- Email: [email protected]
The effects of psychoeducation on thought-action fusion, thought suppression, and responsibility
J. Behav. Ther. & Exp. Psychiat. 41 (2010) 289e296
Contents lists available at ScienceDirect
Journal of Behavior Therapy and Experimental Psychiatry journal homepage: www.elsevier.com/locate/jbtep
The effects of psychoeducation on thought-action fusion, thought suppression, and responsibility Teresa Marino-Carper*, Charles Negy, Gillian Burns, Rachael A. Lunt Department of Psychology, University of Central Florida, P.O. Box 161390, Orlando, FL 32816-1390, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 October 2009 Received in revised form 19 January 2010 Accepted 11 February 2010
The current study examined the effects of a psychoeducational intervention designed to target thoughtaction fusion (TAF) on TAF, thought suppression, and responsibility cognitions. 139 undergraduate students (25 male; 114 female) who were relatively high in TAF with respect to their peers served as participants. Immediately following intervention, individuals who had received psychoeducation regarding TAF reported significantly lower morality TAF scores than individuals who had received psychoeducation regarding thoughts in general and individuals in the control group. At the two-week follow-up assessment, the likelihood TAF scores of those who had received psychoeducation regarding TAF were significantly lower than those of the control group. In addition, the group that received psychoeducation regarding TAF was the only group that did not experience a significant increase in thought suppression from baseline to post-intervention, and was also the only group to experience an increase in both frequency of and belief in low-responsibility thoughts from baseline to follow-up. Implications are discussed. Published by Elsevier Ltd.
Keywords: Thought-action fusion Psychoeducation Intrusive thoughts Cognitive intervention
1. Introduction The tendency to fuse one's thoughts with overt behaviors is a cognitive bias thought to underlie a variety of psychopathologies (Berle & Starcevic, 2005; Rassin, Diepstraten, Merckelbach, & Muris, 2001; Shafran & Rachman, 2004; Shafran, Teachman, Kerry, & Rachman, 1999). This bias, referred to as thought-action fusion (TAF), is believed to be comprised of two distinct elements (Shafran, Thordarson, & Rachman, 1996). The likelihood component of TAF refers to the belief that experiencing an intrusive and distressing thought can increase the probability that the event will actually occur, causing harm to oneself or others. The second component, morality, refers to the belief that thinking about an action is ethically equivalent to actually performing the action. Rachman (1993) first introduced the concept of TAF in the context of its occurrence in obsessive-compulsive disorder (OCD), and later concluded that TAF increases one's susceptibility for developing OCD (Rachman & Shafran, 1999). Subsequent literature has collectively suggested that TAF is a fairly extensive construct that may be present in myriad disturbances, including Generalized Anxiety Disorder, Post-Traumatic Stress Disorder, Panic Disorder, Social Phobia, eating disorders, and depression (Berle & Starcevic,
* Corresponding author. Tel.: þ1 617 909 6317; fax: þ1 407 823 5862. E-mail address: [email protected] (T. Marino-Carper). 0005-7916/$ e see front matter Published by Elsevier Ltd. doi:10.1016/j.jbtep.2010.02.007
2005; Rassin, Diepstraten et al., 2001; Shafran & Rachman, 2004; Shafran et al., 1999). However, TAF is not exclusive to people with mental health problems, and is prevalent in non-clinical populations as well (Zucker, Craske, Barrios, & Holguin, 2002). An inflated sense of personal responsibility has long been thought to play an important role in the onset and maintenance of OCD (Salkovskis & Kirk, 1989). In order to operationalize the construct of dysfunctional responsibility and clarify its role in obsessional thoughts, Rachman et al. (1995) developed the Responsibility Appraisal Questionnaire (RAQ), a 36-item measure designed to assess responsibility outside of the OCD realm. Using a sample of undergraduate students, the authors found a fourfactor solution for the RAQ, suggesting the existence of four discrete components of responsibility: responsibility for harm, social responsibility, a positive attitude towards responsibility, and TAF. The authors then administered the RAQ and measures of obsessivecompulsive symptoms, depression, guilt, and obsessional beliefs to a different sample of undergraduate students. The results of this second analysis corroborated the four-factor solution, suggesting that TAF is a component of exaggerated responsibility. Additionally, TAF was found to be significantly related to obsessive-compulsive symptoms, symptoms of depression, guilt, and obsessional beliefs. These correlations remained significant even after controlling for the effects of depression symptoms. Smári and Hólmsteinsson investigated the relationships among TAF, responsibility, intrusive thoughts, thought suppression, and
290
T. Marino-Carper et al. / J. Behav. Ther. & Exp. Psychiat. 41 (2010) 289e296
obsessive-compulsive symptoms among undergraduate students. The results indicated that both responsibility attitudes and thought suppression mediated the relationship between intrusive thoughts and obsessive-compulsive symptoms. Interestingly, the researchers also found that controlling for TAF in the mediation analyses led to strikingly similar results to those obtained when controlling for responsibility, suggesting that TAF is highly related to responsibility. These results are consistent with those obtained by Marino, Lunt, and Negy (2008), who, using a structural equation model, found that an inflated sense of responsibility significantly predicted TAF. Individuals often engage in cognitive avoidance to cope with the distress of an intrusive thought (Rassin, 2001). Rassin, Muris, Schmidt, and Merckelbach (2000) attempted to determine directionality in the relationship among TAF, thought suppression, and obsessive-compulsive symptoms using structural equation modeling with a sample of undergraduate students. The results indicated a model in which TAF leads to attempts at thought suppression, which in turn, predicts more obsessive-compulsive symptoms. These results support those found by Rassin, Merckelbach, Muris, and Spaan (1999) that suggested TAF may play a causal role in the development of intrusive thoughts. Rassin (2001) empirically examined the influence of TAF and thought suppression on the development of obsession-like intrusions in undergraduate students. TAF was experimentally induced for each individual via the Sentence Completion Task (Rachman, Shafran, Mitchell, Trant, & Teachman, 1996). The Sentence Completion Task involves asking the participant to identify a loved one, subsequently presenting the participant with a form on which the sentence “I hope _______ is in a car accident” is typed, and then asking the individual to rewrite the sentence with their loved one's name in the blank. Afterwards, the participant is asked to visualize the event occurring. Half the participants in Rassin's study were asked to suppress thoughts of the accident while sitting quietly for five minutes, whereas the other half were explicitly told not to try to suppress any thoughts of the accident during this time. Contrary to prediction, those who were instructed to suppress did not have significantly higher anxiety scores or report an increased number of intrusive thoughts. Moreover, they spent less time thinking about the accident and rated the writing of the sentences as less morally wrong than did participants in the non-suppression group. These findings are discordant with the ‘white bear effect’ found by Wegner et al. (1987) that states that attempts at thought suppression are counterproductive insofar as they tend to result in a greater number of intrusions. Rassin attempted to reconcile this discrepancy by arguing that the paradoxical effects of thought suppression may not surpass the effects of TAF. The author further noted that even those in the non-suppression group reported having attempted to suppress their thoughts, suggesting that TAF may inevitably induce thought suppression. Rassin, Diepstraten et al. (2001) found that cognitive-behavioral interventions significantly reduced TAF in inpatients diagnosed with anxiety disorders. This demonstration that TAF is amenable to change sparked an interest in therapeutic interventions for TAF. The effectiveness of a psychoeducational intervention aimed at minimizing the endorsement of TAF was first examined by Zucker et al. (2002). Undergraduates who scored relatively high on the Thought Action Fusion e Revised scale (Shafran et al., 1996) were randomly assigned to be in an experimental group or a control group. All participants were given measures to assess TAF and current anxiety. The experimental group heard a psychoeducational message regarding TAF that emphasized the normality and irrelevance of intrusive thoughts. Those in the control group heard a message regarding stress, its consequences, and techniques to alleviate it. After the manipulation, participants again completed the measure
assessing TAF to assess for change in endorsement. Next, the participants completed the aforementioned Sentence Completion Task used by Rassin (2001) to induce TAF, and subsequently rated their current levels of anxiety. The researchers found a significant interaction effect between group and point of assessment. Analyses of the simple main effects showed that TAF endorsement significantly decreased from pre-task to post-task in the experimental group, but not in the control group. For anxiety that was assessed by a visual analogue scale, there was also a significant interaction effect between group and time, suggesting that although both groups experienced an increase in anxiety from pre-task to posttask, the control group experienced a significantly higher increase in anxiety than the experimental group. An analysis of simple effects for this interaction revealed no significant differences between the two groups in anxiety at baseline, but there was a trend towards significance for the control group to have higher anxiety than the experimental group at post-task. For anxiety that was assessed by a state anxiety measure, there was no significant interaction effect between group and time. There was, however, a significant effect of time, meaning that the anxiety levels of both groups significantly increased from pre-task to post-task. Zucker et al. concluded that these results suggest that mere psychoeducation can impact individuals' appraisals of a disturbing, intrusive thought. Zucker, Craske, Blackmore, and Nitz (2006) followed this work by investigating the effects of a cognitivebehavioral workshop in individuals with subclinical OCD. The findings suggested that those who attended the workshop had lower average TAF scores both one month and five months after the intervention, suggesting that the effects of the intervention demonstrated longevity. Despite the growing body of evidence suggesting that TAF underlies myriad psychological disorders, there is a paucity of research into effective methods for targeting this cognitive bias. In addition, given that thought suppression and inflated responsibility are thought to contribute to the maintenance of TAF, it is important to determine the extent to which a TAF intervention will also decrease these influencing constructs. As such, the purpose of this study was to extend Zucker et al.'s (2002) work by examining the relative effects of a brief intervention designed to specifically target TAF on not only TAF itself, but also the closely related phenomena of thought suppression and responsibility. In addition, the current study aimed to examine the stability of these effects by assessing the outcome variables both immediately following the intervention and after a period of two weeks. The interventions were in the form of a psychoeducational message that referred to either TAF, intrusive thoughts alone, or general stress. The messages regarding TAF and general stress were taken directly from Zucker et al.'s investigation. The intervention novel to the current study was the psychoeducation that discussed the presence of intrusive thoughts in a non-evaluative manner, and was included to help dismantle Zucker et al's findings that the invention explicitly targeting TAF lowered TAF endorsement. By including a condition that discussed the presence of intrusive thoughts more generally, our goal was to ascertain if merely attending to and being more aware of one's thoughts could lower endorsements of the dependent variables, or if it is the information regarding the irrationality of the thoughts that is the critical component. The current study was designed to add to the literature in several ways. First, given that TAF is prevalent in myriad psychopathologies, it is important to examine the extent to which the effects of a TAF intervention are stable in individuals with high levels of TAF in general, as opposed to only within the context of subclinical OCD. Secondly, by including a condition in which participants receive psychoeducation about intrusive thoughts in a non-evaluative manner, the current study might elucidate the
T. Marino-Carper et al. / J. Behav. Ther. & Exp. Psychiat. 41 (2010) 289e296
