Two different types of obsession: autogenous obsessions and reactive obsessions

Behaviour Research and Therapy 41 (2003) 11–29 www.elsevier.com/locate/brat

Two different types of obsession: autogenous obsessions and reactive obsessions H.-J. Lee *, S.-M. Kwon Department of Psychology, Seoul National University, San 56-1, Sillim-Dong, Kwanak-Gu, Seoul 151 742, South Korea Accepted 8 October 2001

Abstract We propose that obsessions are categorized into two subtypes, i.e. autogenous obsessions and reactive obsessions, which are different in terms of identifiability of their evoking stimuli, subjective experiences, contents, and subsequent cognitive processes. Autogenous obsessions tend to come abruptly into consciousness without identifiable evoking stimuli, which are perceived as ego-dystonic and aversive enough to be repelled, and include sexual, aggressive, and immoral thoughts or impulses. On the other hand, reactive obsessions are evoked by identifiable external stimuli, which are perceived as relatively realistic and rational enough to do something toward the stimuli, and include thoughts about contamination, mistake, accident, asymmetry, loss, etc. Through three empirical studies, we confirmed the differences between the two types of obsessional intrusion in their frequency, subjective experiences, subsequent appraisal and control strategy. In particular, autogenous obsessions led to high appraisal on ‘control over thought’ and ‘importance of thought’ and frequent use of ‘avoidant control strategies’, while reactive obsessions linked with high appraisal on ‘responsibility’ and frequent use of ‘confrontational control strategies’. These findings are expected to provide a basis for classifying and explaining the heterogeneous phenomena of obsessive– compulsive disorder.  2003 Elsevier Science Ltd. All rights reserved. Keywords: Autogenous obsessions; Reactive obsessions; Cognitive-behavioral theory; Obsessive–compulsive disorder; Appraisal; Control strategy

* Corresponding author. Fax: +82-2-744-7241. E-mail address: [email protected] (H.-J. Lee).

0005-7967/03/$ - see front matter  2003 Elsevier Science Ltd. All rights reserved. PII: S 0 0 0 5 - 7 9 6 7 ( 0 1 ) 0 0 1 0 1 - 2

12

H.-J. Lee, S.-M. Kwon / Behaviour Research and Therapy 41 (2003) 11–29

1. Introduction The DSM-IV (American Psychiatric Association, 1994) defines obsessions as recurrent and persistent thoughts, impulses, or images (1) that are experienced as intrusive and inappropriate and that cause marked anxiety or distress, (2) that are not simply excessive worries about reallife problems, and (3) that make the person attempt to ignore or suppress them, or to neutralize them with some other thoughts or actions. The most common obsessions include repeated thoughts about contamination, repeated doubts, needs to have things in a particular order, aggressive or horrific impulses, and sexual imagery. In clinical settings, however, patients with obsessive–compulsive disorder (OCD) report obsessions with a variety of contents and qualities. These clinical observations imply that obsessions are heterogeneous phenomena, suggesting a need for differentiating them into more homogeneous subtypes. Let us consider the following two kinds of obsessions which are frequently reported by OCD patients: (a) repeated thoughts about perverted sexual intercourse with their parent or blasphemous activity of killing God; and (b) repeated thoughts about leaving the gas stove on behind to be a big fire or hand with dirty germs to be contaminated. People with the former thoughts tend to have difficulties in reporting the evoking stimuli (e.g. come abruptly into mind or ‘out of blue’), feel the thoughts unrealistic and threatening to their moral nature, and try to erase or expel the thoughts from their consciousness. On the other hand, people with the latter thoughts tend to easily identify their evoking stimuli or situations, feel the thoughts possibly realistic, and try to do something for preventing the possible negative consequences. These observations make us propose that obsessions are categorized into two subtypes, i.e. ‘autogenous obsessions’ and ‘reactive obsessions’, which are different in terms of identifiability of their evoking stimuli, relationship between evoking stimuli and thought contents, content characteristics, subjective experiences, and subsequent cognitive processes. Autogenous obsessions tend to come abruptly into consciousness without identifiable evoking stimuli, which are perceived as ego-dystonic and aversive to be repelled, and include sexual, aggressive, aversive, or immoral thoughts, images, or impulses. Even though the evoking stimuli or situations can be identified, they are likely to be connected with the content of the evoked thoughts in associational, symbolic, or less logical ways. On the other hand, reactive obsessions tend to be evoked by identifiable external stimuli, are perceived as relatively realistic and rational to do something toward the stimuli, and include thoughts, concerns, or doubts about contamination, mistake, accident, asymmetry, loss, etc.. Also, their evoking stimuli or situations are usually connected with the content of the evoked thoughts in realistic and logical ways. In other words, they are not only the trigger of intrusive thoughts but also the problematic situations that should be corrected by compulsion in order to prevent or avoid expected negative consequences. The differential characteristics of these two obsessions are summarized in Table 1. From the cognitive-behavioral perspective, autogenous obsessions and reactive obsessions are expected to be quite different in their subsequent cognitive processes. Many studies on obsessions (Amir, Cashman, & Foa, 1997; Freeston, Ladouceur, Provencher, & Blais, 1995; Freeston, Ladouceur, Thibodeau, & Gagnon, 1991, 1992; Freeston & Ladouceur, 1993; Purdon & Clark, 1994a, 1994b; Wells & Davies, 1994) have shown that appraisal and control strategy are the crucial factors playing determinant roles in escalating intrusive thoughts into pathological obsessions. The International Obsessive–compulsive Cognition Working Group (OCCWG, 1997)

Relatively difficult Less close Symbolical and associational Thoughts of sexual contents Thoughts of aggressive or violent acts Thoughts of blasphemy or immoral things Thought of aversive scene or situation High High High Usually covert and avoidant Pure obsession, rumination, etc.

Identifiability of evoking stimuli

Relationship of thought content with the evoking stimuli

Typical examples

Ego-dystonicity

Perceived irrationality

Distinguishability from automatic thought

Subsequent

Compulsion

Autogenous obsessions

Table 1 Distinguishable characteristics between autogenous obsessions and reactive obsessions

of of of of

contamination or dirt doubt about mistake or accident disorder, disarray, or dissymmetry loss of important things

Washing, checking, arranging, hoarding, etc.

