Behaviour Research and Therapy 38 (2000) 439±455

www.elsevier.com/locate/brat

Appraisal and control of sexual and non-sexual intrusive

thoughts in university students
David A. Clark a,*, Christine Purdon b, E. Sandra Byers a
a
Department of Psychology, University of New Brunswick, Fredericton, NB Canada E3B 6E4
b
Department of Psychology, University of Waterloo, Waterloo, Ontario, Canada N2L 3G1

Abstract

This study examined di€erences in the appraisal and thought control strategies associated with the
perceived control of unwanted sexual and non-sexual intrusive thoughts. Eleven appraisal dimensions,
subjective physiological arousal and 10 thought control strategies were measured in 171 university
students who were administered the Revised Obsessive Intrusions Inventory-Sex Version, a self-report
measure of unwanted intrusive thoughts. Thought-action fusion (TAF) likelihood was a signi®cant
unique predictor of the perceived controllability of respondents' most upsetting sexual and non-sexual
intrusive thought. Moreover greater subjective physiological arousal was a signi®cant predictor of
reduced control over sexual intrusions, whereas worry that one might act on an intrusive thought and
greater e€ort to control the intrusion were signi®cant unique predictors of the control of non-sexual
intrusive thoughts. Various thought control strategies were more often used in response to non-sexual
than sexual cognitions. The results are discussed in terms of the di€erential role of various appraisal
processes in the control of unwanted sexual and non-sexual thoughts. 7 2000 Elsevier Science Ltd. All
rights reserved.

Keywords: Obsessive±compulsive disorder; Intrusive thoughts; Cognitive theory; Thought control; Thought-action
fusion

* Corresponding author. Fax: +1-506-453-4505.

E-mail address: clark@unb.ca (David A. Clark).

0005-7967/00/$ - see front matter 7 2000 Elsevier Science Ltd. All rights reserved.
PII: S 0 0 0 5 - 7 9 6 7 ( 9 9 ) 0 0 0 4 7 - 9
440 D.A. Clark et al. / Behaviour Research and Therapy 38 (2000) 439±455

1. Introduction

Unwanted intrusive thoughts were ®rst de®ned by Rachman (1981) as repetitive thoughts,
images or impulses that are unacceptable, uncontrollable, have an internal origin and interrupt
ongoing activity. Although they occur naturally in the general population (Rachman & de
Silva, 1978; Parkinson & Rachman, 1981; Salkovskis & Harrison, 1984; Clark & de Silva,
1985; Freeston, Ladoucer, Thibodeau & Gagnon, 1991; Freeston & Ladouceur, 1993; Purdon
& Clark, 1993), cognitive intrusions with ego-dystonic content, such as thoughts of situations
or actions that involve personally revolting acts of sex, aggression, harm, dirt, disease and
contamination, have been of particular interest to researchers of obsessive±compulsive
disorders (OCD) because of their similarity to the content of clinical obsessions. Two recent
cognitive theories of obsessions, one proposed by Salkovskis (1985, 1989, 1996) and Salkovskis,
Richards and Forrester (1995) and the other by Rachman (1997, 1998), assert that clinical
obsessions have their origin in the normal unwanted intrusive thoughts found in the general
population. What distinguishes the unwanted cognitive intrusions of the non-clinical person
from the clinical obsessions of the individual su€ering from OCD is the meaning or appraisal
associated with the intrusion. However the cognitive theories of Salkovskis, Rachman and
others di€er in which appraisal processes they consider central to the pathogenesis of
obsessions (Clark & Purdon, 1993; Salkovskis, 1996; Obsessive Compulsive Cognitions
Working Group, 1997; Rachman, 1993, 1997).
Salkovskis (1996), for example, considers two processes to be critical in the escalation of
normal intrusions into clinical obsessions; (a) the interpretation of intrusive thoughts as an
indication of responsibility for harm or its prevention toward self or others and (b) the
subsequent e€ort to neutralize this in¯ated sense of responsibility through direct thought
control strategies. Rachman (1997, 1998) views the catastrophic misinterpretation of the
personal signi®cance of unwanted intrusive thoughts as the appraisal process most likely to
cause an increase in the range and threatening nature of the intrusion. A speci®c process that
can in¯ate the signi®cance of an intrusion, referred to as thought-action fusion (TAF), is a
cognitive bias involving an interpretation of the unwanted thought as morally or realistically
equivalent to its behavioral manifestation (Rachman, 1993; Shafran, Thordarson & Rachman,
1996). Previously we have argued that appraisals about the importance of controlling an
unwanted intrusion and the perceived consequences of failing to exercise complete thought
control are critical in the development of persistent intrusive thoughts and obsessions (Clark,
1989; Clark & Purdon, 1993). The Obsessive Compulsive Cognitions Working Group (1997)
have described a number of other beliefs and appraisals that may be important in
distinguishing obsessions including overimportance, overestimation of threat, intolerance of
uncertainty, perfectionism and ®xity of beliefs.
A number of empirical studies have examined the relationship between the appraisal of
unwanted intrusive thoughts and their frequency. One of the most consistent ®ndings to
emerge from this literature is that the less control individuals believe they have over an
intrusive thought, the more frequent it intrudes into consciousness (Parkinson & Rachman,
1981; Salkovskis & Harrison, 1984; Clark & de Silva, 1985; Niler & Beck, 1989; Purdon &
Clark, 1994a). Other appraisal dimensions found to be associated with an increased frequency
and/or uncontrollability of unwanted intrusions include emotional intensity, unpleasantness,
 D.A. Clark et al. / Behaviour Research and Therapy 38 (2000) 439±455 441

