufficient merit and

importance to warrant full review.

Early Career Investigator Award This award is open to papers where the first author on the

accepted papers is within 7 years of their PhD. By endorsing candidature for the

annual Early Career Investigator Award, your manuscript will be reviewed by the

Associate Editors/Editor-in-Chief for an annual award for the most highly rated

paper. The winner will be announced in print, and will have the option of being

spotlighted (photo and short bio).

The CONSORT guidelines (http://www.consort-statement.org/?) need to be followed for

protocol papers for trials; authors should present a flow diagramme and attach with

69

their cover letter the CONSORT checklist. For meta-analysis, the PRISMA

(http://www.prisma-statement.org/?) guidelines should be followed; authors

should present a flow diagramme and attach with their cover letter the PRISMA

checklist. For systematic reviews it is recommended that the PRISMA guidelines are

followed, although it is not compulsory.

Contact details Any questions regarding your submission should be addressed to the Editor

in Chief: Professor Michelle G. Craske Department of Psychology 310 825-8403

Email: brat@psych.ucla.edu BEFORE YOU BEGIN Ethics in publishing Please see our

information pages on Ethics in publishing and Ethical guidelines for journal

publication. Human and animal rights If the work involves the use of human

subjects, the author should ensure that the work described has been carried out in

accordance with The Code of Ethics of the World Medical Association (Declaration

of Helsinki) for experiments involving humans; Uniform Requirements for

manuscripts submitted to Biomedical journals. Authors should include a statement

in the manuscript that informed consent was obtained for experimentation with

human subjects. The privacy rights of human subjects must always be observed.

AUTHOR INFORMATION PACK 17 Jul 2016 www.elsevier.com/locate/brat 6

All animal experiments should comply with the ARRIVE guidelines and should be carried

out in accordance with the U.K. Animals (Scientific Procedures) Act, 1986 and

associated guidelines, EU Directive 2010/63/EU for animal experiments, or the

National Institutes of Health guide for the care and use of Laboratory animals (NIH

Publications No. 8023, revised 1978) and the authors should clearly indicate in the

manuscript that such guidelines have been followed. Conflict of Interest All authors

70

are requested to disclose any actual or potential conflict of interest including any

financial, personal or other relationships with other people or organizations within

three years of beginning the submitted work that could inappropriately influence,

or be perceived to influence, their work. See also

http://www.elsevier.com/conflictsofinterest. The Conflict of Interest form can be

found at: http://ees.elsevier.com/brat/img/COI.pdf . And for further information,

please view the following link:

http://service.elsevier.com/app/answers/detail/a_id/286/supporthub/publishing .

Submission declaration Submission of an article implies that the work described has

not been published previously (except in the form of an abstract or as part of a

published lecture or academic thesis or as an electronic preprint, see 'Multiple,

redundant or concurrent publication' section of our ethics policy for more

information), that it is not under consideration for publication elsewhere, that its

publication is approved by all authors and tacitly or explicitly by the responsible

authorities where the work was carried out, and that, if accepted, it will not be

published elsewhere including electronically in the same form, in English or in any

other language, without the written consent of the copyright-holder. Changes to

authorship Authors are expected to consider carefully the list and order of authors

before submitting their manuscript and provide the definitive list of authors at the

time of the original submission. Any addition, deletion or rearrangement of author

names in the authorship list should be made only before the manuscript has been

accepted and only if approved by the journal Editor. To request such a change, the

Editor must receive the following from the corresponding author: (a) the reason for

71

the change in author list and (b) written confirmation (e-mail, letter) from all

authors that they agree with the addition, removal or rearrangement. In the case of

addition or removal of authors, this includes confirmation from the author being

added or removed. Only in exceptional circumstances will the Editor consider the

addition, deletion or rearrangement of authors after the manuscript has been

accepted. While the Editor considers the request, publication of the manuscript will

be suspended. If the manuscript has already been published in an online issue, any

requests approved by the Editor will result in a corrigendum. Article transfer service

This journal is part of our Article Transfer Service. This means that if the Editor feels

your article is more suitable in one of our other participating journals, then you may

be asked to consider transferring the article to one of those. If you agree, your

article will be transferred automatically on your behalf with no need to reformat.

Please note that your article will be reviewed again by the new journal. More

information. Copyright Upon acceptance of an article, authors will be asked to

complete a 'Journal Publishing Agreement' (see more information on this). An e-

mail will be sent to the corresponding author confirming receipt of the manuscript

together with a 'Journal Publishing Agreement' form or a link to the online version

of this agreement.

Subscribers may reproduce tables of contents or prepare lists of articles including abstracts

for internal circulation within their institutions. Permission of the Publisher is

required for resale or distribution outside the institution and for all other derivative

works, including compilations and translations. If excerpts from other copyrighted

works are included, the author(s) must obtain written permission from the

72

copyright owners and credit the source(s) in the article. Elsevier has preprinted

forms for use by authors in these cases.

For open access articles: Upon acceptance of an article, authors will be asked to complete

an 'Exclusive License Agreement' (more information). Permitted third party reuse of

open access articles is determined by the author's choice of user license.

Author rights

AUTHOR INFORMATION PACK 17 Jul 2016 www.elsevier.com/locate/brat 7

As an author you (or your employer or institution) have certain rights to reuse your work.

More information. Elsevier supports responsible sharing Find out how you can share

your research published in Elsevier journals. Role of the funding source You are

requested to identify who provided financial support for the conduct of the

research and/or preparation of the article and to briefly describe the role of the

sponsor(s), if any, in study design; in the collection, analysis and interpretation of

data; in the writing of the report; and in the decision to submit the article for

publication. If the funding source(s) had no such involvement then this should be

stated. Funding body agreements and policies Elsevier has established a number of

agreements with funding bodies which allow authors to comply with their funder's

open access policies. Some funding bodies will reimburse the author for the Open

Access Publication Fee. Details of existing agreements are available online. Open

access This journal offers authors a choice in publishing their research:

Open access • Articles are freely available to both subscribers and the wider public with

permitted reuse. • An open access publication fee is payable by authors or on their

behalf, e.g. by their research funder or institution. Subscription • Articles are made

73

available to subscribers as well as developing countries and patient groups through

our universal access programs. • No open access publication fee payable by

authors.

Regardless of how you choose to publish your article, the journal will apply the same peer

review criteria and acceptance standards.

For open access articles, permitted third party (re)use is defined by the following Creative

Commons user licenses: Creative Commons Attribution (CC BY) Lets others

distribute and copy the article, create extracts, abstracts, and other revised

versions, adaptations or derivative works of or from an article (such as a

translation), include in a collective work (such as an anthology), text or data mine

the article, even for commercial purposes, as long as they credit the author(s), do

not represent the author as endorsing their adaptation of the article, and do not

modify the article in such a way as to damage the author's honor or reputation.

Creative Commons Attribution-NonCommercial-NoDerivs (CC BY-NC-ND) For non-

commercial purposes, lets others distribute and copy the article, and to include in a

collective work (such as an anthology), as long as they credit the author(s) and

provided they do not alter or modify the article. The open access publication fee for

this journal is USD 3000, excluding taxes. Learn more about Elsevier's pricing policy:

https://www.elsevier.com/openaccesspricing. Green open access Authors can share

their research in a variety of different ways and Elsevier has a number of green

open access options available. We recommend authors see our green open access

page for further information. Authors can also self-archive their manuscripts

immediately and enable public access from their institution's repository after an

74

embargo period. This is the version that has been accepted for publication and

which typically includes author-incorporated changes suggested during submission,

peer review and in editor-author communications. Embargo period: For

subscription articles, an appropriate amount of time is needed for journals to

deliver value to subscribing customers before an article becomes freely available to

the public. This is the embargo period and it begins from the date the article is

formally published online in its final and fully citable form. This journal has an

embargo period of 24 months.

AUTHOR INFORMATION PACK 17 Jul 2016 www.elsevier.com/locate/brat 8

Elsevier Publishing Campus The Elsevier Publishing Campus (www.publishingcampus.com)

is an online platform offering free lectures, interactive training and professional

advice to support you in publishing your research. The College of Skills training

offers modules on how to prepare, write and structure your article and explains

how editors will look at your paper when it is submitted for publication. Use these

resources, and more, to ensure that your submission will be the best that you can

make it. Language (usage and editing services) Please write your text in good English

(American or British usage is accepted, but not a mixture of these). Authors who

feel their English language manuscript may require editing to eliminate possible

grammatical or spelling errors and to conform to correct scientific English may wish

to use the English Language Editing service available from Elsevier's WebShop.

Submission Our online submission system guides you stepwise through the process

of entering your article details and uploading your files. The system converts your

article files to a single PDF file used in the peer-review process. Editable files (e.g.,

75

Word, LaTeX) are required to typeset your article for final publication. All

correspondence, including notification of the Editor's decision and requests for

revision, is sent by e-mail. Submit your article Please submit your article via

http://ees.elsevier.com/brat/ PREPARATION Article structure Subdivision -

unnumbered sections Divide your article into clearly defined sections. Each

subsection is given a brief heading. Each heading should appear on its own separate

line. Subsections should be used as much as possible when crossreferencing text:

refer to the subsection by heading as opposed to simply 'the text'. Appendices If

there is more than one appendix, they should be identified as A, B, etc. Formulae

and equations in appendices should be given separate numbering: Eq. (A.1), Eq.

(A.2), etc.; in a subsequent appendix, Eq. (B.1) and so on. Similarly for tables and

figures: Table A.1; Fig. A.1, etc. Essential title page information • Title. Concise and

informative. Titles are often used in information-retrieval systems. Avoid

abbreviations and formulae where possible. • Author names and affiliations. Please

clearly indicate the given name(s) and family name(s) of each author and check that

all names are accurately spelled. Present the authors' affiliation addresses (where

the actual work was done) below the names. Indicate all affiliations with a

lowercase superscript letter immediately after the author's name and in front of the

appropriate address. Provide the full postal address of each affiliation, including the

country name and, if available, the e-mail address of each author. • Corresponding

author. Clearly indicate who will handle correspondence at all stages of refereeing

and publication, also post-publication. Ensure that the e-mail address is given and

that contact details are kept up to date by the corresponding author. •

76

Present/permanent address. If an author has moved since the work described in the

article was done, or was visiting at the time, a 'Present address' (or 'Permanent

address') may be indicated as a footnote to that author's name. The address at

which the author actually did the work must be retained as the main, affiliation

address. Superscript Arabic numerals are used for such footnotes. Abstract A

concise and factual abstract is required with a maximum length of 200 words. The

abstract should state briefly the purpose of the research, the principal results and

major conclusions. An abstract is often presented separately from the article, so it

must be able to stand alone. For this reason, References should be avoided, but if

essential, then cite the author(s) and year(s). Also, non-standard or uncommon

abbreviations should be avoided, but if essential they must be defined at their first

mention in the abstract itself. Graphical abstract Although a graphical abstract is

optional, its use is encouraged as it draws more attention to the online article. The

graphical abstract should summarize the contents of the article in a concise,

pictorial form designed to capture the attention of a wide readership. Graphical

abstracts should be submitted as a

AUTHOR INFORMATION PACK 17 Jul 2016 www.elsevier.com/locate/brat 9

separate file in the online submission system. Image size: Please provide an image with a

minimum of 531 × 1328 pixels (h × w) or proportionally more. The image should be

readable at a size of 5 × 13 cm using a regular screen resolution of 96 dpi. Preferred

file types: TIFF, EPS, PDF or MS Office files. You can view Example Graphical

Abstracts on our information site. Authors can make use of Elsevier's Illustration

and Enhancement service to ensure the best presentation of their images and in

77

accordance with all technical requirements: Illustration Service. Highlights

Highlights are mandatory for this journal. They consist of a short collection of bullet

points that convey the core findings of the article and should be submitted in a

separate editable file in the online submission system. Please use 'Highlights' in the

file name and include 3 to 5 bullet points (maximum 85 characters, including

spaces, per bullet point). You can view example Highlights on our information site.

Keywords Immediately after the abstract, provide a maximum of 6 keywords, to be

chosen from the APA list of index descriptors. These keywords will be used for

indexing purposes. Abbreviations Define abbreviations that are not standard in this

field in a footnote to be placed on the first page of the article. Such abbreviations

that are unavoidable in the abstract must be defined at their first mention there, as

well as in the footnote. Ensure consistency of abbreviations throughout the article.

Acknowledgements Collate acknowledgements in a separate section at the end of

the article before the references and do not, therefore, include them on the title

page, as a footnote to the title or otherwise. List here those individuals who

provided help during the research (e.g., providing language help, writing assistance

or proof reading the article, etc.). Formatting of funding sources List funding sources

in this standard way to facilitate compliance to funder's requirements:

Funding: This work was supported by the National Institutes of Health [grant numbers xxxx,

yyyy]; the Bill & Melinda Gates Foundation, Seattle, WA [grant number zzzz]; and

the United States Institutes of Peace [grant number aaaa].

It is not necessary to include detailed descriptions on the program or type of grants and

awards. When funding is from a block grant or other resources available to a

78

university, college, or other research institution, submit the name of the institute or

organization that provided the funding.

If no funding has been provided for the research, please include the following sentence:

This research did not receive any specific grant from funding agencies in the public,

commercial, or not-for-profit sectors. Shorter communications This option is

designed to allow publication of research reports that are not suitable for

publication as regular articles. Shorter Communications are appropriate for articles

with a specialized focus or of particular didactic value. Manuscripts should be

between 3000-5000 words, and must not exceed the upper word limit. This limit

includes the abstract, text, and references, but not the title page, tables and figures.

Artwork Electronic artwork General points • Make sure you use uniform lettering

and sizing of your original artwork. • Embed the used fonts if the application

provides that option. • Aim to use the following fonts in your illustrations: Arial,

Courier, Times New Roman, Symbol, or use fonts that look similar. • Number the

illustrations according to their sequence in the text. • Use a logical naming

convention for your artwork files. • Provide captions to illustrations separately. •

Size the illustrations close to the desired dimensions of the published version. •

Submit each illustration as a separate file.

AUTHOR INFORMATION PACK 17 Jul 2016 www.elsevier.com/locate/brat 10

A detailed guide on electronic artwork is available. You are urged to visit this site; some

excerpts from the detailed information are given here. Formats If your electronic

artwork is created in a Microsoft Office application (Word, PowerPoint, Excel) then

please supply 'as is' in the native document format. Regardless of the application

79

used other than Microsoft Office, when your electronic artwork is finalized, please

'Save as' or convert the images to one of the following formats (note the resolution

requirements for line drawings, halftones, and line/halftone combinations given

below): EPS (or PDF): Vector drawings, embed all used fonts. TIFF (or JPEG): Color or

grayscale photographs (halftones), keep to a minimum of 300 dpi. TIFF (or JPEG):

Bitmapped (pure black & white pixels) line drawings, keep to a minimum of 1000

dpi. TIFF (or JPEG): Combinations bitmapped line/half-tone (color or grayscale),

keep to a minimum of 500 dpi. Please do not: • Supply files that are optimized for

screen use (e.g., GIF, BMP, PICT, WPG); these typically have a low number of pixels

and limited set of colors; • Supply files that are too low in resolution; • Submit

graphics that are disproportionately large for the content. Tables Please submit

tables as editable text and not as images. Tables can be placed either next to the

relevant text in the article, or on separate page(s) at the end. Number tables

consecutively in accordance with their appearance in the text and place any table

notes below the table body. Be sparing in the use of tables and ensure that the data

presented in them do not duplicate results described elsewhere in the article.

Please avoid using vertical rules. References Citation in text Please ensure that

every reference cited in the text is also present in the reference list (and vice versa).

Any references cited in the abstract must be given in full. Unpublished results and

personal communications are not recommended in the reference list, but may be

mentioned in the text. If these references are included in the reference list they

should follow the standard reference style of the journal and should include a

substitution of the publication date with either 'Unpublished results' or 'Personal

80

communication'. Citation of a reference as 'in press' implies that the item has been

accepted for publication. Web references As a minimum, the full URL should be

given and the date when the reference was last accessed. Any further information,

if known (DOI, author names, dates, reference to a source publication, etc.), should

also be given. Web references can be listed separately (e.g., after the reference list)

under a different heading if desired, or can be included in the reference list.

Reference management software Most Elsevier journals have their reference

template available in many of the most popular reference management software

products. These include all products that support Citation Style Language styles,

such as Mendeley and Zotero, as well as EndNote. Using the word processor plug-

ins from these products, authors only need to select the appropriate journal

template when preparing their article, after which citations and bibliographies will

be automatically formatted in the journal's style. If no template is yet available for

this journal, please follow the format of the sample references and citations as

shown in this Guide. Users of Mendeley Desktop can easily install the reference

style for this journal by clicking the following link: http://open.mendeley.com/use-

citation-style/behaviour-research-and-therapy When preparing your manuscript,

you will then be able to select this style using the Mendeley plugins for Microsoft

Word or LibreOffice. Reference style Text: Citations in the text should follow the

referencing style used by the American Psychological Association. You are referred

to the Publication Manual of the American Psychological Association, Sixth Edition,

ISBN 978-1-4338-0561-5, copies of which may be ordered online or APA Order

81

Dept., P.O.B. 2710, Hyattsville, MD 20784, USA or APA, 3 Henrietta Street, London,

WC3E 8LU, UK.

AUTHOR INFORMATION PACK 17 Jul 2016 www.elsevier.com/locate/brat 11

List: references should be arranged first alphabetically and then further sorted

chronologically if necessary. More than one reference from the same author(s) in

the same year must be identified by the letters 'a', 'b', 'c', etc., placed after the year

of publication. Examples: Reference to a journal publication: Van der Geer, J.,

Hanraads, J. A. J., & Lupton, R. A. (2010). The art of writing a scientific article.

Journal of Scientific Communications, 163, 51–59. Reference to a book: Strunk, W.,

Jr., & White, E. B. (2000). The elements of style. (4th ed.). New York: Longman,

(Chapter 4). Reference to a chapter in an edited book: Mettam, G. R., & Adams, L. B.

(2009). How to prepare an electronic version of your article. In B. S. Jones, & R. Z.

Smith (Eds.), Introduction to the electronic age (pp. 281–304). New York: E-

Publishing Inc. Reference to a website: Cancer Research UK. Cancer statistics

reports for the UK. (2003). http://www.cancerresearchuk.org/

aboutcancer/statistics/cancerstatsreport/ Accessed 13.03.03. Video Elsevier accepts

video material and animation sequences to support and enhance your scientific

research. Authors who have video or animation files that they wish to submit with

their article are strongly encouraged to include links to these within the body of the

article. This can be done in the same way as a figure or table by referring to the

video or animation content and noting in the body text where it should be placed.

All submitted files should be properly labeled so that they directly relate to the

video file's content. In order to ensure that your video or animation material is

82

directly usable, please provide the files in one of our recommended file formats

with a preferred maximum size of 150 MB. Video and animation files supplied will

be published online in the electronic version of your article in Elsevier Web

products, including ScienceDirect. Please supply 'stills' with your files: you can

choose any frame from the video or animation or make a separate image. These will

be used instead of standard icons and will personalize the link to your video data.