Usually overt and confrontational

Low

Low

Low

Thoughts Thoughts Thoughts Thoughts

Closer Realistic and logical

Relatively easy

Reactive obsessions

H.-J. Lee, S.-M. Kwon / Behaviour Research and Therapy 41 (2003) 11–29 13

14

H.-J. Lee, S.-M. Kwon / Behaviour Research and Therapy 41 (2003) 11–29

defined appraisal as the ways in which meaning is given to a specific event such as the occurrence of an intrusion. The OCCWG suggested the three important appraisal dimensions on intrusive thoughts, i.e. ‘control over thought’, ‘importance of thought’, and ‘responsibility’, developing the Interpretation of Intrusion Inventory (III). These faulty appraisals on intrusive thoughts make obsessions persist and aggravate (Rachman 1997, 1998; Salkovskis 1985, 1989). Control strategies have been used as the same meaning with the neutralizing responses which patients with OCD covertly or overtly use in order to reduce the intrusive thoughts and (or) related emotional discomfort (Purdon & Clark, 1994a, 1994b; Freeston & Ladouceur, 1997). They have been considered to play a paradoxical role of increasing the salience of the intrusions, thereby contributing to maintaining obsessions (Salkovskis 1985, 1989). According to the cognitivebehavioral theory of OCD, intrusive thoughts are stimuli which initiate these appraisal and control strategies. We propose that the content of intrusive thoughts may play an important role in directing specific appraisal and control strategy, leading to different types of obsessive–compulsive symptoms. Autogenous obsessions and reactive obsessions are expected to be different in their subsequent appraisal and control strategies. In appraisal, autogenous obsessions may lead to high appraisal on control over thought and importance of thought because they include immoral contents, ego-dystonicity, and (or) moral fusion. On the other hand, reactive obsessions may lead to high appraisal on responsibility which means belief that people have central power on evoking or preventing negative results. In control strategies, autogenous obsessions may lead to avoidant ways of controlling the thoughts themselves (e.g. thought stopping, distractive activity, counterimaging, self-punishment, etc.), while reactive obsessions lead to confrontational actual control behaviors in order to put things back to safe or desired states (e.g. washing, checking, arranging, hoarding, etc.). A series of empirical study were carried out to verify the differentiation of the two subtypes of obsession. Study 1 aims at confirming whether various obsessions can be grouped into the two subtypes in terms of their frequency and subjective experiences. Studies 2 and 3 were designed to examine the differences between autogenous obsessions and reactive obsessions in their subsequent appraisal and control strategies. 2. Study 1 Most people experience intrusive thoughts which may lead to obsessions or compulsions (Clark, 1992; Freeston et al. 1991, 1992; Purdon and Clark, 1993). If autogenous obsessions and reactive obsessions are different in various aspects as proposed in the Table 1, people are expected to show different responses to the two types of obsessional intrusion in the frequency and subjective experiences. In Study 1, we tried to confirm whether various intrusive thoughts could be differentiated into two subgroups through factor analysis. 2.1. Methods 2.1.1. Participants Three hundred and seventy undergraduate students (194 men and 176 women) in the introductory psychology class of three universities in Seoul area participated in Study 1 as a course

H.-J. Lee, S.-M. Kwon / Behaviour Research and Therapy 41 (2003) 11–29

15

requisite. Their age is 20.95 years on the average (SD=2.31) and there was no sex difference in age. 2.1.2. Measures 2.1.2.1. Revised obsessional intrusion inventory (ROII) The ROII was developed to evaluate the experience of the intrusive thought, image, and impulse in normal adults (Purdon & Clark, 1993). The part I of the ROII presents 52 items of obsessions and asks participants to rate how frequently they experience each of the thoughts on a 7-point Likert scale (0: never; 1: once or twice ever; 2: a few times a year; 3: once or twice a month; 4: once or twice a week; 5: daily; 6: frequently during the day). The part II asks participants to choose one intrusive thought which has been experienced most disturbingly so far and answer the questions of subjective experiences, appraisal and control strategies about the selected thought on a 7-point Likert scale (0: never; 2: sometimes; 4: frequent; 6: always). In this study, 10 questions of subjective experiences about the selected thought (e.g. unpleasantness, unacceptability, probability, guilt feeling, importance of controlling the thought, worry over the thought’s realization, probability of the thoughts realization, difficulty of removing the thought, etc.) were used. Through several studies (Purdon & Clark, 1993, 1994a, 1994b), it was found out the ROII has adequate reliability and validity as a measure of intrusive thoughts relevant to obsessive–compulsive phenomenon. 2.2. Results and discussion The participants were randomly divided into two halves for the purpose of cross-sample validation. Exploratory factor analysis (EFA) was carried out on the data from the first half of participants (N=185) with unweighted least-squares extraction and direct quartimin rotation. The number of factor to be extracted was constrained as two, because we proposed that various intrusive thoughts can be grouped into two subtypes. The results of factor analysis show that the ROII items can be clearly grouped into two factors which explained 35% of the total variance of the ROII. As shown in Table 2, the intrusive thoughts with high factor loading to the factor 1 (thoughts of violent or impulsive activities, thoughts of impulsive acting out, and thoughts of sexual activities, or aversive or embarrassing scenes) reflect autogenous obsessions, whereas those with high factor loading to the factor 2 (thoughts of doubting mistake and thoughts of doubting contamination or dirts) reflect reactive obsessions. Our data also showed similar factor structure between male and female participants. It suggests that various intrusive thoughts can be grouped into the two subtypes, i.e. autogenous obsessions and reactive obsessions. In order to confirm the adequacy of the two factor structure, we performed confirmatory factory analysis (CFA) on the data from the other half of participants (N=185) with Analysis of Moment Structures, Version 4.0 (AMOS) (Arbuckle & Wothke, 1999). However, the ROII includes too many items (i.e. 41 items for the first factor and 11 items for the second factor) in terms of the proposed number of factor and the sample size to apply the conventional way of CFA to our data. Browne & Cudeck (1993) have indicated that the size of estimation error becomes larger as the number of parameter increases with a given sample size. Therefore, in order to diminish estimation error, we applied a modified procedure of CFA to the proposed two factor model by making composite scores out of the ROIIs 52 items. That is, we reduced the number of parameter by randomly dividing the autogenenous obsessions (41 items of the first factor) and the reactive

16

H.-J. Lee, S.-M. Kwon / Behaviour Research and Therapy 41 (2003) 11–29

Table 2 Results of Exploratory Factor Analysis on the ROII Intrusive thought s of ROII