disapproval, guilt, unacceptability, belief that the intrusive thought might be acted upon in real
life (TAF-Likelihood), worry that the intrusive thought might be acted upon (TAF-Worry),
responsibility and belief that the intrusion may indicate something meaningful about one's
personality (Parkinson & Rachman, 1981; Salkovskis & Harrison, 1984; Clark & de Silva,
1985; Niler & Beck, 1989; Freeston, Ladouceur, Thibodeau & Gagnon, 1992; Purdon & Clark,
1994a,b; Clark & Claybourn, 1997). The importance of appraisals in the frequency and
controllability of intrusion has also been replicated in samples of OCD patients (see Rachman
& de Silva, 1978; Calamari & Janeck, 1996).
Salkovskis (1985, 1989, 1996) also implicated thought control strategies in the escalation of
normal intrusions into clinical obsessions. Attempts to neutralize the sense of in¯ated
responsibility or the distress associated with the intrusion (Rachman, 1997, 1998) may
inadvertently increase the salience and ultimately the frequency of the intrusion. A number of
empirical studies have shown that certain thought control strategies, such as attentive focus on
the intrusion, escape/avoidance, reassurance seeking and thought stopping, may be maladaptive
approaches that lead to a signi®cant increase in the frequency and emotional distress associated
with the intrusion (Freeston et al., 1991, 1992; Freeston & Ladouceur, 1993; Purdon & Clark,
1994b). However Calamari & Janeck (1996) found no signi®cant di€erences between OCD
patients and non-clinical controls in whether they used `do nothing', `attentive thinking' or
`escape/avoidance' in response to their unwanted intrusive thought or obsession. Overall, then,
it is not clear what role thought control strategies play in the persistence and intensity of
unwanted intrusive thoughts relative to the appraisal dimensions.
One issue that has not been addressed in the literature on unwanted intrusive thoughts is
whether the content of the thought might in¯uence how it is appraised and the type of control
response used to suppress the cognition. To date cognitive-clinical research on intrusive
thoughts and obsessions have assumed non-speci®city in the appraisal and control of intrusive
thoughts regardless of their thematic content. In other words, it is assumed that the same
appraisal processes and control strategies operate across diverse types of intrusive thoughts.
However this assumption may not be tenable, especially when comparing ego-dystonic and
ego-syntonic intrusive thoughts (Clark & Purdon, 1995). Recently Clark & Claybourn (1997)
found that appraisals of personal signi®cance uniquely predicted ego-dystonic intrusions,
whereas a focus on the consequences of negative events and the intensity of one's distress were
more characteristic of ego-syntonic intrusive thoughts.
We also question whether the appraisal and control of intrusive thoughts dealing with
unacceptable sexual themes may di€er in important ways from the interpretation and control
of non-sexual obsessive content dealing with aggression, dirt, disease and contamination. In a
principal component analysis of the 13 symptom categories of the Yale-Brown Obsessive
Compulsive Scale Symptom Checklist, Leckman et al. (1997) found that the aggressive, sexual
and religious obsessions loaded together on the same factor and were distinct from symmetry,
cleanliness and hoarding obsessions and compulsions. This suggests that distinct symptom
dimensions may exist within OCD that could have di€erent implications for the pathogenesis
of the disorder (Leckman, 1998). At the very least these results indicate that sexual obsessions
should be distinguished from dirt, contamination and disease obsessions.
Not only may sexual obsessions and intrusive thoughts be distinct from other types of
obsessions, but it is possible that individuals respond to this thought content di€erently than
442 D.A. Clark et al. / Behaviour Research and Therapy 38 (2000) 439±455

dirt, contamination, disease and aggression intrusions. For example, the symptoms of anxiety
and of sexual arousal are very similar. If the emotional arousal associated with sexual intrusive
thoughts is misinterpreted as a sign of sexual arousal rather than anxiety, this would reinforce
individuals' concern that they might act on the thought, which might promote greater e€ort to
suppress a sexual as opposed to a non-sexual intrusive thought.
Second individuals have a longer history of having sexual fantasies that they then act upon
than they do with intentional fantasies of dirt, contamination or aggression. If unwanted
sexual intrusive thoughts are interpreted as closely linked to sexual fantasies, then individuals
may have greater concerns that they will act on their sexual rather than non-sexual intrusions.
Certainly the boundary between ego-syntonic and ego-dystonic sexual thoughts may be less
distinct than with non-sexual obsessive-like intrusions. Gil (1990) found that 50% of a
conservative Christian young adult sample rated their sexual fantasies as `morally ¯awed and
unacceptable'. Child sexual abuse survivors report more sexual fantasies involving being forced
to engage in sexual activity (Briere, Smiljanich & Henschel, 1994). In an initial paper based on
a portion of the data set not included in the present article, Byers et al. (1998) found that men
reported more unwanted, deviant sexual intrusive thoughts with aggressive themes and were
more sexually aroused by these intrusions than were women. A more positive disposition
towards sexuality and more frequent sexual daydreaming were associated with an increased
frequency of unwanted sexual intrusive thoughts, although anxiety over sexuality was related
to a more negative appraisal (i.e. guilt, disapproval, avoidance) of sexual intrusive thoughts.
These ®ndings, then, suggest that not all intrusions are the same and that unwanted intrusive
thoughts of a sexual nature may be experienced di€erently from non-sexually related obsessive
intrusions.
The present study compared appraisal processes and control strategies used in response to
individuals' most upsetting sexual and non-sexual intrusive thought. Three questions guided
this research. Given the gender di€erences reported in our ®rst analysis of the frequency of
di€erent types of unwanted sexual intrusive thoughts (Byers et al., 1998), do men and women
di€er in the appraisal and control of their most upsetting sexual and non-sexual intrusive
thought? Are certain appraisal processes and control strategies uniquely characteristic of sexual
versus non-sexual intrusive thoughts? Does a personality predisposition to respond positively to
sexual cues in¯uence how one appraises their sexual intrusive thoughts? Rachman (1998) has
argued that the thought content that an individual ®nds most upsetting will be determined in
part by what is most important in the individual's value system. It is unknown whether
individuals who generally respond positively to sexual cues or those who have a more negative
evaluative and a€ective response to sexual cues will be more or less upset by the intrusion of
unacceptable, unwanted sexual thoughts and images.

2. Method

2.1. Participants and procedure

A sample of predominantly ®rst year undergraduates (N = 171) consisting of 113 women

and 58 men with a mean age of 19.5 years (S.D.=3.2) were administered a battery of
 D.A. Clark et al. / Behaviour Research and Therapy 38 (2000) 439±455 443

questionnaires that included measures of intrusive thoughts, sexuality, obsessionality and mood
state. The questionnaires were administered in counterbalanced order to the entire group in
one session after participants had signed an informed consent form. All participants were
enrolled in an introductory psychology course at a Canadian university and received bonus
points towards their course grade. Most of the participants in this study were sexually active
with 80% of the women and 76% of the men reporting that they had experienced sexual
intercourse. For those who had engaged in intercourse, the average age of ®rst intercourse for
both men and women was 16 years (S.D.=1.9) and the average number of intercourse partners
was 3.4 for men and 3.0 for women. Because of the research questions posed, this paper
focuses primarily on the appraisal and control ratings of the Revised Obsessive Intrusions
Inventory-Sex Version. Findings from the other measures in the questionnaire packet were
reported in Byers et al. (1998). The ®nal sample on which the following analyses are based
included only respondents who completed appraisal ratings on their most upsetting non-sexual
and/or sexual intrusive thought (n for men=55; n for women=100). However the actual N will
vary across analyses because of listwise deletion of missing values.

