- Email: [email protected]
The potentiating effect of disgust sensitivity on the relationship between disgust propensity and mental contamination
Journal of Obsessive-Compulsive and Related Disorders 6 (2015) 114–119
Contents lists available at ScienceDirect
Journal of Obsessive-Compulsive and Related Disorders journal homepage: www.elsevier.com/locate/jocrd
Clinical report
The potentiating effect of disgust sensitivity on the relationship between disgust propensity and mental contamination Rosemond Travis, Thomas A. Fergus n Baylor University, USA
art ic l e i nf o
a b s t r a c t
Article history: Received 5 March 2015 Received in revised form 25 June 2015 Accepted 29 June 2015 Available online 2 July 2015
Disgust is important to mental contamination, a contamination fear that arises in the absence of physical contact with a perceived contaminant. Researchers have distinguished between disgust propensity, defined as one's general tendency to experience disgust, and disgust sensitivity, defined as one's negative appraisal of the experience of disgust. Based upon speculations that disgust sensitivity may amplify the experience of disgust propensity on disgust-relevant outcomes, this study examined the interaction of disgust propensity and disgust sensitivity in relation to mental contamination among a community sample of adults located in the United States recruited through Amazon's Mechanical Turk (N¼ 478). The results suggest that disgust sensitivity potentiates the effect of disgust propensity on mental contamination. The interactive effect was robust to the effects of negative affect and broader contamination fears. These results indicate that mental contamination is particularly strong among individuals with concurrently high disgust propensity and disgust sensitivity. Implications and future directions are explored. & 2015 Elsevier Inc. All rights reserved.
Keywords: Disgust Disgust propensity Disgust sensitivity Mental contamination
1. Introduction A common trigger for psychological distress comes from a desire to keep one's self or environment clean. This desire for cleanliness is often perpetuated by a fear of contamination (Rachman, 2004). As noted by Rachman (2004), fear of contamination is a near-universal experience and is defined as a fear related to coming into contact with a person or item that is believed to be dirty, whether this contact is direct (e.g., as when one touches garbage) or indirect (e.g., as when one touches an object that trash may have once touched). Those who have a severe fear of contamination tend to either try to limit their contact with potentially unclean people or items and/or engage in cleansing behavior after they come into contact with a perceived contaminant (Rachman, 2004). Whereas contact contamination originates from physical contact with a perceived unclean stimulus, contamination can arise from merely observing or thinking about something unclean, immoral, or undesirable (Rachman, 2004). As described by Rachman (2004), contamination that occurs in the absence of physical contact with a contaminant is called mental contamination. Mental contamination differs from contact contamination along a n Correspondence to: Department of Psychology & Neuroscience, Baylor University, Waco, TX 76798, USA. Fax: þ 1 254 710 3033. E-mail address: [email protected] (T.A. Fergus).
http://dx.doi.org/10.1016/j.jocrd.2015.06.007 2211-3649/& 2015 Elsevier Inc. All rights reserved.