M

SD

Thoughts of violent or impulsive activities 1. Driving into window 0.61 0.86 2. Running car off road 1.18 1.11 3. Hitting animals, people 1.09 1.14 4. Swerving into traffic 1.07 1.03 5. Smashing into objects 0.72 0.99 6. Slitting wrist/throat 1.15 1.08 7. Cutting off finger 1.00 1.09 8. Jumping off high place 1.91 1.24 9. Fatally pushing stranger 0.88 1.03 10. Fatally pushing friend 0.42 0.85 11. Jumping — train, car 1.79 1.35 12. Pushing Stranger — train, car 0.71 0.99 13. Pushing family — train, car 0.36 0.77 14. Hurting strangers 0.92 1.14 15. Insulting stranger 1.18 1.14 16. Bumping into people 0.78 0.90 17. Insulting authority 1.41 1.22 18. Insulting family 1.11 1.19 19. Hurting family 0.79 1.05 20. Choking family member 0.25 0.55 21. Stabbing family member 0.34 0.71 Thoughts of doubting mistake 22. Heat/stove on, accident 2.18 1.32 23. Home unlocked, intruder 1.91 1.36 24. Taps left on, flood 1.16 1.19 Thoughts of impulsive acting out 25. Swearing in public 0.76 1.00 26. Breaking wind in public 0.82 1.05 27. Throwing something 0.61 0.91 28. Causing a public scene 0.98 1.07 29. Scratching car paint 1.14 1.00 30. Breaking window 1.20 1.02 31. Wrecking something 1.31 1.10 32. Shoplifting 1.00 1.06 33. Grabbing money 0.95 0.99 34. Holding up bank teller 0.42 0.71 Thoughts of sexual activities, or aversive or embarrassing scenes 35. Sex with unacquainted person 1.31 1.10 36. Sex with authority figure 0.83 1.07 37. Fly/blouse undone 1.29 1.23 38. Kissing authority figure 0.81 1.04 39. Exposing myself 0.58 0.84 40. Acts against sex preference 0.68 0.96 41. Authority figure naked 0.83 1.01 42. Strangers naked 1.06 1.16

Factor 1

Factor 2

0.388 0.265 0.420 0.413 0.335 0.405 0.323 0.382 0.527 0.474 0.395 0.646 0.516 0.514 0.581 0.692 0.572 0.528 0.581 0.541 0.628

0.098 0.250 0.231 0.173 0.185 0.197 0.164 0.319 0.186 0.188 0.140 0.031 0.137 0.017 ⫺0.029 ⫺0.085 ⫺0.022 0.046 ⫺0.017 ⫺0.042 ⫺0.011

0.176 0.176 0.184

0.426 0.451 0.376

0.620 0.354 0.651 0.510 0.553 0.572 0.622 0.518 0.414 0.486

⫺0.069 0.217 ⫺0.068 0.069 0.054 0.197 0.088 0.125 0.222 0.130

0.641 0.680 0.613 0.637 0.634 0.393 0.641 0.720

⫺0.134 ⫺0.179 0.080 ⫺0.060 ⫺0.137 ⫺0.014 ⫺0.102 ⫺0.182 (continued on next page)

H.-J. Lee, S.-M. Kwon / Behaviour Research and Therapy 41 (2003) 11–29

17

Table 2 (continued) Intrusive thought s of ROII 43. Sex in public 44. Disgusting sex act Thoughts of doubting contamination or dirts 45. Catching STD (sexual transmitted disease) 46. Contamination, door-knobs 47. Contamination, phones 48. Fatal disease, stranger 49. Transmitting fatal disease 50. Everything away 51. All dust off floor 52. Dirt in unseen places Eigenvalue

M

SD

Factor 1

Factor 2

1.05 1.18

1.13 1.22

0.655 0.602

⫺0.131 ⫺0.053

0.99 1.52 1.59 1.03 1.13 1.11 1.35 1.91

1.27 1.47 1.51 1.07 1.14 1.29 1.38 1.52

0.027 ⫺0.153 ⫺0.022 ⫺0.045 ⫺0.095 ⫺0.011 0.00 ⫺0.021 13.545

0.514 0.734 0.633 0.729 0.542 0.615 0.686 0.710 7.431

obsessions (11 items of the second factor), respectively, into three subgroups (see Fig. 1). When a factor consists of items believed to be unidimensional, it is supposed to make no significant difference to randomly divide those items into three to five subgroups to make an appropriate number of indicators in the analysis of covariance structure (S.H. Hong, personal communication, 2001). The internal consistency of the 41 items of the first factor and the 11 items of the second factor was 0.95 and 0.89, respectively. Fit indices include the c2, the Goodness of Fit Index (GFI), the Adjusted Goodness of Fit Index (AGFI) (Jo¨ reskog & So¨ rbom, 1996), the Comparative Fit Index (CFI) (Bentler, 1990), the Tucker–Lewis Index (TLI) (Bentler & Bonett, 1980), and the Root Mean Square Error of Approximation (RMSEA) (Browne & Cudeck, 1993). The results of CFA showed that the fit of the two factor model was good, c2(8, N=185)=10.14,

Fig. 1. Two factor model tested by CFA. aROII-A: the ROIIs items belonging to autogenous obsessions as a result of EFA; bROII-R: the ROIIs items belonging to reactive obsessions as a result of EFA. (*Items constituting each indicator: ROII-A1 — 1, 4, 7, 10, 13, 16, 19, 25, 28, 31, 34, 37, 40, 43/ROII-A2 — 2, 5, 8, 11, 14, 17, 20, 26, 29, 32, 35, 38, 41, 44/ROII-A3 — 3, 6, 9, 12, 15, 18, 21, 27, 30, 33, 36, 39, 42/ROII-R1 — 22, 45, 48, 51/ROII-R2 — 23, 46, 49, 52/ROII-R3 — 24, 47, 50).

18

H.-J. Lee, S.-M. Kwon / Behaviour Research and Therapy 41 (2003) 11–29

p=0.26; GFI=0.98; AGFI=0.96; CFI=0.99; TLI=0.99; RMSEA=0.04, indicating that there were no significant differences between the predicted and observed covariance matrices. Purdon & Clark (1993) had reported one factor structure (sex/aggression themes) of the ROII for men and two factor structure (sex/aggression and dirt/contamination themes) for women through exploratory factor analysis. One factor model as an alternative model was also tested with CFA in order to compare the fit indices with those of the two factor model. One factor model failed to show the adequate indices of goodness-of-fit, c2(9, N=185)=417.71, p=0.00; GFI=0.60; AGFI=0.06; CFI=0.68; TLI=0.47; RMSEA=0.50. We also confirmed that gender difference did not change the preponderance of the two factor model over the one factor model. The two factor model was consistently supported with the good fit indices both in men, c2(8, N=127)=15.63, p=0.05; GFI=0.96; AGFI=0.90; CFI=0.99; TLI=0.98; RMSEA=0.08 and in women, c2(8, N=58)=8.29, p=0.41; GFI=0.96; AGFI=0.90; CFI=0.99; TLI=0.99; RMSEA=0.02. On the other hand, the one factor model showed relatively poor model fit both in men, c2(9, N=127)=284.56, p=0.00; GFI=0.58; AGFI=0.03; CFI=0.69; TLI=0.48; RMSEA=0.49 and in women, c2(9, N=58)=126.86, p=0.00; GFI=0.60; AGFI=0.07; CFI=0.70; TLI=0.50; RMSEA=0.48. Several CFAs on the two factor model, using the data by different ways of dividing the 41 autogenous items and the 11 reactive items, respectively, into three subgroups, did not make any significant difference to the good model fit of the model. These results of EFA and CFA appear to support the distinction between the autogenous obsessions and the reactive obsessions. The autogenous obsessions and reactive obsessions appeared to be quite different in participants’ ratings on subjective experiences to them. The intrusive thoughts which each participant reported to be most disturbing on the Part II of the ROII are classified into autogenous obsessions or reactive obsessions according to the factor structure. Between group t-tests for the 10 items of the ROII-Part II were computed to examine whether the two types of obsessions are different in subjective experiences. To correct for type I error we used the Bonferroni correction procedure and set the alpha level to p=0.005. As seen in Table 3, autogenous obsessions showed significantly higher ratings on unacTable 3 Differences between the two obsession groups in the part II of ROII (*p⬍0.005; alpha set using the Bonferroni correction)

Unpleasantness Guilt Worry thought will come true Difficulty removing thought Unacceptability Likelihood thought will come true Importance of controlling thought Thought portends danger Seriousness of the realized result Urge to avoid triggering situation a

df=368.