extent to which addressing the irrationality of TAF, as opposed to merely increasing metacognitive awareness, is crucial to intervention. Third, investigating moral and likelihood TAF separately allows for the examination of any differential effects of the intervention on these two components. Finally, no study has examined how such an intervention may reduce thought suppression and responsibility cognitions, constructs believed to partially underlie the development and maintenance of TAF. In light of the literature reviewed, we hypothesized that participants in the group exposed to the psychoeducational message on TAF would score significantly lower on measures of both TAF subscales than participants in the group exposed to the psychoeducational message on intrusive thoughts only and participants in the control group, both immediately following the intervention and at the two-week follow-up. It was further postulated that, by proxy of their association with TAF, thought suppression and responsibility would significantly decrease from baseline to post-intervention in the group provided with psychoeducation on TAF, but not in the other two groups, and these decreases would remain at the two-week follow-up period. 2. Methods 2.1. Participants A total of 4376 students enrolled in undergraduate psychology courses were screened for subsequent selection by completing a preliminary informed consent form, demographics questionnaire, and the Thought-Action Fusion Scale e Revised (Shafran et al., 1996; see description below) online in exchange for extra credit towards their respective courses. Of those students, 756 (17.28%) scored at least one standard deviation above the mean of the TAF-R composite scale as established by prior research (i.e., Marino et al., 2008), and were invited to participate in the study. Of the 167 participants who chose to begin the protocol, 15 failed the manipulation check, and 15 (2 of whom also failed the manipulation check) did not complete the Time 2 portion. Thus, the final study sample used in data analysis was composed of 139 (25 male and 114 female) participants. Regarding ethnicity, 73 (52.5%) of the students self-identified as European American, 22 (15.8%) as Latino American, 26 (18.7%) as African American, 11 (7.9%) as Asian American, and 7 (5%) as “other.” With respect to age, 121 (87.1%) of participants were between the ages of 18 and 20, 14 (10.1%) were between 21 and 25, 3 (2.2%) were between 26 and 30, and 1 (0.7%) was between the ages of 31 and 40. 2.2. Materials 2.2.1. Thought-action fusion scale e revised (TAF-R) The TAF-R (Shafran et al., 1996) is a 19-item measure of the extent to which individuals equate their thoughts with actions. Participants respond using a five-point Likert-type scale, with higher scores reflecting a greater proclivity to engage in TAF. In addition to a composite score, ‘moral’ and ‘likelihood’ subscale scores may also be tabulated. The TAF e R has been demonstrated to be psychometrically valid for use in a student population (Shafran et al., 1996). For the current study, the Cronbach alpha coefficients of the TAF-R ranged from 0.88 to 0.94 for the moral subscale, and from 0.95 to 0.97 for the likelihood subscale. Rassin, Merckelbach, Muris, and Schmidt (2001) found significant retest correlations for the TAF-R composite scale, moral subscale, and likelihood subscale (rs ¼ 0.52, 0.54, and 0.51, respectively; all ps < 0.01) among a sample of undergraduate students. However, the mean score at the retest administration was significantly lower than the mean score at baseline, suggesting that TAF may be somewhat unstable in a non-clinical population.
291
2.2.2. Responsibility interpretations questionnaire (RIQ) The RIQ (Salkovskis et al., 2000) is comprised of 22 items designed to measure the frequency of and belief in high responsibly thoughts (e.g., “It would be irresponsible to ignore these thoughts”) and low-responsibility thoughts (e.g., “This is just a thought so it doesn't matter”). The combination of these dimensions yields four subscales, with higher scores on the two subscales that assess high responsibility indicating a greater sense of responsibility, and higher scores on the two subscales that assess low responsibility indicating a lesser sense of responsibility. Salkovskis et al. found the RIQ to an internally consistent instrument that is valid for use with a non-clinical population. For the current study, the Cronbach alpha coefficients of the RIQ ranged from 0.90 to 0.94 for the frequency of high-responsibility thoughts subscale, from 0.83 to 0.90 for the frequency of low-responsibility thoughts subscale, from 0.90 to 0.95 for the belief in high-responsibility thoughts subscale, and from 0.85 to 0.92 for the belief in low-responsibility thoughts subscale. Salkovskis et al. examined the temporal stability of the RIQ over a two-week period, and found favorable test-retest reliability for the subscales measuring the frequency of and belief in high-responsibility thoughts (r ¼ 0.90 and 0.80, respectively), but not for the subscales measuring the frequency of and belief in lowresponsibility thoughts (r ¼ 0.69 and 0.22, respectively). The authors surmised that, given the favorable internal consistency of the two low-responsibility subscales, the lack of stability likely reflects actual instability in the frequency of and belief in lowresponsibility cognitions. 2.2.3. White Bear Suppression Inventory (WBSI) The WBSI (Wegner & Zanakos, 1994) is a 15-item questionnaire designed to assess thought suppression. Participants respond to the items using a five-point Likert-type scale, with a higher score indicating a stronger tendency to engage in thought suppression. The WBSI has been found to have favorable psychometric properties (e.g., Muris, Merckelbach, & Horselenberg, 1996; Smári & Rafnsson, 2001; Spinhoven & van der Does, 1999), and achieved Cronbach alpha coefficients ranging from 0.90 to 0.93. for the current sample. Muris et al. (1996) found that the WBSI demonstrated good stability over a 12-week period, with a test-retest correlation coefficient of 0.80. 2.2.4. Sentence completion form The Sentence Completion Form was used as part of Rachman et al's (1996) aforementioned TAF-inducing Sentence Completion Task. The form consists of the line “I hope _______ is in a car accident”, and a space underneath in which the participant is asked to rewrite the sentence with their loved one's name in the blank. 2.3. Study design and procedure The present design was a 3 (time: baseline, post-manipulation 1, post-manipulation 2) 3 (psychoeducation condition: Group A, Group B, Group C) between-subjects and within-subjects design. In light of our objective of expanding upon the work of Zucker et al. (2002), the current study adopted similar methodology. Individual data collection sessions occurred within the psychology department and lasted approximately a half an hour. Each participant provided informed consent prior to beginning the study. Four individuals elected to terminate participation after being asked to complete the sentence completion task, indicating that doing so would cause marked discomfort. Study procedures were immediately terminated for these individuals, and the researcher verbally assessed the nature and extent of the individual's anxiety. All four participants who chose to terminate denied experiencing excessive distress and declined counseling referrals. At the end of each data
292
T. Marino-Carper et al. / J. Behav. Ther. & Exp. Psychiat. 41 (2010) 289e296
collection session, the researcher verbally assessed the distress level of each participant, all of whom denied experiencing excessive distress and declined counseling referrals. Participants visited the research laboratory on two occasions. Each participant was randomly assigned into one of three groups: a group that received psychoeduation on TAF (Group A; N ¼ 41), a group that received psychoeducation on thoughts and cognitions (Group B; N ¼ 47), and a control group (Group C; N ¼ 51) that received psychoeducation on stress. Participants were asked to complete the RIQ, TAF-R, and WBSI. Next, participants in Group A received psychoeducation regarding intrusive thoughts and TAF in the form of the following digitally-recorded message (taken from Zucker et al., 2002, p. 656): It is estimated that approximately 80e90% of the population has intrusive thoughts. Intrusive thoughts are often unpleasant thoughts that come to your mind out of the blue. For example, you might be driving on a bridge and have a fleeting though about driving off the bridge, or you might have a passing thought about causing harm to a helpless person. Or, you might be in a serious relationship and have unpleasant sexual thoughts about people other than your partner. Often, when people have these bothersome thoughts they feel as though somehow, their thoughts will make the event more likely to happen. This way of thinking is wrong. Why is this way of thinking wrong? By simply thinking about a friend or family member becoming seriously ill, you will not increase the chance that they will become sick. Your thinking has no effects whatsoever on their health. Also unpleasant sexual thoughts alone do not mean you are an evil, bad or immoral person. Remember, it is completely normal to have these types of thoughts. These types of thoughts have no influence over outside events, nor are they a reflection of your character. Participants in Group B received psychoeducation regarding intrusive thoughts in the form of the following digitally-recorded message (developed by the present authors): Most of the time, people have control over their thoughts, and can purposely choose to think about or not think about whatever they want. Sometimes, however, thoughts seem to pop into people's minds without them having intentionally brought on the thought. For example, you may be engaged in a conversation and all of a sudden an irrelevant thought will pop into your mind. These kinds of thoughts that come into your mind out of the blue are called intrusive thoughts. An intrusive thought may be pleasant to the person, or may not cause any emotion at all. At times, intrusive thoughts can also be unpleasant and upsetting. For example, you might be driving on a bridge and have a fleeting thought about driving off the bridge, or you might have a passing thought about causing harm to a helpless person. Or, you might be in a serious relationship and have unpleasant sexual thoughts about people other than your partner. Intrusive thoughts can be very scary and disturbing to the individual. These types of thoughts may cause stress and anxiety. Participants in Group C, the control group, received psychoeducation about stress in the form of the following digitallyrecorded message (taken from Zucker et al., 2002, p. 656): There are many types of situations that cause people to experience increased levels of stress. These situations could range anywhere from driving in heavy traffic to being fired from a job to going away to college for the first time. Maybe people around the world commonly experience some degree of stress on a regular basis. When people feel stressed they might have accompanying physical problems, such as headaches, increased muscle tension and stomachaches. It even could lead to more serious medical problems such
as high blood pressure and ulcers. Many times when people experience these types of physical problems, they don't realize that they could be stress-related. Although stress cannot be fully eliminated from day-to-day life, people should try their best to minimize their amount of stress in their lives and engage in stress management techniques. Stress management can target a variety of lifestyle changes, including increasing exercise and daily activity, eating healthy and establishing regular sleeping patterns. By actively seeking to reduce and better manage stress, people are able to greatly improve the quality of their lives. After listening to the message, all participants completed a manipulation check to ensure that they had attended to and understood the psychoeducation provided. This manipulation check consisted of 3 true/false condition-specific questions regarding the message content (e.g., “It is very common to have unpleasant thoughts come into your mind out of the blue” for Group A). Any incorrectly answered question resulted in a failed manipulation check; of the 15 individuals who failed the manipulation check, 11 answered 1 question incorrectly, 3 answered 2 questions incorrectly, and 1 answered all 3 questions incorrectly. Next, all participants were asked to perform the Sentence Completion Task (Rachman et al., 1996) for the purposes of experimentally inducing TAF. Given that TAF is a trait-like construct that becomes activated in response to a disturbing and intrusive thought, the Sentence Completion Task is often used in the literature as a means of generating the same troubling cognition for all participants, at a uniform point in time during the study protocol. Following the task, all participants again completed the RIQ, TAF-R, and WBSI. The instructions for these measures were adapted to reflect the brief time period between administrations. Two weeks after Time 1, participants returned for the Time 2 procedures. All participants were again asked to perform the sentence completion task followed by the RIQ, TAF-R, and WSBI. Participants were then debriefed and those who were in Groups B and C listened to the psychoeducational message about TAF, for the possibility of therapeutic benefit. 3. Results 3.1. Initial data analysis A multivariate analysis of variance (MANOVA) revealed no significant between-group differences on the baseline measurements of the TAF-R subscales, WBSI, and RIQ subscales (F [14, 230] ¼ 0.66, ns). A one-way analysis of variance (ANOVA) confirmed that there were no significant between-group differences for age (F [2, 136] ¼ 0.04, ns), and a Pearson chi-squared test revealed no significant differences for sex (c2 [2] ¼ 0.75, ns). Of note, the baseline TAF-R mean for the sample used in the current study was 41.20, which is markedly higher than the normative TAF-R means established by Shafran et al. (1996) of 29.21 for individuals with clinical levels of obsessionality and 16.82 for an undergraduate population. Thus, it appears that the screening procedure employed in the current study was effective method of recruiting a sample relatively high in TAF. 3.2. Effects of interventions on TAF The first hypothesis predicted that those in Group A would score significantly lower on measures of both TAF subscales than those in Groups B and C, both immediately following the intervention and at the two-week follow-up. To test this hypothesis, the morality and likelihood subscales of the TAF-R were analyzed separately using a 3 3 mixed model univariate analysis of variance (ANOVA;
T. Marino-Carper et al. / J. Behav. Ther. & Exp. Psychiat. 41 (2010) 289e296
293
group time). A significant interaction effect was found for morality TAF (F [4, 258] ¼ 3.86, p < 0.01, partial h2 ¼ 0.06). To examine between-group differences at the post-intervention and follow-up assessments of morality TAF, post-hoc pairwise comparisons were conducted using the Bonferroni correction. These analyses revealed that individuals in Group A had significantly lower scores at post-intervention than individuals in both Group B (Dx ¼ 5:43, p < 0.05) and Group C (Dx ¼ 5:92, p < 0.05). No significant differences were observed at the two-week followup for morality TAF. The results of this interaction are depicted in Fig. 1. There was also a significant interaction effect for likelihood TAF (F [4, 262] ¼ 4.39, p < 0.01, partial h2 ¼ 0.06). Post-hoc analyses revealed no significant between-group differences at post-intervention for likelihood TAF. However, at follow-up, Group A had significantly lower scores than Group C (Dx ¼ 4:75, p < 0.01). The results of this interaction are depicted in Fig. 2.