For more detailed instructions please visit our video instruction pages. Note: since

video and animation cannot be embedded in the print version of the journal, please

provide text for both the electronic and the print version for the portions of the

article that refer to this content. Supplementary material Supplementary material

can support and enhance your scientific research. Supplementary files offer the

author additional possibilities to publish supporting applications, high-resolution

images, background datasets, sound clips and more. Please note that such items are

published online exactly as they are submitted; there is no typesetting involved

(supplementary data supplied as an Excel file or as a PowerPoint slide will appear as

such online). Please submit the material together with the article and supply a

concise and descriptive caption for each file. If you wish to make any changes to

supplementary data during any stage of the process, then please make sure to

provide an updated file, and do not annotate any corrections on a previous version.

Please also make sure to switch off the 'Track Changes' option in any Microsoft

Office files as these will appear in the published supplementary file(s). For more

detailed instructions please visit our artwork instruction pages. AudioSlides The

journal encourages authors to create an AudioSlides presentation with their

83

published article. AudioSlides are brief, webinar-style presentations that are shown

next to the online article on ScienceDirect. This gives authors the opportunity to

summarize their research in their own words and to help readers understand what

the paper is about. More information and examples are available. Authors of this

journal will automatically receive an invitation e-mail to create an AudioSlides

presentation after acceptance of their paper. Interactive plots This journal enables

you to show an Interactive Plot with your article by simply submitting a data file.

Full instructions. Submission checklist The following list will be useful during the

final checking of an article prior to sending it to the journal for review. Please

consult this Guide for Authors for further details of any item. Ensure that the

following items are present:

AUTHOR INFORMATION PACK 17 Jul 2016 www.elsevier.com/locate/brat 12

One author has been designated as the corresponding author with contact details: • E-mail

address • Full postal address All necessary files have been uploaded, and contain: •

Keywords • All figure captions • All tables (including title, description, footnotes)

Further considerations • Manuscript has been 'spell-checked' and 'grammar-

checked' • References are in the correct format for this journal • All references

mentioned in the Reference list are cited in the text, and vice versa • Permission has

been obtained for use of copyrighted material from other sources (including the

Internet) Printed version of figures (if applicable) in color or black-and-white •

Indicate clearly whether or not color or black-and-white in print is required. For any

further information please visit our Support Center. AFTER ACCEPTANCE Online

proof correction Corresponding authors will receive an e-mail with a link to our

84

online proofing system, allowing annotation and correction of proofs online. The

environment is similar to MS Word: in addition to editing text, you can also

comment on figures/tables and answer questions from the Copy Editor. Web-based

proofing provides a faster and less error-prone process by allowing you to directly

type your corrections, eliminating the potential introduction of errors. If preferred,

you can still choose to annotate and upload your edits on the PDF version. All

instructions for proofing will be given in the e-mail we send to authors, including

alternative methods to the online version and PDF. We will do everything possible

to get your article published quickly and accurately. Please use this proof only for

checking the typesetting, editing, completeness and correctness of the text, tables

and figures. Significant changes to the article as accepted for publication will only be

considered at this stage with permission from the Editor. It is important to ensure

that all corrections are sent back to us in one communication. Please check carefully

before replying, as inclusion of any subsequent corrections cannot be guaranteed.

Proofreading is solely your responsibility. Offprints The corresponding author will, at

no cost, receive a customized Share Link providing 50 days free access to the final

published version of the article on ScienceDirect. The Share Link can be used for

sharing the article via any communication channel, including email and social

media. For an extra charge, paper offprints can be ordered via the offprint order

form which is sent once the article is accepted for publication. Both corresponding

and co-authors may order offprints at any time via Elsevier's Webshop.

Corresponding authors who have published their article open access do not receive

a Share Link as their final published version of the article is available open access on

85

ScienceDirect and can be shared through the article DOI link. Elsevier NIH Policy

Statement As a service to our authors, Elsevier will deposit to PubMed Central

(PMC) author manuscripts on behalf of Elsevier authors reporting NIH funded

research. This service is a continuation of Elsevier's 2005 agreement with the NIH

when the NIH introduced their voluntary 'Public Access Policy'. Please see the full

details at:

http://www.elsevier.com/wps/find/authorsview.authors/nihauthorrequest (this

site also includes details on all other funding body agreements).

Elsevier facilitates author response to the NIH voluntary posting request (referred to as the

NIH "Public Access Policy", see http://www.nih.gov/about/publicaccess/index.htm)

by posting the peerreviewed author's manuscript directly to PubMed Central on

request from the author, 12 months after formal publication. Upon notification

from Elsevier of acceptance, we will ask you to confirm via email (by e-mailing us at

NIHauthorrequest@elsevier.com) that your work has received NIH funding and that

you intend to respond to the NIH policy request, along with your NIH award number

to facilitate processing. Upon such confirmation, Elsevier will submit to PubMed

Central on your behalf a version of your manuscript that will include peer-review

comments, for posting 12 months after

AUTHOR INFORMATION PACK 17 Jul 2016 www.elsevier.com/locate/brat 13

86

Appendix B: Recruitment for the study, email and social media advert.

Psychology Study: The impact of recalling unpleasant memories on our emotions.

I am a Trainee Clinical Psychologist undertaking some research and am currently recruiting participants. The whole study is conducted online and will take no more than 30 minutes. If you are over the age of 18 and can speak English then you are eligible to take part. If you follow the link below you will be taken to a page to give you some more information about the study and show you how to take part. This study is completely anonymous and you can withdraw at any-time.

Many thanks.

87

Appendix C: Ethical approval

88

89

Appendix D= MC Induction- task instructions

Now please close your eyes and form a mental image of this memory in your own mind for

two minutes. Imagining it as if it were happening again, through your own eyes and

focussing on any negative emotions you experienced at the time.

90

Once the two minutes is up you will be taken to the next page.

Appendix E= MC, Visual Analogue Scales

91

Appendix F: Writing task instructions

Control TaskNow I would like you to write about how you spend an average week day. I would like a description of a ‘normal’ day where nothing out of the ordinary happens. This might include:

What time you normally get up

What you usually eat for breakfast

Whether you go to work

How you normally travel to work

What you would usually have for lunch

What you would usually have for dinner.

It may help to take a minute to think about this memory before writing it down.  When you are ready you please write a description of this in the box below. This is not a timed task and you can take as long as you want. However, you will not be able to advance to the next page until two minutes has passed.

Self-esteem TaskNow I would like you to write about a time when you felt worthwhile, valued and empowered. Examples may include but are not limited to:

A time when you were praised for doing well at work. A time when you achieved a goal you had set yourself. A time when you helped another person in some way. A time when you did something for the benefit of others.

These examples are given as a guideline only. You can recall a memory that is different to these examples. The main thing though is that it is a memory of a time when you felt worthwhile, valued or empowered

It may help to take a minute to think about this memory before writing it down.  When you are ready you please write a description of this in the box below. Include

92

any positive emotions or thoughts you had at the time. This is not a timed task and you can take as long as you want. However, you will not be able to advance to the next page until two minutes has passed.

Self-Compassion task:In relation to the memory you described earlier about of a time when you felt humiliated, ashamed or betrayed, or harmed in some way by someone else. I want you to write about this event again but expressing kindness and compassion towards yourself and your feelings about it. It might help to think about how you would express concern for a friend who had undergone the same experience. For example, if your friend had been cheated on by a partner how would you help them feel better and show them care and understanding?

It may help to take a minute to think about this before writing it down.  When you are ready please write a description of this in the box below. Include any positive emotions or thoughts you may have. This is not a timed task and you can take as long as you want. However, you will not be able to advance to the next page until two minutes has passed.

93

Appendix G: Information sheet

Information Sheet

Title: The impact of recalling unpleasant memories on our emotions I would like to invite you to take part in my study looking at the impact of recalling unpleasant memories on our emotions, how long this impact lasts and how we can overcome this. Please take the time to read the following information before deciding whether you would like to take part in this study. Who is conducting the research?My name is Phoebe Horrocks, a Trainee Clinical Psychologist at the University of Surrey. This research forms part of my doctoral training and is supervised by Dr Laura Simonds, a Chartered Psychologist and Lecturer at the University of Surrey.

Am I eligible to take part?Anyone aged over 18 years of age is eligible to take part in this research. You will also need to be able to think of a time or event where you felt humiliated, ashamed or betrayed or have been harmed by someone else in some way. If you are likely to become very upset or distressed when thinking about this time, you are advised NOT to take part. If you have any questions and would like more information before taking part, please email me: p.horrocks@surrey.ac.uk

What will I have to do?You will be asked to complete an online survey. During the survey you will be asked to complete some questions. You will then be asked to think of a time or event where you felt humiliated, ashamed or betrayed or have been harmed by someone else in some way. You will be given more guidance as to the type of memory to think about, but the memory you choose will be your choice. You will be asked to think about this memory for two minutes, after which you will be asked to complete some further questions. In the second part of the study you will be asked to think about a different memory or the same memory in a different way and then asked to repeat the short questions. Once you have started the survey it is important to complete it in one sitting as it cannot be saved and returned to at a later date.

94

The survey will take approximately 30 minutes to complete.

Do I have to take part?Participation in this research is entirely voluntary. You are under no obligations to take part and have a right to withdraw from the study at any point up until completion of the online survey. If you choose to withdraw you will be given the option as to whether the researcher can use the information you have submitted up until that point for the research. If you choose ‘no’ your data will be deleted. If you choose ‘yes’ your data will be kept anonymously. After completion of the study or clicking ‘yes’ after withdrawing, as all of the data is anonymised, I will be unable to remove your responses, but you will not be identifiable from your data. To withdraw from the research during the survey simply click ‘no’ at the bottom of the page when asked if you wish to continue. What will happen to my data?At no point will the survey ask for your name or any identifiable information so there will be no way to identify you or your responses. The research data will be handled by the researcher and shared with the supervisor. In line with the Data Protection Act 1998, all data will be securely stored. It will be stored for at least 10 years from the end date of the study. The study should be completed by September 2016. It is usual practice for researchers to publish their findings in professional journals so that research can be shared within the profession. Again, there will be no way you can be identified in any publications. What are the benefits and downsides of taking part in this research?Whilst there may be no direct benefits to you taking part in this research you will be contributing to the development of the field of Clinical Psychology and enhancing how we understand and respond to certain events and situations. This will be beneficial in developing ways to support people’s wellbeing.

I do not anticipate any significant risks to you in taking part in this study. However, you will be asked to think about a time someone hurt or upset you which may involve some uncomfortable thoughts and emotions. If you are likely to become significantly distressed doing this, then please DO NOT take part. If you find some of the questions too upsetting or personal then you do not have to answer them and can leave the survey at any point. As we are aware of the possibility that some

95

people may be caused upset upon completion of this research, you will receive a full debriefing at the end of the survey and will be signposted to appropriate sources of support if you feel that you need to discuss things further. 

Thank you for taking the time to read this information sheet. 

If you would like to continue to take part in the research then select the arrow at the bottom of the page. If you have decided not to take part, then please close the

browser. Who can I contact about this research? Researcher:Phoebe Horrocks, Trainee Clinical Psychologist, PsychD Clinical Psychology ProgrammeSchool of Psychology, University of Surrey, Guildford, GU2 7XHEmail: p.horrocks@surrey.ac.ukSupervisor: Dr. Laura Simonds, Lecturer, School of Psychology, University of Surrey. Email: l.simonds@surrey.ac.uk. Tel: 01483 686936

96

Appendix H: Consent form

Consent form

I have read and understood the Information Screen. I have been given an

explanation of the nature, purpose, and likely duration of the study, and of

what I will be expected to do.

I have been advised that participation involves thinking of a memory that

might be upsetting. I have been informed that sources of information and

support will be provided that I can contact after the study if necessary.

I have been given the researcher’s details and have had the opportunity to

contact them and to ask questions about the study.

I understand that I can withdraw from the study whilst I am taking part if I no

longer wish to continue.

I understand that if I withdraw part way through the study I will be given the

option of whether the data I have submitted to this point can be used by the

researcher or not.

I understand that I cannot withdraw my data once I have completed the study

and submitted my data.

I understand that the researcher will be writing about the results of the study

and may publish them. I am happy for the researcher to do this on the

understanding that I cannot be identified.

97

I understand that all personal data relating to participants is held and

processed in the strictest confidence, and in accordance with the Data

Protection Act 1998.

I confirm that I have read and understood the above and freely consent to

participating in this study. I have been given adequate time to consider my

participation and agree to comply with the instructions and restrictions of the

study.

By clicking the arrow below you consent to taking part in this study. If you do not

wish to take part, close your web browser now

Appendix I: De-brief

De-briefThank you for participating in this research. The survey is nearly complete. Below is

some further  information about  the nature of the study. 

Post-survey Information Sheet You were told on the information sheet that this research was investigating the impact of recalling unpleasant memories on our emotions, how long this lasts and how we can overcome it. This is true, this research wanted to investigate factors that help to reduce something called ‘Mental Contamination’. Previous research has found that recalling unpleasant memories can lead people to experiencing “mental contamination”, where you might feel internally dirty and have an increase of negative emotions including: shame, guilt, sadness, fear and anxiety. This study was interested in the feelings that people have after they remember a time that someone else made them feel bad. We are studying this so we can help develop interventions for people who have difficult experiences in their lives. You were allocated to one of 3 conditions as we were trying to find out if writing about an event in particular ways can reduce any negative feelings associated with the memory.

1. The control group- who had to write about an average day.2. The cognitive restructuring group- who had to write about a time they felt

valued, empowered and worthwhile.3. The compassionate group- who had to write about the original memory but

from a compassionate perspective.

98

As we asked you to remember some bad event we also then asked you to remember something positive happening to you. It is normal for all of us to have both positive and negative experiences and also for us to have difficult emotions when we think of bad times. We don’t expect any negative emotions to last a long time but if this study has made you want to find out more information about mental health or if you have difficult thoughts and feelings then please see the websites below.

 Mental Health Charities:SANE: http://www.sane.org.uk/ or helpline: 0845 767 8000MIND: http://www.mind.org.uk/OCD UK: http://www.ocduk.org/ - online support and discussion forumsOCD Action: http://www.ocdaction.org.uk/ or helpline: 0845 390 6232 Alternatively, you can also contact your GP to discuss any concerns or questions that you may have about mental health. If you have any further questions about the research then please do not hesitate to contact Phoebe on p.horrocks@surrey.ac.uk. 

Thank you again for you participation.

Appendix J: Participation drop out and exclusion for MC induction

99

100

197 participants accessed the survey online.

146 participants consented to take part in the survey

2 participants dropped out after giving consent, before going further

35 participants dropped out after reading MC induction task instructions

109 participants completed pre MC measures and MC induction task

4 participants dropped out before completing the post MC measures

11 participants were removed due to timing errors

1 participant was removed due to poor task adherence

93 participant data was analysed after MC induction

task

Appendix K: Normality checks for MC 1 and MC 2 scores across all indices.Internal dirtiness:

Pre MC induction Internal dirtiness:

101

Post MC Induction internal dirtiness

102

103

Pre MC Induction: Anxiety

104

105

Post MC induction: anxiety

106

107

Pre MC Induction: Shame

108

109

Post MC Induction: Shame

110

111

Pre MC Induction: Guilt

112

113

Post MC Induction: Guilt

114

115

Pre MC Induction: Fear

116

117

Post MC induction: Fear

118

119

Pre MC Induction: Sadness:

120

121

Post MC Induction: Sadness

122

123

Pre MC Induction: Humiliation

124

125

Post MC Induction: Humiliation

126

127

128

Appendix L: Normality tests for all MC indexes (MC 1-pre induction and MC 2- post induction)

Tests of Normality

Kolmogorov-Smirnova Shapiro-Wilk

Statistic df Sig. Statistic df Sig.

Pre_internal_dirty .274 90 .000 .703 90 .000

Pre_Anxious .139 90 .000 .901 90 .000

Pre_shame .262 90 .000 .678 90 .000

Pre_guilty .230 90 .000 .753 90 .000

Pre_Fear .250 90 .000 .715 90 .000

Pre_Sad .264 90 .000 .663 90 .000

Pre_Humiliated .288 90 .000 .620 90 .000

Post_1_internal_dirty .240 90 .000 .785 90 .000

Post_1_Anxious .123 90 .002 .941 90 .000

Post_1_Shame .146 90 .000 .890 90 .000

Post_1_Guilt .199 90 .000 .844 90 .000

Post_1_Fear .203 90 .000 .800 90 .000

Post_1_Sad .114 90 .006 .921 90 .000

Post_1_Humiliated .119 90 .003 .914 90 .000

a. Lilliefors Significance Correction

129

Appendix M: Participation drop out and exclusion at writing tasks

130

93 Participants were randomised into the three writing groups.

14 dropped out before completing the writing tasks

79 participants completed the writing tasks

ControlN=30

CognitiveN=20 Compassion

N=29

Control0 removed due to task

adherence

Cognitive2 removed due to task

adherence (time).

Compassion4 removed due to task adherence. (1- time, 3,

writing)

ControlFinal total analysed:

N=30

CognitiveFinal total analysed:

N=18 CompassionFinal total analysed:

N=25

Appendix N: Examples from the writing tasks

Control:

“On a normal weekday, my alarms start going off around 6:30 and I get up anytime in the next 45 minutes. I get ready quite fast as we have a casual dress code at work. Makeup only takes me about ten minutes. I commute to work via bus and train which is jst under an hour. Depends on delays. It can sometimes be very crowded! Days when you get a seat are the best! We get free breakfast at work so I normally take my laptop to the cafe upstairs and check emails while eating with my team. The day can be good r bad based on those emails! I spend most of my day doing admin tasks, working on computers and dealing with support requests from other colleagues. It can be a very busy job but I enjoy it when I know what I'm doing and feel in control of the schedule. Lck of scheduling is really challenging for me. Lunch is in the cafe upstairs again, usually with the same team, we're quite good friends. I get to bring lunch from home but if I don't we have a lot to choose from in the cafe. Afternoon same sort of tasksas morning. I sometimes meet old school or uni friends for dinner in the evening, or if not travel home an hour and make myself dinner. I like to cook, particularly Mexican or Italian food. My housemates and I get along really well so sometimes we watch V together while we eat or other times we all keep to our separate rooms and Netflix. I shower at night as I'm not a morning person. Depending on how hooked I am in whatever is on Netflix, I'll go to bed between 10-11.”

“On a normal work day my alarm goes off about 7am, I snooze until bout 7.15am. Get up and have a shower, get dressed and dry my hair and do my makeup. I don't normally eat breakfast but have been trying to recently - Special K. I grab my lunch out of the ridge. I walk to work at about 8.10am and get there about 8.40am. For lunch I normally have the leftovers from the night before or make a couscous salad - I normally eat it at my desk. I usually leave work about 6.30pm and have all sorts of things for diner from all over the world (my boyfriend normally cooks and he's pretty good) - thai curry, moroccan stew, chilli etc. We normally watch TV if we don't have any plans and go to bed about 10.30/11pm.”