2.2. Measures

2.2.1. Revised Obsessive Intrusions Inventory-Sex Version (ROII-v2)

This is a 66-item self-report questionnaire constructed to assess the frequency, appraisal and
perceived control of non-sexual and sexual unwanted intrusive thoughts, images or impulses.
The current version of the instrument is based on the Revised Obsessional Intrusions Inventory
(ROII) developed by Purdon and Clark (1993, 1994a). The ROII-v2 consists of two parts. In
the ®rst part of the measure, participants are provided with a list of 20 non-sexual intrusive
thought statements dealing with themes of harm, aggression, accidents, dirt, disease and
contamination. The non-sexual items were taken from previous versions of the ROII.
Participants rate the frequency of each item on a scale ranging from I never had this thought (0)
to I have this thought frequently during the day (6). A ROII non-sexual total score is calculated
as the sum of respondents' frequency ratings across the 20 intrusive thought statements. The
ROII total score based on the original version of the instrument has acceptable internal
consistency as well as high concurrent and discriminant validity, emerging as a unique
predictor of obsessional symptomatology but not depressive or anxious symptoms (Purdon &
Clark, 1993, 1994a, 1994b).
After completing the checklist, participants were asked to select their most upsetting non-
sexual intrusive thought from the items they had experienced at least `rarely' (i.e. rating of 1 or
more). The most frequently selected intrusive thoughts were ``leaving the house without doing
something important to prevent burglary or accident'', ``when using a sharp object that I will
slit my wrist or throat or otherwise harm myself'' or ``saying something rude or insulting to
others''.
Participants then rated their evaluation of this thought on 11 appraisal dimensions using a
®ve point Likert scale. The appraisal ratings were: (a) thought-action fusion (TAF)-morality
(``how strong is your feeling that having this thought/image/impulse is just as bad as acting
upon it?''), (b) TAF-likelihood (``how likely it is that the thought/image/impulse itself will come
true in real life?''), (c) TAF-worry (``when this thought/image/impulse comes into your head,
444 D.A. Clark et al. / Behaviour Research and Therapy 38 (2000) 439±455

how much do you worry that you might act on it or that it might otherwise happen in real
life?''), (d) responsibility (``how responsible do you feel for letting this thought/image/impulse
come into your head?''), (e) threat (``to what extent does having this thought/image/impulse
signify harm/danger to yourself or others?''), (f) avoidance (``how strong is your desire to avoid
situations that might `trigger' this thought/image/impulse?''), (g) control (``how dicult is it for
you to get rid of this thought/image/impulse once it comes into your head?''), (h) resistance
(``how hard do you try to get rid of this thought/image/impulse?''), (i) importance (``how
important it is that you rid your mind of this thought/image/impulse?''), (j) disapproval (``how
much do you disapprove of this thought/image/impulse?'') and guilt (``how guilty do you feel
for having let this thought/image/impulse come into your head?''). An additional item was
constructed for the current version of the ROII to assess subjective arousal associated with
respondents' most upsetting intrusion. Individuals were asked to rate the degree to which they
experienced seven possible physiological reactions; increased heart rate, breathing, sweaty
palms, ¯ushing skin, upset stomach, tingling or other sensations in the genitals and other signs
of sexual arousal (e.g. erection). Ratings on these seven variables were summed to provide a
perceived arousal score. In the present study the internal consistency for both the non-sexual
(a=82) and the sexual (a=0.85) perceived arousal scales were within the acceptable range. Of
the total sample of 171 students, 131 (77%) indicated which non-sexual thought was their most
upsetting intrusion.
As a ®nal feature of the assessment of non-sexual intrusive thoughts, participants were
asked to rate how often they utilized 10 di€erent thought control strategies in order to
rid their mind of the upsetting intrusive thought. The thought control strategies are
identical to the ROII control item statements presented in Purdon and Clark (1994b).
In the second part of the ROII-v2, participants rated how frequently they experienced 20
sexual intrusive thoughts, images or impulses on the same 7-point frequency rating scale. The
sexual intrusive items re¯ected themes of unwanted, possibly even disgusting, sexual behavior,
thoughts and impulses. The items were drawn from sexually oriented items from the ROII,
interviews conducted with undergraduate students, studies on sexual fantasy, and item
evaluations by a group of sex researchers (for more details on item development see Byers et
al., 1998). A ROII sexual total score was calculated by summing over the frequency ratings for
the 20 items. Byers et al., 1998 reported Cronbach's alpha of 0.92. for the ROII sexual total
score. In addition, they failed to ®nd any signi®cant di€erences between men and women on
the ROII-v2 non-sexual or sexual total scores.
As in the ®rst part of the ROII-v2, participants selected their most upsetting sexual intrusive
thought from the items they had experienced at least `rarely' (i.e. rating of 1 or more). The
most frequently selected sexual intrusive thoughts were `of being sexually victimized', `having
sex in public' or ` engaging in a sexual act with someone who is unacceptable to me because
they have authority over me'. In addition participants were asked to rate their evaluation,
perceived physiological arousal and the control strategies associated with their most upsetting
sexual intrusive thought using identical rating scales to those described for the non-sexual
intrusive thought. One hundred and nine individuals (64%) completed the appraisal ratings on
the second part of the ROII-v2 by recording their most upsetting sexual intrusive thought.
 D.A. Clark et al. / Behaviour Research and Therapy 38 (2000) 439±455 445

2.2.2. Sexual Opinion Survey (SOS; Fisher, Byrne, White & Kelley, 1988)
The ®ve item short form of the SOS was administered to assess participants' tendency to
respond to sexual cues in a positive or negative evaluative and a€ective manner. Fisher et al.
(1988) reported that the ®ve item SOS correlated 0.71 and 0.82 for women and men,
respectively, with the original 21 item SOS. However, in the present study the Cronbach a was
only 0.52 indicating that the short form of the SOS may not be a stable measure of positive or
negative disposition to sexual cues.

2.2.3. Obsessive Thoughts Checklist (OTC; Bouvard, Mollard, Cottraux & Guerin, 1989)
This is a 28 item instrument designed to assess the frequency of obsessional thinking.
Bouvard et al. (1989) reported good reliability, convergent and discriminant validity for the
French version of the OTC. In the current study the English version of the OTC was used and
it had an a=0.94.

3. Results

3.1. Descriptive statistics

Table 1 presents means and standard deviations on the 11 appraisal ratings and the
perceived physiological arousal total score of participants' most distressing non-sexual and
sexual intrusive thought, as well as the SOS and OTC total scores. As can be seen there was a
considerable degree of consistency in the range of participants' ratings across the various
appraisal dimensions. Individuals rated their most distressing sexual and non-sexual intrusive
thought within the `not at all' to `somewhat' range. Thus we can conclude that most students
were not very upset by their intrusive thoughts, nor did they believe that the thoughts were
particularly signi®cant or would result in dire consequences. As a result they did not expend
much e€ort in trying to control the thoughts.
None of the paired t-tests comparing the appraisals of sexual and non-sexual intrusive
thoughts was statistically signi®cant, with or without Bonferroni adjustment. Independent t-
tests with Bonferroni adjustment set at p < 0.001 failed to reveal any signi®cant di€erences
between men and women in their appraisal of non-sexual or sexual intrusive thoughts.
Independent t-tests indicated that men scored signi®cantly higher on the SOS (t(153)=4.40, p
< 0.001) than women, although there were no signi®cant gender di€erences on the OTC.
Table 2 presents the zero-order correlations on the total sample amongst the 11 appraisal
ratings and the perceived physiological arousal total score (labelled arousal) for non-sexual and
sexual intrusive thoughts as well as the SOS and OTC. Most of the correlations were within
the low to moderate range indicating that each of the ratings assessed fairly distinct appraisal
dimensions. Interestingly the three ratings that dealt with various aspects of thought-action
fusion were not highly intercorrelated suggesting that thought-action fusion may be a complex,
fairly heterogeneous construct. Responsibility was most highly correlated with TAF-moral and
avoidance of situations that would trigger non-sexual intrusions, whereas responsibility
exhibited its strongest correlation with TAF-moral and the threat associated with sexual
intrusive thoughts. Control of non-sexual intrusions was most strongly correlated with TAF-
446 D.A. Clark et al. / Behaviour Research and Therapy 38 (2000) 439±455