number of qualities, including that mental contamination, unlike contact contamination, can develop due to reflection upon certain thoughts, memories, and images (see Fairbrother, Newth, & Rachman, 2005, for a review). Moreover, mental contamination, unlike contact contamination, does not require a tangible external source (Fairbrother et al., 2005). As such, Fairbrother et al. (2005) noted that it may be difficult for an individual to identify the source of mental contamination or the location of perceived dirtiness, as mental contamination can be generated from an internal process. Although they share overlap, mental contamination and contact contamination are distinguishable (Coughtrey, Shafran, Knibbs, & Rachman, 2012) and, thus, studies have focused specifically upon improving our understanding of mental contamination as its own distinct construct (e.g., Badour, Feldner, Blumenthal, & Bujarski, 2013; Badour, Ojserkis McKay, & Feldner, 2014; Coughtrey, Shafran, & Rachman, 2014; Elliott & Radomsky, 2009; Fairbrother et al., 2005; Herba & Rachman, 2007; Radomsky & Elliott, 2009; Radomsky, Rachman, Shafran, Coughtrey, & Barber, 2014). A consistent finding across these prior studies is an association between disgust and mental contamination, which provides support for the viewpoint that contamination concerns originate as a result of disgust or anticipated exposure to stimuli that elicit disgust (Olatunji, Cisler, McKay, & Phillips, 2010). Disgust is a basic emotion that has traditionally been thought to arise as the result of a need to beware of potentially harmful contact substances; however, its application is also seen in the
R. Travis, T.A. Fergus / Journal of Obsessive-Compulsive and Related Disorders 6 (2015) 114–119
repulsion of morally or socially disgusting acts (Olatunji & Sawchuk, 2005). Researchers have distinguished between two disgust-related vulnerability factors: (a) disgust propensity, defined as one's general tendency to experience disgust, and (b) disgust sensitivity, defined as one's negative appraisal of the experience of disgust (van Overveld, De Jong, Peters, Cavanagh & Davey, 2006). Extensions of these definitions propose that disgust propensity is a predisposition toward a specific type of negative affect (i.e., disgust) and disgust sensitivity represents fear of experiencing disgust (Olatunji et al., 2010). Empirical support for the distinctiveness of disgust propensity and disgust sensitivity comes from research finding that they are structurally distinguishable (Olatunji, Cisler, Deacon, Connolly, & Lohr, 2007; Fergus & Valentiner, 2009; van Overveld et al., 2006) and evidence a unique pattern of associations with criterion variables (Cisler, Olatunji, & Lohr, 2009; Fergus & Valentiner, 2009; Olatunji et al., 2007, 2010). As noted, prior studies have supported an association between disgust and mental contamination. However, prior studies examining this association have often used single-item indicators of disgust (e.g., Rachman, Radomsky, Elliott, & Zysk, 2012) or measures that only target disgust propensity (e.g., Radomsky & Elliott, 2009). In fact, only one known published study has examined how disgust sensitivity relates to mental contamination. In that study, Badour et al. (2013) found that disgust sensitivity shared a positive association with mental contamination. A notable limitation of Badour et al.'s (2013) study is that they isolated the relation between disgust sensitivity and mental contamination rather than simultaneously examining how both disgust propensity and disgust sensitivity relate to mental contamination. Isolating relations between each disgust vulnerability factor is a notable limitation because “disgust propensity and sensitivity may interact and predict disgust-related psychopathological complaints” (van Overveld et al., 2006, p. 1242). Disgust sensitivity would conceptually be expected to serve as the moderator of the relationship between disgust propensity and outcomes, as van Overveld et al. (2006) noted that the construct of disgust sensitivity parallels Reiss's (1987) concept of the fear of anxiety (i.e., anxiety sensitivity). According to Reiss (1987), fear of anxiety enhances the discomfort level of anxiety and, thus, serves as an amplifying factor. Consistent with this possibility, research has found that fear of anxiety moderates the relationship between the propensity to experience anxiety and outcomes (e.g., Dixon, Stevens, & Viana, 2014). Providing evidence that disgust sensitivity may similarly serve as an amplifying factor, Engelhard, Olatunji, and de Jong (2011) found that disgust sensitivity moderates the relationship between disgust reactions experienced during a traumatic event and posttraumatic stress symptoms. We propose that the moderating effect of disgust sensitivity extends to mental contamination. Clinical observations made by Coughtrey, Shafran, Lee, and Rachman (2013) indicate that the misappraisal of negatively valenced emotions is related to mental contamination, such that individuals misappraise emotions as a sign of having done something wrong and, consequently, experience an internal sense of dirtiness. Further, disgust evokes physiological changes in heart rate and skin conductance, indicating parasympathetic nervous system response (Cisler, Olatunji, Sawchuk, & Lohr, 2008). It is possible that individuals who are marked by heightened fear of disgust (i.e., disgust sensitivity; Olatunji et al., 2010) may be more likely to misinterpret those reactions as a sign of internal dirtiness. Indeed, as reviewed, mental contamination typically does not involve an external contaminant, but occurs as a result of internal cues (Fairbrother et al., 2005). As such, and paralleling the amplifying nature of the fear of anxiety (Dixon et al., 2014; Reiss, 1987), disgust sensitivity may contribute to misappraisals of disgust reactions and, thus, strengthen the
115
association between the frequency of disgust reactions (i.e., disgust propensity) and mental contamination. If our speculation is tenable, disgust sensitivity should moderate the association between disgust propensity and mental contamination. If an interaction between disgust propensity and disgust sensitivity was supported, we next sought to examine its robustness by controlling for negative affect, which is a correlate of disgust and contamination fears (e.g., Cisler et al., 2009). We also controlled for the effects of contamination fears more broadly to account for overlap with contamination fears that extend beyond mental contamination. Study findings were expected to contribute to research seeking to identify factors that may help account for the experience of mental contamination (e.g., Herba & Rachman, 2007).