Self-esteem:“I was asked to talk at a work conference in front of about 60 senior clients and a couple of senior stakeholders from my company. To my slight suprise I absolutely nailed the presentation, I think maybe having a microphone and a proper auditorium actually helped. I recieved lots of possotive feedback for it and the work

131

i was presenting and the work has been held up as an example to the rest of the company. It even got called out by our main client in a weekly meeting how well i had done.”

“I was on a long weekend trip in Wales and I was with some people who were really into tombstoning. I had never done this before and I was really scared at the prospect of doing it. Because I wanted to impress the people I was with, I decided to give it a o. We had to swim across a freezing cold lagoon and on the way over I was really nervous. As we climbed the cliff, we finally got to the 25ft jump. As I walked close to the end, I was terrified and had lots of adrenaline running through me. I am not a masive fan of heights and I don't like feeling scared. After the firs person had jumped, I ended up just closing my eyes, breathing and jumped as I opened my eyes. Once I had landed in the sea and everything was fine, I felt really pleased with myself. I don’t often feel brave enough to do this sort of thing. I was also able to share that pride with the other members of the group, who seemed impressed and proud that I had managed to do my first ever jump. It left me feeling quite exhilarated.”

Self-Compassion:“My husband didn't cheat on me because I wasn't good enough or pretty enough. Perhaps he felt intimidated by my new business taking off. I wasn't a bad person to him and I always supported him and loved him unconditionally. He was in a dark place with his depression and made an error in judgement. He didn't do it to consciously hurt me.”

“This situation was out of your control. The other person involved after badly towards you and there was nothing you could do to change things. You did everything you could to try and make amends and save the friendship. In fact you probably tried harder tan most people would have. You coped really well given the difficult situation you were put in and the fact you were able to put it behind you and move on despite how upsetting it was for you demonstrates how maturely you dealt with the situation. I think most people would have felt pretty angry if they were in your position test you managed to hold that anger in when trying to resolve the situation.”

132

Appendix O: Distribution of age across the three groups

133

134

Key: Cognitive= Self-esteem task

135

136

Appendix P: Distribution of time spent on writing tasks in the three groups

Control:

137

Self-esteem:

138

139

Compassion:

140

141

Tests of Normality

Writing Condition

Kolmogorov-Smirnova Shapiro-Wilk

Statistic df Sig. Statistic df Sig.

Control Timing-#QuestionText,

TimingPageSubmit#.204 30 .003 .750 30 .000

Cognitive Timing-#QuestionText,

TimingPageSubmit#.227 18 .014 .833 18 .005

Compassion Timing-#QuestionText,

TimingPageSubmit#.143 25 .200* .911 25 .033

*. This is a lower bound of the true significance.

a. Lilliefors Significance Correction

142

Key: Cognitive=self-esteem

Appendix Q: : Normal distribution tests for word count across the three conditions

Control:

143

Cognitive:

144

145

Compassion:

146

147

Tests of Normality

Writing Condition

Kolmogorov-Smirnova Shapiro-Wilk

Statistic df Sig. Statistic df Sig.

Word_Count Control .195 30 .005 .831 30 .000

Cognitive .211 18 .034 .838 18 .005

Compassion .181 25 .033 .835 25 .001

a. Lilliefors Significance Correction

148

Appendix R: Distribution for MC change scores on all MC indices for all conditions

Internal dirtiness:

149

150

Anxiety:

151

152

153

Shame:

154

155

156

Guilt:

157

158

159

Fear:

160

161

162

Sadness:

163

164

165

Humiliation:

166

167

168

169

Appendix S: Normality tests for MC change score- all conditions and MC indices.

Tests of Normality

Writing Condition

Kolmogorov-Smirnova Shapiro-Wilk

Statistic df Sig. Statistic df Sig.

Control Anxious_Change .217 30 .001 .863 30 .001

Shame_Change .209 30 .002 .875 30 .002

Guilt_Change .249 30 .000 .732 30 .000

Fear_Change .253 30 .000 .681 30 .000

Sad_Change .148 30 .092 .912 30 .017

Humiliated_Change .146 30 .100 .913 30 .017

Cognitive Anxious_Change .118 18 .200* .936 18 .249

Shame_Change .166 18 .200* .948 18 .398

Guilt_Change .311 18 .000 .841 18 .006

Fear_Change .285 18 .000 .827 18 .004

Sad_Change .230 18 .013 .827 18 .004

Humiliated_Change .143 18 .200* .913 18 .097

Compassion Anxious_Change .107 25 .200* .979 25 .858

Shame_Change .151 25 .145 .941 25 .158

Guilt_Change .208 25 .007 .897 25 .016

Fear_Change .242 25 .001 .740 25 .000

Sad_Change .203 25 .009 .872 25 .005

Humiliated_Change .181 25 .034 .882 25 .008

*. This is a lower bound of the true significance.

a. Lilliefors Significance Correction

170

Appendix T: Levene’s test for MC change scores

Test of Homogeneity of Variances

Levene Statistic df1 df2 Sig.

Dirtiness_change 7.939 2 70 .001

Anxious_Change 4.950 2 70 .010

Shame_Change .307 2 70 .737

Guilt_Change .741 2 70 .480

Fear_Change .974 2 70 .383

Sad_Change .662 2 70 .519

Humiliated_Change .964 2 70 .386

171

MRP PROPOSAL

6289478

3,000 words

Title: Investigating factors associated with mental contamination.

172

1.1 Background and Rationale:

Mental contamination (MC) is a relatively new psychological concept which has

been gaining more research focus in recent years. Mental contamination,

conceptualised by Rachman, is defined as an internal feeling of dirtiness that can

arise without physical contact with a contaminate and is accompanied by negative

emotions such as: anxiety, disgust, fear, shame, guilt and revulsion and can produce

an urge to wash (Rachman 1994, 2004, 2006). MC sits within Rachman’s theory of

fear of contamination which includes both contact contamination and MC (Rachman

2004, 2006). Contact contamination arises after physical contact with something

perceived as soiled, impure, infectious or harmful and produces a similar reaction as

MC (Rachman 2004). Although contact contamination has traditionally been given

more focus in research more attention is beginning to be given to MC.

Rachman’s (2006) theory suggests that MC can arise from a physical violation (e.g.

sexual assault), a mental violation (e.g. feeling ashamed, degraded, humiliated,

betrayed) or self-contamination (unwanted thoughts, images or memories). The

source of mental contamination is usually human: a dirty, dangerous or immoral

human which can include the self (Rachaman 2006).

Clinically MC has been observed in obsessive compulsive disorder (OCD), Post

traumatic stress (PTSD) and in victims of sexual assault (Oluntaji, Elwood, Williams

& Lohr 2008; Warnock-Parkes, Salkovskis & Rachman 2008; Fairbrother &

Rachman 2004). MC is conceptually linked to OCD because of the role fear of

contamination plays in the disorder, over half of OCD sufferers have contamination

fears (Calamari, Weigartz, Riemann, Cohen, Greer et al 2004). Studies of OCD

sufferers found that over half had clinical levels of MC (Coughtrey et al 2012). MC

may be a distinct phenomenon in OCD which needs exploring and focussing on to

173

possibly elucidate the high relapse rates and treatment resistant nature often found in

OCD (Ponniah, Magiati & Hollon 2013). Furthermore, the emotions associated with

MC are also present in a number of other depressive and anxiety related conditions

suggesting MC has the potential to be a trans-diagnostic factor across other

psychological conditions (Fergus, Valentiner, McGrath & Jencius 2010; Sangmoon,

Ryan, Rendall 2011; Ille, Schoggl, Kapfhammer, Arendasy, Sommer & Schienle

2014). This highlights the importance of understanding the concept of MC,

consequences of MC and individual vulnerabilities to MC.

Research has found support for the concept of MC and that it can be induced through

scenarios where a moral transgression takes place. A number of studies induced MC

using a non-consensual kiss paradigm where female participants were asked to listen

to an audio tape and imagine being kissed at a party against their wishes (Fairbrother,

Newth & Rachman 2005; Herba & Rachman 2007; Elliott & Radomsky 2009, 2012).

In these studies women in the non-consensual conditions reported higher feelings of

internal dirtiness, urge to wash and negative emotions associated with MC. Similar

findings have been shown when men were asked to imagine perpetrating a non-

consensual kiss (Rachman, Radomsky, Elliott & Zysk 2012). These studies can been

criticised for lacking ecological validity and recent studies have found MC can also

be induced through autobiographical recall of moral transgressions (Coughtrey,

Shafran & Rachman 2014).

Research has begun to look at the factors associated with mental contamination, the

most explored variables in the research literature include: obsessive-compulsiveness,

PTSD symptoms, disgust, anxiety and cognitive biases. These factors have been

found to be related to some aspects of MC but relationships have not been

consistently found across all studies and indices of MC (Herba & Rachman 2007;

174

Radomsky & Elliott 2009; Elliott & Radomsky 2013; Cougle et al 2008; Coughtrey

et al 2012; Badour e al 2013a; Badour et al 2013b; Carreresi et al 2013; Fairbrother

& Rachman 2004; Olantunji et al 2008). Most studies explored MC by inducing MC

through the non-consensual kiss paradigm and investigating predictive variables.

This study will use a different paradigm to explore MC and investigate possible

associated variables. Coughtrey et al (2014) used an autobiographical memory recall

paradigm which has increased ecological validity and enabled men and women to be

included in the study. However, they did not have a control group and only measured

anxiety as the emotional component of MC. This study will use their ‘victim

memory’ task to induce MC and add a control group. Also, this study will extend the

emotional components of MC measured by also measuring disgust, shame, anger,

guilt and sadness as these are conceptually linked to MC and have been found in

other MC induction studies (Rachman 2006; Radomsky & Elliott 2009; Elliott &

Radomsky 2013; Rachman et al 2012).

This study will explore additional variables as predictors of MC responses these will

include: attribution style, moral disgust sensitivity, appraisals and shame and guilt

proneness. Attributional style is a cognitive personality variable that influences how

individuals assess events that happen to them (Dykema et al 1996). Attributional

style is measured in terms of internal or external, stable or unstable and global or

specific attributions, Research has found that an internal, stable and global attribution

style is related to depression, anxiety and PTSD after assault (Clark, Watson &

Mineka 1994, Luten, Ralph & Mineka 1997, Feiring, Taska & Chen 2002). Given

the cognitive nature of MC attributional style may be related to individual

differences in experiencing MC. Also, many of the emotional reactions associated

with MC have been found to be effected by attributional style (Tracy & Robins

175

2006). This will also extend findings by Elliott & Radomsky (2013) and Radomsky

& Elliott (2009) that individuals specific appraisals of responsibility and violation of

the non-consensual kiss predicted some indices of MC. Responsibility and violation

appraisals of individuals memories generated in this study will also be measured and

appraisals of the betrayal experienced will be added as Rachman et al (2012) found

adding elements of betrayal increased feelings of MC.

Previous studies have explored disgust sensitivity as a predictor of MC and found it

predicted some aspects of MC but this was not consistent across studies (Elliott &

Radomsky 2013; Radomsky & Elliott 2009 & Herba & Rachman 2007). This may be

because the paradigm used to induce MC varied slightly between studies and because

the disgust measured used focussed on physical disgust sensitivity. This study will

look specifically at moral disgust sensitivity because of its link with moral

transgressions and MC (Ille et al 2014).

Shame and guilt are associated with MC but shame and guilt proneness have not

been looked at as predictive factors of MC and research has shown a link between

OCD and shame and guilt proneness so it will be interesting to explore this in MC

(Fergus, Valentiner, McGrath & Jencius 2010).

Understanding the factors associated with MC could have important clinical

implications in terms of identifying individuals who may be at risk for MC and

developing assessment and interventions.

1.2 Research questions:

The main aim of the study is to explore the predictors of feelings of MC responses in

the context of autobiographical memories of a moral transgression. In particularly

this study is interested in whether attributional style, appraisals, moral disgust

sensitivity and guilt and shame proneness predict feelings of mental contamination in

176

terms of feelings of general dirtiness, feelings of internal dirtiness, urge to wash and

emotional reactions. The study will also explore the emotional reactions associated

with mental contamination and the relationship to predictor variables.

The study has four main hypotheses:

1. The experimental group will report higher feelings of MC than the control

group.

2. The experiment group will report higher emotional reactions on all measures

3. All predictor measures will be related to MC scores.

4. Attributional style will predict MC scores.

2.0 Method:

2.1 Design:

This study will use an experimental between subjects design with a control group and

an experimental group. There will be a number of variables measured in the study.

Mental contamination will be measured in terms of feelings of general dirtiness,

feelings of internal dirtiness, urge to wash and emotional responses including:

anxiety, disgust, shame, anger, guilt and sadness. These measures are in line with

measures of mental contamination used in previous research. The emotional reaction

measures of MC will be completed pre and post in order to have a baseline for this.

The other indices of MC will only be completed after the experiment in order to

avoid priming participants’ responses to their memories.

After the experiment participants in the experimental condition will also be asked to

rate the appraisal of their memories on three measures: responsibility, violation and

betrayal ratings.

177

Prior to the experiment participants will also be asked to complete measures of

predictor variables which will include: attribution style, moral disgust, guilt

proneness and shame proneness.

2.2 Participants:

Total sample size= 66.

For hypothesis one, previous studies looking at the effect of inducing mental

contamination via autobiographical victim moral transgression memories have

presented effect sizes in the range of d=0.62-1.55 (Coughtrey et al 2014). Assuming

we want to have a power of 0.8 to detect an effect of 0.62, 1 tailed with alpha=0.05

using an independent samples t-test a priori calculation using G*Power 3.1.7 (Faul,

Erdfelder, Lang & Buchner, 2007) suggested we need to obtain a sample size of 66

with equal numbers in the control and experimental group.

For the other hypotheses, assuming we have a power of 0.8 to detect an effect of

0.15, 1 tailed with an alpha of 0.05 using a multiple regression analysis a priori

calculation using G*Power 3.1.7 (Faul et al 2007) suggested we need to obtain a

sample size of 43.

Inclusion and exclusion criteria:

This study will use a non-clinical sample of English speaking male and female

University students over the age of 18. Exclusion criteria would include not being

able to understand English, as assessment tools will be in English.

Recruitment:

The sample will be recruited from the University of Surrey’s student population, they

will be recruited via posters around campus and via e-mails.

Expected response rates:

178

As I am sampling from a large University population (13,576) with quite open

inclusion and exclusion criteria I would expect to be able to recruit the desired

sample size. Previous research in this area which have used student samples have

recruited sample sizes ranging from sixty to two hundred and fifty seven (Berman et

al 2012; Chen et al 2013; Coughtrey et al 2014a; Coughtrey et al 2014b; Cougle et al

2008; Elliott & Radomsky 2013; Elliott & Radomsky 2009; Elliott & Radomsky

2012; Fairbrother et al 2005; Herba & Rachman 2007; Ishikawa et al 2014; Lee et al

2013; Rachman et al 2012). My research will be conducted online and a meta-

analysis of response rates to online studies found a mean response rate of 34.6% with

no missing data (Cook et al 2000).

2.3 Measures:

Predictor variables

Attributional style:

The Attributional Style Questionnaire (ASQ; Peterson et al 1982) is a 12 item self-

report measure which assesses attributions for negative and positive hypothetical

events. Participants are asked to write down one possible cause for these events and

rate on a likert scale of 1 to 7 how internal, stable and global that cause is. Scores can

be calculated for each dimension and an overall ASQ score can be calculated. It has

acceptable internal consistency (α=.75) and test-retest reliability (r=.70) (Peterson et

al 1982). The author of this measure has been contacted.

Moral disgust sensitivity:

The moral disgust sub-scale from The Three Domain Disgust Scale (TDDS; Tybur,

Lieberman & Griskeviaivs 2009). This is a 7 item self-report measure which asks

participants to rate how disgusting items are on a 7 point scale. This measure has

179

been shown to have good internal consistency (α=.84) and test-retest reliability (r.64)

(Tybur et al 2009, Olatunji et al 2012). Appendix B.

Shame and guilt proneness:

The Personal Feelings Questionnaire Two (PFQ-2 Harder & Zalma 1990). This is a

16 item self-report measure with a shame 10 item subscale and a guilt 6 item

subscale. Participants are asked to rate how often they experience each item on a

likert scale from 0- never experience the feeling to 4 experience it continuously or

almost continuously. It has good internal consistency for both subscales (guilt α=.72,

shame α=.78) and test re-test reliability (guilt r=.85, shame r=.91) (Harder & Zalma

1990). The authors of this measure have been contacted for access to the measure.

Appraisals:

Adapted from Elliott & Radomsky’s (2013) study Visual Analogue Scales (VAS)

will be used to assess appraisals of participants’ memories. Participants will be asked

“How responsible did you feel for the events that occurred in the situation you

described?”, “Did you feel violated in the situation you described” and “Did you feel

betrayed in the situation you described?”. Answers will be rated on a 100mm scale

with 0 representing “not at all” and 100 representing “completely”.

Mental Contamination:

Mental contamination will be measured in the same way it was measured in

Coughtrey et al’s (2014) study, through VAS scales. Participants will answer the

following questions on a 100 mm scale ranging from 0=not at all to 100=extremely:

“How dirty do you feel inside your body?”, “How dirty do you feel in general?” and

“How strong is your urge to wash?”.

Emotional reactions:

180

Pre and post the study participants will be asked to rate on a VAS scale how anxious,

disgusted, ashamed, angry, guilty and sad participants feel.

2.4 Procedure:

The study will take place online, so participants will be recruited via e-mail or poster

adverts and asked to follow an online link. The e-mail and posters will give them

some information about the study and the online link will give them some more.

First participants will read the information sheet about the study and will be asked to

give their consent to take part and told they may withdraw from the study at any

time. If participants want to continue to take part they will be asked some

demographic data including: sex, age and ethnicity.

Participants will then be asked to fill in the pre experimental measures including:

attributional style, moral disgust sensitivity, shame and guilt proneness and the VAS

measures of emotions. After this participants will be randomised to either the

experimental or the control group.

In the mental contamination condition participants will be asked to recall a time they

were the victim of a moral transgression. They will be asked to think of a time when

they had felt humiliated, ashamed or betrayed, or a time that they had received bad

news or been harmed in some way (victim memory, Coughtrey et al 2014).

They will be asked to form a mental image of this memory in their mind for two

minutes imagining it as if it were happening again, through their own eyes as if they

were there and focussing on the negative emotions that they experienced at the time.

After the two minutes participants will be asked to write down a description of their

memory.

181

Then participants will complete the time 2 VAS measures of emotion, the mental

contamination measures of feelings of dirtiness and urge to wash and answer

questions about the appraisal of their memories.

After the task participants will read a de-brief and if they found the task distressing

contact details of the University wellbeing centre will be given out.

The control group will fill in the same pre and post measures but will be asked to

remember a neutral task they have performed that day for example: brushing their

teeth or eating their breakfast. They will then have the same instructions as the

experimental group: to form a mental image of this memory in their mind for two

minutes imagining it as if it were happening again, through their own eyes.