Autogenous obsessions Reactive obsessions M SD M SD

ta

2.23 1.81 1.54 1.47 2.08 1.22 2.37 1.53 4.00 2.98

⫺1.963 6.771* ⫺4.334* ⫺2.765 3.169* ⫺6.406* 5.560* 2.802* 3.542* 0.736

1.50 1.61 1.26 1.22 1.68 1.27 1.62 1.45 0.66 0.68

2.64 0.59 2.29 1.95 1.40 2.36 1.37 0.99 3.17 0.81

1.53 1.16 1.27 1.35 1.27 1.37 1.13 1.33 1.83 1.81

H.-J. Lee, S.-M. Kwon / Behaviour Research and Therapy 41 (2003) 11–29

19

ceptability of the thought, guilt feeling, and importance of controlling the thought than reactive obsessions. On the other hand, reactive obsessions showed significantly higher ratings on worry over the thought’s realization and probability than autogenous obsessions. It suggests that autogenous obsessions are experienced as more unacceptable, guilty, and dangerous, whereas reactive obsessions are experienced as more probable, realistic, and likely to be realized. These results indicate that autogenous obsessions and reactive obsessions could be differentiated in terms of their subjective experiences as well as frequency. 3. Study 2 Cognitive appraisal and control strategy on intrusive thoughts are the crucial factors in escalating intrusive thoughts into pathological obsessions. We proposed that autogenous obsessions and reactive obsessions are quite different in their appraisal and control strategy. In appraisal, autogenous obsessions are expected to lead to high appraisal on control over thought and importance of thought, whereas reactive obsessions may lead to high appraisal on responsibility. In control strategies, autogenous obsessions are expected to lead to avoidant ways of controlling the thoughts themselves, while reactive obsessions lead to confrontational actual control behaviors in order to put things back to safe or desired states. Study 2 was performed to examine these differences between autogenous and reactive obsessions in appraisal and control strategy. 3.1. Methods 3.1.1. Participants Two hundred and forty four undergraduate students (118 men and 126 women) in the introductory psychology class in the Seoul National University participated in Study 2 as a course requisite. Their age is 21.0 years on the average (SD=2.29) and there was no sex difference in age. 3.1.2. Measures and procedure 3.1.2.1. Revised Obsessional Intrusion Inventory In order to differentiate the two subtype intrusive thoughts, the part I of the ROII (Purdon & Clark, 1993) was used. Participants were asked to choose one intrusive thought which have been experienced most disturbingly so far among 52 items of obsessions. The most disturbing thought was classified as autogenous obsession or reactive obsession according to the factor structure in Study 1. The part II of the ROII was designed and modified by us in order to examine appraisal and control strategy on the selected thought, containing 12 items for appraisal and 12 items for control strategy which ask participants to rate on a 7-point scale (0: never; 2: sometimes; 4: frequent; 6: always). The appraisal section with 12 items was constructed to measure three dimensions of appraisal (i.e. control over thought, importance of thought, and responsibility) mainly based on the study of Obsessive Compulsive Cognitions Working Group (1997) and their measure (i.e. The Interpretation of Intrusion Inventory). The four items for measuring each of the three appraisal dimensions follow: control over thought “I should not be thinking this kind of thing. Having this intrusive thought means I’m out of control. Because I can’t control this thought, I am a weak person. If I don’t control this unwanted thought, something bad is bound to happen.”; importance of thought

20

H.-J. Lee, S.-M. Kwon / Behaviour Research and Therapy 41 (2003) 11–29

“Thinking this thought could make it happen. Because I’ve had this thought, I must want it to happen. Thinking this thought is as bad as causing it to happen or doing it actually. Having this thought means I am a weird, abnormal, or terrible person.”; and responsibility “Because I’ve thought of bad things that might happen, I must act to prevent them. If I don’t do something about this intrusive thought, it will be my fault if something terrible happens. It is wrong not to do something about this unwanted thought. If I ignore this thought, I could be responsible for serious harm”. The items come from the factor structure of our previous study (Lee & Kwon, 1999) in which the internal consistency (Cronbach’s alpha) of the 12 items was 0.85. The internal consistencies of the items for measuring control over thought, importance of thought, and responsibility were 0.84, 0.68, and 0.88, respectively. The 12 items for control strategy were mainly chosen from the part II of the ROII (Purdon & Clark, 1993, 1994a, 1994b) and added by us to measure the three types of control strategy, i.e. confrontational strategy, neutral avoidant strategy, and negative avoidant strategy. The four items for measuring each of the three types of control strategy follow: confrontational strategy (taking action to put things back to safe or desired state, analyzing the thought rationally, seeking information or talking about this thought, seeking reassurance from others); negative avoidant strategy (worrying or thinking about other negative things, self-blaming or self-punishment, saying ‘stop’ to oneself, praying or other religious ways); and neutral avoidant strategy (diverting attention by engaging oneself in something else, avoidance from anxious objects or situations, thinking about other pleasant things, counterimaging). The internal consistency (Cronbach’s alpha) of the 12 items was 0.77. A factor analysis on the 12 items confirmed the three factor structure reflecting confrontational strategy, negative avoidant strategy, and neutral avoidant strategy each of which has four items with the internal consistency, 0.64, 0.66, and 0.70, respectively (Lee & Kwon, 1999). 3.2. Results and discussions Among 244 participants, 166 persons selected one of the autogenous obsessions as the most disturbing thought, while 78 persons selected one of reactive obsessions. Their appraisal and control strategy on the most disturbing intrusive thoughts were compared between these two groups. As seen in Table 4, the two groups showed significant differences in all the dimensions of appraisal and control strategy. As expected, autogenous obsession group showed high ratings on control over thought and importance of thought, while reactive obsession group show high ratings on responsibility. In control strategy, autogenous obsession group preferred the two avoidant types, whereas reactive obsession group chose the confrontational type. These results indicate that autogenous obsessions and reactive obsessions lead people to the different ways of appraisal and control strategy. Because autogenous obsessions are perceived to be unacceptable, immoral and guilty, people are likely to attach more importance and significance to the thoughts themselves and feel more pressure to control them. However, they have difficulties not only in identifying the evoking stimuli or situations but also in finding specific behavioral methods for changing external environments related to the autogenous obsessions. Therefore, they tend to prefer or employ avoidant control strategies by expelling the thoughts themselves from consciousness, turning to religious or magical method, or punishing themselves. On the other