frequency of low-responsibility thoughts, results yielded a significant interaction effect (F [3.77, 248.89] ¼ 3.07, p < 0.05, partial h2 ¼ 0.04). Post-hoc analyses were conducted using the Bonferroni correction to examine within-group differences for all four subscales between baseline and post-intervention, and baseline and follow-up. For belief in low-responsibility thoughts, individuals in Group A experienced a significant increase from baseline to follow-up (Dx ¼ 15:24, p < 0.01), whereas individuals in Groups B and C did not. For frequency of low-responsibility thoughts, individuals in Group A experienced a significant increase from baseline to follow-up (Dx ¼ 0:82, p < 0.001), as well as from post-intervention to follow-up (Dx ¼ 0:55, p < 0.01), whereas individuals in Groups B and C did not. Cell means and standard deviations for all subscales of the RIQ are provided in Table 1.
3.3. Effects of Interventions on thought suppression and responsibility
The finding that individuals receiving psychoeducation regarding TAF were less likely to endorse morality-related TAF beliefs immediately following the intervention than individuals in the control group is consistent with previous work (i.e., Zucker et al., 2002). Consistent with our hypothesis, psychoeducation regarding TAF was also superior to psychoeducation regarding thoughts in general for morality-related TAF beliefs. Thus, it appears that identification and explanation of the irrationality of these thoughts is a vital component of the intervention. These findings were not present at follow-up, suggesting that the effects of providing TAF psychoeducation on the morality component of TAF diminish over time. Surprisingly, there were no differences among the groups in likelihood TAF immediately following the intervention. However, at follow-up, the individuals who had received psychoeducation regarding TAF were significantly less likely to endorse likelihoodrelated TAF beliefs than individuals in the control group. It may, therefore, be the case that, to be effective, the message that experiencing an intrusive thought does not increase the probability of the thought actually occurring needs to be paired with evidence (i.e., two-weeks elapsing without the participant's loved one being involved in a car accident). However, the significant group differences suggest that this evidence alone is not as effective at reducing TAF-likelihood endorsement as it is when paired with psychoeducation regarding TAF. These findings suggest that the effects of this brief intervention have some degree of longevity on the likelihood component of TAF. Although previous research has demonstrated the immediate effects of a brief intervention on TAF endorsement (Zucker et al., 2002) as well as longer-term effects of a more intensive
The second hypothesis predicted that the post-intervention and follow-up RIQ and WBSI scores would be significantly lower than baseline for Group A, but not for Groups B or C. To examine changes in thought suppression, a 3 3 mixed model univariate ANOVA (group time) was conducted using the WBSI as the DV. Due to a sphericity violation, the Greenhouse-Geisser correction was used in the analysis (Katz, 2006). Results revealed no significant interaction effect (F [3.56, 234.60] ¼ 0.43, ns). Given that a priori hypotheses justify examination of post-hoc tests, multiple comparison tests were conducted using the Bonferroni correction to examine within-group differences between baseline and postintervention, and baseline and follow-up. For both Groups B and C, WBSI scores significantly increased from baseline to post-intervention (both ps < 0.05), although this was not the case for participants in Group A. No significant differences from baseline to follow-up assessment on the WBSI were found for any of the groups. Cell means and standard deviations for the WBSI are provided in Table 1. To examine changes in responsibility, four separate mixed model univariate ANOVAs (group time) were conducted using each of the four components of the RIQ as the DV. The GreenhouseGeisser correction was used to correct for a sphericity violation in the analyses concerning frequency of low-responsibility thoughts and belief in high-responsibility thoughts. No significant interaction effect was found for frequency of high responsibility thoughts (F [4, 238] ¼ 0.21, ns), or belief in either high (F [3.79, 221.59] ¼ 0.13, ns) or low (F [4, 264] ¼ 0.85, ns) responsibility thoughts. For
4. Discussion
Fig. 1. Interaction between intervention group and time of assessment on TAF-moral. Note. Standard errors for each mean are represented in parenthesis.
294
T. Marino-Carper et al. / J. Behav. Ther. & Exp. Psychiat. 41 (2010) 289e296
Fig. 2. Interaction between intervention group and time of assessment on TAF-likelihood. Note. Standard errors for each mean are represented in parenthesis.
intervention on TAF endorsement (Zucker et al., 2006), the current study is the first to establish that the effects of a brief intervention can be sustained for at least a two-week period. Moreover, although two weeks is a relatively short period of time, these results have potential clinical significance. Given that many psychotherapy sessions occur on a weekly basis, therapeutic interventions aimed at dismantling the faulty cognitive assumptions underlying TAF may reduce some portion of TAF for the intersession duration. The fact that the effects of the intervention on the morality and likelihood components of TAF displayed different patterns was somewhat surprising. It appears that the impact of the psychoeducation regarding TAF on morality diminished over time, but the effects on likelihood strengthened over time. Again, it may be that the two-week interval was beneficial for TAF-likelihood insofar as the intervention message was supported by the participants' experience (i.e., they received evidence that the thought regarding the loved one in the car accident did not cause the event to occur). Conversely, the lack of a comparable form of concrete evidence supporting the idea that such thoughts do not make one immoral may help explain why the effects of the intervention on TAF-moral weakened over time. The overall superiority of the TAF psychoeducation over the other two interventions may be due in part to its approximation to a cognitive intervention. In particular, the message explicitly highlights the irrationality of TAF, challenges the cognitive errors in both the morality (i.e., “unpleasant sexual thoughts alone do not mean you are an evil, bad or immoral person”) and likelihood (i.e., “by simply thinking about a friend or family member becoming seriously ill, you will not increase the chance that they will become sick ”) components, and may thus encourage participants to restructure their own cognitions. Therefore, it is uncertain the extent to which the psychoeducative quality of message, as opposed to its similarity to cognitive-based intervention, accounts for the variance in TAF-R scores. Since the psychoeducation regarding TAF was more effective than the psychoeducation regarding intrusive thoughts in general for reducing TAF-morality in short-term and TAF-likelihood in the longer term, it may be that cognitive-based interventions are more appropriate for TAF than interventions that emphasize awareness of one's thoughts. Clearly, it cannot be inferred from the current study that such mindfulness-type approaches are inferior to cognitive-based interventions at reducing TAF; however, the present study does offer some evidence that merely being attending to one's thoughts and being exposed to a non-evaluative description of intrusive thoughts does not significantly impact either component of TAF (although the individuals in Group B may have benefitted from the intervention if it had also highlighted the
importance of accepting these intrusive thoughts in a non-judgmental manner, a technique that is often emphasized in mindfulness-type interventions). With respect to thought suppression, although we predicted that only those who received psychoeducation regarding TAF would experience a decrease from baseline to post-intervention, participants in that group were the only ones who did not experience an increase. Given that thought suppression often occurs following an intrusion (Rassin, 2001; Rassin et al., 2000), it may be the case that those who received the psychoeducation regarding either thoughts in general or stress were actively suppressing thoughts of the sentence they had just written. Conversely, the psychoeducation regarding TAF may have served as a buffer against the increase in thought suppression. If it is the case that thought suppression increases the frequency of intrusive thoughts, the decrease in susceptibility to engaging in thought suppression may facilitate a decrease in the frequency of intrusive, disturbing thoughts. The current findings offer some substantiation for Rassin's (2001) theory that the experience of an intrusive thought Table 1 Means and standard deviations of WBSI and RIQ scores by group and time. Variable
Group
Baseline
Post-intervention
Follow-up
M WBSIa
A B C
51.13a 51.55a 50.20a
SD
M
SD
M
SD
10.94 10.98 11.98
52.18a 53.77b 52.40b
11.02 11.94 12.86
51.93a 53.43a,b 51.06a,b
10.79 11.89 11.65
RIQ-FHb
A B C
1.08a 1.03a 1.11a
0.73 0.74 0.81
1.15a 1.11a 1.19a
0.92 0.89 0.87
0.92a 0.98a 0.95a
0.72 0.85 0.80
RIQ-FLc
A B C
1.74a 1.98a 1.68a
0.93 1.14 1.08
2.00a 2.07a 1.83a
1.12 1.03 1.17
2.55b 2.13a 1.85a
1.02 1.10 1.11
RIQ-BHd
A B C
24.04a 23.19a 23.69a
17.87 17.38 18.67
27.73a 27.77a 29.80a
26.17 24.58 23.71
19.47a 21.06a 21.64a
19.78 21.66 21.38
RIQ-BLe
A B C
49.39a 54.95a 40.83a
25.77 27.09 24.61
54.96a,b 57.18a 46.02a
31.80 29.22 29.47
64.63b 60.43a 48.60a
29.34 29.39 30.32
Note. N (listwise) ¼ 1301; 1222; 1353; 1204; 1355. Means within the same group not sharing a common superscript letter are significantly different at p < 0.05. a WBSI ¼ White Bear Suppression Inventory (Wegner & Zanakos, 1994). RIQ ¼ Responsibility Interpretations Questionnaire (Salkovskis et al., 2000). b Subscales ¼ frequency of high-responsibility thoughts. c Subscales ¼ frequency of low-responsibility thoughts. d Subscales ¼ belief in high-responsibility thoughts. e Subscales ¼ belief in low-responsibility thoughts.