worry and resistance, whereas control of sexual intrusive thoughts was most highly correlated
with TAF-likelihood. Appraisals of both non-sexual and sexual intrusive thoughts had higher
correlations with the OTC total score than the SOS total score, suggesting that appraisals of
both types of intrusions have more relevance to obsessionality than they do with attitudes and
values toward sexuality.

Table 1
Means and standard deviations on appraisal ratings of non-sexual and sexual intrusive thoughts by gender. SOS=-
Sexual Opinion Survey; OTC=Obsessive Thoughts Checklist. Total sample size for the SOS and OTC was 158 and
155, respectively

ROII appraisal ratings Men (n = 42) Women (n = 75) Total (n = 117)

M S.D. M S.D. M S.D.

Non-sexual intrusions
TAF-moral 1.57 0.99 1.81 1.01 1.73 1.01
TAF-likelihood 1.60 0.80 1.80 0.97 1.73 0.92
TAF-worry 1.81 0.86 2.08 1.09 1.98 1.02
Responsibility 2.26 1.01 2.35 0.97 2.32 0.98
Threat 1.67 0.95 1.95 1.16 1.85 1.10
Avoidance 1.88 1.06 2.44 1.24 2.24 1.21
Control 1.86 0.98 1.97 1.07 1.93 1.03
Resistance 2.26 1.04 2.79 1.15 2.60 1.14
Importance 2.19 0.83 2.68 1.19 2.50 1.10
Disapproval 2.88 1.23 3.23 1.38 3.10 1.34
Guilt 2.05 1.06 2.39 1.00 2.26 1.03
Perceived arousal 8.10 5.11 9.23 5.44 8.82 5.33

Sexual intrusions
TAF-moral 1.57 0.83 1.83 1.07 1.74 0.99
TAF-likelihood 1.81 1.31 1.56 0.89 1.65 1.06
TAF-worry 1.67 1.03 1.89 1.12 1.81 1.09
Responsibility 2.26 1.13 2.21 1.19 2.23 1.16
Threat 1.57 1.06 1.61 1.09 1.60 1.08
Avoidance 2.00 1.21 2.20 1.22 2.13 1.21
Control 1.90 1.14 1.88 1.06 1.89 1.09
Resistance 2.40 1.08 2.55 1.27 2.50 1.20
Importance 2.17 1.10 2.55 1.33 2.41 1.26
Disapproval 2.50 1.11 3.23 1.49 2.97 1.41
Guilt 2.05 0.99 2.47 1.20 2.32 1.14
Perceived arousal 8.67 6.07 8.80 6.13 8.75 6.08

Other measures
SOS Total Score 17.04 5.39 13.14 5.22 14.56 5.56
OTC Total Score 24.15 18.22 25.62 17.26 25.29 17.68
 D.A. Clark et al. / Behaviour Research and Therapy 38 (2000) 439±455 447

3.2. Appraisal of non-sexual and sexual intrusive thoughts

The zero-order correlations suggest that some di€erences may be apparent in the appraisals
associated with the perceived control of non-sexual and sexual intrusive thoughts in university
students. To explore this further, separate hierarchical regression analyses were conducted with
the perceived control (i.e. ``How dicult is it for you to get rid of this thought/image/impulse
once it comes into your head?'') of respondents' most upsetting non-sexual and sexual intrusive
thought as dependent variables. Gender was entered alone on the ®rst step, the SOS and OTC
total scores on the second step and the 10 appraisal ratings and physiological arousal total
score entered simultaneously on the third step. Table 3 presents a summary of the regression
statistics.
As can be seen, gender was not a signi®cant predictor of the perceived control of non-sexual
or sexual intrusive thoughts. However on the second step the SOS and OTC did account for a
signi®cant amount of variance in the perceived control of sexual, but especially non-sexual,
intrusive thoughts. This was due almost entirely to the OTC, indicating that the perceived
control of sexual as well as non-sexual unwanted intrusive thoughts is related more to a
predisposition for obsessionality than a predisposition toward sexuality.
On the third step, the 10 appraisal ratings accounted for a signi®cant amount of additional
variance in the prediction of the perceived control of non-sexual and sexual intrusive thoughts.
Di€erences did emerge at the level of individual predictors. TAF-Likelihood emerged as an
important unique predictor of the perceived control of both non-sexual but especially sexual
intrusive thoughts. TAF-worry and resistance were also signi®cant predictors of the control of

Table 2
Correlations amongst the ROII appraisal ratings of non-sexual and sexual intrusive thoughts. Coecients above the
diagonal are appraisals of non-sexual intrusive thought whereas coecients in bold below the diagonal are for sexual
intrusive thought appraisals. For non-sexual intrusions listwise deletion resulted in n = 145 and signi®cance levels of
r > 0.16, p < 0.05; r > 0.24, p < 0.01. Listwise deletion for sexual intrusive thought appraisal resulted in n = 125
and signi®cance levels of r > 0.17, p < 0.05; r > 0.25, p < 0.01. SOS=Sexual Opinion Survey; OTC=Obsessive
Thoughts Checklist