2. Method 2.1. Participants The sample consisted of 478 adults located in the United States recruited through Amazon's Mechanical Turk (MTurk), an online crowdsourcing website. Recruitment was limited to MTurk workers over 18 years of age and worker specifications included requiring participants to have internet protocol (IP) addresses located in the United States. Methods to improve MTurk data quality are important and have been of interest to researchers (Paolacci & Chandler, 2014). Although “catch” questions are sometimes used in an attempt to improve data quality, Paolacci and Chandler (2014) recommend not using such questions because they “have high measurement error, rely on the questionable assumption that measured attentiveness is constant throughout the task, and may tap into correlated traits rather than state-level differences in attentiveness” (p. 186). We followed Paolacci and Chandler's (2014) recommendation and sought to improve data quality by restricting MTurk worker approval ratings, as research has found that “catch” questions do not improve data quality above and beyond recruiting MTurk workers with approval ratings above 95% (Peer, Vosgerau, & Acquisti, 2014). Worker specifications in the present study included restricting participation to MTurk workers who had approval ratings above 95% (following Peer et al., 2014). The mean age of the sample was 33.5 years (SD ¼ 12.5; ranging from 18 to 75) and respondents predominantly self-identified as female (58.8%). In terms of racial identification, 79.5% of the sample self-identified as White, 7.1% as Asian, 5.7% as Black, 3.6% as multi-racial, 3.4% as Latino, and 0.8% as American Indian. A majority of the sample reported receiving a two-year college degree or higher (59.7%) and as currently employed at least parttime (67.8%). 2.2. Measures 2.2.1. Vancouver Obsessional Compulsive Inventory-Mental Contamination Scale (VOCI-MC; Radomsky et al., 2014) The VOCI-MC is a 20-item measure that assesses mental contamination fears on a trait level (e.g., I often feel dirty or contaminated even though I haven't touched anything dirty) using a 5-point scale (ranging from 0 to 4). The items of the VOCI-MC initially were developed for inclusion in the VOCI (Thordarson et al., 2004), but were not formally included in the published measure. The VOCI-MC shares a strong correlation with a measure of contact contamination (rs ranging from .61–.76) and that correlation is stronger than its correlation with general distress (rs ranging from .12–.41; Radomsky et al., 2014). Further, the VOCI-MC continues to relate to criterion variables after controlling for the
116
R. Travis, T.A. Fergus / Journal of Obsessive-Compulsive and Related Disorders 6 (2015) 114–119
effects of general distress and contact contamination. Higher scores indicate greater mental contamination. 2.2.2. Disgust Propensity and Sensitivity Scale-Revised (DPSS-R; van Overveld et al., 2006) The original DPSS-R consisted of 16 items that assessed disgust propensity (e.g., I experience disgust) and disgust sensitivity (e.g., It scares me when I feel faint), respectively, on a 5-point scale (ranging from 1 to 5). Olatunji et al. (2007) identified four DPSS-R items that evidenced poor factorial stability and, following their recommendations, Fergus and Valentiner (2009) examined a 12item DPSS-R that removed those four items. Fergus and Valentiner (2009) found that the disgust propensity and disgust sensitivity scales of a 12-item DPSS-R, consisting of six items each, shared a strong intercorrelation (r ¼.59) and small-to-moderate correlations with disgust-relevant criterion measures (rs ranging from .28–.39). Fergus and Valentiner's (2009) 12-item DPSS-R was used in this study. Higher scores on the DPSS-R indicate greater trait levels of disgust propensity and disgust sensitivity, respectively.
full. Compensation was $1 (USD), an amount consistent with the precedence for compensation given to MTurk workers completing prior studies of similar length (Buhrmester et al., 2011).