T. Marino-Carper et al. / J. Behav. Ther. & Exp. Psychiat. 41 (2010) 289e296
may spontaneously result in a desire to suppress it. However, these results also imply that this tendency may be somewhat statedependent, as those individuals who had just received information regarding the irrationality of TAF did not engage in a significantly greater amount of thought suppression in response to the disturbing thought, whereas those in the other two groups did. The lack of significant deviation in baseline and follow-up WBSI scores may suggest that thought suppression is resistant to such a brief intervention. Additionally, this finding demonstrates that the increase in thought suppression that occurred in Groups B and C immediately following the intervention was relatively short-lived. This pattern is consistent with the habituation response; although the initial disturbing thought provokes an urge to suppress the thought, a repeated presentation elicits less of this response. Because engagement in TAF elicits thought suppression (Rassin, 2001), the analysis was re-conducted controlling for the effects of TAF. However, the findings were consistent even after controlling for TAF, suggesting that the intervention had a direct effect on thought suppression. Clinically, it may, therefore, be useful to educate individuals who tend to engage in dysfunctional thought suppression about the importance of distinguishing thoughts from corresponding behaviors. Given that the current results imply that the effects of such psychoeducation lack longevity, such psychoeducation would likely need to be presented repeatedly to be beneficial. For low-responsibility thoughts, individuals in Group A experienced a significant increase in both frequency of and belief in these thoughts from baseline to follow-up, whereas individuals in the other groups did not. Given that the message to which individuals in Group A were exposed contained many explicit low-responsibility statements (e.g., “Your thinking has no effects whatsoever on [your friend or family member's] health”), it would make sense that these individuals would tend to report an increase in the frequency of and belief in these thoughts. It is noteworthy that an increase in thoughts of low-responsibility was found for Group A, but not a decrease in thoughts of high-responsibility. These findings suggest that high-responsibility thoughts may be resistant to brief interventions, which is consistent with Salkovskis et al.'s (2000) postulation that high-responsibility thoughts are more stable than low-responsibility thoughts. However, it may be the case that high and low-responsibility thoughts are not mutually exclusive, and that the individuals in Group A utilized low-responsibility thoughts as a way to counteract, or neutralize, high-responsibility thoughts. It also is of interest that the increase in low-responsibility thoughts for Group A was not present at the post-intervention assessment. Given that there was no significant difference between baseline and post-intervention scores for either frequency of or belief in lowresponsibility thoughts (a finding corroborated by the significant increase found from the post-intervention to follow-up scores for frequency of low-responsibility thoughts), it may be that the message was internalized. Thus, TAF psychoeducation may not have immediate therapeutic benefit on responsibility cognitions, as individuals may need time to internalize and contemplate the information. In consideration of previous work having found that inflated responsibility influences TAF (e.g., Marino et al., 2008), the current findings suggest that psychoeducation may impact TAF both directly and indirectly via an increase the belief in and frequency of corrective responsibility cognitions. 4.1. Future directions Further research into the clinical translations of the current findings is warranted. In particular, future studies investigating interventions for TAF can extend upon these findings by examining the relative effectiveness of different psychotherapeutic interventions on TAF (e.g., an intervention based in cognitive theory vs. an
295
intervention based in mindfulness theory). Further work is also needed regarding the durability of these findings. It would be especially informative and clinically relevant to examine the longer-term effects of a TAF intervention incorporated into the general psychotherapy for individuals for whom TAF is an underlying symptom. Additionally, given that TAF is thought to play a role in a variety of Axis I disorders, there is a need to investigate the extent to which TAF may manifest differently among different psychological disturbances. For example, little is known about the extent to which the relative salience of the two components of TAF, morality and likelihood, vary depending on the larger psychological context. It may be that the optimal type of TAF intervention will differ based on the nature and content of the TAF. Finally, given that TAF falls on a normal distribution, research aimed at determining the point at which it becomes psychologically impairing will likely inform intervention strategies. 4.2. Limitations The present findings must be interpreted in the context of their limitations. In light of the absence of a clinical cut-off for TAF-R scores, the point at which this cognitive bias becomes problematic is vague. Additionally, no data was collected regarding participants' reactions to the psychoeducational messages. Given that the psychoeducation regarding TAF was somewhat more reassuring in nature than the other two interventions, it may have been perceived as being more positively toned or supportive, which could have accounted for some of its efficacy. It is also important to note that the range of the TAF-R scores was truncated given that only those who scored on the high end of the TAF distribution were invited to participate in the study. Restricted variance can reduce the power to detect between-group differences, and thus the results regarding the impact of the interventions on TAF may have been more robust, albeit less clinically relevant, had the range not been constricted. Another limitation concerns the fact that our online collection of TAF-R scores only allowed us access to composite scale scores; thus we cannot be certain that the proportion of participants that scored comparatively high on TAF-likelihood did not significantly differ from the proportion that scored comparatively high on TAF-moral. If these proportions were unequal, this may have some explanatory power regarding the finding that TAF-moral was instantly receptive to the psychoeducation regarding TAF, whereas TAF-likelihood was not. Finally, a more gender-balanced sample would have increased the external validity of the findings. References Berle, D., & Starcevic, V. (2005). Thought-action fusion: review of the literature and future directions. Clinical Psychology Review, 25, 263e284. Katz, M. (2006). Multivariable analysis: A practical guide for clinicians. New York: Cambridge University Press. Marino, T., Lunt, R., & Negy, C. (2008). Thought-action fusion: a comprehensive analysis using structural equation modeling. Behaviour Research and Therapy, 46, 845e853. Muris, P., Merckelbach, H., & Horselenberg, R. (1996). Individual differences in thought suppression. The White Bear Suppression Inventory: factor structure, reliability, validity, and correlates. Behavior Research and Therapy, 34, 501e513. Rachman, S. (1993). Obsessions, responsibility, and guilt. Behaviour Research and Therapy, 31, 149e154. Rachman, S., Thordarson, D., Shafran, R., & Woody, S. (1995). Perceived responsibility: Structure and significance. Behaviour Research and Therapy, 33, 779e784. Rachman, S., & Shafran, R. (1999). Cognitive distortions: thought-action fusion. Clinical Psychology and Psychotherapy, 6, 80e85. Rachman, S., Shafran, R., Mitchell, D., Trant, J., & Teachman, B. (1996). How to remain neutral: an experimental analysis of neutralization. Behaviour Research and Therapy, 34, 889e898. Rassin, E. (2001). The contribution of thought-action fusion and thought suppression in the development of obsession-like intrusions in normal participants. Behaviour Research and Therapy, 39, 1023e1032.
296
T. Marino-Carper et al. / J. Behav. Ther. & Exp. Psychiat. 41 (2010) 289e296
Rassin, E., Diepstraten, P., Merckelbach, H., & Muris, P. (2001). Thought-action fusion and thought suppression in obsessive-compulsive disorder. Behaviour Research and Therapy, 39, 757e764. Rassin, E., Merckelbach, H., Muris, P., & Schmidt, H. (2001). The thought-action fusion scale: further evidence for its reliability and validity. Behaviour Research and Therapy, 39, 537e544. Rassin, E., Merckelbach, H., Muris, P., & Spaan, V. (1999). Thought-action fusion as a causal factor in the development of intrusions. Behaviour Research and Therapy, 37, 231e237. Rassin, E., Muris, P., Schmidt, H., & Merckelbach, H. (2000). Relationships between thought-action fusion, thought suppression and obsessive-compulsive symptoms: a structural equation modeling approach. Behaviour Research and Therapy, 38, 889e897. Salkovskis, P., & Kirk, J. (1989). Obsessional disorders. In K. Hawton, P. Salkovskis, J. Kirk, & D. Clark (Eds.), Cognitive behaviour therapy for psychiatric problems: A practical guide (pp. 129e168). New York: Oxford University Press. Salkovskis, P., Wroe, A., Gledhill, A., Morrison, N., Forrester, E., Richards, C., et al. (2000). Responsibility and interpretations are characteristic of obsessive compulsive disorder. Behaviour Research and Therapy, 38, 347e372. Shafran, R., & Rachman, S. (2004). Thought-action fusion: a review. Journal of Behavior Therapy and Experimental Psychiatry, 35, 87e107.
Shafran, R., Teachman, B., Kerry, S., & Rachman, S. (1999). A cognitive distortion associated with eating disorders: thought-shape fusion. British Journal of Clinical Psychology, 38, 167e179. Shafran, R., Thordarson, D., & Rachman, S. (1996). Thought-action fusion in obsessive compulsive disorder. Journal of Anxiety Disorders, 10(5), 379e391. Smári, J., & Rafnsson, F. D. (2001). Chronic thought suppression and obsessionality: the relationships between the White Bear Suppression Inventory and two inventories of obsessive-compulsive symptoms. Personality and Individual Differences, 30, 159e165. Spinhoven, P., & van der Does, W. (1999). Thought suppression, dissociation, and psychopathology. Personality and Individual Differences, 27, 877e886. Wegner, D., Schneider, D., Carter, S., & White, T. (1987). Paradoxical effects of thought suppression. Journal of Personality and Social Psychology, 53(1), 5e13. Wegner, D., & Zanakos, S. (1994). Chronic thought suppression. Journal of Personality and Social Psychology, 62(4), 615e640. Zucker, B., Craske, M., Barrios, V., & Holguin, M. (2002). Thought action fusion: can it be corrected? Behaviour Research and Therapy, 40, 653e664. Zucker, B., Craske, M., Blackmore, M., & Nitz, A. (2006). A cognitive behavioral workshop for subclinical obsessions and compulsions. Behaviour Research and Therapy, 44(2), 289e304.