After the two minutes participants will be asked to write down a description of their

memory.

They will then fill in the time two VAS measures of emotion and the mental

contamination measures.

2.5 Ethical Considerations:

During this research participants in the experimental group will be asked to recall a

negative memory designed to induce feelings of MC which is associated with

unpleasant emotions. Coughtrey et al’s 2014 who used this paradigm to induce MC

found that feelings of MC generally decayed after three minutes so these feelings

would not necessarily last a long time. Another issue is that participants will be asked

to recall a moral transgression they have experienced and the wording is quite

opening including a time they have felt ‘harmed in some way’. This could bring up

difficult memories for people

Before taking part in the experiment participants will be told that the study may

induce some unpleasant but short lived feelings so they can decide whether to take

182

part or not. Participants will also be made aware that they can withdraw from the

study at any time. Participants will be given a full de-brief at the end of the

experiment and information of where they can get further support if the study

brought up anything difficult for them. As the study will be conducted on Surrey

University students details of the University well-being service will be given out if

they feel they need further support.

An application to the ethics committee of the faculty of arts and human sciences will

be submitted.

3.0 Proposed data analysis:

Hypothesis one: In order to check the experimental manipulation was successful in

evoking MC an independent sample t-tests will be carried out to look for a difference

between the control and experimental group on ratings of feelings of general

dirtiness, internal feelings of dirtiness and urge to wash.

Hypothesis two: To investigate the emotional reactions associated with inducing

mental contamination a 2 x 2 repeated measures anova will be performed. This will

enable the study to look at the difference between pre and post measures of emotions

in both groups to see the effect of both time and condition.

Hypothesis three and four: To explore predictors of mental contamination analysis

will focus on the experimental condition only. Correlations and regression analyses

will be performed for each index of mental contamination: feelings of general

dirtiness, feelings of internal dirtiness, urge to wash and emotional reactions. If the

correlational analysis reveals a relationship then a multiple linear regression will be

used to look at the predictors of each index of MC.

4.0 Involving/Consulting Interested Parties:

183

This study will be using a non-clinical sample of University students to explore the

phenomenon of MC. It will be useful to consult this population in order to assess

whether the online materials are user friendly and garner opinion on the experimental

paradigm. It would also be useful to consult this population on the best way to recruit

participants to the study.

5.0 Contingency Plan:

This study involves inducing mental contamination and looking at predictor

variables. There are a number of measures for participants to complete and if there is

a high incompletion rate for the study then it would be difficult to obtain the data

needed. If it were difficult to obtain data then a correlational design could be used.

This would involve using the same sample group but instead of inducing feelings of

mental contamination, symptoms of mental contamination would be measured via

the VOCI-MC (Radomsky 2005) appendix C. Participants would then be asked to fill

in the same predictor measures: attributional style, moral disgust, guilt proneness and

shame proneness. These measures could also be completed online and would be

presented in a counter-balance order. A correlational analysis could then be

performed to assess any relationship between the MC and the variables. This would

still add to the literature around mental contamination and variables associated with

it.

6.0 Dissemination strategy:

Previous research in this area has been published in the Journal of Behaviour and

Experimental Psychiatry and the journal of Behaviour research and Therapy so this

research could be submitted to these peer review journals.

184

References

Badour, C. L., Feldner, M. T., Blumenthal, H., & Bujarski, S. J. (2013a). Examination of

increased mental contamination as a potential mechanism in the association between

disgust sensitivity and sexual assault-related posttraumatic stress. Cognitive Therapy

and Research, 37 (4), 697-703.

Badour, C. L., Feldner, M. T., Babson, K. A., Blumenthal, H., & Dutton, C. E. (2013b).

Disgust, mental contamination, and posttraumatic stress: Unique relations following

sexual versus non-sexual assault. Journal of Anxiety Disorders, 27(1), 155-162.

Calamari, J. E., Weigartz, P.S., Riemann, B.C., Cohen, R. J., Greer, A., Jacobi, D. M., Jahn,

S.C., & Carmin, C. (2004). Obsessive-compulsive disorder subtypes: An attempted

replication and extension of a symptom based taxonomy. Behaviour Research and

Therapy, 42, 647-670.

Carraresi, C., Bulli, F., Melli, G., & Stopani, E. (2013). Mental contamination in OCD: Its

role in the relationship between disgust propensity and fear of contamination.

Clinical Neuropsychiatry, 10(3),13-19.

Chen, Z., Teng, F., & Zhang, H. (2013). Sinful flesh: Sexual objectification threatens

women's moral self. Journal of Experimental Social Psychology, 49(6), 1042-1048.

185

Clark, L., Watson, D., & Mineka, S. (1994). Temperament, Personality, and the mood and

anxiety disorders. Journal of Abnormal Psychology, 10 (1), 103-116.

Cook, C., Heath, F.,& Thompson, R. (2000). A meta-analysis of response rates in web or

internet based surveys. Educational and Psychological Measurement, 60(6), 821-

836.

Coughtrey, A. E., Shafran, R., Knibbs, D., & Rachman, J. (2012). Mental contamination in

obsessive-compulsive disorder. Journal of Obsessive-Compulsive and Related

Disorders, 1(4), 244-250.

Coughtrey, A. E., Shafran, R., Lee, M., & Rachman, S. J. (2012). It's the feeling inside my

head: A qualitative analysis of mental contamination in obsessive-compulsive

disorder. Behavioural and Cognitive Psychotherapy,40(2), 163-173.

Coughtrey, A. E., Shafran, R., & Rachman, S. J. (2014a). The spontaneous decay and

persistence of mental contamination: An experimental analysis. Journal of Behavior

Therapy and Experimental Psychiatry, 45(1), 90-96.

Coughtrey, A. E., Shafran, R., & Rachman, S. J. (2014b). The spread of mental

contamination. Journal of Behavior Therapy and Experimental Psychiatry, 45(1),

33-38.

Cougle, J. R., Lee, H., Horowitz, J. D., Wolitzky-Taylor, K. B., & Telch, M. J. (2008). An

exploration of the relationship between mental pollution and OCD symptoms.

Journal of Behavior Therapy and Experimental Psychiatry,39(3), 340-353.

186

Dykema, J., Bergbower, K., Doctora, J., & Peterson, C. (1996). An Attributional Style

Questionnaire for general use. Journal of Psychoeducational Assessment, 14, 100-

108.

Elliott, C. M., & Radomsky, A. S. (2009). Analyses of mental contamination: Part I,

experimental manipulations of morality. Behaviour Research and Therapy, 47(12),

995-1003.

Elliott, C. M., & Radomsky, A. S. (2012). Mental contamination: The effects of imagined

physical dirt and immoral behaviour. Behaviour Research and Therapy, 50(6), 422-

427.

Elliott, C. M., & Radomsky, A. S. (2013). Meaning and mental contamination: Focus on

appraisals. Clinical Psychologist, 17(1), 17-25.

Erdfelder, E., Faul, F. & Buchner, A. (1996). GPower: A general power analysis program.

Behavior Research Methods, Instruments, & Computers, 28, 1-11.

Fairbrother, N., & Rachman, S. (2004). Feelings of mental pollution subsequent to sexual

assault. Behaviour Research and Therapy, 42(2), 173-189.

Fairbrother, N., Newth, S. J., & Rachman, S. (2005). Mental pollution: Feelings of dirtiness

without physical contact. Behaviour Research and Therapy, 43(1), 121-130.

187

Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G*Power 3: A flexible statistical

power analysis for the social, behavioral, and biomedical sciences. Behavior

Research Methods, 39, 175-191..

Feiring, C., Taska, L. & Chen, K. (2002). Trying to understand why horrible things happen:

attribution, shame and symptom development following sexual abuse. Child

Maltreatment, 7, 25-39.

Fergus, T. A., Valentiner, D. P., McGrath, P. B., & Jencius, S. (2010). Shame and guilt

proneness: relationships with anxiety disorder symptoms in a clinical sample.

Journal of Anxiety Disorders, 24, 811-815.

Harder, D., & Zalma, A. (1990). Two Promising Shame and Guilt Scales: A Construct

Validity Comparison. Journal of Personality Assessment, 5, 729-745.

Herba, J. K., & Rachman, S. (2007). Vulnerability to mental contamination. Behaviour

Research and Therapy, 45(11), 2804-2812.

Ille, R., Schoggl, H., Kapfhammer, H., Arendasy, M., Sommer, M., &Schienle, A. (2014).

Self-disgust in mental disordes- symptom related or disorder-specific?

Comprehensive Psychiatry, 55, 938-943,

Ishikawa, R., Kobori, O., Komuro, K. & Shimizu, E. (2014). Comparing the roles of

washing and non-washing behaviour in the reduction of mental contamination.

Journal of Obsessive-Compulsive and Related Disorders, 3(1), 60-64.

188

Lee, M., Shafran, R., Burgess, C., Carpenter, J., Millard, E., & Thorpe, S. (2013). The

induction of mental and contact contamination. Clinical Psychologist, 17(1), 9-16.

Luten, A., Ralph, J. & Mineka, S. (1997). Pessimistic attributional style: Is it specific to

depression versus anxiety versus negative affect? Journal of Behaviour Research

Therapy, 35 (8), 703-719.

Olatunji, B. O., Elwood, L. S., Williams, N. L., & Lohr, J. M. (2008). Mental pollution and

PTSD symptoms in victims of sexual assault: A preliminary examination of the

mediating role of trauma-related cognitions. Journal of Cognitive Psychotherapy,

22(1), 37-47.

Olatunji, B., Adams, T., Ciesielski, B., David, B., Sarawgi, S., & Broman-Fulks, J. (2012).

The three domains of disgust scale: Factor Structure, Psychometric Properties and

conceptual limitation. Assessment, 19, 2, 205-225.

Peterson, C., Semmel, A., Von Baeyer, C., Abramson, L., Metalsky, G. & Seligman, M.

(1982). The attributional style questionnaire. Cognitive therapy Research, 6, 287-

299.

Ponniah, K., Magiati, I., Hollon, S. (2013). An update on the efficacy of psychological

treatments for obsessive-compulsive disorder in adults. Journal of Obsessive-

Compulsive and Related Disorders, 2 (2), 207-218.

Rachman, S. (1994). Pollution of the mind. Behaviour Research and Therapy, 32(3), 311-

314.

189

Rachman, S. (2004). Fear of Contamination. Behaviour Research and Therapy, 42, 1227-

1255.

Rachman, S. (2006). Fear of contamination: Assessment and treatment. Oxford University

Press.

Radomsky, A. S., & Elliott, C. M. (2009). Analyses of mental contamination: Part II,

individual differences. Behaviour Research and Therapy, 47(12), 1004-1011.

Rachman, S. (2010). Betrayal: A Psychological Analysis. Behaviour Research and Therapy,

48, 304-311.

Rachman, S., Elliott, C. M., & Zysk, E. (2012). Mental contamination: The perpetrator

effect. Journal of Behavior Therapy and Experimental Psychiatry, 43(1), 587-593.

Radomsky, A. S., Rachman, S., Shafran, R., Coughtrey, A. E., & Barber, K. C. (2014). The

nature and assessment of mental contamination: A psychometric analysis. Journal of

Obsessive-Compulsive and Related Disorders, 13 (2), 181-187.

Sangmoon, K., Ryan, T., & Randall, S. (2011). Shame, guilt and depressive symptoms; a

meta-analytic review. Psychological Bulletin, 137 (1), 68-96.

Tracy, J., & Robins, R. (2006). Appraisal antecedents of shame and guilt: Support for a

theoretical model. Personality and Social Psychology Bulletin, 32 (10), 1339-1351.

190

Tybur, J., Lieberman, D., & Griskevicius, V. (2009). Microbes, mating and morality:

Individual differences in three functional domains of disgust. Journal of Personality

and Social Psychology, 97, 103-122.

Warnock-Parkes, E., Salkovskis, P., & Rachman, J. (2012). When the Problem is Beneath

the Surface in OCD: The Cognitive Treatment of a Case of Pure Mental

Contamination. Behavioural and Cognitive Psychotherapy, 40(4), 383-399.

191

Literature Review

Reviewing the current state of knowledge on the relatively new concept of

Mental Contamination.

URN:6289478

Word Limit: 8000 words

University of Surrey

192

Abstract

Background: Conceptualised by Rachman mental contamination (MC) is an internal

feeling of dirtiness that can arise without physical contact with a contaminant, is

accompanied by negative emotions such as fear and disgust and produces an urge to

wash. MC is a relatively new concept and is conceptually linked to OCD and PTSD

and so it is important to understand this concept fully. This review evaluates the

current state of knowledge on MC.

Method: A search of electronic databases yielded twenty five eligible studies from an

initial pool of six hundred and sixty nine. Included studies were represented in this

review in four groups, studies that: i) validated measures of MC, ii) induced MC and

investigated its nature, iii) investigated correlates of MC and iv) evaluated

interventions for MC.

Results: The studies reviewed generally supported and extended the theory of MC

and the variables associated with it. The main finding was that MC can be induced

without physical contact with a contaminant and through: physical violations, moral

violations or self-contamination.

Conclusions: Studies are beginning to explore the concept of MC contamination and

its relation to other psychological disorders. Factors associated with inducing MC

supported the theory of MC but more research in clinical populations is needed.

Variables that are related to vulnerabilities to experiencing MC need to be explored

further in order to make firmer conclusions. Furthermore, more research on the

maintenance factors of MC is needed as are more robust measurements of MC that

can be used to assess therapeutic change.

Keywords: Mental contamination, mental pollution

193

Abbreviations: MC=Mental Contamination, MP=Mental Pollution, INE=Internal

negative emotions (such as shame and guilt), ENE=External negative emotions (such

as anger and anxiety)

Statement of Journal Choice

Clinical Psychology Review publishes papers related to research or clinical practice

in clinical psychology including: psychotherapy, psychopathology, behaviour and

cognitive therapies, community mental health and assessment. The journal also

publishes reviews on a variety of topics relevant to the field of clinical psychology.

This journal is aimed at psychologists and clinicians working in the field of mental

health and seeks to publish cutting edge research which will further advance clinical

psychology practice. The impact factor of this journal is high at 6.696 indicating that

articles in this journal are often cited by other authors.

This review fits well with this journal as it is exploring a novel phenomenon in

Clinical Psychology that is hypothesised to underlie other psychological disorders.

Exploring and reviewing the current state of knowledge on MC in this review may be

helpful to improve the effectiveness of assessments and treatments of psychological

difficulties in mental health settings.

194

195

IntroductionThis literature review aims to evaluate current research on the topic of mental

contamination (MC). MC is a relatively new psychological concept and was first

conceptualised by Rachman in 1994 using the term ‘pollution of the mind’. Mental

pollution (MP) is defined as “a sense of internal un-cleanness which can and usually

does arise and persist regardless of the presence or absence of observable dirt”

(Rachman 1994, p.311). This concept first developed from clinical observations of

OCD patients expressing that they could never feel entirely clean even after repeated

washing (Rachman 2004). MP had now been incorporated into the larger concept of

MC which sits within Rachman’s theory of fear of contamination (2004, 2006).

The fear of contamination is an intense feeling of dirtiness that is accompanied by

negative emotions such as shame and disgust and is usually followed by a strong

urge to wash. Fear of contamination includes two categories: contact contamination

and MC (Rachman 2006). Traditionally, more focus has been given to contact

contamination which arises after physical contact with something that is perceived to

be soiled, impure, infectious or harmful (Rachman 2004). Recently, more attention is

being given to Rachman’s theory of mental contamination which is when feelings of

fear of contamination arise without physical contact with a contaminant.

Like contact contamination, mental contamination involves internal feelings of

dirtiness and is associated with a variety of negative emotions including: disgust,

fear, anger, shame, guilt and revulsion (Rachman 2006). Rachman (2006) theorised

that MC can arise from a physical violation (e.g. sexual assault), a mental violation

(e.g. feeling ashamed, degraded, humiliated) or from self-contamination (e.g. having

unwanted or unacceptable thoughts, images or memories). Rachman also described

the phenomenon of ‘morphing’ which is a fear of taking on the characteristics of an

196

undesirable person by looking at or touching them. A central difference between MC

and contact contamination is that MC can occur without physical contact with a

contaminant. In addition, Rachman (2006) hypothesised that the source of

contamination can be different for MC and contact contamination; contact

contamination tends to occur after physical contact with a perceived soiled object or

place and is associated with danger, disease and dirt (e.g. bodily wastes, toxins). In

contrast, although MC can also occur from non-human sources it is proposed that the

source is usually human: a dirty, dangerous, immoral or harmful person (including

the self). The sources of contact contamination are often seen as a threat to most

people (e.g. chemicals) and therefore it is seen as easily transferable to other people

or objects (Tolin, Worhunsky & Maltby 2004). The theory of MC suggests that MC

is unique to the affected person and therefore is not easily transferred to others

(Rachman 2004). Both contact and mental contamination are proposed to evoke an

urge to wash but washing is only effective, in the short term, in contact

contamination where the feelings of dirtiness are limited to a specific, identifiable

bodily location. In MC the feeling of dirtiness is internal and not localised therefore

it does not effectively respond to washing but may respond to mental, neutralising

rituals (Rachman 2006). Although there are differences between MC and CC they are

thought to co-occur and share some overlapping characteristics including: urge to

wash and discomfort (Coughtrey, Shafran, Lee & Rachman 2012).

Factors that contribute to the maintenance of MC are unclear but Rachman (2006)

hypothesised that thought action fusion (TAF) could play a part. TAF is the belief

that just having unpleasant, repugnant thoughts is the moral equivalent of acting on

them and that thinking about something unpleasant increases the likelihood of it

happening (Shafran et al 1996). TAF is postulated to contribute to MC by increasing

197

beliefs about the probability of harm and increasing feelings of guilt or responsibility

(Coughtrey, Shafran & Rachman 2013). Over-estimation of the likelihood and

danger of being contaminated may also play a part as it is hypothesised by Rachman

(2006) to increase and maintain fear of contamination and this has been found for

contact contamination (Green & Teachman 2013).

MC has been observed in Obsessive Compulsive Disorder (OCD), Post traumatic

Stress Disorder (PTSD) and victims of sexual assault giving it important clinical

relevance (Oluntaji, Elwood, Williams & Lohr 2008; Warnock-Parkes, Salkovskis &

Rachman 2008; Fairbrother & Rachman 2004). MC is strongly conceptually linked

to OCD because of the role that fear of contamination plays in the disorder; studies

have found that over half of OCD sufferers have contamination fears (Calamari,

Weigartz, Riemann, Cohen, Greer et al 2004). Furthermore, Coughtrey et al’s (2012)

study found that nearly half of participants with OCD also had clinical levels of MC.

MC is linked to OCD and yet it is not routinely assessed for or treated separately.