3. Results 3.1. Preliminary analyses Descriptive statistics and zero-order correlations among the study variables are presented in Table 1. Both covariates (negative affect and broader contamination fears) significantly correlated with disgust propensity, disgust sensitivity, and mental contamination. Hierarchical multiple linear regressions were used to examine the study predictions. The maximum variance inflation factor (VIF) was 1.91 in the regression models, well below conventional guidelines for indicating problems with multicollinearity (410; Cohen, Cohen, West, & Aiken, 2003). 3.2. Moderating effect of disgust sensitivity
2.2.3. Positive and Negative Affect Schedule (PANAS; Watson, Clark, & Tellegen, 1988) The PANAS asks respondents to indicate to what extent singleword descriptors (e.g., distressed) capture how they felt over the past week on a 5-point scale (ranging from 1 to 5). The negative affect scale of the PANAS – the PANAS scale of interest – consists of 10 items. The negative affect scale has shown strong correlations with other measures of negative affect (rs ranging from .51–.74; Watson et al., 1988). Higher scores indicate greater negative affect. 2.2.4. The Dimensional Obsessive-Compulsive Scale (DOCS; Abramowitz et al., 2010) The DOCS is a 20-item measure that assesses the severity of obsessive-compulsive symptoms using a 5-point scale (ranging from 0 to 4). The contamination scale – the DOCS scale of interest in this study – asks respondents to rate the severity of contamination fears (e.g., Thoughts or feelings that you are contaminated because you came into contact with (or were nearby) a certain object or person) along five aspects. These aspects pertain to the amount of time occupied by intrusive thoughts and neutralizing behavior, engagement in avoidance behavior, associated distress, interference in daily living, and attempts to control intrusive thoughts and refrain from engaging in neutralizing behavior. Although the contamination scale includes examples of contact contamination for respondents to consider when responding to items, it does not force respondents to only consider that type of contamination. Nonetheless, the contamination scale shares strong correlations with measures of contact contamination (rs ranging from .57–.88; Abramowitz et al., 2010) and seemed adequate for the purpose of controlling for the effects of contamination fears that extend beyond mental contamination. Higher scores indicate greater contamination fears. 2.3. Procedure Participants were recruited using MTurk, an internet-based platform that allows individuals to request the completion of jobs (e.g., survey completion) for monetary compensation. Respondents completing surveys through MTurk have been found to produce high quality data and are more demographically diverse than American undergraduate samples (Buhrmester, Kwang, & Gosling, 2011; Paolacci & Chandler, 2014; Shapiro, Chandler, & Mueller, 2013). The present research was approved by the local institutional review board. Participants were required to provide electronic consent and there was no penalty for study withdrawal. Upon completion of the study, participants were debriefed and paid in
Moderation analyses were completed following the recommendations of Aiken and West (1991), which included meancentering the statistical predictors in Step 1 to examine main effects and examining moderation by including the product of the centered predictors (interaction term) in Step 2 of a hierarchical multiple linear regression. Regression results not including the covariates are shown in Table 2. As shown, disgust propensity and disgust sensitivity evidenced main effects in relation to mental contamination in Step 1 of the regression model. However, as predicted, disgust propensity and disgust sensitivity evidenced an interaction in Step 2. We next used Hayes's (2013) PROCESS macro to a provide 95% confidence interval (CI) for the interaction term. PROCESS included running the analyses with 1000 bootstrapped samples. A 95% CI not containing zero is indicative of a significant effect. The 95% CI for the interaction term in relation to mental contamination did not contain zero, 95% CI [0.02, 0.10]. The magnitude of the interaction was small (Cohen's f2 ¼.02). 3.3. Covariates and simple effects We examined whether the observed interactive effect was attributable to overlap with negative affect (PANAS-Negative Affect) or broader contamination fears (DOCS-Contamination). The interaction remained significant while controlling for both covariates, β ¼.09, t ¼2.69, p o.01, 95% CI [0.01, 0.07]. Graphs and simple regression equations (simple effects) were used next to further
Table 1 Descriptive statistics and zero-order correlations. Variable
Mean SD
1
1. DPSS-R-Disgust Propensity 2. DPSS-R-Disgust Sensitivity 3. VOCI-Mental Contamination 4. DOCS-Contamination 5. PANAS-Negative Affect
15.42
4.12
(.85)
12.27
4.50 .65nn (.81)
11.58 12.44 .51nn 2.82 18.50
2
3
4
5
.52nn (.95)
2.91 .44nn .48nn .69nn (.84) 7.91 .31nn .37nn .53nn .40nn (.92)
Note. N ¼ 478. nn p o .01 (two-tailed). Cronbach's alpha values listed in parentheses along the diagonal. DPSS-R ¼ Disgust Propensity and Sensitivity Scale-Revised; VOCI ¼Vancouver Obsessional Compulsive Inventory; DOCS ¼ Dimensional ObsessiveCompulsive Scale; PANAS ¼Positive and Negative Affect Schedule.