Contents lists available at ScienceDirect
Journal of Behavior Therapy and Experimental Psychiatry journal homepage: www.elsevier.com/locate/jbtep
The effects of psychoeducation on thought-action fusion, thought suppression, and responsibility Teresa Marino-Carper*, Charles Negy, Gillian Burns, Rachael A. Lunt Department of Psychology, University of Central Florida, P.O. Box 161390, Orlando, FL 32816-1390, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 October 2009 Received in revised form 19 January 2010 Accepted 11 February 2010
The current study examined the effects of a psychoeducational intervention designed to target thoughtaction fusion (TAF) on TAF, thought suppression, and responsibility cognitions. 139 undergraduate students (25 male; 114 female) who were relatively high in TAF with respect to their peers served as participants. Immediately following intervention, individuals who had received psychoeducation regarding TAF reported significantly lower morality TAF scores than individuals who had received psychoeducation regarding thoughts in general and individuals in the control group. At the two-week follow-up assessment, the likelihood TAF scores of those who had received psychoeducation regarding TAF were significantly lower than those of the control group. In addition, the group that received psychoeducation regarding TAF was the only group that did not experience a significant increase in thought suppression from baseline to post-intervention, and was also the only group to experience an increase in both frequency of and belief in low-responsibility thoughts from baseline to follow-up. Implications are discussed. Published by Elsevier Ltd.
Keywords: Thought-action fusion Psychoeducation Intrusive thoughts Cognitive intervention
1. Introduction The tendency to fuse one's thoughts with overt behaviors is a cognitive bias thought to underlie a variety of psychopathologies (Berle & Starcevic, 2005; Rassin, Diepstraten, Merckelbach, & Muris, 2001; Shafran & Rachman, 2004; Shafran, Teachman, Kerry, & Rachman, 1999). This bias, referred to as thought-action fusion (TAF), is believed to be comprised of two distinct elements (Shafran, Thordarson, & Rachman, 1996). The likelihood component of TAF refers to the belief that experiencing an intrusive and distressing thought can increase the probability that the event will actually occur, causing harm to oneself or others. The second component, morality, refers to the belief that thinking about an action is ethically equivalent to actually performing the action. Rachman (1993) first introduced the concept of TAF in the context of its occurrence in obsessive-compulsive disorder (OCD), and later concluded that TAF increases one's susceptibility for developing OCD (Rachman & Shafran, 1999). Subsequent literature has collectively suggested that TAF is a fairly extensive construct that may be present in myriad disturbances, including Generalized Anxiety Disorder, Post-Traumatic Stress Disorder, Panic Disorder, Social Phobia, eating disorders, and depression (Berle & Starcevic,
* Corresponding author. Tel.: þ1 617 909 6317; fax: þ1 407 823 5862. E-mail address: [email protected] (T. Marino-Carper). 0005-7916/$ e see front matter Published by Elsevier Ltd. doi:10.1016/j.jbtep.2010.02.007
2005; Rassin, Diepstraten et al., 2001; Shafran & Rachman, 2004; Shafran et al., 1999). However, TAF is not exclusive to people with mental health problems, and is prevalent in non-clinical populations as well (Zucker, Craske, Barrios, & Holguin, 2002). An inflated sense of personal responsibility has long been thought to play an important role in the onset and maintenance of OCD (Salkovskis & Kirk, 1989). In order to operationalize the construct of dysfunctional responsibility and clarify its role in obsessional thoughts, Rachman et al. (1995) developed the Responsibility Appraisal Questionnaire (RAQ), a 36-item measure designed to assess responsibility outside of the OCD realm. Using a sample of undergraduate students, the authors found a fourfactor solution for the RAQ, suggesting the existence of four discrete components of responsibility: responsibility for harm, social responsibility, a positive attitude towards responsibility, and TAF. The authors then administered the RAQ and measures of obsessivecompulsive symptoms, depression, guilt, and obsessional beliefs to a different sample of undergraduate students. The results of this second analysis corroborated the four-factor solution, suggesting that TAF is a component of exaggerated responsibility. Additionally, TAF was found to be significantly related to obsessive-compulsive symptoms, symptoms of depression, guilt, and obsessional beliefs. These correlations remained significant even after controlling for the effects of depression symptoms. Smári and Hólmsteinsson investigated the relationships among TAF, responsibility, intrusive thoughts, thought suppression, and
290
T. Marino-Carper et al. / J. Behav. Ther. & Exp. Psychiat. 41 (2010) 289e296
obsessive-compulsive symptoms among undergraduate students. The results indicated that both responsibility attitudes and thought suppression mediated the relationship between intrusive thoughts and obsessive-compulsive symptoms. Interestingly, the researchers also found that controlling for TAF in the mediation analyses led to strikingly similar results to those obtained when controlling for responsibility, suggesting that TAF is highly related to responsibility. These results are consistent with those obtained by Marino, Lunt, and Negy (2008), who, using a structural equation model, found that an inflated sense of responsibility significantly predicted TAF. Individuals often engage in cognitive avoidance to cope with the distress of an intrusive thought (Rassin, 2001). Rassin, Muris, Schmidt, and Merckelbach (2000) attempted to determine directionality in the relationship among TAF, thought suppression, and obsessive-compulsive symptoms using structural equation modeling with a sample of undergraduate students. The results indicated a model in which TAF leads to attempts at thought suppression, which in turn, predicts more obsessive-compulsive symptoms. These results support those found by Rassin, Merckelbach, Muris, and Spaan (1999) that suggested TAF may play a causal role in the development of intrusive thoughts. Rassin (2001) empirically examined the influence of TAF and thought suppression on the development of obsession-like intrusions in undergraduate students. TAF was experimentally induced for each individual via the Sentence Completion Task (Rachman, Shafran, Mitchell, Trant, & Teachman, 1996). The Sentence Completion Task involves asking the participant to identify a loved one, subsequently presenting the participant with a form on which the sentence “I hope _______ is in a car accident” is typed, and then asking the individual to rewrite the sentence with their loved one's name in the blank. Afterwards, the participant is asked to visualize the event occurring. Half the participants in Rassin's study were asked to suppress thoughts of the accident while sitting quietly for five minutes, whereas the other half were explicitly told not to try to suppress any thoughts of the accident during this time. Contrary to prediction, those who were instructed to suppress did not have significantly higher anxiety scores or report an increased number of intrusive thoughts. Moreover, they spent less time thinking about the accident and rated the writing of the sentences as less morally wrong than did participants in the non-suppression group. These findings are discordant with the ‘white bear effect’ found by Wegner et al. (1987) that states that attempts at thought suppression are counterproductive insofar as they tend to result in a greater number of intrusions. Rassin attempted to reconcile this discrepancy by arguing that the paradoxical effects of thought suppression may not surpass the effects of TAF. The author further noted that even those in the non-suppression group reported having attempted to suppress their thoughts, suggesting that TAF may inevitably induce thought suppression. Rassin, Diepstraten et al. (2001) found that cognitive-behavioral interventions significantly reduced TAF in inpatients diagnosed with anxiety disorders. This demonstration that TAF is amenable to change sparked an interest in therapeutic interventions for TAF. The effectiveness of a psychoeducational intervention aimed at minimizing the endorsement of TAF was first examined by Zucker et al. (2002). Undergraduates who scored relatively high on the Thought Action Fusion e Revised scale (Shafran et al., 1996) were randomly assigned to be in an experimental group or a control group. All participants were given measures to assess TAF and current anxiety. The experimental group heard a psychoeducational message regarding TAF that emphasized the normality and irrelevance of intrusive thoughts. Those in the control group heard a message regarding stress, its consequences, and techniques to alleviate it. After the manipulation, participants again completed the measure
assessing TAF to assess for change in endorsement. Next, the participants completed the aforementioned Sentence Completion Task used by Rassin (2001) to induce TAF, and subsequently rated their current levels of anxiety. The researchers found a significant interaction effect between group and point of assessment. Analyses of the simple main effects showed that TAF endorsement significantly decreased from pre-task to post-task in the experimental group, but not in the control group. For anxiety that was assessed by a visual analogue scale, there was also a significant interaction effect between group and time, suggesting that although both groups experienced an increase in anxiety from pre-task to posttask, the control group experienced a significantly higher increase in anxiety than the experimental group. An analysis of simple effects for this interaction revealed no significant differences between the two groups in anxiety at baseline, but there was a trend towards significance for the control group to have higher anxiety than the experimental group at post-task. For anxiety that was assessed by a state anxiety measure, there was no significant interaction effect between group and time. There was, however, a significant effect of time, meaning that the anxiety levels of both groups significantly increased from pre-task to post-task. Zucker et al. concluded that these results suggest that mere psychoeducation can impact individuals' appraisals of a disturbing, intrusive thought. Zucker, Craske, Blackmore, and Nitz (2006) followed this work by investigating the effects of a cognitivebehavioral workshop in individuals with subclinical OCD. The findings suggested that those who attended the workshop had lower average TAF scores both one month and five months after the intervention, suggesting that the effects of the intervention demonstrated longevity. Despite the growing body of evidence suggesting that TAF underlies myriad psychological disorders, there is a paucity of research into effective methods for targeting this cognitive bias. In addition, given that thought suppression and inflated responsibility are thought to contribute to the maintenance of TAF, it is important to determine the extent to which a TAF intervention will also decrease these influencing constructs. As such, the purpose of this study was to extend Zucker et al.'s (2002) work by examining the relative effects of a brief intervention designed to specifically target TAF on not only TAF itself, but also the closely related phenomena of thought suppression and responsibility. In addition, the current study aimed to examine the stability of these effects by assessing the outcome variables both immediately following the intervention and after a period of two weeks. The interventions were in the form of a psychoeducational message that referred to either TAF, intrusive thoughts alone, or general stress. The messages regarding TAF and general stress were taken directly from Zucker et al.'s investigation. The intervention novel to the current study was the psychoeducation that discussed the presence of intrusive thoughts in a non-evaluative manner, and was included to help dismantle Zucker et al's findings that the invention explicitly targeting TAF lowered TAF endorsement. By including a condition that discussed the presence of intrusive thoughts more generally, our goal was to ascertain if merely attending to and being more aware of one's thoughts could lower endorsements of the dependent variables, or if it is the information regarding the irrationality of the thoughts that is the critical component. The current study was designed to add to the literature in several ways. First, given that TAF is prevalent in myriad psychopathologies, it is important to examine the extent to which the effects of a TAF intervention are stable in individuals with high levels of TAF in general, as opposed to only within the context of subclinical OCD. Secondly, by including a condition in which participants receive psychoeducation about intrusive thoughts in a non-evaluative manner, the current study might elucidate the
T. Marino-Carper et al. / J. Behav. Ther. & Exp. Psychiat. 41 (2010) 289e296
extent to which addressing the irrationality of TAF, as opposed to merely increasing metacognitive awareness, is crucial to intervention. Third, investigating moral and likelihood TAF separately allows for the examination of any differential effects of the intervention on these two components. Finally, no study has examined how such an intervention may reduce thought suppression and responsibility cognitions, constructs believed to partially underlie the development and maintenance of TAF. In light of the literature reviewed, we hypothesized that participants in the group exposed to the psychoeducational message on TAF would score significantly lower on measures of both TAF subscales than participants in the group exposed to the psychoeducational message on intrusive thoughts only and participants in the control group, both immediately following the intervention and at the two-week follow-up. It was further postulated that, by proxy of their association with TAF, thought suppression and responsibility would significantly decrease from baseline to post-intervention in the group provided with psychoeducation on TAF, but not in the other two groups, and these decreases would remain at the two-week follow-up period. 2. Methods 2.1. Participants A total of 4376 students enrolled in undergraduate psychology courses were screened for subsequent selection by completing a preliminary informed consent form, demographics questionnaire, and the Thought-Action Fusion Scale e Revised (Shafran et al., 1996; see description below) online in exchange for extra credit towards their respective courses. Of those students, 756 (17.28%) scored at least one standard deviation above the mean of the TAF-R composite scale as established by prior research (i.e., Marino et al., 2008), and were invited to participate in the study. Of the 167 participants who chose to begin the protocol, 15 failed the manipulation check, and 15 (2 of whom also failed the manipulation check) did not complete the Time 2 portion. Thus, the final study sample used in data analysis was composed of 139 (25 male and 114 female) participants. Regarding ethnicity, 73 (52.5%) of the students self-identified as European American, 22 (15.8%) as Latino American, 26 (18.7%) as African American, 11 (7.9%) as Asian American, and 7 (5%) as “other.” With respect to age, 121 (87.1%) of participants were between the ages of 18 and 20, 14 (10.1%) were between 21 and 25, 3 (2.2%) were between 26 and 30, and 1 (0.7%) was between the ages of 31 and 40. 2.2. Materials 2.2.1. Thought-action fusion scale e revised (TAF-R) The TAF-R (Shafran et al., 1996) is a 19-item measure of the extent to which individuals equate their thoughts with actions. Participants respond using a five-point Likert-type scale, with higher scores reflecting a greater proclivity to engage in TAF. In addition to a composite score, ‘moral’ and ‘likelihood’ subscale scores may also be tabulated. The TAF e R has been demonstrated to be psychometrically valid for use in a student population (Shafran et al., 1996). For the current study, the Cronbach alpha coefficients of the TAF-R ranged from 0.88 to 0.94 for the moral subscale, and from 0.95 to 0.97 for the likelihood subscale. Rassin, Merckelbach, Muris, and Schmidt (2001) found significant retest correlations for the TAF-R composite scale, moral subscale, and likelihood subscale (rs ¼ 0.52, 0.54, and 0.51, respectively; all ps < 0.01) among a sample of undergraduate students. However, the mean score at the retest administration was significantly lower than the mean score at baseline, suggesting that TAF may be somewhat unstable in a non-clinical population.