H.-J. Lee, S.-M. Kwon / Behaviour Research and Therapy 41 (2003) 11–29

21

Table 4 Differences in appraisal and control strategies between the two obsession groups (*p⬍0.02; alpha set using the Bonferroni corrections to appraisal section and control strategy section, respectively)

Appraisal Control over thought Importance of thought Responsibility Control strategy Confrontational Negative avoidant Neutral avoidant a b

Autogenous obsessions SD Ma

Reactive obsessions M SD

tb

2.12 1.76 1.17

1.31 1.10 1.10

0.84 0.63 2.09

0.87 0.66 1.33

9.018* 9.996* ⫺5.702*

1.18 1.25 2.18

0.97 1.12 1.19

1.96 0.41 1.39

1.05 0.59 1.10

⫺5.649* 7.606* 4.945*

Average of the rating scores on 4 questions in each dimension of appraisal and control strategy. df=242.

hand, people with reactive obsessions perceive them to be more realistic and probable, feel more responsibility for preventing the possibly dangerous outcomes, and prefer confrontational control strategies because they cannot only identify the evoking stimuli or situations easily but also find specific and actual coping behaviors (e.g. washing, checking, arranging, hoarding, etc.) for putting things back to safe or desired states in external environments. There was a significant difference between the two obsession groups in their experienced frequency, t(242)=⫺5.34, p⬍0.01. The average frequency of autogenous obsessions was 2.80 (SD=1.16) and the average frequency of reactive obsessions was 3.67 (SD=1.24). In order to examine whether the differences in frequency of obsession are associated with cognitive appraisal and control strategy as well as subjective experiences on obsessions, the additional analysis was carried out. From the autogenous group and the reactive group, we selected participants who reported their experienced frequency of obsessions to be equal to or more than ‘once or twice a week’. There were 40 participants from the autogenous obsession group and 39 participants from the reactive obsession group. The average frequency of obsessional thoughts was 4.50 (SD=0.68) in the selected autogenous group and 4.74 (SD=0.72) in the selected reactive group, showing no significant difference between the two groups, t(77)=⫺1.55, p=0.13. However, the two groups were quite different in cognitive appraisal (control over thought, importance of thought, responsibility) and control strategy (confrontational, negative and neutral avoidant strategies) as well as subjective experiences (guilt, unacceptability, importance of controlling thought, etc.) on their obsessional thoughts. These results were almost the same as those of Table 4, suggesting that the differentiation between the two types of obsessional thoughts could not be accounted for by the difference of frequency. 4. Study 3 Study 2 showed that autogenous obsessions and reactive obsessions lead to the different ways of subsequent cognitive appraisal and control strategy. What then makes these differences? There

22

H.-J. Lee, S.-M. Kwon / Behaviour Research and Therapy 41 (2003) 11–29

can be several interpretations on the results of Study 2 in which participants were separated into autogenous obsession group and reactive obsession group according to their most disturbing thought and compared in their appraisal and control strategy on the thought. Therefore, it is unclear in whether the results of Study 2 may be due to the content of obsessions itself or due to other possible differences between the two groups in their several characteristics (e.g. their severity of obsessive symptom, personality traits, etc.). In fact, we carried out an additional analysis on the data of Study 2 to exclude the possibility that the differences between the two groups can be due to the difference in their severity of OCD. In this analysis, participants were separated into two groups, i.e. high and low obsessive groups according to their OCD score of SCL-90-R. This analysis showed no differences in the patterns of appraisal and control strategy on intrusive thoughts between the two groups, regardless of their extent of obsessive tendency. We here propose that the differences between autogenous and reactive obsessions in appraisal and control strategy be mainly due to the content of the intrusive thoughts. In other words, an individual can select a different way of appraisal and control strategy according to the content of intrusive thoughts. Study 3 was designed to verify this proposal by presenting all the participants with the same typical examples of the intrusive thoughts of autogenous or reactive subtypes and comparing the pattern of appraisal and control strategy between autogenous obsessions and reactive obsessions. 4.1. Methods 4.1.1. Participants One hundred and thirty one undergraduate students (64 men and 67 women) in the introductory psychology class of the Seoul National University participated in Study 3 as a course requisite. Their age is 21.0 years on the average (SD=2.53) and there was no sex difference in age. 4.1.2. Measures and procedure 4.1.2.1. Appraisal and control strategies questionnaire (ACSQ) This questionnaire was developed by us to evaluate a person’s appraisal and control strategy on various intrusive thoughts. The ACSQ provides eight kinds of intrusive thoughts with different contents including: (1) sexuality, (2) aggression and violence, (3) blasphemy and immorality, (4) aversive scenes or situations, (5) contamination, (6) mistakes or accidents, (7) asymmetry or disorder, and (8) loss of important informations or things. The former four thoughts represent the autogenous obsession and the latter four thoughts represent the reactive obsession. Each of the intrusive thoughts is rated by participants in terms of appraisal and control strategy on it. In the ACSQ, general content and three typical examples about each of the intrusive thoughts are provided on the top of page. In the case of sexuality, the stimulus thoughts are presented as follows. Thoughts of sexual content (example): —thought of doing perverted sexual behaviors with my mother (or father); —thought of doing aggressive sexual behavior with family member, friends, or unfamiliar persons; —thought of showing my sexual organ to many people in subway.

H.-J. Lee, S.-M. Kwon / Behaviour Research and Therapy 41 (2003) 11–29

23

Then, participants were asked to rate 12 appraisal questions on 7-point Likert scale (0: never, 2: sometimes, 4: frequent, 6: always) with the instruction “If these thoughts come into your consciousness repeatedly, how do you think about them?”. The 12 appraisal questions for evaluating each of three appraisal dimensions, i.e. control of thoughts, importance of thoughts, and responsibility come from the part II of the ROII used in Study 2. Lastly, participants were asked to rate six questions about control strategy on 7-point Likert scale (0: never, 2: sometimes, 4: frequently, 6: always) with the instruction “If these thoughts come into your consciousness repeatedly, how frequently will you use the following coping methods?”. The six questions include three concerning confrontational control strategies (i.e. seeking information from or talking about the thought with others, taking an action to put things back to safe or desired state, and analyzing the thought rationally and deliberating whether it is reasonable) and three concerning avoidant control strategies (i.e. devoting oneself to other thought or activities to distract attention from the thought, saying stop to oneself, and self-reassuring that the thought is meaningless). To reduce the pressure given to participants by the burdensome task of evaluating eight content domains, we presented only six control strategies which were reported to be used pervasively against obsessive intrusion by previous studies and our data (Freeston & Ladouceur, 1997; Lee & Kwon, 1999). Each of the intrusive thoughts and related questions appeared on a separate page. The four autogenous obsessions and four reactive obsessions were given one by one in the alternating order. The ACSQ was administered in group and took about 30 min to complete 4.2. Results and discussion Table 5 shows the mean and standard deviation of rating scores of the three appraisal dimensions in the two types of obsession as well as each content of obsessions. Rating scores of Table 5 Mean scores of the three appraisal dimensions in the two subtypes of obsession