Appraisal ratings 1 2 3 4 5 6 7 8 9 10 11 12 13 14

(1) TAF-moral 0.06 0.15 0.49 0.10 0.43 0.08 0.23 0.33 0.19 0.35 0.25 ÿ0.13 0.33
(2) TAF-likely 0.03 0.32 0.19 0.17 0.03 0.36 0.04 0.00 ÿ0.16 ÿ0.07 0.20 0.04 0.18
(3) TAF-worry 0.38 0.26 0.21 0.33 0.24 0.51 0.43 0.23 0.10 0.25 0.34 ÿ0.02 0.42
(4) Responsibility 0.44 0.07 0.30 0.19 0.39 0.26 0.24 0.31 0.25 0.27 0.32 ÿ0.01 0.24
(5) Threat 0.44 0.29 0.43 0.45 0.34 0.28 0.30 0.31 0.21 0.05 0.31 0.04 0.18
(6) Avoidance 0.29 0.06 0.38 0.23 0.33 0.16 0.29 0.52 0.26 0.28 0.33 ÿ0.10 0.24
(7) Control 0.30 0.48 0.33 0.15 0.34 0.09 0.41 0.16 0.06 0.13 0.31 0.04 0.45
(8) Resistance 0.42 0.03 0.49 0.26 0.30 0.50 0.24 0.56 0.37 0.39 0.24 ÿ0.15 0.22
(9) Importance 0.44 0.09 0.53 0.36 0.38 0.54 0.19 0.70 0.53 0.40 0.20 ÿ0.08 0.09
(10) Disapproval 0.46 ÿ0.03 0.42 0.37 0.40 0.53 0.20 0.56 0.75 0.60 0.12 ÿ0.07 0.03
(11) Guilt 0.56 ÿ0.13 0.38 0.22 0.36 0.37 0.22 0.54 0.46 0.57 0.13 ÿ0.13 0.15
(12) Arousal 0.28 0.28 0.34 0.26 0.37 0.20 0.36 0.16 0.29 0.20 0.22 0.13 0.47
(13) SOS total ÿ0.18 ÿ0.10 ÿ0.06 ÿ0.02 0.03 ÿ0.12 ÿ0.14 ÿ0.12 ÿ0.10 ÿ0.15 ÿ0.19 0.07 ÿ0.02
(14) OTC total 0.39 0.20 0.26 0.17 0.31 0.12 0.26 0.19 0.20 0.12 0.15 0.47 ÿ0.08
448 D.A. Clark et al. / Behaviour Research and Therapy 38 (2000) 439±455

non-sexual intrusions, whereas perceived physiological arousal was the only other signi®cant
predictor of perceived control of sexual intrusions. These results indicate that although
thought-action fusion bias may be important in the perceived control of both types of intrusive
thoughts, resistance and worry also play a prominent role in the control appraisals for non-
sexual intrusions whereas perceived physiological arousal is more important to the perceived
control of sexual intrusive thoughts.

Table 3
Hierarchical regression analysis predicting perceived control of respondent's most upsetting non-sexual and sexual
intrusive thought from ROII appraisal ratings. For non-sexual intrusions n = 145; for sexual intrusions n = 125. Sig-
ni®cance levels p < 0.05, p < 0.01, p < 0.001

Variables R 2 change F change Beta sr2 t-value

Perceived control of non-sexual intrusions

Step 1: gender 0.00 0.21
Step 2: 0.20 17.91
SOS total 0.06 0.00 0.81
OTC total 0.45 0.20 5.94
Step 3: 0.45 5.34
TAF-moral ÿ0.15 0.01 1.82
TAF-likelihood 0.22 0.04 2.95
TAF-worry 0.20 0.02 2.36
Responsibility 0.13 0.01 1.62
Threat 0.05 0.00 0.69
Avoidance 0.01 0.00 0.95
Resistance 0.30 0.05 3.48
Importance ÿ0.05 0.00 0.52
Disapproval ÿ0.01 0.00 0.92
Guilt ÿ0.02 0.00 0.27
Perceived arousal 0.01 0.00 0.91

Perceived control of sexual intrusions

Step 1: gender 0.00 0.36
Step 2: 0.09 6.35
SOS total ÿ0.16 0.02 1.77
OTC total 0.26 0.07 2.98
Step 3: 0.31 5.30
TAF-moral 0.15 0.01 1.41
TAF-likelihood 0.41 0.12 4.76
TAF-worry 0.08 0.00 0.82
Responsibility ÿ0.07 0.00 0.80
Threat 0.06 0.00 0.62
Avoidance ÿ0.14 0.01 1.52
Resistance 0.18 0.01 1.57
Importance ÿ0.23 0.02 1.69
Disapproval 0.18 0.01 1.46
Guilt 0.07 0.00 0.65
Perceived arousal 0.20 0.03 2.19
 D.A. Clark et al. / Behaviour Research and Therapy 38 (2000) 439±455 449

Perceived control, though, may not only be determined by how it is appraised. According to
cognitive models the actual thought control strategies one employs should predict degree of
control over the intrusion. We turn now to consider the role of control strategies in the self-
rated suppression of sexual and non-sexual intrusive thoughts.

3.3. Control strategies associated with non-sexual and sexual intrusions

Table 4 presents means and standard deviations on the self-rated frequency with which
participants used various control strategies in response to their most upsetting non-sexual or
sexual intrusive thought. A 2  2 repeated measures MANOVA was conducted on the 10
control strategies with gender (men versus women) as the between-group factor and type of
intrusion (non-sexual versus sexual) as the within-group factor. Although the gender  type of
intrusion interaction was non-signi®cant, the main e€ects for Gender (Wilks' Lambda=0.77,
F(10, 107)=3.14, p < 0.01) and type of intrusion (Wilks' Lambda=0.75, F(10, 107)=3.60, p <
0.001) were signi®cant. Univariate F-tests revealed that women used cognitive distraction (F(1,
116)=10.86, p < 0.001), cognitive restructuring (F(1, 116)=18.62, p < 0.001), reassurance

Table 4
Means and standard deviations on control strategies used with non-sexual and sexual intrusive thoughts by gender

ROII control strategies Men (n = 43) Women (n = 75) Total (n = 118)

M S.D. M S.D. M S.D.

Non-sexual intrusions
Cognitive distraction 1.79 1.17 2.49 1.06 2.24 1.14
Behavioural neutralization 1.67 1.06 1.76 1.09 1.73 1.08
Behavioural distraction 2.19 1.16 2.55 1.00 2.42 1.07
Cognitive neutralization 1.88 1.31 1.93 1.21 1.92 1.24
Cognitive restructuring 1.79 1.32 2.51 1.16 2.25 1.26
Reassurance seeking 0.81 1.16 1.61 1.46 1.32 1.41
Thought stopping 1.14 1.26 1.80 1.17 1.56 1.24
Do nothing 1.47 1.12 0.99 1.25 1.16 1.22
Say a prayer 0.65 1.04 0.95 1.14 0.84 1.11
Self-reassurance 1.26 1.09 1.93 1.27 1.69 1.25