Widely used clinical OCD measures, such as the Yale and Brown Obsessive

Compulsive Scale, do not assess for MC. If MC is a distinct phenomenon in OCD

then it may be that not assessing and treating it properly contributes to the high

relapse rate and treatment resistant nature often found in OCD (Ponniah, Magiati &

Hollon 2013). For instance, studies have found that OCD sufferers with cleaning

compulsions respond less well to CBT and fear of contamination is thought to

underpin cleaning compulsions (Coehlo & Whittal 2001). If MC is a distinct

phenomenon in OCD or other psychological disorders it may need specific

treatments tailored towards it; therefore it is important to understand the concept,

maintenance factors associated with it and individual differences in vulnerability to

it.

198

Given that MC is a relatively new but potentially important concept, which has only

begun to garner research interest in the last ten years, a review will be useful to

obtain a clearer picture of the current state of knowledge on MC and identify any

knowledge gaps.

199

MethodSearch strategy

The following electronic databases were searched in February 2014: Web of

Knowledge, PsycINFO and Scopus. As MC is a relatively new concept no time

restrictions were placed on the search or any other data limiters. The search used the

following combination of search terms: “mental pollution” or “mental contam*” or

“imag* contam*” or “imag* pollut*”, or “mental contagion*” or “transformation

obsession*” or “morphing obsession*”.

Eligibility Criteria

Studies were only included if they had operationalized MC in the study. They were

considered to have operationalized MC if they assessed MC using a validated scale

or had used an experimental method to directly induce MC and subsequently used

relevant indices to assess it. This meant that papers looking at the theory of MC but

not operationalizing it were not included. Studies were also excluded if they were not

relevant to Psychology (e.g. a large number of studies that came up in the search

referred to environmental pollution or contamination). Studies were also excluded if

they had a different definition of MC to the one described in the Introduction above.

Only studies published in peer reviewed journals were included so, dissertations,

books and book reviews were excluded. Studies using both clinical and non-clinical

samples were included.

Data Analysis:

Following the screening process (see Figure 1) twenty five studies were retained for

the current review. These studies are represented in four categories for the purpose of

this review: i) studies that looked at validating measures of MC; ii) studies which

induced MC and investigated the nature of MC; iii) studies that investigated the

200

correlates or predictors of MC; and iv) studies that looked at interventions for MC.

Some studies fell into more than one category as they investigated more than one

aspect of MC. Studies in this review were critically apprised using the HTA quality

assessment criteria (Kmet et al 2004)

201

Potentially eligible records identified through database searches: (N=669)Web of Knowledge N=284PsycINFO N=66Scopus N=319

Potentially eligible records (N=464)

Exclusion of duplicate records (N=205)

Excluded: Did not meet criteria after screening (titles and abstracts). (N=423)

Not relevant to Psychology

Described a different concept of MC

No identified author Not available in the

English language

Full text articles assessed for eligibility (N=41)

Exclusion of articles after screening(N=16):

Did not operationalize MC (N=6)

Article not relevant (N=3)

Dissertation (N=1) Book or book review

(N=3) Theory paper (N=3)

Articles included in the review (N=25)

ResultsDescription of included studiesDetails of included studies can be seen in Table 1. Studies included in this review

were published between 2004 and 2014 and fifteen were published since 2010. The

total sample size across all studies was 2,974 participants with an age range of

approximately 11-65; not all studies reported the age range. Across all studies the

majority of participants were female (79%), 13 studies used female only samples,

one study used a male only sample and the rest used mixed samples. Only five

studies reported the ethnicity of their participants. Of these 60.5%-85% identified

themselves as White/Caucasian. Most studies recruited undergraduate university

students (N=16), followed by OCD clinical samples (N=6), history of traumatic

events (N=2) and mixed OCD and undergraduate (N=1). The majority of the studies

were conducted in either the UK (N=7), Canada (N=9) or the USA (N=5). The

remaining studies were conducted in Italy, Germany, China or Japan.

i) Self-report measures of MC

MC has been operationalized through self-report measures that tap into propensity to

experience MC with higher scores indicating greater propensity. The psychometric

properties of two measures were investigated in this review: the Vancouver

Obsessions and Compulsions Inventory-Mental Contamination (VOCI-MC)

(Radomsky, Rachman, Shafran, Coughtrey & Barber 2014) and the Mental Pollution

Questionnaire (MPQ) (Cougle, Horowitz, Wolitzk-Taylor & Telch 2008). The design

and results of these studies are displayed in Table 2. The VOCI-MC is a measure of

symptoms of mental contamination asking participants to rate items on a five point

Likert scale producing one overall score. Radomsky et al (2014) found the VOCI-

MC detected the presence of MC and reported good reliability across the sample

types used. The measure was found to have good validity: it correlated with other

202

measures of OCD, and discriminated between OCD contact contamination symptoms

and MC symptoms.

203

Table 1. Continued

/ Data not reported

Table 1.Continued

Authors/Location Age range Mean Age (SD) N Female (%) SampleBadour et al (2013a) USA / 28.18 (13.93) 40 100% History of traumatic event as defined by criterion A of DSM-IV for PTSDBadour et al (2013b) USA / 32.34 (13.55) 38 100% History of traumatic sexual assault as defined by DSM-IV

Berman et al (2012) USA / 19.46 (2.75) 264 73% Undergraduate sample affiliated to a religious group

Carraresi et al (2013) Italy / 32.6 (9.6) 83 44.00% Clinical- OCD symptoms (as identified through semi-structured interview by MHP)

Chen et al (2013) China / Exp 1: 20.4 (.86)Exp 2: 30.59 (12.52) Exp 3: 33.78 (11.35)

Exp 1: 40Exp2: 56Exp 3:65

100% Non-clinical sample- Undergrads Hong Kong and US

Coughtrey et al (2012) UK / Study 1: 34.40 (11.43) Study 2: 33.39 (10.89)

Study 1: 177 Study 2: 54 (from

study 1)

Study 1: 73% Study 2: 65%

Clinical: Study 1: Diagnosis of OCD by GP or MHP Study 2: diagnosis of OCD via structured interview

Coughtrey et al (2013) UK 18-65 years 28.83 (8.54) 12 58% Clinical Sample OCD including MC

Coughtrey et al (2014a) UK Exp 1: 18-44 Exp 2: 18-38

Exp 1: 22.60 (5.33) Exp 2: 20.53 (4.30)

Exp 1: 40 Exp 2: 60

Exp 1: 70% Exp 2: 82%

Non Clinical- Exp 1: Undergraduates Exp 2: undergraduates scoring >10 on VOCI-MC

Coughtrey et al (2014b) UK 18-38 20.53 (4.30) 60 82% Non-Clinical: Undergraduates scoring >10 on VOCI-MC

Cougle, J et al 2008 USA Study 1: 18-43 Study 2: 17-27 Study 3: 19-29

Study 1: 19.45 (5.3) Study 2: 19.45 (5.3) Study 3: 19.45 (5.3)

Study 1: 218 Study 2: 257 Study 3: 84

Study 1: 61% Study 2: 72% Study 3: 75

Non-Clinical: Undergraduates

Elliott & Radomsky (2013) Canada 18-44 21.59(4.01) 59 100% Undergraduates

Elliott & Radomsky (2009) Canada 17-48 22.86 (4.46) 148 100% Non Clinical: Heterosexual Undergraduates

Elliott & Radomsky (2012) Canada 18-55 22.70 (5.29) 140 100% Non Clinical: Undergraduates

Fairbrother & Rachman (2004)Canada 16-49 24.5 (/) 50 100% Non Clinical: Undergraduates who's experienced an unwanted sexual experience (as defined by Canadian law)

Fairbrother et al (2005) Canada 15-38 20.51 (3.17) 121 100% Non Clinical: Undergraduates

Herba & Rachman (2007) Canada / 20.73 (4.73) 120 100% UndergraduatesIshikawa et al (2014) Japan 18-25 18.36 (2.31) 48 100% Non Clinical: Japanese Undergraduates no history mental health problems

Jung & Steil (2012) Germany 25-51 38(/) 2 100% Clinical sample- suffered childhood sexual abuse, current PTSD and feelings of being contaminated

Lee et al (2013) UK 18-57 22.25 (8.22) 60 83% Non Clinical: Undergraduates

Olatunji et al (2008) USA / 19.78 (1.37) 48 100% Undergraduate victims of sexual assault

Rachman et al (2012) Canada Exp 1: 18-26 Exp 2: 18-36 Exp 3: 19-43 Exp 4:17-52

Exp1: 20.36 (1.63) Exp 2: 20.36 (2.93) Exp 3: 21.53 (4.95) Exp 4: 22.75 (5.45

Exp 1: 39 Exp 2: 40 Exp 3: 40 Exp 4: 40

0% Non Clinical: Undergraduates

Radomsky & Elliott (2009) Canada 18-43 23.30(4.77) 70 100% Undergraduates

Radomsky et al (2014) Canada / Group A: 36.13 (10.99) Group B: 48.81 (14.86) Group C: 38.13 (14.45) Group D: 22.45 (4.48)

A:30 B:27 C:24

D:410

A: 57% B: 44% C:62% D:86%

Clinical and non-clinical: A + B: Clinical- OCD diagnosis (via ADIS-IV) C: Clinical- other anxiety disorder D: Non-Clinical undergraduates

Volz and Heyman (2007) UK 11-17 18.22 (/) 9 11% Clinical: OCD with transformation obsessions

Warnock-parkes et al (2012) UK 40s / 1 100% Clinical Sample- OCD including MC

Table 1: Details of included studies

However, there was no report of test-retest reliability and the Cronbach’s alpha was

quite high (α.97) which could suggest a redundancy of items in the measure. More

research is needed to see if the VOCI-MC can be reliably used as a measure of

therapeutic change (Radomsky et al 2014).

The MPQ is a measure of symptoms of mental pollution asking participants to rate

items on a 7 point Likert scale. It measures two factors of mental pollution: MPQ-

Wash (MPQ-W) assesses washing rituals performed in response to perceived mental

pollution, and MPQ-Ideation (MPQ-I) assesses sense of inward contamination not

connected to washing behaviour. In a factor analysis Cougle et al (2008) found the

two factors accounted for 56.2% of the variance and the overall mean had good

internal consistency (α=.86) and test re-test reliability (r.88). However, the test re-

test reliability was only based on a sample of 18 students. Furthermore, the study

used an undergraduate sample which limits the application of the MPQ in a clinical

setting without further assessment of its properties.

The two studies above were the only studies in this review that explored the

psychometric properties of measures of MC. However, another commonly used

measure in the studies reviewed here was the Mental Contamination Report (MCR)

which was used in the experimental studies which induced MC. The MCR (Herba &

Rachman 2007) measures two indices of MC- feelings of dirtiness and urge to wash.

The construct validity of this version is uncertain as it does not include measures of

negative emotions thought to accompany MC. Elliott & Radomsky (2009) amended

this measure to include indices of internal negative emotions (INE) such as shame

and guilt and external negative emotions (ENE) such as anger, anxiety and disgust.

The validity and reliability of this measure is not well reported in the studies

204

reviewed here and there has been no factor analysis reported for this measure. For

details of measures of MC used across studies in this review see Table 3.

205

Table 2. Details and results of psychometric studies of MC self-report measures

StudyMC measure(s) Aspect of MC

measuredNo. of items Reliability Validity Results Factor

analysisConclusions

Radomsky et al (2014)

1.VOCI-MC 1. Symptoms of MC 1. 20 1. α.93 to α.97

Convergent: VOCI-MC TO CSS r.61- 87, VOCI-MC to CTAF r.45, (all p<0.001) Discriminant Validity: VOCI-MC discriminated between those with OCD contamination fears and other groups

VOCI-MC Predicted unique variance in OCD. ß.344, p<0.001

no Acceptable psychometric properties. MC coherent concept and measurable. Detects presence of MC

Cougle, J et al 2008

MPQ Internal feelings of dirtiness. Two subscales- MPQ-Wash= Internal dirtiness attempted to be relieved via washing. MPQ-Ideation=internal dirtiness separate to washing

Originally 11 items

α.86, test-retest r.88 p<0.0001

  MPQ-W accounted for 32.7% of variance and MPQ-I 29.5%

8 items factor loading above .5

Reliable and Valid measure of Mental Pollution

VOCI-MC=Vancouver Obsessions and Compulsion Inventory- Mental Contamination Subset; CSS=Contamination Sensitivity Scale; CTAF=Contamination based Thought Action Fusion; MPQ=Mental Pollution Questionnaire, OCD= Obsessive Compulsive Disorder, MC= Mental Contamination.

206

Study Measure of MC Reliability and Validity

Experimental Studies

Chen et al (2013) Likert Scales. Exp 1: One item 'I am dirty', Exp 2 and 3: 3 items- I feel dirty, I feel polluted, I feel contaminated

Exp 1: +, Exp 2: α.90, Exp 3: α.77

Coughtrey et al (2014a) Exp 1: Visual Analogue scales, 4 indices: Dirty on inside, dirty in general, urge to wash, anxiety Exp 2: VOCI-MC (Rachman 2006)

Exp 1: not reported Exp 2: not reported

Coughtrey et al (2014b) VOCI-MC (Rachman 2006) Not reported for this study Elliott & Radomsky (2009) MCR (developed for this study) amended from Herba & Rachman (2007) added INE and ENE Urge to wash: α.91 INE: α.91 ENE: α.90 Feelings of dirtiness: +

Elliott & Radomsky (2012) MCR (Elliott & Radomsky 2009) Urge to wash: α.92 INEα.90 ENE α.88 Feelings of dirtiness: +

Fairbrother & Rachman (2004) MPI (Fairbrother 2004). SARA (3 items MP) MPI-not reported. SARA-α.84

Fairbrother et al (2005) USEQ- designed for this study not reported for this study

Ishikawa et al (2014) MCR, translated to Japanese (Ishikawa et al 2013) Not reported for this study.

Lee et al (2013) VAS: internal feelings of dirtiness and urge to wash. not stated

Rachman et al (2012) Likert Scales: feelings of dirtiness, urges to wash and negative emotions not stated

Correlational Studies

Badour et al (2013a) MCR (Herba & Rachman 2007) Urge to wash α.86-88, Feelings of dirtiness: +

Badour et al (2013b) SARA (Fairbrother & Rachman 2004) 3 items rate sexual assault related MC α.80

Berman et al (2012) MPQ (Cougle et al 2008) MPQ-Wash α.86 MPQ-Ideation α.85

Carraresi et al (2013) VOCI-MC (Rachman 2006) Translated into Italian .85 ≤ αs ≤ .95, test-retest r.88

*Coughtrey et al (2012) VOCI-MC (Rachman 2006) α.94

Cougle et al (2008) MPQ (Cougle et al 2008) α.86, r.88

Elliott & Radomsky (2013) MCR (Elliott & Radomsky 2009, 2012) Urge to wash: α. 81, INE: α.86, ENE: α.79, Feeligs of dirtiness:+

Herba & Rachman (2007) MCR- developed for this study Urge to wash α.89, Feelings of dirtiness: +

Olatunji et al (2008) MPQ (Fairbrother & Rachman 2004) and SARA α.73. Concurrent Validity- r.47, p<.02

Radomsky & Elliott (2009) MCR (Elliott & Radomsky 2009) amended Herba & Rachman added INE and ENE Urge to wash α.91 INE α.91 ENE α.90 Feelings of dirtiness α?

Intervention studies

207

Table 3. Measures of Mental Contamination used across studies reviewed

Coughtrey et al (2013) B VOCI-MC not reported for this study

Jung & Steil (2012) Feeling of being Contaminated on 4 visual analogue scales (0-100): 1. Intensity 2. Vividness 3. Uncontrollability 4. Distress

not reported

Warnock-parkes et al (2012) VOCI-MC not reported

Volz and Heyman (2007) NA NA

208

+= Single item only, INE= Internal negative emotions, ENE= External negative emotions, MCR=Mental Contamination Report (Herba & Rachamn 2007, Elliott et al 2009), MPQ=Mental Pollution Questionnaire (Cougle et al 2008), VOCI-MC= Vancouver Obsessions and Compulsions Inventory- Mental Contamination Scale (Rachman 2006), MPI= Mental Pollution Interview (Fairbrother et al 2004), SARA= Sexual Assault and Rape Appraisals (Fairbrother & Rachman 2004), USEQ=Unwanted Sexual Experience Questionnaire (Fairbrother et al 2005), VAS=Visual Analogue Scales.

ii) Induction of MC

MC was operationalized in ten studies by inducing it through various experimental

paradigms and then assessing changes in state MC. Typical indices assessed

included: feelings of dirtiness, urge to wash and negative emotions associated with

MC such as disgust and anxiety. Four different paradigms were used across the

studies to induce MC: non-consensual kiss victim (N=4), non-consensual kiss

perpetrator (N=1), autobiographical memory recall (N=3), objectification (N=1) and

using imagined or real vomit to induce MC (N=1). For more details see Table 4. All

of the studies used non-clinical undergraduate student populations but one only

included undergraduates who had, had a previous unwanted sexual experience

(Fairbrother & Rachman 2004). For detailed results from these studies please see

Table 5.

The first study to investigate inducing MC was Fairbrother & Rachman (2004) who

investigated the presence of mental pollution following sexual assault. 70% of the

sample reported feelings of mental pollution after a sexual assault. Recalling the

most distressing moment of the assault induced MP significantly more than recalling

a happy memory on all aspects of MP measured (p<0.01). However, it is difficult to

interpret the strength of the difference between experiment and control conditions as

effect sizes were not reported. A strength of this study is that it used both self report

and interview measures of MC. However, the within-subject design could have led to

‘carry over effects’ as no counterbalancing was reported

Four studies in this review built on this work and explored conditions needed for

inducing MC in the normal population (Fairbrother, Newth & Rachman 2005; Herba

& Rachman 2007; Elliott & Radomsky 2009, 2012). All of these studies explored

whether MC could be evoked by an imagined non-consensual kiss. The studies asked

209

female participants to listen to an audio recording and imagine they were the woman

receiving the kiss. All the studies randomly assigned participants to experimental or

control conditions to help control for confounding variables and all conducted

randomization checks. However, none of the studies reported on the method of

randomisation used. All the studies also conducted manipulation checks to assess

ease and vividness of imagining the scenarios. All four studies found that an

imagined non-consensual kiss induced significantly more feelings of MC than an

imagined consensual kiss on all indices of MC measured (p<0.01- <0.001).

Two studies used recordings to induce MC which included information about the

man in the non-consensual condition being physically dirty (Fairbrother et al 2005;

Herba & Rachman 2007). Both studies found that women in the non-consensual

condition felt significantly stronger feelings of MC than the controls (p<0.001 to

p<0.01). Herba & Rachman (2007) used their MCR scale and found a large effect for

feelings of dirtiness and urge to wash (d=2.00 to d=2.14).Both studies randomised

participants but neither used experimenters blind to the conditions so results may be

subject to experimenter effects. Also, in Fairbrother et al’s (2005) study all

participants first listened to the consensual tape before being randomised so this may

have meant participants became aware of the aims of the experiment.

The audiotapes used in these two studies have been criticised because they

simultaneously manipulated more than one construct of MC; the man was both

physically dirty and immoral because he engaged in a non-consensual kiss. Elliott &

Radomsky (2009, 2012) attempted to separate out these constructs to explore the

situational variables needed to evoke MC.