Contents of obsessions

Control over thought SD Ma

Importance of thought M SD

Responsiblity M SD

Autogenous obsessions 1. Sexual 2. Aggressive/violence 3. Blasphemy/immoral 4. Aversive image

2.27 2.43 2.10 2.52 2.05

1.19 1.32 1.32 1.64 1.42

1.79 1.68 1.97 1.99 1.51

0.93 1.15 1.10 1.22 1.23

1.41 1.41 1.40 1.64 1.18

1.08 1.19 1.14 1.57 1.42

Reactive obsessions 5. Contamination/contagion 6. Mistake/accident 7. Disorder/asymmetry 8. Loss of information

1.11 1.10 1.28 1.02 1.06

0.84 1.07 1.18 0.98 1.06

0.75 0.72 0.74 0.73 0.81

0.64 0.77 0.76 0.85 0.75

1.66 1.35 2.09 1.43 1.78

1.16 1.21 1.51 1.40 1.53

a

Average of the rating scores on four questions in each dimension of appraisal.

24

H.-J. Lee, S.-M. Kwon / Behaviour Research and Therapy 41 (2003) 11–29

appraisal over four content domains in each obsession type were averaged. First, there were significant differences in appraisal and control strategies in response to the two obsession types. Compared to reactive obsessions, autogenous obsessions elicited higher ratings on control over thought, t(130)=12.27, p⬍0.01, and importance of thought, t(130)=13.48, p⬍0.01. On the other hand, reactive obsessions elicited higher ratings on responsibility, t(130)=2.46, p⬍0.05, than autogenous obsessions. In response to autogenous obsessions, control over thought and importance of thought were rated as higher than responsibility, t(130)=9.21, p⬍0.01 and t(130)=4.60, p⬍0.01. On the other hand, in response to reactive obsessions, responsibility was rated as higher than control over thought and importance of thought, t(130)=5.21, p⬍0.01 and t(130)=9.80, p⬍0.01. Table 6 shows the mean and standard deviation of rating scores of the two types of control strategy in two types of obsession as well as each content of obsessions. First, there were significant differences in control strategy in response to the two obsession types. Autogenous obsessions elicited more frequent selection of the avoidant control strategies than reactive obsessions, t(130)=8.89, p⬍0.01. On the other hand, reactive obsessions elicited more frequent selection of the confrontational control strategies than autogenous obsessions, t(130)=9.64, p⬍0.01. In response to autogenous obsessions, the avoidant control strategies were selected more frequently than the confrontational control strategies, t(130)=12.35, p⬍0.01. On the other hand, in response to reactive obsessions, the confrontational control strategies were selected more frequently than the avoidant control strategies, t(130)=3.54, p⬍0.01. The results of Study 3 show that autogenous obsession and reactive obsession are clearly different in their appraisal and control strategy. As expected, autogenous obsessions led into high ratings on control over thought and importance of thought, and frequent use of thought control strategies of avoidance type, while reactive obsessions led into high ratings on responsibility and frequent use of relatively actual control strategies of confrontational type. The pattern of appraisal and Table 6 Mean scores of the two control strategy types in the two subtypes of obsession

Contents of obsessions

Avoidant control strategy SD Ma

Confrontational control strategy M SD

Autogenous obsessions 1. Sexual 2. Aggressive/violence 3. Blasphemy/immoral 4. Aversive image

2.82 3.02 2.56 2.66 3.04

1.23 1.25 1.35 1.53 1.50

1.46 1.35 1.48 1.53 1.49

1.04 1.13 1.14 1.23 1.30

Reactive obsessions 5. Contamination/contagion 6. Mistake/accident 7. Asymmetry/disorder 8. Loss of information

1.91 2.20 1.79 1.89 1.77

1.15 1.34 1.43 1.42 1.38

2.32 2.22 2.25 2.29 2.51

1.07 1.25 1.32 1.25 1.32

a

Average of the rating scores on 3 questions about each type of control strategy.

H.-J. Lee, S.-M. Kwon / Behaviour Research and Therapy 41 (2003) 11–29

25

control strategy was the same in each content domain of the two obsession types. Thoughts of sexual, aggressive, blasphemous contents, or aversive scene in autogenous obsessions shared the pattern of their subsequent appraisal and control strategies, the same was true of thoughts of contamination, mistake, disorder or loss in reactive obsessions. These results not only support the distinction of the two subtypes of obsessions but also suggest that differences between autogenous and reactive obsessions in subsequent appraisal and control strategy are under the significant influence of the content or meaning of the thoughts. 5. General discussion In this study, we proposed that obsessions can be categorized into the two subtypes, i.e. autogenous obsessions and reactive obsessions, which are different in terms of identifiability of their evoking stimuli, contents, and subsequent cognitive processes. Study 1 showed that various intrusive thoughts could be clearly differentiated into two subtypes in their frequency and subjective experiences such as unacceptability, guilty feeling, probability, and importance of controlling the thought. Study 2 confirmed that autogenous and reactive obsessions lead to different ways of subsequent appraisal and control strategy on them. Autogenous obsessions led to high appraisal on control over thought and importance of thought, and frequent use of avoidant control strategies, while reactive obsessions linked with high appraisal on responsibility and frequent use of ‘confrontational control strategies’. Study 3 showed that an individual tends to select a different way of appraisal and control strategy according to the content characteristics of intrusive thoughts. These findings suggest that OCD can be divided into two subtypes which are quite different in their developing and maintaining mechanism as well as clinical features. Individuals with autogenous obsessions tend to perceive the thoughts to be ego-dystonic and irrational, attempt to expel or suppress them from consciousness, and frequently employ avoidant control strategies by covert, magical or superstitious compulsive behaviors. On the other hand, individuals with reactive obsessions tend to believe the thoughts to be relatively rational and realistic, although they frequently or superficially describe their thoughts as being irrational and absurd to clinicians. Therefore, they devote themselves to subsequent actual coping behaviors for preventing the unwanted possible consequences of the obsessive thoughts rather than for expelling the thoughts themselves. These two types of obsession seem to be maintained and exacerbated through different psychological mechanisms. Thought suppression and anxiety reduction have been regarded as important maintaining processes of obsessions. Intentional attempt to suppress unwanted thoughts were already known well with its paradoxical effect (Freeston et al., 1997; Salkovskis & Campbell, 1994; Trinder and Salkovskis, 1994; Wegner 1989, 1994; Wegner, Schneider, Carter, & White, 1987). Anxiety reduction through compulsive behavior and its result of negative reinforcement have been traditionally regarded as important factors making compulsion endure and blocking the opportunity for disconfirming faulty appraisal of obsession (Rachman 1997, 1998; Salkovskis 1985, 1989). We believe that thought suppression is involved in the maintenance of autogenous obsession, whereas anxiety reduction is related to the maintenance of reactive obsession, although both mechanisms can be involved in the maintenance of both two types of obsession. Individuals with autogenous obsessions attempt to suppress the thoughts in order to expel them from consciousness, but experience more intrusive obsessions due to the paradoxical rebound effect of