R. Travis, T.A. Fergus / Journal of Obsessive-Compulsive and Related Disorders 6 (2015) 114–119
Table 2 Regression results examining moderating effect of disgust sensitivity without covariates. Variable
VOCI-Mental Contamination ΔR2
Step 1 DP DS Step 2 DPxDS
Step 1
Step 2
β
t
β
t
.30 .32
6.01nn 6.42nn
.30 .28
6.15nn 5.31nn
.12
2.84nn
nn
.32
.01nn
Note: N ¼478. nn p o.01 (two-tailed). DP¼ Disgust Propensity and DS ¼ Disgust Sensitivity scales of Disgust Propensity and Sensitivity Scale-Revised. VOCI ¼ Vancouver Obsessional Compulsive Inventory.
Fig. 1. The potentiating effect of disgust sensitivity (DS) on the relationship between disgust propensity (DP) and mental contamination.
investigate the significant interaction term (following Aiken & West, 1991). Two simple regression equations were constructed (using þ1 SD from the mean score on the disgust sensitivity scale of the DPSS-R) for each model to depict the interaction effect. To plot these equations, two values of the disgust propensity scale of the DPSS-R (þ1 SD from the mean) were substituted into the equations. Simple effects indicated that disgust propensity shared an association with mental contamination at high (β ¼.39, t¼ 6.64, p o.01, 95% CI [0.83, 1.52]) and low (β ¼.22, t¼ 3.88, p o.01, 95% CI [0.33, 1.01]) disgust sensitivity scores. The significant omnibus interaction term in the earlier reported regression analysis indicated that the slopes of the two simple effects significantly differed from one another (Aiken & West, 1991). As such, disgust propensity shared a significantly stronger association with mental contamination at high, relative to low, disgust sensitivity scores. Simple effects are depicted in Fig. 1.
4. Discussion The results of this study supported predictions that disgust sensitivity impacts the association between disgust propensity and mental contamination, with the observed interactive effect being robust to the effects of negative affect and broader contamination fears. The pattern of the interactive effect indicated that the association between disgust propensity and mental contamination was potentiated by disgust sensitivity, such that the association was strongest at high disgust sensitivity scores. It is important to note that simple effects indicated that disgust propensity still
117
shared a relationship with mental contamination at low disgust sensitivity scores. One implication of the pattern of associations is that disgust sensitivity is not necessary for disgust to relate to mental contamination; rather, it strengthens the relationship of disgust propensity and mental contamination. Disgust sensitivity may not be necessary for disgust propensity to relate to mental contamination because individuals who frequently experience disgust reactions (i.e., reflecting disgust propensity; van Overveld et al., 2006) may have greater opportunities to misappraise disgust as a sign of internal dirtiness, thereby still being associated with mental contamination (Coughtrey et al., 2013). Disgust sensitivity may potentiate the likelihood that individuals misappraise disgust reactions, as individuals who have heightened fear of disgust may be more likely to find disgust reactions intolerable and become preoccupied with the implications of the reactions (e.g., “I will never be clean again”; Badour et al., 2013). As noted by Badour et al. (2013), the preoccupation with disgust reactions are linked to greater mental contamination, distress (including even greater disgust), and maladaptive coping (e.g., washing). Future longitudinal and experimental studies are needed to explicate whether this explanation for the observed associations is tenable. While disgust and mental contamination are often experienced in the general population, these results may also be applicable to those who frequently experience disgust and mental contamination. Fear of contamination is a hallmark of certain types of obsessive-compulsive disorder (OCD) and this fear can include mental contamination (Rachman, 2004). Preliminary research indicates that mental contamination shares a unique association with obsessive-compulsive symptoms and is elevated among individuals with contamination-based OCD (Radomsky et al., 2014). Prior research has found that disgust shares a robust association with mental contamination (e.g., Badour et al., 2014) and contamination-based obsessive-compulsive symptoms (e.g., Olatunji et al., 2010). Coughtrey et al. (2013) examined the effectiveness of treating OCD with intervention strategies geared toward modifying cognitive aspects of mental contamination. The fear of disgust was not a direct target of intervention. If future studies continue to support the potentiating effect of disgust sensitivity in relation to mental contamination, interventions for OCD designed to reduce mental contamination may consider targeting fear of disgust. The present results indicate that mitigating fear of disgust may reduce mental contamination even among individuals who experience frequent disgust reactions, which may ultimately have implications for the treatment of OCD. As discussed, there