Elliott et al (2009) explored how the im/morality of the man influenced MC. In their

audio recordings the man was described as physically attractive and gave some

210

information about his morality to participants before the kiss. They found that a non-

consensual kiss evoked more MC than a consensual kiss, regardless of the morality

of the man, and a main effect of the morality of the man (p<0.05) with a small effect

size (ɳ2=.04) was also found. This was significant for all four indices on the MCR

with effect sizes ranging from medium to large (ɳ2=.20-.53). This study

demonstrated that a non-consensual kiss can evoke feelings of MC even in a man

described as moral and a consensual kiss can evoke MC if it is from an immoral

person. This study had experimenters blind to the conditions and randomised

participants to conditions.

Elliott et al (2012) used the same paradigm but manipulated the cleanliness of the

man not his morality. They found that a non-consensual kiss evoked feelings of MC

on all indices, regardless of whether the man was clean or not (p<0.05). They also

found a main effect of the cleanliness of the man (p<0.001) with a large effect size

(pɳ2=.37). A consensual kiss by a man described as dirty evoked significantly

greater feelings of MC than a consensual kiss from a clean man on all indices.

Lee, Shafran, Burgess, Carpenter, Millard et al (2013) further extended these

findings using a novel paradigm of imagining and describing a bucket of vomit

(mental condition) or seeing and handling a bucket of vomit (contact condition).

Both conditions induced feelings of internal dirtiness and urge to wash, ranging from

a small to large effect size (ɳ2=.11 to ɳ2=.66). This supports Elliott et al’s (2012)

findings that physical dirt separate from an immoral act can induce MC and extends

this to show physical dirt in the absence of a human source can induce MC. An

advantage of the paradigm used here was that they could use both male and female

participants, unlike the non-consensual kiss paradigm, however only ten men were

included in the study.

211

212

Table 4: Experimental studies: methods used to induce MCStudy Design Paradigm Used Methodology

Chen et al (2013)

Experimental, between group design

Objectification Exp 1: Female participants spoke to men in internet chat room and were told they were being watched over web cam. 1. Objectification condition: Received comments solely on physical appearance. 2. Control- comments on general character. Exp 2: Participants first recalled and wrote down a past interaction with a person of the opposite sex. Those in the objectification condition recalled an experience in which they felt that the other person only focused on their physical appearance. Control condition recalled an experience in which the other person responded appropriately during the interaction.

Coughtrey et al (2014) A.

Experimental Autobiographical memory recall

Exp 1: Participants asked to verbally recall personal memories in response to 10 contamination cue words: disgust, humiliate, shame, violated, degraded, dirty, betrayed, contaminated, impure, immoral. Exp 2: Induced MC using 5 tasks: Perpetrator memory, Victim memory, Recalling unwanted thoughts, Imagining wearing a sweater belonging to an immoral person or a hat belonging to an alcoholic. Then they were randomised to re-evoke condition (re-evoke MC 20 times), re-evoke and wash condition or control condition (re-evoke once then do nothing).

Coughtrey et al (2014) B

Experimental, within subject design

Autobiographical memory recall

Same as experiment 2 above for inducing MC. Asked to transfer MC to a neutral pencil either through contact or non-contact and asked to spread to 12 other neutral pencils. Came back after 30 mins and re-rated MC of pencil 12.

Elliott & Radomsky (2009)

Experimental, between group design

Non-consensual Kiss- Morality of man

4 conditions: Audio recordings that described a scenario at a party. 1. Control- physically attractive, moral man + consensual kiss. 2. Attractive, moral man + non-consensual kiss 3. Moral man + non-consensual kiss 4. Immoral man + non-consensual kiss

Elliott et al (2012)

Experimental Non-consensual Kiss-Dirtiness of man

4 conditions: Audio recordings that described scenario at party (same as above). 1. Clean man + consensual kiss 2. Clean man + non-consensual kiss 3. Dirty man+ Consensual kiss 4. Dirty man + non-consensual kiss

Fairbrother & Rachman (2004)

Experimental, within subject design

Recall of sexual assault Induced MC by asking women to bring to mind the most distressing moment of their assault for 20 seconds. Then asked to rate on scale 0-100 feelings of anxiety, distress, dirty on inside, strength of their urge to wash. Also recorded washing behaviours in break after imagined exposure. Control= first asked to imagine happy/pleasant memory and make same ratings.

Fairbrother et al (2005)

Experimental Non-consensual Kiss-immoral and dirty man

Four conditions: 3 non-consensual, 1 consensual. First, all participants listened to the consensual tape. Next, participants were randomly assigned to listen to either the consensual tape again, or one of the three versions of the non-consensual tape. 1. In the contamination condition, the man is also dirty and smelly. 2. and 3. defeat conditions, the participant is told she feels immobilized, trapped, helpless, under his control and unable to get away.

Ishikawa et al (2014)

Experimental Non-consensual Kiss Listened to audio recordings from Elliott & Radomsky (2009)- only non-consensual kiss recordings. Then randomly assigned to two groups- washing and no washing. Washing group led to bathroom and asked to wash hands and drink water. Non- washing asked to stay in room and not wash. MCR given out before task, after audio recording, after washing or non-washing behaviour (5 mins after time 1) and then again 15 minutes later.

213

Lee et al (2013) Experimental Vomit 2 conditions. 1. Contact contamination- asked to move a bowl of (fake) vomit from the floor to the table, then sit next to the bowl thinking about the appearance and smell of the vomit for 1 minute. 2. Mental contamination- asked to imagine a bowl of vomit for the same duration, specifically to imagine how the vomit looked and smelt. Participants were then asked to describe the look and smell of the vomit (either physically present or imagined) in three words each and say how it had made them feel in order to fully engage in the task.

Rachman et al (2012)

Experimental Non-consensual Kiss- perpetrator

Listen to audio recording of a party: 1. Control- imagine kissing a female consensual. 2. experimental- imagine giving a female a non-consensual kiss. Exp 1: Narrators voice female. Exp 2: narrators voice male plus social repercussions. Exp 3: Voice still male, still social repercussions and also betrayal (women is friend’s sister, blames women).

214

Table 5. Results from experimental studies inducing MCStudy How defined MC Results

Chen et al (2013) Exp 1

1 item: Perceived dirtiness on likert scale

Objectified condition significantly higher perceived dirtiness than control: p<.01 ɳ2.27

Chen et al (2013)Exp 2

3 items: Contaminated, polluted, dirty Objectified condition significantly higher perceived contamination than control: p<.0011 ɳ2.17

Chen et al (2013)Exp 3

3 items: Contaminated, polluted, dirty Significant association between perceived responsibility and perceived contamination: p<.01, b.59

Coughtrey et al (2014) A. Exp 1

General distress, internal dirtiness, urge to wash, anxiety

Significant increase in scores after inducing MC: General distress (d1.07), Internal dirtiness (d1.24), urge to wash (d.79) and anxiety (d1.29). All p<0.001

Coughtrey et al (2014) A. Exp 2

General distress, internal dirtiness, urge to wash, anxiety

General distress: Control group showed greater decrease over time than re-evoke only condition (d.1.59) and re-evoke and wash condition (d.72). General distress increased from trial 1 to 20 for re-evoke only (d.1.31) and re-evoke and wash (d.87). Internal dirtiness: Control group significantly greater decrease over time than re-evoke only (d.81) and wash (d1.00). Re-evoke only condition significantly higher at trial 20 than 1 (d.1.00) but significantly higher in re-evoke and wash condition (p.06). Urge to wash: Control significantly greater decrease than re-evoke (d.1.19) and wash (d.69). Significantly higher at trial 20 for re-evoke only condition (d.87) but not others. Anxiety: control group significantly greater decrease than re-evoke only (d.1.49) and wash (d.1.29). Trial 20 sig higher than 1 for re-evoke only (d.1.98) and wash (d 1.11) conditions.

Coughtrey et al (2014) B.

VOCI-MC MC transferred to neutral object in both conditions: sig increase (p<0.001) in contamination ratings of pencil from baseline to initial rating. Contact: d1.09 No contact: d.71. MC spread between neutral objects in both conditions: pencil 12 sig higher than base line (p<0.001-<0.05) Contact: d.42 No contact: d.32 Degradation: For both conditions no sig diff in contamination ratings of pencil 30 mins later(p>0.05).

Elliott & Radomsky (2009)

Feelings of dirtiness, urge to wash, INE, ENE and washing behaviours

1. Main effect of desirability of kiss (consensual vs. non) p<0.001 ɳ2=0.34 greater MC in non-consensual conditions.2. Main effect of immorality of man p<0.05, ɳ2=0.04. 3. Feelings of dirtiness: significant groups differences (p<0.001 ɳ2=0.26): No sig diffs between two non-consensual conditions (moral or immoral), Non-consensual conditions scored significantly higher than Consensual Immoral condition (CI) and CI higher than Consensual Moral condition (CM). 4. Urge to wash: Significant group differences (p<0.001, ɳ2=0.32): pattern same as feelings of dirtiness. 5. INE- significant group differences( p<0.001 ɳ2=0.20)- CM significantly less than all other conditions. 6. ENE-significant group diffs (p<0.001 ɳ2=0.53): No significant difference between two non-consensual conditions but significantly higher scores than CI and CI sig more than CM. 7.

Elliott et al (2012) Feelings of dirtiness, urge to wash, INE, ENE and washing behaviours

1. A main effect of desirability of the kiss (consensual vs non-consensual) p < .001, partial η2 = .39 2. A main effect of physicality of the man (dirty or clean) p < .001, partial η2 = .37 3. Feelings of dirtiness: Significant group differences (p < .001, partial η2 = .54): Non-Consensual Physically Dirty(NCPD)= highest feelings of dirtiness. Urges to wash: significant group differences (p < .001, partial η2 = .37.) NCPD highest. Negative Internal Emotions: significant group differences (p < .001; partial η2 = .27): Consensual Physically Clean (CPC) lowest scores. External negative emotions: significant group differences (p < .001, partial η2 = .61.) NCPD greatest.

Fairbrother & Rachman (2004)

Mental pollution: Anxiety, distress, dirty on the inside, urge to wash

Significant difference between happy memory recollection and assault recollection on all items- anxiety, distress, dirty on inside, strength of their urge to wash (p<0.01

Fairbrother et al (2005)

Core mental pollution items, feeling: Dirty or unclean, dirty on the outside, dirty on the inside, dirty non-physical terms

Participants in the non-consensual condition reported feeling significantly more feelings of mental pollution than control on all indices: dirty or unclean, dirty on the outside, dirty on the inside and dirty in non-physical terms. All p<0.001.

215

*Herba & Rachman (2007)

Dirtiness and urge to wash Participants in non-consensual condition significantly stronger feelings of dirtiness (p<.01, d=2.00) and urge to wash (p<.01, d=2.14). 27% women in NC condition engaged in a form of washing. 0% from consensual

Ishikawa et al (2014)

Feelings of dirtiness, urge to wash, INE, ENE

No sig diffs in MC between washing and non-washing groups. Was a significant main effect of time for both groups- p<0.001, partial ɳ2=.45. Both groups MC reduced over time for all indices of MC. Apart from INE which took 20 mins to reduce, didn't reduce after 5 mins like other indices.

Lee et al (2013) Disgust, feelings of dirtiness, internal feelings of dirtiness, urge to wash, anxiety, washing behaviour.

In both conditions contamination was induced p<0.05. All MC indices increased from baseline after carrying fake vomit and after imagining carrying vomit, although there was a greater increase in the contact conditions. Feelings of dirtiness: p<0.001, pɳ2=.32, Disgust: p<0.001, pɳ2=.66, internal feelings of dirtiness: p<0.001, pɳ2=.11, Anxiety: p<0.001, pɳ2=.17, Urge to wash: p<0.001, pɳ2=.16, Washing behaviours: no significant difference between groups

Rachman et al (2012) Exp 1:

Feelings of dirtiness, urge to wash, negative emotions

Anxiety: time x condition interaction (p<0.005) Anxiety significant increase in non-consensual condition (NC) but not in control, Same for disgust and shame ( p<0.001). Dirtiness: NC sig higher than Consensual Condition (CC) (p<0.005), Urge to wash: No sig diffs

Rachman et al (2012) Exp 2:

Feelings of dirtiness, urge to wash, negative emotions

Anxiety: time x condition interaction (p<0.005) Anxiety increased in NC group after experiment but not in the control group. Same for disgust, shame and guilt (p<0.05). Dirtiness: NC sig increased from baseline after experiment CC didn't (p<0.005). Urge to wash: No sig diffs.

Rachman et al (2012) Exp 3

Feelings of dirtiness, urge to wash, negative emotions

Anxiety: time x condition interaction (p=0.06): Anxiety increased after exp in NC group but not control (CC). Same for disgust, shame, guilt, anger and sadness (p<0.05). Dirtiness: NC sig increased more than CC (p<0.001) from baseline. Urge to wash: NC sig higher than CC p<0.001.

Rachman et al (2012) Exp 4

Feelings of dirtiness, urge to wash, negative emotions

Anxiety: time x condition interaction p<0.05 ɳ2=.12. NC sig increased CC didn't. NC sig higher ratings of shame, guilt, distress, disgust (p<0.001). Feelings of dirtiness: NC sig more (p<0.001). Urge to wash: NC sig more than CC (p<0.005).

INE=Internal Negative Emotions, ENE=External Negative Emotions, CC= Consensual Condition, NC= Non-consensual condition

216

Table 5. Continued

The following three studies extended research on MC using different paradigms to

induce MC (Rachman, Elliott & Zysk 2012; Chen, Teng & Zhang 2013; Coughtrey,

Shafran & Rachman 2014a).

Chen et al (2013) used two experimental paradigms to investigate whether sexual

objectification could induce MC. They found that receiving objectifying comments

via a web chat led Chinese women to feel significantly higher levels of dirtiness than

the control group (p<.01) with a large effect size (ɳ2=.27). In a group of American

women recalling a past experience of objectification, significantly higher feelings of

MC were reported than in the control group (all ps<.001) with a large effect size

(ɳ2=.17). Experiment one did not require participants to imagine a scenario and the

second paradigm used real life recollections both of which improves the external

validity of the study. However, experiment one does not report on how participants

were assigned to the experimental or control conditions and does not report any

checks to see if groups were comparable on core variables. Therefore, confounding

variables may have influenced results. Also, the first experiment only measured MC

on one indices, perceived dirtiness, so does not measure the whole construct of MC.

Coughtrey et al’s (2014a) first experiment showed that MC could be induced by

recalling autobiographical memories of events related to accusations, betrayal, harm

and threat. All four indices of MC significantly increased after recalling the

memories (all p<0.001). Over the four indices the effect size range was d.=79 to

d=1.24 indicating a large effect size. This study used memories of real life scenarios

rather than imagined events and used a male and female sample increasing the

external validity of the study. However, there was no control group so feelings of

MC may have increased over time without the manipulation.

217

Rachman et al (2012) used the kiss paradigm in an all-male sample to explore if MC

could be induced in the perpetrator of the kiss. Four interconnected experiments were

carried out which manipulated the recordings played to participants (see Table 4). In

all experiments, internal feelings of dirtiness and negative emotions were

significantly higher in the non-consensual ‘perpetrator’ group than in the control (all

p<0.005). However, significant differences in urge to wash were only found when

elements of betrayal were added to the recordings in experiments three and four

(p<0.001). Results study showed MC could be evoked by one’s own unacceptable

imagined behaviour and that adding social repercussions and elements of betrayal

increased MC feelings. This study included an independent replication of experiment

three in another laboratory and found similar results increasing the external validity

of this study. However, all four experiments used quite small sample sizes (N=39-40)

and no power analysis was reported. There is also a question of experimenter effects

because the study did not use investigators blind to the condition.

In addition to investigating the factors involved in inducing MC studies have also

looked at the decay and spread of MC. Two studies investigated the spontaneous

decay of MC and properties that might affect this using different paradigms

(Ishikawa, Kobori, Komuro & Shimizu 2014; Coughtrey et al 2014a). Both found

feelings of MC spontaneously decayed after a short time (p<0.001). This took 5

minutes in Ishikawa et al’s (2014) study for all MC indices, measured by the MCR,

except INE, such as shame and guilt, which took 20 minutes. In Coughtrey et al’s

(2014a) study MC, measured by VAS, returned to baseline on all indices after 3

minutes; this study did not include measures of INE. Ishikawa et al (2014) included a

washing group where participants were asked to wash their hands after the induction

of MC; no effect of washing on levels of MC decay was found. Coughtrey et al

218

(2014 A) had a re-evoke and a re-evoke and washing condition, both of which

prevented a return to baseline on all indices of MC. The highest levels of MC were

found in the re-evoke without washing group suggesting that repeated triggers may

be the cause of ongoing MC. Both studies randomised participants to groups and did

randomisation checks. However, neither study controlled whether participants in the

control conditions engaged in any internal neutralising behaviours to reduce feelings

of MC. A strength of the paradigm used in Coughtrey et al’s (2014a) study was that

it used five different scenarios to evoke MC and then selected the one that was the

most sensitive for each participant for the decay task (see Table 4). This recognised

the idiosyncratic nature of MC ensuring the induction of MC was meaningful for

each participant, increasing ecological validity.

In order to investigate whether MC can spread, Coughtrey, Shafran & Rachman

(2014b) induced MC and asked participants to transfer it to a neutral pencil and then

to further spread it across twelve neutral pencils either with or without direct physical

contact. MC was transferred to a neutral pencil both through physical contact (72%)

and without physical contact (48%). There was a significant increase in

contamination ratings of the pencil (p<0.001) and a large effect in the contact

condition (d=1.09) and medium effect in the non-contact condition (d=.71). Feelings

of MC were also spread across pencils: significantly higher levels of MC were

reported for pencil 12 than baseline ratings in both conditions (p<0.05) with a small

to medium effect size (contact d=.42, non-contact d=.32). There was no degradation

of MC on pencil 12 after 30 minutes. This study used undergraduates with

contamination fears so may not be generalizable beyond this. This study used a

within-subject design and included counterbalancing to minimise fatigue effects.

iii) Correlates and predictors of MC

219

Ten studies explored the relationship of MC to a variety of other variables, seven of

the studies measured MC in their sample without inducing it and three studies used

an experimental paradigm to induce MC and then measured it. Four measures of MC

were used across the studies: a version of the MCR (N=4), the VOCI-MC (N=2), the

MPQ (N=3) and the Sexual Assault and Rape Appraisals (SARA) (N=1). A variety

of factors related to MC were investigated, the most common factors were:

obsessive-compulsiveness (N=7), disgust (N=7), anxiety (N=4) and PTSD symptoms

(N=4). For correlations and regression co-efficents see Tables 6 and 7

MC and OCD symptoms

One of the most common variables investigated across studies was obsessive-

compulsiveness which was included in six studies but was measured in different

ways across them (Herba & Rachman 2007; Cougle et al 2008; Radomsky & Elliott

2009; Coughtrey et al 2012; Elliott & Radomsky 2013; Carraresi, Bulli, Melli &

Stopani 2013). It was measured in terms of: obsessions and compulsions (N=4),

contact contamination (N=3) and sensitivity to contamination (N=1). In all studies

obsessive-compulsiveness significantly, positively correlated with some aspect of

MC with effect sizes ranging from .25 to .61 indicating a small to large effect size.