Sexual intrusions
Cognitive distraction 1.58 1.10 2.12 1.19 1.92 1.18
Behavioural neutralization 1.49 1.14 1.55 1.18 1.53 1.16
Behavioural distraction 2.05 1.15 2.23 1.23 2.16 1.20
Cognitive neutralization 1.91 1.11 1.81 1.25 1.85 1.20
Cognitive restructuring 1.12 1.16 1.99 1.27 1.67 1.29
Reassurance seeking 0.65 1.02 1.08 1.34 0.92 1.25
Thought stopping 0.95 1.25 1.48 1.34 1.29 1.33
Do nothing 1.12 1.26 1.04 1.20 1.07 1.22
Say a prayer 0.67 1.06 0.79 1.08 0.75 1.07
Self-reassurance 0.93 0.91 1.61 1.42 1.36 1.30
450 D.A. Clark et al. / Behaviour Research and Therapy 38 (2000) 439±455

seeking from others (F(1, 116)=7.58, p < 0.01), thought stopping (F(1, 116)=7.77, p < 0.01)
and self-reassurance (F(1, 116)=11.22, p < 0.001) more than men. Further analysis on the
main e€ect of type of intrusion indicated that cognitive distraction (F(1, 116)=7.76, p < 0.01),
behavioural distraction (F(1, 116)=4.00, p < 0.05), cognitive restructuring (F(1, 116)=16.99, p
< 0.001), reassurance seeking from others (F(1, 116)=10.07, p < 0.01), thought stopping (F(1,
116)=5.05, p < 0.05), and self-reassurance (F(1, 116)=7.19, p < 0.01) were used signi®cantly
more often in response to participants' most upsetting non-sexual than sexual intrusion. These
®ndings indicate that di€erent types of upsetting intrusive thoughts may be associated with the
use of di€erent control strategies.
Separate hierarchical regression analyses were performed on the perceived control of non-
sexual and sexual intrusive thoughts with the 10 control strategies entered as predictors.
Gender was entered on the ®rst step, followed by the SOS and OTC total scores on the second
step and then the 10 control strategies on the ®nal step. Together the control strategies on the
®nal step did not account for a signi®cant amount of additional variance in the perceived
control of non-sexual (R 2 change=0.08, F(10, 130)=1.40, ns) or sexual (R 2 change=0.10,
F(10, 111)=1.42, ns) intrusive thoughts. Thus there was no signi®cant association between the
type of control strategies employed by participants and the perceived controllability of their
most distressing non-sexual or sexual intrusive thoughts.

4. Discussion

We began this study by proposing three questions about the experience of sexual and non-
sexual intrusive thoughts. In response to the ®rst question we failed to ®nd any signi®cant
di€erences in how university student men and women appraised their most upsetting sexual or
non-sexual intrusive thought. Over the 11 appraisal dimensions assessed in this study,
respondents generally scored in the `not at all' to `mild' range indicating that they engaged in
relatively healthy and adaptive interpretations of their unwanted, sometimes bizarre, and
deviant intrusive thoughts. However gender di€erences did emerge when it came to the
execution of thought control strategies, with women tending to engage more often in a variety
of thought control e€orts to suppress their unwanted cognitive intrusions. We can only
speculate why women might engage in more thought control responses than men. Possibly
women are more introspective or cognizant of how they respond to unwanted sexual and non-
sexual intrusive thoughts and so produced higher ratings on the ROII-v2 control items. Or it
may be that young women actually use thought control strategies more often than men but
despite this response style di€erence they perceive themselves to be no more successful at
suppressing their unwanted cognitive intrusions than men. At this time we do not know how
robust this ®nding is, given that Freeston et al. (1991) failed to ®nd any gender di€erences in
the frequency with which respondents reported using their 10 thought control strategies.
Nevertheless it may be advisable for researchers to run separate analyses on men and women
when investigating thought control strategies to unwanted intrusive thoughts.
A second research question posed for this study was whether di€erent appraisal and
thought control strategies are associated with unwanted and unacceptable sexual intrusive
thoughts as opposed to other types of obsessional intrusions that deal with aggression,
 D.A. Clark et al. / Behaviour Research and Therapy 38 (2000) 439±455 451

harm, dirt, contamination and disease. Although there were no signi®cant di€erences
between sexual and non-sexual intrusions in the extent to which respondents engaged in
various appraisal processes, nonetheless the results of the regression analyses indicated that
certain appraisal processes were more prominent in the perceived control of sexual as
opposed to non-sexual intrusions. In addition to TAF-likelihood, TAF-worry and
resistance were signi®cant unique predictors of the perceived controllability of non-sexual
intrusive thoughts, whereas subjective physiological arousal was uniquely associated with
the perceived control of sexual intrusive thoughts. As well, inspection of the zero-order
correlations amongst the appraisal variables indicated that TAF-moral had higher
correlations with the other appraisal ratings for the sexual than for the non-sexual
intrusive thoughts. Thus, the evaluation of a sexual intrusive thought as morally wrong
was associated with a tendency to evaluate the thought more negatively on a number of
other appraisal dimensions.
It is not surprising that sexual intrusive thoughts that resulted in greater physiological
arousal were perceived as more dicult to control. It may be that unwanted sexual
thoughts that elicit greater physiological and sexual arousal are considered more signi®cant
or salient and so are more dicult to dismiss from conscious awareness. TAF-worry, or
the concern that one might act on an intrusive thought, and resistance or greater e€ort in
trying to dismiss an intrusive thought, were associated with increased diculty in
controlling non-sexual intrusive thoughts. These results, then, indicate that di€erent
appraisal processes may be most salient with the control of non-sexual and sexual
intrusive thoughts. Unwanted intrusive thoughts with a sexual connotation will be more
dicult to control if individuals believe that they could act on the thought (TAF-
likelihood) and if the thought is perceived as more arousing. Unwanted intrusive thoughts
of aggression, harm, dirt, contamination and disease will be more dicult to control if
individuals believe that they could act on the thought (TAF-likelihood), if they are overly
concerned or worried that they could carry out the thought in real life (TAF-worry) and
if they try very hard to suppress the thought (resistance).
Contrary to our expectations, results from the MANOVA revealed that thought control
responses were more often used in response to non-sexual than sexual intrusive thoughts. Even
though in the present study there was no evidence that young adults experience more sexual
than non-sexual intrusive thoughts, it would appear that less active e€orts at thought control
are expended with sexual intrusive thoughts. Possibly respondents felt there was little use in
trying to suppress sexual intrusive thoughts, or may be the thought control strategies listed on
the ROII-v2 were more relevant for non-sexual than sexual intrusions. Another possibility is
that the sexual arousal associated with unwanted sexual intrusions was evaluated more
positively than the anxious arousal associated with non-sexual intrusions, thereby leading to
less e€ort to consciously suppress the sexual thought.
The negative ®ndings from the regression analyses indicated that there was no signi®cant
association between thought control strategies and perceived controllability of non-sexual or
sexual intrusive thoughts. In an earlier study Purdon and Clark (1994b) found that control
strategies accounted for a small but signi®cant amount of variance in perceived controllability
of intrusive thoughts. In sum it is clear that the strategy one chooses in an attempt to control
an intrusive thought is less important in determining its actual controllability than how the
452 D.A. Clark et al. / Behaviour Research and Therapy 38 (2000) 439±455