26

H.-J. Lee, S.-M. Kwon / Behaviour Research and Therapy 41 (2003) 11–29

thought suppression. Or they tend to employ avoidant control strategies by covert, magical or superstitious compulsive behaviors (e.g. counting numbers, praying, undoing the thought) because they have difficulties in finding meaningful relationships of the thought contents with evoking situations and overt actual coping methods toward external environments. On the other hand, individuals with reactive obsessions attempt to employ overt and actual compulsive behaviors (e.g. washing, checking, arranging, hoarding) for preventing the unwanted possible consequences of the thoughts. These compulsive behaviors are maintained not only by their reinforcing effect of anxiety reduction but also by blocking the opportunity for disconfirming irrationality of obsessional thoughts. These arguments suggest that OCD can be divided into the two subtypes which are mainly originated from autogenous obsessions and reactive obsessions. Thus far, a lot of attention for categorizing OCD into subtypes has been given to the subtypes of washing, checking, hoarding, and arranging according to the characteristics of compulsive behaviors (Antony, Downie, & Swinson, 1998; Rasmussen & Tsuang, 1986). And pure obsessional type has also been treated as a subtype which has no overt compulsion (Freeston, Ladouceur, Gagnon, & Thibodeau, 1997; Freeston & Ladouceur, 1999; Salkovskis, 1999; Zohar & Pato, 1991). In relation to these subtypes, autogenous obsessions might provide a base for developing pure obsession, while reactive obsessions might lead to the types of OCD which are accompanied by overt responses like washing, checking, arranging, or hoarding according to the contents of thoughts and their connected triggers. Taylor (1999), in his insightful comments on the six beliefs domains of OCD which were suggested by OCCWG (1997), described two hypothetical types of OCD patients. One type of patients overestimate threat, are intolerant of uncertainty, have inflated responsibility, have strong belief about the necessity of detecting and preventing harm from external sources, and are likely to suffer from overt compulsive behaviors. The other type of patients tend to strongly believe in the overimportance of thoughts, have the excessive need to control thoughts, and are likely to suffer from harm-related obsessions. His arguments appear to be very similar to our speculation on the two subtypes of OCD originated from reactive obsessions and autogenous obsessions. Differentiation between the two types of obsession may have implications on therapeutic intervention as well as future investigations on subtypes of OCD. Autogenous obsessions are likely to be difficult to be treated with the traditional way of exposure and response prevention (ERP) mainly based on the reinforcement paradigm, because they are frequently accompanied with few or covert compulsive behaviors which are inappropriate as targets of exposure. This can be one of the reasons why various behavioral treatments including ERP have been proved less successful when they were applied to pure obsession than when applied to overt compulsion (Clark, 1999; Freeston et al., 1997; Mark, 1981). Patients with autogenous obsessions may be better treated with cognitive-behavioral therapy (CBT) rather than with the traditional behavioral techniques, as Whittal and McLean (1999) said “CBT for OCD is in its infancy and promising, particularly for obsessions without overt compulsions% CBT approaches will be effective for some who would have failed at ERP” (p. 395). Autogenous obsessions in which the intrusive thoughts can be more clearly distinguished from the automatic thoughts are expected to respond to CBT approaches that discover and correct faulty appraisal of intrusion. On the other hand, reactive obsessions mostly accompanied with overt compulsion can be effectively dealt with behavioral approaches such as ERP, as supported in many outcome studies (Boersma, Den Hengst, & Dekker,

H.-J. Lee, S.-M. Kwon / Behaviour Research and Therapy 41 (2003) 11–29

27

1976; Emmelkamp & van Kraanen, 1977; Foa & Goldstein, 1978; Foa & Kozak, 1996; Meyer & Levy, 1973). However, cognitive approaches to reactive obsessions are likely to be more limited in their applicability and effects than to autogenous obsessions, because reactive obsessions not only have unclear boundary between their intrusive thoughts and automatic thoughts but also are perceived to be rather realistic and rational by patients. Differentiation between the two types of obsession are closely linked with the controversial issue about relationship between obsession and worry. Compared with obsession, worry is regarded as more verbally oriented, more realistic, and less involuntary (Langlois, Freeston, & Ladouceur, 2000; Turner, Beidel, & Stanley, 1992; Wells and Morrison, 1994). Also, worries are often precipitated by identifiable triggers in common circumstances of daily living, which are related to realistic experiences, and are associated with overt compulsions (Clark & Claybourn, 1997; Tallis & DeSilva, 1992; Wells & Morrison, 1994). The results of the present study indicate that, while autogenous obsessions are clearly different from worries, reactive obsessions are very similar to worries in their contents, appraisal and coping strategies. Tallis & DeSilva (1992) have also suggested that worry is similar to checking symptom in its preventive function. The controversial issues of high comorbidity between OCD and GAD or relationship between obsession and worry may result from the similarities between reactive obsession and worry. In a continuum from obsession to worry, autogenous obsessions are considered to be pure obsessions in an extreme, while reactive obsessions are seen to be hybrids between obsession and worry. Further investigations for clarifying differences between reactive obsession and worry are needed. Lastly, some limitations of the present study should be noted. As Clark & Purdon (1995) pointed out that many studies focusing on the appraisal and coping process of obsessive intrusions have weakness in their measures in terms of reliability and validity, the measures used in the present study are not sufficiently assured in their psychometric properties. The results of this study need to be confirmed in further studies using more validated measures and clinical population of OCD.

Acknowledgements Prof Sehee Hong and Prof Cheongtag Kim are thanked for their valuable comments on data analysis.

References American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: American Psychiatric Association. Amir, N., Cashman, L., & Foa, E. B. (1997). Strategies of thought control in obsessive–compulsive disorder. Behaviour Research and Therapy, 35, 775–777. Antony, M. M., Downie, F., & Swinson, R. P. (1998). Diagnostic issues and epidemiology in obsessive–compulsive disorder. In R. P. Swinson, M. M. Antony, S. Rachman, & M. Richter (Eds.), Obsessive–compulsive disorder; theory, research, and treatment (pp. 3–32). New York: Guilford Press. Arbuckle, J., & Wothke, W. (1999). Amos 4.0 user’s guide. Chicago: Smallwaters Corperation. Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological Bulletin, 107, 238–246.