291
2.2.2. Responsibility interpretations questionnaire (RIQ) The RIQ (Salkovskis et al., 2000) is comprised of 22 items designed to measure the frequency of and belief in high responsibly thoughts (e.g., “It would be irresponsible to ignore these thoughts”) and low-responsibility thoughts (e.g., “This is just a thought so it doesn't matter”). The combination of these dimensions yields four subscales, with higher scores on the two subscales that assess high responsibility indicating a greater sense of responsibility, and higher scores on the two subscales that assess low responsibility indicating a lesser sense of responsibility. Salkovskis et al. found the RIQ to an internally consistent instrument that is valid for use with a non-clinical population. For the current study, the Cronbach alpha coefficients of the RIQ ranged from 0.90 to 0.94 for the frequency of high-responsibility thoughts subscale, from 0.83 to 0.90 for the frequency of low-responsibility thoughts subscale, from 0.90 to 0.95 for the belief in high-responsibility thoughts subscale, and from 0.85 to 0.92 for the belief in low-responsibility thoughts subscale. Salkovskis et al. examined the temporal stability of the RIQ over a two-week period, and found favorable test-retest reliability for the subscales measuring the frequency of and belief in high-responsibility thoughts (r ¼ 0.90 and 0.80, respectively), but not for the subscales measuring the frequency of and belief in lowresponsibility thoughts (r ¼ 0.69 and 0.22, respectively). The authors surmised that, given the favorable internal consistency of the two low-responsibility subscales, the lack of stability likely reflects actual instability in the frequency of and belief in lowresponsibility cognitions. 2.2.3. White Bear Suppression Inventory (WBSI) The WBSI (Wegner & Zanakos, 1994) is a 15-item questionnaire designed to assess thought suppression. Participants respond to the items using a five-point Likert-type scale, with a higher score indicating a stronger tendency to engage in thought suppression. The WBSI has been found to have favorable psychometric properties (e.g., Muris, Merckelbach, & Horselenberg, 1996; Smári & Rafnsson, 2001; Spinhoven & van der Does, 1999), and achieved Cronbach alpha coefficients ranging from 0.90 to 0.93. for the current sample. Muris et al. (1996) found that the WBSI demonstrated good stability over a 12-week period, with a test-retest correlation coefficient of 0.80. 2.2.4. Sentence completion form The Sentence Completion Form was used as part of Rachman et al's (1996) aforementioned TAF-inducing Sentence Completion Task. The form consists of the line “I hope _______ is in a car accident”, and a space underneath in which the participant is asked to rewrite the sentence with their loved one's name in the blank. 2.3. Study design and procedure The present design was a 3 (time: baseline, post-manipulation 1, post-manipulation 2) 3 (psychoeducation condition: Group A, Group B, Group C) between-subjects and within-subjects design. In light of our objective of expanding upon the work of Zucker et al. (2002), the current study adopted similar methodology. Individual data collection sessions occurred within the psychology department and lasted approximately a half an hour. Each participant provided informed consent prior to beginning the study. Four individuals elected to terminate participation after being asked to complete the sentence completion task, indicating that doing so would cause marked discomfort. Study procedures were immediately terminated for these individuals, and the researcher verbally assessed the nature and extent of the individual's anxiety. All four participants who chose to terminate denied experiencing excessive distress and declined counseling referrals. At the end of each data
292
T. Marino-Carper et al. / J. Behav. Ther. & Exp. Psychiat. 41 (2010) 289e296
collection session, the researcher verbally assessed the distress level of each participant, all of whom denied experiencing excessive distress and declined counseling referrals. Participants visited the research laboratory on two occasions. Each participant was randomly assigned into one of three groups: a group that received psychoeduation on TAF (Group A; N ¼ 41), a group that received psychoeducation on thoughts and cognitions (Group B; N ¼ 47), and a control group (Group C; N ¼ 51) that received psychoeducation on stress. Participants were asked to complete the RIQ, TAF-R, and WBSI. Next, participants in Group A received psychoeducation regarding intrusive thoughts and TAF in the form of the following digitally-recorded message (taken from Zucker et al., 2002, p. 656): It is estimated that approximately 80e90% of the population has intrusive thoughts. Intrusive thoughts are often unpleasant thoughts that come to your mind out of the blue. For example, you might be driving on a bridge and have a fleeting though about driving off the bridge, or you might have a passing thought about causing harm to a helpless person. Or, you might be in a serious relationship and have unpleasant sexual thoughts about people other than your partner. Often, when people have these bothersome thoughts they feel as though somehow, their thoughts will make the event more likely to happen. This way of thinking is wrong. Why is this way of thinking wrong? By simply thinking about a friend or family member becoming seriously ill, you will not increase the chance that they will become sick. Your thinking has no effects whatsoever on their health. Also unpleasant sexual thoughts alone do not mean you are an evil, bad or immoral person. Remember, it is completely normal to have these types of thoughts. These types of thoughts have no influence over outside events, nor are they a reflection of your character. Participants in Group B received psychoeducation regarding intrusive thoughts in the form of the following digitally-recorded message (developed by the present authors): Most of the time, people have control over their thoughts, and can purposely choose to think about or not think about whatever they want. Sometimes, however, thoughts seem to pop into people's minds without them having intentionally brought on the thought. For example, you may be engaged in a conversation and all of a sudden an irrelevant thought will pop into your mind. These kinds of thoughts that come into your mind out of the blue are called intrusive thoughts. An intrusive thought may be pleasant to the person, or may not cause any emotion at all. At times, intrusive thoughts can also be unpleasant and upsetting. For example, you might be driving on a bridge and have a fleeting thought about driving off the bridge, or you might have a passing thought about causing harm to a helpless person. Or, you might be in a serious relationship and have unpleasant sexual thoughts about people other than your partner. Intrusive thoughts can be very scary and disturbing to the individual. These types of thoughts may cause stress and anxiety. Participants in Group C, the control group, received psychoeducation about stress in the form of the following digitallyrecorded message (taken from Zucker et al., 2002, p. 656): There are many types of situations that cause people to experience increased levels of stress. These situations could range anywhere from driving in heavy traffic to being fired from a job to going away to college for the first time. Maybe people around the world commonly experience some degree of stress on a regular basis. When people feel stressed they might have accompanying physical problems, such as headaches, increased muscle tension and stomachaches. It even could lead to more serious medical problems such
as high blood pressure and ulcers. Many times when people experience these types of physical problems, they don't realize that they could be stress-related. Although stress cannot be fully eliminated from day-to-day life, people should try their best to minimize their amount of stress in their lives and engage in stress management techniques. Stress management can target a variety of lifestyle changes, including increasing exercise and daily activity, eating healthy and establishing regular sleeping patterns. By actively seeking to reduce and better manage stress, people are able to greatly improve the quality of their lives. After listening to the message, all participants completed a manipulation check to ensure that they had attended to and understood the psychoeducation provided. This manipulation check consisted of 3 true/false condition-specific questions regarding the message content (e.g., “It is very common to have unpleasant thoughts come into your mind out of the blue” for Group A). Any incorrectly answered question resulted in a failed manipulation check; of the 15 individuals who failed the manipulation check, 11 answered 1 question incorrectly, 3 answered 2 questions incorrectly, and 1 answered all 3 questions incorrectly. Next, all participants were asked to perform the Sentence Completion Task (Rachman et al., 1996) for the purposes of experimentally inducing TAF. Given that TAF is a trait-like construct that becomes activated in response to a disturbing and intrusive thought, the Sentence Completion Task is often used in the literature as a means of generating the same troubling cognition for all participants, at a uniform point in time during the study protocol. Following the task, all participants again completed the RIQ, TAF-R, and WBSI. The instructions for these measures were adapted to reflect the brief time period between administrations. Two weeks after Time 1, participants returned for the Time 2 procedures. All participants were again asked to perform the sentence completion task followed by the RIQ, TAF-R, and WSBI. Participants were then debriefed and those who were in Groups B and C listened to the psychoeducational message about TAF, for the possibility of therapeutic benefit. 3. Results 3.1. Initial data analysis A multivariate analysis of variance (MANOVA) revealed no significant between-group differences on the baseline measurements of the TAF-R subscales, WBSI, and RIQ subscales (F [14, 230] ¼ 0.66, ns). A one-way analysis of variance (ANOVA) confirmed that there were no significant between-group differences for age (F [2, 136] ¼ 0.04, ns), and a Pearson chi-squared test revealed no significant differences for sex (c2 [2] ¼ 0.75, ns). Of note, the baseline TAF-R mean for the sample used in the current study was 41.20, which is markedly higher than the normative TAF-R means established by Shafran et al. (1996) of 29.21 for individuals with clinical levels of obsessionality and 16.82 for an undergraduate population. Thus, it appears that the screening procedure employed in the current study was effective method of recruiting a sample relatively high in TAF. 3.2. Effects of interventions on TAF The first hypothesis predicted that those in Group A would score significantly lower on measures of both TAF subscales than those in Groups B and C, both immediately following the intervention and at the two-week follow-up. To test this hypothesis, the morality and likelihood subscales of the TAF-R were analyzed separately using a 3 3 mixed model univariate analysis of variance (ANOVA;
T. Marino-Carper et al. / J. Behav. Ther. & Exp. Psychiat. 41 (2010) 289e296
293
group time). A significant interaction effect was found for morality TAF (F [4, 258] ¼ 3.86, p < 0.01, partial h2 ¼ 0.06). To examine between-group differences at the post-intervention and follow-up assessments of morality TAF, post-hoc pairwise comparisons were conducted using the Bonferroni correction. These analyses revealed that individuals in Group A had significantly lower scores at post-intervention than individuals in both Group B (Dx ¼ 5:43, p < 0.05) and Group C (Dx ¼ 5:92, p < 0.05). No significant differences were observed at the two-week followup for morality TAF. The results of this interaction are depicted in Fig. 1. There was also a significant interaction effect for likelihood TAF (F [4, 262] ¼ 4.39, p < 0.01, partial h2 ¼ 0.06). Post-hoc analyses revealed no significant between-group differences at post-intervention for likelihood TAF. However, at follow-up, Group A had significantly lower scores than Group C (Dx ¼ 4:75, p < 0.01). The results of this interaction are depicted in Fig. 2.