has been interest in comparing and contrasting mental versus contact contamination (Fairbrother et al., 2005). Considered in conjunction with prior findings, the present results may highlight further potential similarities and differences between these two types of contamination concerns. For example, disgust propensity and disgust sensitivity both evidenced independent main effects in relation to mental contamination in the present study and in relation to contact contamination in a prior study completed by Cisler et al. (2009). Such findings support proposals that the tendency to experience disgust reactions and the fear of disgust are important unique factors for contamination concerns (Olatunji & Broman-Fulks, 2009; Olatunji et al., 2010). Contrary to the present results, Cisler et al. (2009) found that disgust propensity and disgust sensitivity did not interact in relation to contact contamination. Cisler et al.'s (2009) findings suggest that general emotional dysregulation, rather than disgust sensitivity, potentiates the effect of disgust propensity on contact contamination. Cisler et al. (2009) interpreted those findings to indicate that difficulties regulating negative emotions in general underlie contact contamination. As such, disgust sensitivity may uniquely contribute to, but not serve as an amplifier for, contact
118
R. Travis, T.A. Fergus / Journal of Obsessive-Compulsive and Related Disorders 6 (2015) 114–119
contamination because contact contamination is marked by difficulties regulating many negative emotions. Alternatively, difficulties regulating disgust may be more relevant for mental contamination, with Elliott and Radomsky (2009) stating that “feelings of disgust are usually experienced alongside feelings of mental contamination” (p. 996). Although it remains to be empirically tested, difficulties regulating disgust may be more important to mental contamination than are difficulties regulating negative emotions more broadly. The study results and above discussion points should be considered with the following study limitations in mind. It is important to note that the magnitude of the observed interactive effect was small in size. Nonetheless, the interactive effect was (a) consistent with the magnitude of interactive effects found in prior studies examining the moderating impact of disgust sensitivity (e.g., Cisler et al., 2009), (b) in a typical range for the magnitude of moderation effects founds in prior self-report based studies (e.g., 1–3% of variance explained; Aiken & West, 1991) and (c) associated with a percentage of variance that can be considered a meaningful amount for moderation analyses (i.e., at least 1%) based upon Monte Carlo studies (Evans, 1985). Moreover, the interactive effect continued to account for unique variance in mental contamination scores even after statistically accounting for negative affect and broader contamination fears. As such, although modest in magnitude, we believe the observed interaction is meaningful and should be interpreted as such. Another limitation is the method by which the sample was obtained; while MTurk users are more demographically diverse than both American undergraduate samples, MTurk users may not be representative of the general population (Buhrmester et al., 2011; Paolacci & Chandler, 2014; Shapiro et al., 2013). Although our use of a community sample was supported by continuous nature of disgust propensity, disgust sensitivity, and contamination concerns (Olatunji & Broman-Fulks, 2009), it is important to replicate these findings among respondents who likely experience consistently higher scores on the mental contamination and disgust measures (e.g., individuals with obsessive-compulsive disorder or who have trauma exposure; Badour et al., 2013; Radomsky et al., 2014). Additionally, our study is limited by the sole use of self-report measures, as the monomethod assessment may have inflated study associations. Our methodology is unlikely to be the reason for the observed interactive effect, though, as correlated measurement error, which can be present as a result of monomethod assessment (Aiken & West, 1991), has not been found to produce spurious interactions and actually seems to attenuate the strength of interactions (Evans, 1985). Finally, although the study variables were modeled in a manner consistent with theory and prior studies, the cross-sectional design precludes causal conclusions related to study variable associations. In sum, disgust sensitivity appears to be an amplifying factor in relation to mental contamination. Given the robust association found between disgust and mental contamination across studies, future directions may include examining how mental contamination relates to other disgust-related phenomena. For example, appraisals of various situations, or even certain people (e.g., outgroup members; Cottrell & Neuberg, 2005), may be related to disgust responses, and, consequently, mental contamination. As such, even broader benefits of elucidating the link between disgust and mental contamination in future studies may be realized.