28

H.-J. Lee, S.-M. Kwon / Behaviour Research and Therapy 41 (2003) 11–29

Bentler, P. M., & Bonett, D. G. (1980). Significance tests and goodness of fit in the analysis of covariance structure. Psychological Bulletin, 88, 588–606. Boersma, K., Den Hengst, S., & Dekker, J. (1976). Exposure and response prevention: a comparison with obsessive– compulsive patients. Behaviour Research and Therapy, 14, 19–24. Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A. Bollen, & J. S. Long (Eds.), Testing structural equation models (pp. 136–162). Newbury Park, CA: Sage. Clark, D. A. (1992). Depressive, anxious and intrusive thoughts in psychiatric inpatients and outpatients. Behaviour Research and Therapy, 30, 93–102. Clark, D. A. (1999). Cognitive behavioral treatment of obsessive–compulsive disorders; A commentary. Cognitive and Behavioral Practice, 6, 408–415. Clark, D. A., & Claybourn, M. (1997). Process characteristics of worry and obsessive intrusive thoughts. Behaviour Research and Therapy, 35, 1139–1141. 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. Emmelkamp, P. M. G., & van Kraanen, J. (1977). Therapist-controlled exposure in vivo; a comparison with obsessive– compulsive patients. Behaviour Research and Therapy, 15, 491–495. Foa, E. B., & Goldstein, A. (1978). Continuous exposure and complete response prevention of obsessive–compulsive disorder. Behaviour Therapy, 9, 821–829. Foa, E. B., & Kozak, M. J. (1996). Psychological treatment for obsessive–compulsive disorder. In M. R. Mavissakalian, & R. F. Prien (Eds.), Long-term treatments of anxiety disorders (pp. 285–309). Washington, DC: APA. 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. (1997). What do patients do with their obsessive thoughts? Behaviour Research and Therapy, 35, 335–348. Freeston, M. H., & Ladouceur, R. (1999). Exposure and response prevention for obsessive thoughts. Cognitive and Behavioral Practice, 6, 362–383. Freeston, M. H., Ladouceur, R., Gagnon, F., & Thibodeau, N. (1997). Cognitive-behavioral treatment of obsessive thoughts: A controlled study. Journal of Consulting and Clinical Psychology, 65, 405–413. Freeston, M. H., Ladouceur, R., Langlois, F., Dugas, M. J., Rheaume, J., & Leger, E. (1997). Intrusive cognitions: specific applications of a generalized model. Communication presented at the Congress of the European Association of Behavioural and Cognitive Therapies, Venice, Italy, September 1997. Freeston, M. H., Ladouceur, R., Provencher, M., & Blais, F. (1995). Strategies used with intrusive thoughts; context, appraisal, mood, and efficacy. Journal of Anxiety Disorders, 9, 201–215. 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. II. Associations with depressive, anxious, and compulsive symptoms. Behaviour Research and Therapy, 30, 263–271. Jo¨ reskog, K. G., & So¨ rbom, D. (1996). LISREL 8: user’s reference guide. Chicago: Scientific Software International. Langlois, F., Freeston, M. H., & Ladouceur, R. (2000). Differences and similarities between obsessive intrusive thoughts and worry in a non-clinical population: study 1. Behaviour Research and Therapy, 38, 157–173. Lee, H.J., Kwon, S.M. Assessment of appraisal and control strategies on obsessive intrusions. Unpublished manuscript, 1999. Mark, I. M. (1981). Review of behavioural psychotherapy, I: Obsessive–compulsive disorders. American Journal of Psychiatry, 138, 584–592. Meyer, V., & Levy, R. (1973). Modification of behavior in obsessive–compulsive disorders. In H. E. Adams, & P. Unikel (Eds.), Issues and trends in behavior therapy (pp. 77–136). Springfield, IL: Charles C Thomas. Obsessive Compulsive Cognitions Working Group (1997). Cognitive assessment of obsessive–compulsive disorder. Behaviour Research and Therapy, 35, 667–681. Purdon, C. L., & Clark, D. A. (1994b). Perceived control and appraisal of obsessional intrusive thoughts: a replication and extension. Behavioural and Cognitive Psychotherapy, 22, 269–285. Purdon, C. L., & 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.

H.-J. Lee, S.-M. Kwon / Behaviour Research and Therapy 41 (2003) 11–29

29

Purdon, C. L., & 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. 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. Rasmussen, S. A., & Tsuang, M. T. (1986). Clinical characteristics and family history in DSM-III obsessive–compulsive disorder. American Journal of Psychiatry, 143, 317–322. Salkovskis, P. M. (1985). Obsessional–compulsive problem: A cognitive-behavioral 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. (1999). Understanding and treating obsessive–compulsive disorder. Behaviour Research and Therapy, 37, S29–S52. Salkovskis, P. M., & Campbell, P. (1994). Thought suppression induces intrusion in naturally occurring negative intrusive thoughts. Behaviour Research and Therapy, 32, 1–8. Tallis, F., & DeSilva, P. (1992). Worry and obsessional symptoms: A correlational analysis. Behaviour Research and Therapy, 30, 103–105. Taylor, S. (1999). Special series: treatment of obsessive-compulsive disorder: progress, prospect, and problems; Introduction. Cognitive and Behavioral Practice, 6, 342–344. Trinder, H., & Salkovskis, P. M. (1994). Personally relevant intrusions outside the laboratory: long-term suppression increases intrusion. Behaviour Research and Therapy, 32, 833–842. Turner, S. M., Beidel, D. C., & Stanley, M. A. (1992). Are obsessional thoughts and worry different cognitive phenomena? Clinical Psychology Review, 12, 257–270. Wegner, D. M. (1989). White bears and other unwanted thoughts. New York: Viking. Wegner, D. M. (1994). Ironic processes of mental control. Psychological Review, 101, 34–52. Wegner, D. M., Schneider, D. J., Carter, S. R., & White, T. L. (1987). Paradoxical effects of thought suppression. Journal of Personality and Social Psychology, 53, 5–13. Wells, A., & Davies, M. (1994). The thought control questionnaire: a measure of individual differences in the control of unwanted thoughts. Behaviour Research and Therapy, 32, 871–878. Wells, A., & Morrison, A. P. (1994). Qualitative dimensions of normal worry and normal obsessions: a comparative study. Behaviour Research and Therapy, 32, 867–870. Whittal, M. L., & McLean, P. D. (1999). CBT for OCD: the rationale, protocol, and challenges. Cognitive and Behavioral Practice, 6, 383–396. Zohar, J., & Pato, M. T. (1991). Diagnostic considerations. In M. T. Pato, & J. Zohar (Eds.), Current treatment of obsessive–compulsive disorder (pp. 1–12). Washington, DC: APA.