frequency of low-responsibility thoughts, results yielded a significant interaction effect (F [3.77, 248.89] ¼ 3.07, p < 0.05, partial h2 ¼ 0.04). Post-hoc analyses were conducted using the Bonferroni correction to examine within-group differences for all four subscales between baseline and post-intervention, and baseline and follow-up. For belief in low-responsibility thoughts, individuals in Group A experienced a significant increase from baseline to follow-up (Dx ¼ 15:24, p < 0.01), whereas individuals in Groups B and C did not. For frequency of low-responsibility thoughts, individuals in Group A experienced a significant increase from baseline to follow-up (Dx ¼ 0:82, p < 0.001), as well as from post-intervention to follow-up (Dx ¼ 0:55, p < 0.01), whereas individuals in Groups B and C did not. Cell means and standard deviations for all subscales of the RIQ are provided in Table 1.
3.3. Effects of Interventions on thought suppression and responsibility
The finding that individuals receiving psychoeducation regarding TAF were less likely to endorse morality-related TAF beliefs immediately following the intervention than individuals in the control group is consistent with previous work (i.e., Zucker et al., 2002). Consistent with our hypothesis, psychoeducation regarding TAF was also superior to psychoeducation regarding thoughts in general for morality-related TAF beliefs. Thus, it appears that identification and explanation of the irrationality of these thoughts is a vital component of the intervention. These findings were not present at follow-up, suggesting that the effects of providing TAF psychoeducation on the morality component of TAF diminish over time. Surprisingly, there were no differences among the groups in likelihood TAF immediately following the intervention. However, at follow-up, the individuals who had received psychoeducation regarding TAF were significantly less likely to endorse likelihoodrelated TAF beliefs than individuals in the control group. It may, therefore, be the case that, to be effective, the message that experiencing an intrusive thought does not increase the probability of the thought actually occurring needs to be paired with evidence (i.e., two-weeks elapsing without the participant's loved one being involved in a car accident). However, the significant group differences suggest that this evidence alone is not as effective at reducing TAF-likelihood endorsement as it is when paired with psychoeducation regarding TAF. These findings suggest that the effects of this brief intervention have some degree of longevity on the likelihood component of TAF. Although previous research has demonstrated the immediate effects of a brief intervention on TAF endorsement (Zucker et al., 2002) as well as longer-term effects of a more intensive
The second hypothesis predicted that the post-intervention and follow-up RIQ and WBSI scores would be significantly lower than baseline for Group A, but not for Groups B or C. To examine changes in thought suppression, a 3 3 mixed model univariate ANOVA (group time) was conducted using the WBSI as the DV. Due to a sphericity violation, the Greenhouse-Geisser correction was used in the analysis (Katz, 2006). Results revealed no significant interaction effect (F [3.56, 234.60] ¼ 0.43, ns). Given that a priori hypotheses justify examination of post-hoc tests, multiple comparison tests were conducted using the Bonferroni correction to examine within-group differences between baseline and postintervention, and baseline and follow-up. For both Groups B and C, WBSI scores significantly increased from baseline to post-intervention (both ps < 0.05), although this was not the case for participants in Group A. No significant differences from baseline to follow-up assessment on the WBSI were found for any of the groups. Cell means and standard deviations for the WBSI are provided in Table 1. To examine changes in responsibility, four separate mixed model univariate ANOVAs (group time) were conducted using each of the four components of the RIQ as the DV. The GreenhouseGeisser correction was used to correct for a sphericity violation in the analyses concerning frequency of low-responsibility thoughts and belief in high-responsibility thoughts. No significant interaction effect was found for frequency of high responsibility thoughts (F [4, 238] ¼ 0.21, ns), or belief in either high (F [3.79, 221.59] ¼ 0.13, ns) or low (F [4, 264] ¼ 0.85, ns) responsibility thoughts. For
4. Discussion
Fig. 1. Interaction between intervention group and time of assessment on TAF-moral. Note. Standard errors for each mean are represented in parenthesis.
294
T. Marino-Carper et al. / J. Behav. Ther. & Exp. Psychiat. 41 (2010) 289e296
Fig. 2. Interaction between intervention group and time of assessment on TAF-likelihood. Note. Standard errors for each mean are represented in parenthesis.
intervention on TAF endorsement (Zucker et al., 2006), the current study is the first to establish that the effects of a brief intervention can be sustained for at least a two-week period. Moreover, although two weeks is a relatively short period of time, these results have potential clinical significance. Given that many psychotherapy sessions occur on a weekly basis, therapeutic interventions aimed at dismantling the faulty cognitive assumptions underlying TAF may reduce some portion of TAF for the intersession duration. The fact that the effects of the intervention on the morality and likelihood components of TAF displayed different patterns was somewhat surprising. It appears that the impact of the psychoeducation regarding TAF on morality diminished over time, but the effects on likelihood strengthened over time. Again, it may be that the two-week interval was beneficial for TAF-likelihood insofar as the intervention message was supported by the participants' experience (i.e., they received evidence that the thought regarding the loved one in the car accident did not cause the event to occur). Conversely, the lack of a comparable form of concrete evidence supporting the idea that such thoughts do not make one immoral may help explain why the effects of the intervention on TAF-moral weakened over time. The overall superiority of the TAF psychoeducation over the other two interventions may be due in part to its approximation to a cognitive intervention. In particular, the message explicitly highlights the irrationality of TAF, challenges the cognitive errors in both the morality (i.e., “unpleasant sexual thoughts alone do not mean you are an evil, bad or immoral person”) and likelihood (i.e., “by simply thinking about a friend or family member becoming seriously ill, you will not increase the chance that they will become sick ”) components, and may thus encourage participants to restructure their own cognitions. Therefore, it is uncertain the extent to which the psychoeducative quality of message, as opposed to its similarity to cognitive-based intervention, accounts for the variance in TAF-R scores. Since the psychoeducation regarding TAF was more effective than the psychoeducation regarding intrusive thoughts in general for reducing TAF-morality in short-term and TAF-likelihood in the longer term, it may be that cognitive-based interventions are more appropriate for TAF than interventions that emphasize awareness of one's thoughts. Clearly, it cannot be inferred from the current study that such mindfulness-type approaches are inferior to cognitive-based interventions at reducing TAF; however, the present study does offer some evidence that merely being attending to one's thoughts and being exposed to a non-evaluative description of intrusive thoughts does not significantly impact either component of TAF (although the individuals in Group B may have benefitted from the intervention if it had also highlighted the
importance of accepting these intrusive thoughts in a non-judgmental manner, a technique that is often emphasized in mindfulness-type interventions). With respect to thought suppression, although we predicted that only those who received psychoeducation regarding TAF would experience a decrease from baseline to post-intervention, participants in that group were the only ones who did not experience an increase. Given that thought suppression often occurs following an intrusion (Rassin, 2001; Rassin et al., 2000), it may be the case that those who received the psychoeducation regarding either thoughts in general or stress were actively suppressing thoughts of the sentence they had just written. Conversely, the psychoeducation regarding TAF may have served as a buffer against the increase in thought suppression. If it is the case that thought suppression increases the frequency of intrusive thoughts, the decrease in susceptibility to engaging in thought suppression may facilitate a decrease in the frequency of intrusive, disturbing thoughts. The current findings offer some substantiation for Rassin's (2001) theory that the experience of an intrusive thought Table 1 Means and standard deviations of WBSI and RIQ scores by group and time. Variable
Group
Baseline
Post-intervention
Follow-up
M WBSIa
A B C
51.13a 51.55a 50.20a
SD
M
SD
M
SD
10.94 10.98 11.98
52.18a 53.77b 52.40b
11.02 11.94 12.86
51.93a 53.43a,b 51.06a,b
10.79 11.89 11.65
RIQ-FHb
A B C
1.08a 1.03a 1.11a
0.73 0.74 0.81
1.15a 1.11a 1.19a
0.92 0.89 0.87
0.92a 0.98a 0.95a
0.72 0.85 0.80
RIQ-FLc
A B C
1.74a 1.98a 1.68a
0.93 1.14 1.08
2.00a 2.07a 1.83a
1.12 1.03 1.17
2.55b 2.13a 1.85a
1.02 1.10 1.11
RIQ-BHd
A B C
24.04a 23.19a 23.69a
17.87 17.38 18.67
27.73a 27.77a 29.80a
26.17 24.58 23.71
19.47a 21.06a 21.64a
19.78 21.66 21.38
RIQ-BLe
A B C
49.39a 54.95a 40.83a
25.77 27.09 24.61
54.96a,b 57.18a 46.02a
31.80 29.22 29.47
64.63b 60.43a 48.60a
29.34 29.39 30.32
Note. N (listwise) ¼ 1301; 1222; 1353; 1204; 1355. Means within the same group not sharing a common superscript letter are significantly different at p < 0.05. a WBSI ¼ White Bear Suppression Inventory (Wegner & Zanakos, 1994). RIQ ¼ Responsibility Interpretations Questionnaire (Salkovskis et al., 2000). b Subscales ¼ frequency of high-responsibility thoughts. c Subscales ¼ frequency of low-responsibility thoughts. d Subscales ¼ belief in high-responsibility thoughts. e Subscales ¼ belief in low-responsibility thoughts.