The largest correlation co-efficient was reported by Coughtrey et al (2012) which

measured both obsessions and compulsions, while Elliott & Radomsky (2013) and

Radomsky & Elliott (2009) who measured contact contamination, reported the

smallest co-efficient.

Cougle et al (2008) conducted two studies using two different measures of obsessive-

compulsiveness (VOCI and OCI-R). Obsessive-compulsiveness was significantly

associated with some aspect of MC, as measured by the MPQ subscales. Effect sizes

ranged from .36 to .56 suggesting a medium to large effect of OCD symptoms on the

220

mental pollution sub-type of MC. However, the discriminant validity of the MPQ has

not been investigated and it is possible the relationship found is because the MPQ is

also measuring obsessive-compulsiveness.

Two studies used the VOCI-MC and investigated obsessive compulsiveness

(Carraresi et al 2013; Coughtrey et 2012). Both studies found a significant

correlation between obsessive compulsiveness and MC with a large effect size range

of .48 to .61. Coughtrey et al (2012) also found a significant relationship between

sensitivity to contamination and MC (r=.56). These two studies had larger effect

sizes (.48 to .61) than the other studies in this group. These were the only studies to

use a clinical OCD sample so this may account for the larger effect sizes found.

The lowest effect size range was found in experimental studies which investigated

contact contamination and MC (.25 to .36) indicating a small to medium relationship

(Herba & Rachman 2007; Radomsky & Elliott 2009; Elliot & Radomsky 2013). All

studies found contact contamination was significantly positively correlated with

feelings of dirtiness (r=.25 to 33) and urge to wash (r.24 to r.25) suggesting a small

to medium effect. The two studies which measured INE found significant

correlations with an effect size range of r=.26 to .33 (Radomsky & Elliott 2009;

Elliott & Radomsky 2013). However, across the studies contact contamination did

not consistently correlate with the other indices of MC (see Table 7).

Contact contamination was entered into a regression analysis; Elliott & Radomsky

(2013) found that contact contamination did not predict any indices of MC. In

contrast, Herba & Rachman (2007) and Radomsky & Elliott (2009) found it was a

significant predictor of feelings of dirtiness (ß range .24 to 1.20) and urge to wash (ß

range .37 to .99). Radomsky & Elliott also found INE (ß.92) and ENE (ß.71) were

221

predicted by contact contamination but these indices were not measured in Herba &

Rachman’s (2007) study.

Cognitive processes and MCTwo studies included variables which are considered to be cognitive biases often

seen in people with OCD: Thought-action fusion (TAF) and inflated responsibility

beliefs (Coughtrey et al 2012; Cougle et al 2008). TAF is the belief that thinking

something immoral is as bad as doing it (TAF-moral) and that thinking something

increases the chance of it happening to the self (TAF-likelihood self) or others (TAF-

likelihood others). There is also TAF-MC thinking about contamination increases the

likelihood of it happening. Two studies examined the relationship between TAF and

MC and both found a significant positive relationship with some aspect of TAF and

MC with an effect size range of r=.27 to .61 indicating a small to large effect

(Coughtrey et al 2012; Cougle et al 2008). The smallest effect size was found by

Cougle et al (2008), between MPQ-Wash and TAF-moral, and the largest effect size

was found by Coughtrey et al (2012) study 1 between VOCI-MC and TAF.

However, in Coughtrey et al’s (2012) second study TAF was not significantly

correlated with MC but TAF-MC was with a medium effect size (r.49). Cougle et al

(2008) found that both indices of the MPQ correlated with all indices of TAF. The

range of correlation co-efficient was .27 to .50 suggesting a small to large effect size

of TAF on the mental pollution sub type of MC. Cougle et al (2008) also found a

significant positive correlation between inflated responsibility beliefs and both

aspects of the MPQ (p<0.001)

MC and Disgust

Six studies investigated the relationship between MC and disgust: five studies

investigated disgust sensitivity (Cougle et al 2008; Herba & Rachman 2007;

222

Radomsky & Elliot 2009; Elliott & Radomsky 2013; Badour et al 2013b) and one

investigated disgust propensity (Carreresi et al 2013). In the latter, disgust propensity

was significantly correlated with MC with an effect size of .29 suggesting a small to

borderline medium effect.

For disgust sensitivity Cougle et al (2008) found no significant correlations with

either indices of the MPQ which suggests disgust sensitivity is not related to MP. It

is worth noting though that while all studies used the same disgust scale, Cougle et al

(2008) used the short form. Four studies found a significant correlation with disgust

sensitivity and MC with a co-efficient range of: r=.23 to .43 suggesting a small to

large effect (Badour et al 2013b; Radomsky & Elliot 2009; Elliot & Radomsky 2013;

Herba & Rachman 2007). The largest co-efficient was found by: Badour et al

(2013b), between disgust and sexual assault related MC in a sample of women with a

history of sexual assault, and Elliot & Radomsky (2009) between disgust and ENE in

a non-clinical sample. The smallest co-efficient was found between INE and disgust

(Elliott & Radomsky 2009).

Of the three studies which used the MCR, disgust sensitivity significantly correlated

with and predicted some aspects of MC in all studies but this was not consistent

across indices and studies. Radomsky & Elliott (2009) found it predicted ENE

(ß=1.54), Elliott & Radomsky (2013) found it predicted urge to wash (ß.35) and

Herba and Rachman (2007) found it predicted feelings of dirtiness (ß.26).

Overall, across the studies disgust did not consistently predict indices of MC which

makes it hard to make conclusions about the relationship between the two.

MC and Guilt

Two studies explored guilt in relation to MC both using the MPQ, (Cougle et al

2008; Berman, Wheaton, Fabricant & Abramowitz 2012). Correlation co-efficients

223

ranged from .16 to .59. The large effect size range is likely due to the different

aspects of guilt that were measured. Cougle et al (2008) found the largest co-efficient

which was between MPQ-Ideation and trait guilt. No relationship was found between

trait guilt and MPQ-Washing.

Berman et al (2012) looked at guilt induction from parents and scores on the MPQ.

Two indices of guilt measured were: guilt induction disparagement (MGI-D) relating

to ratings of parental criticism and blame, and guilt induction self-serving (MGI-SS)

relating to ratings of parents exaggerated sacrifices. Berman et al (2012) found, in a

regression analysis, that guilt disparagement was a significant predictor of the MPQ

washing scale (ß.47) but neither of the guilt subscales predicted MPQ ideation

scores.

MC and anxiety

Three studies examined the relationship between anxiety sensitivity and MC,

measured by the MCR, (Herba & Rachman 2007; Radomsky & Elliott 2009; Elliott

and Radomsky 2013). All studies found anxiety sensitivity correlated with some

aspect of MC; co-efficient range was .25 to .29 suggesting small to borderline

medium effects. The indices of MC that anxiety sensitivity correlated with was not

consistent across the studies. Urge to wash correlated with anxiety in two studies

(Elliott & Radomsky 2013; Radomsky & Elliott 2009), feelings of dirtiness

correlated in one study (Herba & Rachman 2007) and INE and ENE correlated in

one study (Radomsky & Elliott 2009). In a regression analysis only Radomsky &

Elliott (2013) found anxiety sensitivity predicted any aspect of MC: urge to wash,

ß=.35.

Appraisals and MC

224

In their studies Elliott & Radomsky (2013) and Radomsky & Elliott (2009) induced

MC using the non-consensual kiss paradigm. They explored the relationship between

ratings of responsibility for the kiss and perceiving the kiss as a violation and MC.

In a regression analysis both studies found responsibility appraisals was a significant

predictor of INE: ß.37 (Elliott & Radomsky 2013) and ß.78 (Radomsky & Elliott

2009). Elliott & Radomsky (2013) found that responsibility appraisals did not predict

any other indices of MC but Radomsky & Elliott (2009) found that they also

predicted feelings of dirtiness (ß.50) and urge to wash (ß.60). Overall these findings

suggest that Elliott & Radomsky (2013) failed to replicate their previous findings.

Both studies found that violation appraisals were a significant predictor of INE (ß

range .36 to .37) and ENE (ß range .23 to .53). Only Radomsky & Elliott (2009)

found violation appraisals predicted feelings of dirtiness (ß.41) and neither study

found they predicted urge to wash.

PTSD Symptoms and MC

Four studies explored the relationship between PTSD symptoms and MC in samples

who had experienced sexual assault. All found a significant relationship with a

correlation co-efficient range of r.47 to r.66 (Badour et al 2013a; Badour et al 2013b;

Olantunji et al 2008; Fairbrother & Rachman 2004). Suggesting medium to large

effects. Olantunji et al (2008) also found that MC was significantly correlated with

PTSD cognitions (r=.49). In a regression Badour et al (2013b) and Olantunji et al

(2008) found PTSD symptoms predicted MC however, when Olantuji et al (2008)

controlled for PTSD cognitions PTSD symptoms were no longer a significant

predictor. This could suggest that cognitions mediate the relationship between PTSD

and MP. Badour et al (2013a) induced MC by asking participants to recall elements

225

of the traumatic event and found that PTSD symptom severity was associated with an

increase in MC indices in the sexual but not physical assault group.

Other indices correlated with MC

A number of other variables have been assessed infrequently across the studies.

Berman et al (2012) investigated childhood trauma and religion in relation to MC.

For childhood trauma, sexual abuse was a significant individual predictor of MPQ-

Wash (ß.25) and emotional abuse was a significant individual predictor of MPQ-

Ideation (ß.28). For religion, intrinsic-extrinsic motivation for religion was found to

be a significant predictor of MPQ-W only (ß.33). However, the internal consistency

for the measure of this aspect of religion is not strong α=.58. Also, the study only

looked at the Christian faith as religious predictors.

Radomsky & Elliott (2009) and Elliott & Radomsky (2013) investigated neuroticism

and found conflicting results: only Radomsky & Elliott (2009) found neuroticism

was a significant predictor of any MC indices: urge to wash (ß.45). Inconsistent

findings make it difficult to draw conclusions about the relationship between

neuroticism and MC.

226

  OCD Symptoms

Thought Action Fusion

Disgust Guilt Anxiety

Inflated responsibilit

y

Childhood trauma

Religion Neuroticism

Responsibility Appraisals

Violation Appraisals

VOCI-MC (1 MC indices)

No of studies 3 2 1 0 2 0 0 0 0 0

No. correlation coefficients 4 3 1 2

No. significant correlations4, p<0.001

2, p<0.001 1, p<0.001 0

Effect size range .48 to .61 .56 to .61 0.29                

MPQ (2 MC indices)

No of studies 2 1 1 2 0 1 1 1 0 0

No. correlation coefficients 8 5 1 4 2 4 6

No of significant correlations 6, p<0.001 4, p<0.001 03,

p<0.05 2, p<0.001 4, p<0.0012,

p<0.05

No. sig correlated MPQ-Wash 3 2 0 1 2 1

No. sig correlated MPQ-I 3 2 0 2 2 1

Effect size range .36 to .64 .27 to .50 0 .16

to .59   .36 to .48 .13 to .49.13

to .31      

MCR (4 MC indices)

No of studies 3 0 3 0 3 0 0 0 2 2 2

No. correlation coefficients 12 12 12 8 8 8

No. significant correlations 8, p<0.05 6, p<0.055,

p<0.05 3, p<0.05 4, p<0.001 6, p<0.05

No. of studies sig findings 1

Feelings of dirtiness 3 2 1 1 1 1

Urge to wash 2 2 2 1 1 1

INE 2 1 1 1 2 2

ENE 1 1 1 0 0 2

Effect size range .25 to .36   .23 to .43.16

to .59.25

to .29       .22 to .39 .37 to .55 .22 to .49

TOTAL RANGE (All studies) .25 to .64 .27 to .61 .23 to .43.16

to .59.25

to .29 .36 to .48 .13 to .49 .13 to.31 .22 to .39 .31 to .44 .22 to .49

227

Table 6. Correlation co-efficients extracted from studies

 

Contact Contamination (VOCI-CTN)

Disgust sensitivity

Anxiety Responsibility Appraisals

Violation Appraisals

Personality Traits-

Neuroticism (BFI-NI)

MCR (4 indices MC)No. of studies 3 3 3 2 2 2No. of indices of MC predicted 0-4 1 0-1 1-3 2-3 0-1No. of studies predicted Feelings of dirtiness 2 1 0 1 1 0No. of studies predicted Urge to wash 2 1 1 1 0 1No. of studies predicted INE 1 0 0 2 2 0No. of studies predicted ENE 1 1 1 0 2 0

No. of significant predictors 6, p<.01-0.15 3, p <.052,

p<.05 4, p<.001-.05 5, p<0.05 1, p<.05

ß range .24 to 1.20 .26 to 1.54.35

to .37 .37 to .78 .23 to .53 0.45

228

Table 7. Regression co-efficents extracted from studies

iv) Interventions for Mental Contamination

Three studies in this review explored interventions for MC and one presented a case study of

transformation obsessions. Jung et al (2009) used a PTSD clinical sample and the remaining

three studies an OCD clinical sample. In the intervention studies either CBT tailored for

mental contamination (Warnock-Parkes et al 2013; Coughtrey et al 2013) or cognitive

restructuring and imagery modification (CRIM; Jung et al 2009) was used. Although

treatment type differed between the studies there were some similarities in the interventions:

all three studies reported using some form of imagery technique, all used psychoeducation

and looked at cognitive aspects of MC either through meaning of MC, appraisals of MC or

cognitive restructuring. The main differences in the CBT treatment for MC was only one

used exposure with response prevention (Warnock-Parkes et al 2008) although both used

behavioural experiments. The number of sessions also differed; treatment length of the CBT

studies was longer (13 or 20 sessions) whereas CRIM only had one session and a booster

session.

Results show that all studies found a reduction of MC from pre-treatment to post treatment.

Coughtrey et al (2014) reported the largest reduction in scores on the VOCI-MC with a mean

reduction of 33.92 which was significant (p<0.001) and had a large effect size (d=1.42).

Warnock-Parkes et al (2012) reported a reduction of 10 points on the VOCI-MC but did not

report whether this difference was significant. Jung & Steil (2012) found a reduction in

scores across all MC indices from pre-treatment to post treatment but did not report

information on significance levels.

In all studies, although MC appears to have been responsive to the treatment it is difficult to

make conclusions about the effectiveness of interventions given study limitations. None of

the studies had comparison or control groups so it cannot be clear that it was the intervention

which improved MC scores. Also, the studies are based on a small number of cases so cannot

229

be generalised to the wider clinical population. Warnock-Parkes et al’s (2009) study reports

that their participant (N=1) had 6 sessions of normal CBT first and MC measures were only

taken after this. Therefore, it is hard to separate the effect of the two separate interventions on

MC. Although Coughtrey et al (2013) found a significant effect of their intervention, the

external validity of their study is questioned as the treatment was given by expert clinicians

and delivered in a research setting. Furthermore, the measures of MC used by these studies

(VOCI-MC and VAS) lack solid research to establish whether they are reliable measures of

therapeutic change.

Volz & Heyman (2007) were the only study in this literature review which specifically

explored the concept of transformation/morphing obsessions which is a sub-type of MC. In

their case series they presented evidence for the presence of transformation obsessions in an

adolescent population of people with OCD. They found morphing obsessions were linked to

avoidance, rituals and function impairments. They reported that psychoeducation around this

symptom, ERP and cognitive work specifically targeting morphing led to improvements in

the cases described. However, no measures or statistical information was reported.

Discussion:

The main aim of this review was to assess the current body of research on the relatively new

concept of MC. The twenty five studies reviewed here contribute to the knowledge base of

MC The main findings of this review found that MC can be induced without physical contact

with a contaminant through a variety of different paradigms (see Tables 4 & 5) .The

induction studies found that MC can be induced after a physical violation, moral violation or

through self-contamination- repugnant thoughts, images and memories. The studies by Elliott

& Radomsky (2009, 2012) supported the concept that the source of MC can be a dirty or

immoral person and Lee et al (2013) extended this to show a dirty object can also evoke MC.

These findings confirm the theoretical principles of MC hypothesis by Rachman (2006).

230

Rachman’s theory also suggests that MC leaves individuals with feelings of internal dirtiness

and generates negative emotions including: disgust, anxiety, fear, anger, shame, guilt and

revulsion leading to an urge to wash (Rachman 2006). The findings of this review generally

supported these theoretical notions. All studies found that evoking MC lead to increased

feelings of internal dirtiness and studies that measured negative emotions associated with MC

found these were also evoked after inducing. MC was found to lead to an urge to wash in

studies that measured it with the exception of Rachman et al (2012) where MC only led to an

urge to wash when betrayal was added to the paradigm (Elliott & Radomsky 2009, 2012; Lee

et al 2013; Coughtrey et al 2014a) . The reason for this discrepancy is likely to be because

Rachman’s et al (2012) study was focussed on perpetrators while other studies focussed on

victims. This suggests a link between betrayals and urges to wash in perpetrators which is

consistent with Rachman’s (2010) theory but this needs further research.

The review highlights a new element of MC that is beginning to receive attention is the

spread and decay of MC. Rachman’s theory suggests that feelings of MC will remain even

after the source of contamination has gone and will not be responsive to washing behaviours.

Ishikawa et al (2014) found support that MC was not responsive to washing but there is

limited research in this area. In contrast to the theory, Coughtrey et al (2014a) and Ishikawa

et al (2014) found that MC spontaneously decayed quickly; however, Coughtrey et al (2014b)

found no decay after 30 minutes in feelings of MC. The discrepancies in findings could be

due to the fact that Coughtrey et al’s (2014b) study required participants to retain the MC for

longer as they were also transferring MC or it may be that, in the earlier studies internal

neutralising behaviours were taking place. Coughtrey et al’s (2014b) study found MC could

be spread between pencils without a large reduction in the strength of MC supporting

Rachman’s (2006) theory that MC can be spread to similar objects.

231

When exploring factors affecting MC the most examined factors across studies were:

obsessive-compulsiveness, PTSD symptoms, disgust, anxiety and cognitive biases. The

findings of this review showed that these factors were related to aspects of MC but

relationships were not consistently found across studies or indices of MC. Obsessive-

compulsiveness is conceptually linked to MC as it is part of fear of contamination found in

many cases of OCD (Calamari et al 2004). Studies in this review found support for this

relationship but did not tease out how obsessions or compulsions may have related differently

to MC. Contact contamination was looked at separately in three experimental studies and two

found contact contamination predicted feelings of MC which supports Rachman’s (2006)

hypothesis that individuals may possess a general underlying sensitivity to contamination.

However, Radomsky & Elliott (2013) did not find contact contamination predicted MC; this

could be due to the smaller sample size used in this study and no report of whether this was

sufficient to detect a relationship.