thought is appraised. However it is noteworthy that students engaged in more thought control
strategies in response to aggression, harm, dirt, disease and contamination intrusive thoughts
than to unwanted, upsetting sexual intrusive thoughts and images.
In response to the third research question posed for this study, the results were very clear-
cut. Appraisal and control of both sexual and non-sexual intrusive thoughts were related to a
predisposition for obsessional symptoms, as assessed by the OTC, but were not related to a
tendency to respond positively or negatively to sexual cues. Thus individuals with greater
obsessive±compulsive symptomatology were more likely to appraise their intrusive thoughts as
uncontrollable. In this regard sexual as well as non-sexual intrusions appear to be very relevant
to OCD symptoms. Unfortunately, the instrument used to assess predisposition toward
sexuality, the SOS-Short Form, lacks internal coherence and so cannot be considered a reliable
measure of sexuality. Future studies employing better sexuality measures are needed to
determine the role of personality and sexuality in the experience of unwanted sexual intrusive
thoughts.
One ®nding that was particularly noteworthy was the important role that TAF-Likelihood
played in the perceived control of non-sexual, and especially sexual, intrusive thoughts. This is
consistent with our previous research in which belief that the intrusive thought might be acted
upon, worry that one might act on the thought, or likelihood estimates that the intrusive
thought might come true, were all signi®cant unique predictors of the frequency or control of
unwanted intrusive thoughts (Purdon & Clark, 1994a,b). These ®ndings, then, support
Rachman's contention that thought-action fusion is a cognitive bias that may increase the
salience or signi®cance of intrusive thoughts. However we found that TAF-likelihood rather
than TAF-morality was the more signi®cant feature of thought-action fusion which predicts
the controllability of sexual and non-sexual intrusive thoughts. Shafran, Thordarson, and
Rachman (1996, p. 379) de®ned TAF-Likelihood as ``the belief that thinking about an
unacceptable or disturbing event make that event more probable, more likely to happen in
reality'', whereas Rachman (1997, pp. 795±796) de®ned TAF-moral as ``the belief that having a
repugnant unacceptable thought is morally equivalent to carrying out the relevant action''. In
two studies using a multi-item questionnaire of TAF and based on samples of OCD patients,
university students and community adults, Shafran et al. (1996) found that moral and
likelihood items formed salient dimensions of TAF in all samples. Moreover, in the student
sample both moral and likelihood TAF were signi®cantly correlated with the Maudsley
Obsessional Compulsive Inventory (MOCI) Checking subscale after partialling our BDI scores,
whereas for the OCD patients only TAF-likelihood for others was signi®cantly correlated with
MOCI Checking after controlling for depression scores. Shafran et al. (1996) also questioned
whether TAF-moral was less pathological and therefore less relevant to obsessional problems
than TAF-likelihood because in their second study the obsessional patients did not score
signi®cantly higher on TAF-moral than the non-clinical students. Certainly our present results
would support this explanation. TAF-likelihood was a signi®cant unique predictor of the
perceived control of intrusive thoughts whereas TAF-moral was not a signi®cant predictor of
the control of non-sexual or sexual intrusive thoughts. Thus, likelihood estimations may be the
most salient feature of thought-action fusion bias for understanding the uncontrollability of
sexual and non-sexual intrusive thoughts.
Rachman and his colleagues have also argued that thought-action fusion and perceived
 D.A. Clark et al. / Behaviour Research and Therapy 38 (2000) 439±455 453

responsibility are closely related (Shafran et al., 1996; Rachman, 1998). TAF is proposed as an
internal trigger leading to an in¯ated sense of responsibility for one's intrusive thoughts. As
can be seen from Table 2, TAF-moral and responsibility were strongly correlated for both non-
sexual and sexual intrusive thoughts. However, TAF-likelihood, which is considered the more
likely candidate in in¯ated responsibility (Shafran et al., 1996), showed a minimal association
with responsibility ratings. As well, responsibility failed to emerge as a signi®cant unique
predictor of the perceived control of sexual or non-sexual intrusive thoughts. This negative
®nding for responsibility is consistent with our earlier study (Purdon & Clark, 1994b) but
contrary to research using questionnaire (Freeston et al., 1992; RheÂaume, Freeston, Duga,
Letarte & Ladouceur, 1995) or experimental manipulations (Ladouceur et al., 1995; Lopatka &
Rachman, 1995; Ladouceur, RheÂaume & Aublet, 1997) which report a signi®cant link between
perceived responsibility and obsessional symptoms. It may be that our responsibility question
did not capture the key feature of in¯ated responsibility which is the belief that one has pivotal
in¯uence to provoke or prevent undesirable outcomes (RheÂaume, Ladouceur, Freeston &
Letarte, 1995). Rachman, Thordarson, Shafran and Woody (1995) failed to ®nd signi®cant
correlations between in¯ated responsibility and MOCI subscale scores in a large sample of
students prompting the researchers to conclude that the relationship between in¯ated
responsibility and obsessional problems may be more situation-speci®c and idiosyncratic than
previously thought. Thus, despite the centrality of in¯ated responsibility in Salkovskis' model,
the role of excessive responsibility in the persistence and uncontrollability of cognitive
intrusions remains uncertain.
One important limitation of the present study was our exclusive reliance on a non-clinical
college student sample. It could be argued that given the di€erences in sexual experiences
between young people and older adults, the generalizability of the present results are limited.
Also the non-clinical nature of the sample resulted in a restriction in the scoring range on the
appraisal ratings to the lower end of the distribution. However, Shafran et al. (1996) found
that their university student samples scored midway between an OCD clinical sample and a
group of older community adults on measures of TAF, obsessionality and depressive
symptoms. Gibbs (1996) in her review of research on analogue versus clinical OCD concluded
that non-clinical individuals with obsessive compulsive symptoms present have the same type
of symptomatology as OCD clinical patients but to a lesser severity. Rachman & de Silva,
(1978) noted di€erences between `normal' and abnormal obsessions that appear to be a matter
of degree rather than kind. Thus there is no empirical evidence to date that individuals
su€ering with obsessive compulsive disorder o€er qualitatively di€erent interpretations to their
obsessions than non-clinical individuals do to their unwanted intrusions. Although replication
is needed on clinical samples, the present results suggest that an important area of future
research is to compare the di€erential appraisal and thought control strategies that may be
associated with di€erent types of intrusive thoughts. Such research may advance our
understanding of the di€erential pathogenesis of clinical obsessions from the development of
deviant sexual thoughts and fantasies.
454 D.A. Clark et al. / Behaviour Research and Therapy 38 (2000) 439±455