T. Marino-Carper et al. / J. Behav. Ther. & Exp. Psychiat. 41 (2010) 289e296
may spontaneously result in a desire to suppress it. However, these results also imply that this tendency may be somewhat statedependent, as those individuals who had just received information regarding the irrationality of TAF did not engage in a significantly greater amount of thought suppression in response to the disturbing thought, whereas those in the other two groups did. The lack of significant deviation in baseline and follow-up WBSI scores may suggest that thought suppression is resistant to such a brief intervention. Additionally, this finding demonstrates that the increase in thought suppression that occurred in Groups B and C immediately following the intervention was relatively short-lived. This pattern is consistent with the habituation response; although the initial disturbing thought provokes an urge to suppress the thought, a repeated presentation elicits less of this response. Because engagement in TAF elicits thought suppression (Rassin, 2001), the analysis was re-conducted controlling for the effects of TAF. However, the findings were consistent even after controlling for TAF, suggesting that the intervention had a direct effect on thought suppression. Clinically, it may, therefore, be useful to educate individuals who tend to engage in dysfunctional thought suppression about the importance of distinguishing thoughts from corresponding behaviors. Given that the current results imply that the effects of such psychoeducation lack longevity, such psychoeducation would likely need to be presented repeatedly to be beneficial. For low-responsibility thoughts, individuals in Group A experienced a significant increase in both frequency of and belief in these thoughts from baseline to follow-up, whereas individuals in the other groups did not. Given that the message to which individuals in Group A were exposed contained many explicit low-responsibility statements (e.g., “Your thinking has no effects whatsoever on [your friend or family member's] health”), it would make sense that these individuals would tend to report an increase in the frequency of and belief in these thoughts. It is noteworthy that an increase in thoughts of low-responsibility was found for Group A, but not a decrease in thoughts of high-responsibility. These findings suggest that high-responsibility thoughts may be resistant to brief interventions, which is consistent with Salkovskis et al.'s (2000) postulation that high-responsibility thoughts are more stable than low-responsibility thoughts. However, it may be the case that high and low-responsibility thoughts are not mutually exclusive, and that the individuals in Group A utilized low-responsibility thoughts as a way to counteract, or neutralize, high-responsibility thoughts. It also is of interest that the increase in low-responsibility thoughts for Group A was not present at the post-intervention assessment. Given that there was no significant difference between baseline and post-intervention scores for either frequency of or belief in lowresponsibility thoughts (a finding corroborated by the significant increase found from the post-intervention to follow-up scores for frequency of low-responsibility thoughts), it may be that the message was internalized. Thus, TAF psychoeducation may not have immediate therapeutic benefit on responsibility cognitions, as individuals may need time to internalize and contemplate the information. In consideration of previous work having found that inflated responsibility influences TAF (e.g., Marino et al., 2008), the current findings suggest that psychoeducation may impact TAF both directly and indirectly via an increase the belief in and frequency of corrective responsibility cognitions. 4.1. Future directions Further research into the clinical translations of the current findings is warranted. In particular, future studies investigating interventions for TAF can extend upon these findings by examining the relative effectiveness of different psychotherapeutic interventions on TAF (e.g., an intervention based in cognitive theory vs. an
295
intervention based in mindfulness theory). Further work is also needed regarding the durability of these findings. It would be especially informative and clinically relevant to examine the longer-term effects of a TAF intervention incorporated into the general psychotherapy for individuals for whom TAF is an underlying symptom. Additionally, given that TAF is thought to play a role in a variety of Axis I disorders, there is a need to investigate the extent to which TAF may manifest differently among different psychological disturbances. For example, little is known about the extent to which the relative salience of the two components of TAF, morality and likelihood, vary depending on the larger psychological context. It may be that the optimal type of TAF intervention will differ based on the nature and content of the TAF. Finally, given that TAF falls on a normal distribution, research aimed at determining the point at which it becomes psychologically impairing will likely inform intervention strategies. 4.2. Limitations The present findings must be interpreted in the context of their limitations. In light of the absence of a clinical cut-off for TAF-R scores, the point at which this cognitive bias becomes problematic is vague. Additionally, no data was collected regarding participants' reactions to the psychoeducational messages. Given that the psychoeducation regarding TAF was somewhat more reassuring in nature than the other two interventions, it may have been perceived as being more positively toned or supportive, which could have accounted for some of its efficacy. It is also important to note that the range of the TAF-R scores was truncated given that only those who scored on the high end of the TAF distribution were invited to participate in the study. Restricted variance can reduce the power to detect between-group differences, and thus the results regarding the impact of the interventions on TAF may have been more robust, albeit less clinically relevant, had the range not been constricted. Another limitation concerns the fact that our online collection of TAF-R scores only allowed us access to composite scale scores; thus we cannot be certain that the proportion of participants that scored comparatively high on TAF-likelihood did not significantly differ from the proportion that scored comparatively high on TAF-moral. If these proportions were unequal, this may have some explanatory power regarding the finding that TAF-moral was instantly receptive to the psychoeducation regarding TAF, whereas TAF-likelihood was not. Finally, a more gender-balanced sample would have increased the external validity of the findings. References Berle, D., & Starcevic, V. (2005). Thought-action fusion: review of the literature and future directions. Clinical Psychology Review, 25, 263e284. Katz, M. (2006). Multivariable analysis: A practical guide for clinicians. New York: Cambridge University Press. Marino, T., Lunt, R., & Negy, C. (2008). Thought-action fusion: a comprehensive analysis using structural equation modeling. Behaviour Research and Therapy, 46, 845e853. Muris, P., Merckelbach, H., & Horselenberg, R. (1996). Individual differences in thought suppression. The White Bear Suppression Inventory: factor structure, reliability, validity, and correlates. Behavior Research and Therapy, 34, 501e513. Rachman, S. (1993). Obsessions, responsibility, and guilt. Behaviour Research and Therapy, 31, 149e154. Rachman, S., Thordarson, D., Shafran, R., & Woody, S. (1995). Perceived responsibility: Structure and significance. Behaviour Research and Therapy, 33, 779e784. Rachman, S., & Shafran, R. (1999). Cognitive distortions: thought-action fusion. Clinical Psychology and Psychotherapy, 6, 80e85. Rachman, S., Shafran, R., Mitchell, D., Trant, J., & Teachman, B. (1996). How to remain neutral: an experimental analysis of neutralization. Behaviour Research and Therapy, 34, 889e898. Rassin, E. (2001). The contribution of thought-action fusion and thought suppression in the development of obsession-like intrusions in normal participants. Behaviour Research and Therapy, 39, 1023e1032.
296
T. Marino-Carper et al. / J. Behav. Ther. & Exp. Psychiat. 41 (2010) 289e296
Rassin, E., Diepstraten, P., Merckelbach, H., & Muris, P. (2001). Thought-action fusion and thought suppression in obsessive-compulsive disorder. Behaviour Research and Therapy, 39, 757e764. Rassin, E., Merckelbach, H., Muris, P., & Schmidt, H. (2001). The thought-action fusion scale: further evidence for its reliability and validity. Behaviour Research and Therapy, 39, 537e544. Rassin, E., Merckelbach, H., Muris, P., & Spaan, V. (1999). Thought-action fusion as a causal factor in the development of intrusions. Behaviour Research and Therapy, 37, 231e237. Rassin, E., Muris, P., Schmidt, H., & Merckelbach, H. (2000). Relationships between thought-action fusion, thought suppression and obsessive-compulsive symptoms: a structural equation modeling approach. Behaviour Research and Therapy, 38, 889e897. Salkovskis, P., & Kirk, J. (1989). Obsessional disorders. In K. Hawton, P. Salkovskis, J. Kirk, & D. Clark (Eds.), Cognitive behaviour therapy for psychiatric problems: A practical guide (pp. 129e168). New York: Oxford University Press. Salkovskis, P., Wroe, A., Gledhill, A., Morrison, N., Forrester, E., Richards, C., et al. (2000). Responsibility and interpretations are characteristic of obsessive compulsive disorder. Behaviour Research and Therapy, 38, 347e372. Shafran, R., & Rachman, S. (2004). Thought-action fusion: a review. Journal of Behavior Therapy and Experimental Psychiatry, 35, 87e107.
Shafran, R., Teachman, B., Kerry, S., & Rachman, S. (1999). A cognitive distortion associated with eating disorders: thought-shape fusion. British Journal of Clinical Psychology, 38, 167e179. Shafran, R., Thordarson, D., & Rachman, S. (1996). Thought-action fusion in obsessive compulsive disorder. Journal of Anxiety Disorders, 10(5), 379e391. Smári, J., & Rafnsson, F. D. (2001). Chronic thought suppression and obsessionality: the relationships between the White Bear Suppression Inventory and two inventories of obsessive-compulsive symptoms. Personality and Individual Differences, 30, 159e165. Spinhoven, P., & van der Does, W. (1999). Thought suppression, dissociation, and psychopathology. Personality and Individual Differences, 27, 877e886. Wegner, D., Schneider, D., Carter, S., & White, T. (1987). Paradoxical effects of thought suppression. Journal of Personality and Social Psychology, 53(1), 5e13. Wegner, D., & Zanakos, S. (1994). Chronic thought suppression. Journal of Personality and Social Psychology, 62(4), 615e640. Zucker, B., Craske, M., Barrios, V., & Holguin, M. (2002). Thought action fusion: can it be corrected? Behaviour Research and Therapy, 40, 653e664. Zucker, B., Craske, M., Blackmore, M., & Nitz, A. (2006). A cognitive behavioral workshop for subclinical obsessions and compulsions. Behaviour Research and Therapy, 44(2), 289e304.