The conceptual link between PTSD symptoms after sexual assault and MC was supported by

studies in this review (Fairbrother et al 2004; Oltunaji et al 2008; Badour et al 2013a; Badour

et al 2013b). This lends support to the theory that the two are related and MC may be a

vulnerability or maintenance factor in PTSD (Fairbrother & Rachman 2004). In addition,

Oltunaji et al (2008) found that negative trauma-related cognitions mediated this relationship;

this needs to be explored further.

This review found that MC was related to negative emotions including: anxiety, disgust and

guilt however, inconsistent results were found across studies. In the three experimental

studies anxiety and disgust sensitivity did not consistently correlate with or predict the same

indices of MC (Herba & Rachman 2007; Elliott & Radomsky 2009; Radomsky & Elliott

2013). These discrepancies may be due to the fact that the source of MC was different for all

studies (dirty and immoral man, immoral man or dirty), which could suggest that the impact

232

of emotional sensitivities on MC varies depending on the source of MC. The experimental

studies used an imagined event to induce MC and negative emotions may be higher for real

life violations. In terms of disgust, none of the studies measured moral disgust which is often

triggered by moral violations so may be further linked to MC (Ille, Kapfhammer, Arendasy,

Sommer et al 2014). Guilt was also found to be associated with MC in two studies but it is

difficult to make conclusions here.

Cognitive biases associated with MC were commonly explored through appraisals of the

experimental situations and TAF. In OCD it is the interpretations of one’s intrusive thoughts

that leads to distress and compulsive behaviours (Salkvoskis 2000; Shafran et al 1996) and it

is hypothesised that interpretation may also play a role in MC. Across two experimental

studies responsibility and violation appraisals were related to MC but did not consistently

predict the same indices of MC (Elliott & Radomsky 2009; Radomsky & Elliott 2013). The

discrepancies in findings may be due to the different sources used to evoke MC. In both

studies it is unclear whether interpretations led to MC or whether feelings of MC led to

negative appraisals. Findings from this review suggest some relationship between TAF and

MC however, further investigation is needed to draw conclusions. It may be that TAF leads

to MC, MC leads to TAF or that there is a third variable which influences them both.

More studies are needed to elucidate which aspects of MC are related to the different

variables and also to explore whether how MC is induced effects its relationship with other

variables.

Limitations:

The studies inducing MC lack external validity in that they were all conducted in an artificial

setting with undergraduate, volunteer samples which limits how far the results can be

generalised. However, the laboratory setting and use of randomisation did allow for good

control of variables in some studies. The non-consensual kiss paradigm involved an imagined

233

hypothetical situation which again lacks ecological validity; however, the use of memory

recall paradigms in some studies improves on this. There was also a high proportion of

females used in the studies making results difficult to generalise to the whole population and

MC in men needs further exploration. The perpetrator study involved just men so these

findings cannot be extended to women perpetrators. Across the induction studies none used a

clinical sample so there is a gap in the literature around evoking MC in clinical samples.

The studies exploring factors associated with MC cannot establish causality and it is difficult

to know whether the variables cause MC or experiencing MC leads to them. Studies that

induced MC and then looked at predictors used hypothetical, imagined events so lack

ecological validity. The self-report measures used are also subject to social desirability

effects and the small number of clinical samples used lacks generalizability to psychological

disorders.

Another limitation across the studies was the measures of MC used as not all reported robust

validity and reliability findings: the inter-item reliability reported for measures was good but

evidence was lacking on test-retest reliability to ascertain whether the measures used showed

stability over time. The review itself is limited as it does not include any grey material

making the results subject to publication bias.

Implications and recommendations

MC has the potential to be a trans-diagnostic factor across other psychological disorders. It

has been found in this review to relate to OCD and PTSD. The evidence that MC can be

induced through perceived moral transgressions by others or the self and can be evoked and

re-evoked by thoughts and memories adds further support to this idea. Furthermore, the

emotions found to be associated with MC are also evident in a number of depressive and

anxiety related conditions (Fergus et al 2010; Sangmoon, Ryan, Randall 2011; Ille et al

2014). If MC is a distinct phenomenon in some psychological disorders then it is important

234

to continue to understand the mechanisms and vulnerabilities of MC and develop specific

interventions; treatment studies reviewed here begun to investigate this but more robust

evidence is needed for this.

This review highlighted a lack of robust findings on the mechanisms of MC in relation to the

cognitive biases and appraisals involved in experiencing MC and the spread of MC. In terms

of the emotions associated with MC this also needs further exploration; shame is purported to

be associated with MC but this was not examined in these studies. It is hypothesized that the

misinterpretation or internalisation of this emotional arousal activates and maintains MC and

this has not been explored thoroughly in the research reviewed here (Coughtrey et al 2012). It

would also be useful to understand the differing or similar mechanisms involved in

experiencing MC for victims and perpetrators of MC. Furthermore, as MC leads to an urge to

wash and can be spread in the same way as contact contamination then it may be underlying

compulsive washing in people with OCD particularly if it can be transferred without physical

contact. It would therefore be important to explore the extent of the transference of MC and

whether it can it be spread to other non-similar objects or to people. The review also showed

that there is a lack of research with clinical populations on the induction of MC and factors

associated with it, research is needed to explore this further. More ecologically valid research

is also required; it would be interesting to see if the effects of the perpetrator study would

translate to real memories of assaults. The review also highlighted a lack of research on the

MC subtype of morphing obsessions; Volz & Heyman (2007) described this as leading to

functional impairment so research should begin to focus on this aspect of MC. Finally, a

measure of MC that has been validated in clinical populations and can be used as a measure

of therapeutic change is needed in order to accurately assess MC in psychological disorders.

235

References

Badour, C. L., Feldner, M. T., Blumenthal, H., & Bujarski, S. J. (2013a). Examination of increased

mental contamination as a potential mechanism in the association between disgust sensitivity

and sexual assault-related posttraumatic stress. Cognitive Therapy and Research, 37 (4), 697-

703.

Badour, C. L., Feldner, M. T., Babson, K. A., Blumenthal, H., & Dutton, C. E. (2013b). Disgust,

mental contamination, and posttraumatic stress: Unique relations following sexual versus

non-sexual assault. Journal of Anxiety Disorders, 27(1), 155-162.

236

Berman, N. C., Wheaton, M. G., Fabricant, L. E., & Abramowitz, J. S. (2012). Predictors of mental

pollution: The contribution of religion, parenting strategies, and childhood trauma. Journal of

Obsessive-Compulsive and Related Disorders, 1(3), 153-158.

Calamari, J. E., Weigartz, P.S., Riemann, B.C., Cohen, R. J., Greer, A., Jacobi, D. M., Jahn, S.C., &

Carmin, C. (2004). Obsessive-compulsive disorder subtypes: An attempted replication and

extension of a symptom based taxonomy. Behaviour Research and Therapy, 42, 647-670.

Carraresi, C., Bulli, F., Melli, G., & Stopani, E. (2013). Mental contamination in OCD: Its role in the

relationship between disgust propensity and fear of contamination. Clinical Neuropsychiatry,

10(3),13-19.

Chen, Z., Teng, F., & Zhang, H. (2013). Sinful flesh: Sexual objectification threatens women's moral

self. Journal of Experimental Social Psychology, 49(6), 1042-1048.

Cohelo, J. S., & Whittal, M. L. (2001). Are subtypes of OCD differentially responsive to treatment?.

Vancouver, Canada: World Congress of Behavioural and Cognitive Therapies

Coughtrey, A. E., Shafran, R., Knibbs, D., & Rachman, J. (2012). Mental contamination in

obsessive-compulsive disorder. Journal of Obsessive-Compulsive and Related Disorders,

1(4), 244-250.

Coughtrey, A. E., Shafran, R., Lee, M., & Rachman, S. J. (2012). It's the feeling inside my head: A

qualitative analysis of mental contamination in obsessive-compulsive disorder. Behavioural

and Cognitive Psychotherapy,40(2), 163-173.

237

Coughtrey, A. E., Shafran, R., & Lee, M. (2013). The treatment of mental contamination: A case

series. Cognitive and Behavioral Practice 20(2), 221-231.

Coughtrey, A. E., Shafran, R., & Rachman, S. J. (2013). Imagery in mental contamination: A

questionnaire study. Journal of Obsessive-Compulsive and Related Disorders, 2(4), 385-390.

Coughtrey, A. E., Shafran, R., & Rachman, S. J. (2014a). The spontaneous decay and persistence of

mental contamination: An experimental analysis. Journal of Behavior Therapy and

Experimental Psychiatry, 45(1), 90-96.

Coughtrey, A. E., Shafran, R., & Rachman, S. J. (2014b). The spread of mental contamination.

Journal of Behavior Therapy and Experimental Psychiatry, 45(1), 33-38.

Cougle, J. R., Lee, H., Horowitz, J. D., Wolitzky-Taylor, K. B., & Telch, M. J. (2008). An

exploration of the relationship between mental pollution and OCD symptoms. Journal of

Behavior Therapy and Experimental Psychiatry,39(3), 340-353.

Elliott, C. M., & Radomsky, A. S. (2009). Analyses of mental contamination: Part I, experimental

manipulations of morality. Behaviour Research and Therapy, 47(12), 995-1003.

Elliott, C. M., & Radomsky, A. S. (2012). Mental contamination: The effects of imagined physical

dirt and immoral behaviour. Behaviour Research and Therapy, 50(6), 422-427.

238

Elliott, C. M., & Radomsky, A. S. (2013). Meaning and mental contamination: Focus on appraisals.

Clinical Psychologist, 17(1), 17-25.

Fairbrother, N., & Rachman, S. (2004). Feelings of mental pollution subsequent to sexual assault.

Behaviour Research and Therapy, 42(2), 173-189.

Fairbrother, N., Newth, S. J., & Rachman, S. (2005). Mental pollution: Feelings of dirtiness without

physical contact. Behaviour Research and Therapy, 43(1), 121-130.

Fergus, T. A., Valentiner, D. P., McGrath, P. B., & Jencius, S. (2010). Shame and guilt proneness:

relationships with anxiety disorder symptoms in a clinical sample. Journal of Anxiety

Disorders, 24, 811-815.

Green, J., & Teachman, B. (2013) Predictive validity of explicit and implicit overestimation in

contamination fear. Journal of Obsessive-Compulsive and Related Disorders, 2, 1-8.

Herba, J. K., & Rachman, S. (2007). Vulnerability to mental contamination. Behaviour Research

and Therapy, 45(11), 2804-2812.

Ille, R., Schoggl, H., Kapfhammer, H., Arendasy, M., Sommer, M., &Schienle, A. (2014). Self-

disgust in mental disordes- symptom related or disorder-specific? Comprehensive Psychiatry,

55, 938-943,

239

Ishikawa, R., Kobori, O., Komuro, K. & Shimizu, E. (2014). Comparing the roles of washing and

non-washing behaviour in the reduction of mental contamination. Journal of Obsessive-

Compulsive and Related Disorders, 3(1), 60-64.

Jung, K., & Steil, R. (2012). The feeling of Being Contaminated in Adult Survivors of Childhood

Sexual Abuse and its treatment via a two-session program of cognitive restructuring and

imagery modification. A case study. Behaviour Modification, 36(1): 67-86.

Kmet, L., Lee, R., & Cook, L. (2004). Standard quality assessment criteria for evaluating primary

research papers from a variety of fields. HTA Initiative, 02 (13)

Lee, M., Shafran, R., Burgess, C., Carpenter, J., Millard, E., & Thorpe, S. (2013). The induction of

mental and contact contamination. Clinical Psychologist, 17(1), 9-16.

Olatunji, B. O., Elwood, L. S., Williams, N. L., & Lohr, J. M. (2008). Mental pollution and PTSD

symptoms in victims of sexual assault: A preliminary examination of the mediating role of

trauma-related cognitions. Journal of Cognitive Psychotherapy, 22(1), 37-47.

Ponniah, K., Magiati, I., Hollon, S. (2013). An update on the efficacy of psychological treatments for

obsessive-compulsive disorder in adults. Journal of Obsessive-Compulsive and Related

Disorders, 2 (2), 207-218.

Rachman, S. (1994). Pollution of the mind. Behaviour Research and Therapy, 32(3), 311-314.

Rachman, S. (2004). Fear of Contamination. Behaviour Research and Therapy, 42, 1227-1255.

240

Rachman, S. (2006). Fear of contamination: Assessment and treatment. Oxford University Press.

Rachman, S., Elliott, C. M., & Zysk, E. (2012). Mental contamination: The perpetrator effect.

Journal of Behavior Therapy and Experimental Psychiatry, 43(1), 587-593.

Radomsky, A. S., & Elliott, C. M. (2009). Analyses of mental contamination: Part II, individual

differences. Behaviour Research and Therapy, 47(12), 1004-1011.

Radomsky, A. S., Rachman, S., Shafran, R., Coughtrey, A. E., & Barber, K. C. (2014). The nature

and assessment of mental contamination: A psychometric analysis. Journal of Obsessive-

Compulsive and Related Disorders, 13 (2), 181-187.

Sangmoon, K., Ryan, T., & Randall, S. (2011). Shame, guilt and depressive symptoms; a meta-

analytic review. Psychological Bulletin, 137 (1), 68-96.

Salkovskis, P. M., Wroe, A. L., Gledhill, A., Morrison, N., Forrester, E., Richards, C., Reynold, M.,

& Thorpe, S. (2000). Responsibility attitudes and interpretations are characteristic of

obsessive compulsive disorder. Behaviour Research and Therapy, 38, 347-372.

Shafran, R., Thordarson, M. & Rachman, S. (1996). Thought-Action Fusion in obsessive compulsive

disorder. Journal of Anxiety Disorders, 10, 379-391.

Tolin, D., Worhunsky, P., & Maltby, N. (2004). Sympathetic magic in contamination related OCD.

Journal of Behaviour Therapy and Experimental Psychiatry, 35, 193-206.

241

Volz, C., & Heyman, I. (2007). Case series: Transformation obsession in young people with

obsessive-compulsive disorder (OCD). Journal of the American Academy of Child &

Adolescent Psychiatry, 46(6), 766-772.

Warnock-Parkes, E., Salkovskis, P., & Rachman, J. (2012). When the Problem is Beneath the

Surface in OCD: The Cognitive Treatment of a Case of Pure Mental Contamination.

Behavioural and Cognitive Psychotherapy, 40(4), 383-399.

242

Overview of Clinical experience:

Year one:

My adult placement was split between a community mental health team and an IAPT service.

The community team saw adults with severe and enduring mental health difficulties and I

provided one on one therapy for people with a range of difficulties including: bi-polar,

personality disorders psychosis, depression and anxiety. One to one therapy generally

followed the CBT model but also included some ACT work. I also conducted Psychology

assessments and neuropsychology assessments.

On this placement I also worked on two inpatient wards in a family therapy clinic that was

run on the wards. This used systemic therapy in a consultation model to work with inpatients

and their families. I saw people with a range of difficulties and developed my systemic

therapy skills. I presented some teaching to the nursing staff on the ward on basic CBT and

behavioural activation.

On the IAPT part of my placement I worked with adults with predominantly difficulties with

depression and anxiety using CBT.

Year two:

Older adult: For my older adult placement I worked in a memory assessment service which

aimed for early diagnosis of dementia and to provide psychological support to people with

dementia and their families.

Extensive and detailed experience completing neuropsychological assessment within the memory clinic, assessing neurodegenerative conditions. Profiles have included frontal lobe dementia, vascular dementia and early stages of Alzheimer’s disease.

I saw carers of people with dementia individually for CBT based work around stress, anxiety

and uncertainty. I also co-facilitated a reflective practice group for the dementia advisors in

the team.

I also co-facilitated a number of groups for people with dementia and their carers. The

memory management group was a psycho-education and CBT group for carers and people

243

with dementia. The Living well with dementia group was an adjustment group for people

with dementia. I helped to develop the Carer’s wellbeing group which was based on

elements of psychoeducation, CBT, ACT and compassion focused work. I also conducted a

small research project to evaluate this group.

Learning disabilities: I worked in a learning disability health team working with people witha LD, carers, staff teams, care homes and other services. The work included:

Assessing and providing adapted psychological interventions for adults with a LD Neuropsychological diagnostic assessments of clients with a suspected LD Consultation work with care homes and day centres who were working with clients

with behaviour that challenged. Training to care home staff on LD and bereavement.

Year 3:

Child placement:

I worked in a Child and Adolescent Mental Health team and the work involved:

o Assessing and providing interventions for families and children with anxiety, depression, developmental disorders (Autism Spectrum Disorders), self-harm, separation anxiety, OCD and PTSD.

o Consulting with schools.o Providing training to schools on self-harm in adolescents. o Neuropsychological assessments: WAIT and WISC.o CBT approaches combined with narrative principles

Specialist placement:

I worked in a Child and Young Persons Community Eating Disorder Team. The work

involved assessing and providing interventions for children and adolescents with Anorexia

and Bulimia. I used a CBT-E approach to work individually with adolescents to help them

overcome bulimia. I worked with other professionals using the Family Based Therapy

Approach to help young people with Anorexia. Also worked with young people who had co-

morbidity’s including depression and trichotillomania using a CBT approach.

244

Table of assessments:

Year

one

assessments:

Year two assessments:

ASSESSMENT TITLE

Professional Issues Essay

In relation to yourself critically explore the statement that clinical psychologists should ‘move away from psychiatric diagnoses….which have significant conceptual and empirical limitations and develop alternative approaches which recognize the centrality of the complex range of life

245

ASSESSMENT TITLE

WAIS-III Short report of WAIS-III data and practice administration

Practice Case Report Cognitive Behavioural Model for a young male suffering from generalized anxiety disorder

Problem Based Learning – Reflective Account

PBL Reflective Account

Major Research Project Literature Review

Reviewing the current state of knowledge on the relatively new concept of Mental Contamination

Adult – Case Report 1 A cognitive behavioural model for a young man presenting with symptoms of worry and anxiety.

Adult – Case Report 2 A cognitive behavioural model for a woman presenting with symptoms of depression.

Major Research Project Proposal

Investigating factors associated with mental contamination

experiences..’ (BPS time for a paradigm shift).

Problem Based Learning – Reflective Account

PBL Reflective Account year two

People with Learning Disabilities/Child and Family/Older People – Case Report

Neuropsychology assessment of a woman in her late sixties with suspected dementia.

Personal and Professional Learning Discussion Groups – Process Account

Personal and Professional Learning Discussion Group Process Account

People with Learning Disabilities/Child and Family/Older People – Oral Presentation of Clinical Activity

Working with challenging behavior in a person with a learning disability and involving the wider system.

Year III Assessments ASSESSMENT TITLE

Service-Related Project Evaluation of a carers, of people with dementia, wellbeing group in a memory assessment service

Major Research Project Empirical Paper

Comparing the effectiveness of brief writing tasks in reducing feelings of mental contamination.

Personal and Professional Learning – Final Reflective Account

On becoming a clinical psychologist: A retrospective, developmental, reflective account of the experience of training

Child and Family/People with Learning Disabilities/Older People/Specialist – Case Report

Family Based Therapy for an adolescent girl with Anorexia.

246