References

Bouvard, M., Mollard, E., Cottraux, J., & Guerin, J. (1989). Etude preliminaire d'une liste de pensees obsedantes.
Validation et analyse factorielle. L'Encephale, 15, 351±354.
Briere, J., Smiljanich, K., & Henschel, D. (1994). Sexual fantasies, gender, and molestation history. Child Abuse and
Neglect, 18, 131±137.
Byers, E. S., Purdon, C., & Clark, D. A. (1998). Sexual intrusive thoughts of college students. The Journal of Sex
Research 35, 359±369.
Calamari, J. E., & Janeck, A. S. (1996). Response to negative intrusive thoughts: appraisal and response di€erences
in obsessional patients. Unpublished manuscript, Department of Psychology, Finch University of Health
Sciences/The Chicago Medical School, Chicago, IL.
Clark, D. A. (1989). A schema-control model of negative thoughts. Paper presented at the World Congress of
Cognitive Therapy, Oxford, UK.
Clark, D. A., & Claybourn, M. (1997). Process characteristics of worry and obsessive intrusive thoughts. Behaviour
Research and Therapy, 35, 1139±1141.
Clark, D. A., & de Silva, P. (1985). The nature of depressive and anxious thoughts: distinct or uniform phenomena?
Behaviour Research and Therapy, 23, 383±393.
Clark, D. A., & Purdon, C. (1993). New perspectives for a cognitive theory of obsessions. Australian Psychologist,
28, 161±167.
Clark, D. A., & Purdon, C. L. (1995). The assessment of unwanted intrusive thoughts: a review and critique of the
literature. Behaviour Research and Therapy, 33, 967±976.
Fisher, W. A., Byrne, D., White, L. A., & Kelley, K. (1988). Erotophobia±erotophilia as a dimension of personality.
The Journal of Sex Research, 25, 123±151.
Freeston, M. H., & Ladouceur, R. (1993). Appraisal of cognitive intrusions and response style: replication and
extension. Behaviour Research and Therapy, 31, 185±191.
Freeston, M. H., Ladouceur, R., Thibodeau, N., & Gagnon, F. (1991). Cognitive intrusions in a non-clinical
population. I. Response style, subjective experience, and appraisal. Behaviour Research and Therapy, 29, 585±597.
Freeston, M. H., Ladouceur, R., Thibodeau, N., & Gagnon, F. (1992). Cognitive intrusions in a non-clinical
population. Part II. Associations with depressive, anxious, and compulsive symptoms. Behaviour Research and
Therapy, 30, 263±271.
Gibbs, N. A. (1996). Non-clinical populations in research on obsessive-compulsive disorder: a critical review.
Clinical Psychology Review, 16, 729±773.
Gil, V. (1990). Sexual fantasy experiences and guilt among conservative Christians: an exploratory study. The
Journal of Sex Research, 27, 629±638.
Ladouceur, R., RheÂaume, J., & Aublet, F. (1997). Excessive responsibility in obsessional concerns: a ®ne-grained
experimental analysis. Behaviour Research and Therapy, 35, 423±427.
Ladouceur, R., RheÂaume, J., Freeston, M. H., Aublet, F., Jean, K., Lachance, S., Langlois, F., & de Pokomandy-
Mortin, K. (1995). Experimental manipulations of responsibility: an analogue test for models of obsessive±
compulsive disorder. Behaviour Research and Therapy, 33, 937±946.
Leckman, J. F. (1998). Personal communication. September.
Leckman, J. F., Grice, D. E., Boardman, J. et al. (1997). Symptoms of obsessive±compulsive disorder. American
Journal of Psychiatry, 154, 911±917.
Lopatka, C., & Rachman, S. (1995). Perceived responsibility and compulsive checking: an experimental analysis.
Behaviour Research and Therapy, 33, 673±684.
Niler, E. R., & Beck, S. J. (1989). The relationship among guilt, dysphoria, anxiety and obsessions in a normal
population. Behaviour Research and Therapy, 27, 213±220.
Obsessive Compulsive Cognitions Working Group (1997). Cognitive assessment of obsessive±compulsive disorder.
Behaviour Research and Therapy, 35, 667±681.
Parkinson, L., & Rachman, S. (1981). Part II. The nature of intrusive thoughts. Advances in Behaviour Research and
Therapy, 3, 101±110.
Purdon, C., & Clark, D. A. (1993). Obsessive intrusive thoughts in nonclinical subjects. Part I. Content and relation
with depressive, anxious and obsessional symptoms. Behaviour Research and Therapy, 31, 713±720.
 D.A. Clark et al. / Behaviour Research and Therapy 38 (2000) 439±455 455

Purdon, C., & Clark, D. A. (1994a). Obsessive intrusive thoughts in nonclinical subjects. Part II. Cognitive
appraisal, emotional response and thought control strategies. Behaviour Research and Therapy, 32, 403±410.
Purdon, C., & Clark, D. A. (1994b). Perceived control and appraisal of obsessional intrusive thoughts: a replication
and extension. Behavioural and Cognitive Psychotherapy, 22, 269±285.
Rachman, S. (1981). Part I. Unwanted intrusive cognitions. Advances in Behaviour Research and Therapy, 3, 89±99.
Rachman, S. (1993). Obsessions, responsibility and guilt. Behaviour Research and Therapy, 31, 149±154.
Rachman, S. (1997). A cognitive theory of obsessions. Behaviour Research and Therapy, 35, 793±802.
Rachman, S. (1998). A cognitive theory of obsessions: elaborations. Behaviour Research and Therapy 36, 385±401.
Rachman, S., & de Silva, P. (1978). Abnormal and normal obsessions. Behaviour Research and Therapy, 16, 233±
248.
Rachman, S., Thordarson, D. S., Shafran, R., & Woody, S. R. (1995). Perceived responsibility: structure and
signi®cance. Behaviour Research and Therapy, 33, 779±784.
RheÂaume, J., Freeston, M. H., Dugas, M. J., Letarte, H., & Ladouceur, R. (1995). Perfectionism, responsibility and
obsessive±compulsive symptoms. Behaviour Research and Therapy, 33, 785±794.
RheÂaume, J., Ladouceur, R., Freeston, M. H., & Letarte, H. (1995). In¯ated responsibility in obsessive compulsive
disorder: validation of an operational de®nition. Behaviour Research and Therapy, 33, 159±169.
Salkovskis, P. M. (1985). Obsessional±compulsive problems: a cognitive-behavioural analysis. Behaviour Research
and Therapy, 23, 571±583.
Salkovskis, P. M. (1989). Cognitive-behavioural factors and the persistence of intrusive thoughts in obsessional
problems. Behaviour Research and Therapy, 27, 677±682.
Salkovskis, P. M. (1996). Cognitive-behavioural approaches to the understanding of obsessional problems. In R. M.
Rapee, Current controversies in the anxiety disorders (pp. 103±133). New York: Guilford Press.
Salkovskis, P. M., & Harrison, J. (1984). Abnormal and normal obsessions: a replication. Behaviour Research and
Therapy, 22, 549±552.
Salkovskis, P. M., Richards, H. C., & Forrester, E. (1995). The relationship between obsessional problems and
intrusive thoughts. Behavioural and Cognitive Psychotherapy, 23, 281±299.
Shafran, R., Thordarson, D. S., & Rachman, S. (1996). Thought-action fusion in obsessive compulsive disorder.
Journal of Anxiety Disorders, 10, 379±391.