References Abramowitz, J. S., Deacon, B. J., Olatunji, B. O., Wheaton, M. G., Berman, N. C., Losardo, D., et al. (2010). Assessment of obsessive-compulsive symptom
dimensions: development and evaluation of the Dimensional Obsessive-Compulsive Scale. Psychological Assessment, 22, 180–198. Aiken, L. S., & West, S. G. (1991). Multiple regression: testing and interpreting interactions. Thousand Oaks, CA: Sage. Badour, C. L., Feldner, M. T., Blumenthal, H., & Bujarski, S. J. (2013). Examination of increased mental contamination as a potential mechanism in the association between disgust sensitivity and sexual assault-related posttraumatic stress. Cognitive Therapy and Research, 37, 697–703. Badour, C. L., Ojserkis, R., McKay, D., & Feldner, M. T. (2014). Disgust as a unique affective predictor of mental contamination following sexual trauma. Journal of Anxiety Disorders, 28, 704–711. Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazon's Mechanical Turk: a new source of inexpensive, yet high quality, data? Perspectives on Psychological Science, 6, 3–5. Cisler, J. M., Olatunji, B. O., & Lohr, J. M. (2009). Disgust sensitivity and emotion regulation potentiate the effect of disgust propensity on spider fear, blood-injection-injury fear, and contamination fear. Journal of Behavior Therapy and Experimental Psychiatry, 40, 219–229. Cisler, J. M., Olatunji, B. O., Sawchuk, C. N., & Lohr, J. M. (2008). Specificity of emotional maintenance processes among contamination fears and blood–injection–injury fears. Journal of Anxiety Disorders, 22, 915–923. Cohen, J., Cohen, P., West, S. G., & Aiken, L. A. (2003). Applied multiple regression/ correlation analysis for the behavioral sciences (3rd ed.). Hillsdale, NJ: Lawrence Erlbaum. Cottrell, C. A., & Neuberg, S. L. (2005). Different emotional reactions to different groups: a sociofunctional threat-based approach to “prejudice.”. Journal of Personality and Social Psychology, 88, 770–789. Coughtrey, A. E., Shafran, R., Knibbs, D., & Rachman, S. J. (2012). Mental contamination in obsessive-compulsive disorder. Journal of Obsessive-Compulsive and Related Disorders, 1, 244–250. Coughtrey, A. E., Shafran, R., Lee, M., & Rachman, S. J. (2013). The treatment of mental contamination: a case series. Cognitive and Behavioral Practice, 20, 221–231. Coughtrey, A. E., Shafran, R., & Rachman, S. J. (2014). The spread of mental contamination. Journal of Behavior Therapy and Experimental Psychiatry, 45, 33–38. Dixon, L. J., Stevens, E. N., & Viana, A. G. (2014). Anxiety sensitivity as a moderator of the relationship between trait anxiety and illicit substance use. Psychology of Addictive Behaviors, 28, 1284–1289. Elliott, C. M., & Radomsky, A. S. (2009). Analyses of mental contamination: Part I, experimental manipulations of morality. Behaviour Research and Therapy, 47, 995–1003. Engelhard, M., Olatunji, B. O., & de Jong, P. J. (2011). Disgust and the development of posttraumatic stress among soldiers deployed to Afghanistan. Journal of Anxiety Disorders, 25, 58–63. Evans, M. G. (1985). A Monte Carlo study of the effects of correlated method variance in moderated multiple regression analysis. Organizational Behavior and Human Decision Processes, 36, 305–323. Fairbrother, N., Newth, S. J., & Rachman, S. (2005). Mental pollution: feelings of dirtiness without contact contact. Behaviour Research and Therapy, 43, 121–130. Fergus, T. A., & Valentiner, D. P. (2009). The disgust propensity and sensitivity scalerevised: an examination of a reduced-item version. Journal of Anxiety Disorders, 23, 703–710. Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: a regression-based approach. New York, NY: Guilford. Herba, J. K., & Rachman, S. (2007). Vulnerability to mental contamination. Behaviour Research and Therapy, 45, 2804–2812. Olatunji, B. O., & Broman-Fulks, J. J. (2009). Latent structure of aversion: taxometric exploration. Journal of Anxiety Disorders, 23, 87–92. Olatunji, B. O., Cisler, J. M., Deacon, B. J., Connolly, K., & Lohr, J. M. (2007). The disgust propensity and sensitivity scale–revised: psychometric properties and specificity in relation to anxiety disorder symptoms. Journal of Anxiety Disorders, 21, 918–930. Olatunji, B. O., Cisler, J., McKay, D., & Phillips, M. L. (2010). Is disgust associated with psychopathology? Emerging research in the anxiety disorders. Psychiatry Research, 175, 1–10. Olatunji, B. O., Moretz, M. W., Wolitzky-Taylor, K. B., McKay, D., McGrath, P. B., & Ciesielski, B. G. (2010). Disgust vulnerability and symptoms of contaminationbased OCD: descriptive tests of incremental specificity. Behavior Therapy, 41, 475–490. Olatunji, B. O., & Sawchuk, C. N. (2005). Disgust: characteristic features, social manifestations, and clinical implications. Journal of Social and Clinical Psychology, 24, 932–962. Paolacci, G., & Chandler, J. (2014). Inside the turk: understanding Mechanical Turk as a participant pool. Current Directions in Psychological Science, 23, 184–188. Peer, E., Vosgerau, J., & Acquisti, A. (2014). Reputation as a sufficient condition for data quality on Amazon Mechanical Turk. Behavior Research Methods, 46, 1023–1031. Rachman, S. (2004). Fear of contamination. Behaviour Research and Therapy, 42, 1227–1255. Rachman, S., Radomsky, A. S., Elliott, C. M., & Zysk, E. (2012). Mental contamination: the perpetrator effect. Journal of Behavior Therapy and Experimental Psychiatry, 43, 587–593. Radomsky, A. S., & Elliott, C. M. (2009). Analyses of mental contamination: Part II, individual differences. Behaviour Research and Therapy, 47, 1004–1011.
R. Travis, T.A. Fergus / Journal of Obsessive-Compulsive and Related Disorders 6 (2015) 114–119
Radomsky, A. S., Rachman, S., Shafran, R., Coughtrey, A. E., & Barber, K. C. (2014). The nature and assessment of mental contamination: a psychometric analysis. Journal of Obsessive-Compulsive and Related Disorders, 3, 181–187. Reiss, S. (1987). Theoretical perspectives on the fear of anxiety. Clinical Psychology Review, 7, 585–596. Shapiro, D. N., Chandler, J., & Mueller, P. A. (2013). Using Mechanical Turk to study clinical populations. Clinical Psychological Science, 1, 213–220. Thordarson, D. S., Radomsky, A. S., Rachman, S., Shafran, R., Sawchuk, C. N., & Hakstian, A. R. (2004). The Vancouver Obsessional Compulsive Inventory (VOCI). Behaviour Research and Therapy, 42, 1289–1314.
119
van Overveld, W. J. M., De Jong, P. J., Peters, M. L., Cavanagh, K., & Davey, G. C. L. (2006). Disgust propensity and disgust sensitivity: separate constructs that are differentially related to specific fears. Personality and Individual Differences, 41, 1241–1252. Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of Positive and Negative Affect: The PANAS Scales. Journal of Personality and Social Psychology, 54, 1063–1070.
Contents lists available at ScienceDirect
Journal of Obsessive-Compulsive and Related Disorders journal homepage: www.elsevier.com/locate/jocrd
Clinical report
The potentiating effect of disgust sensitivity on the relationship between disgust propensity and mental contamination Rosemond Travis, Thomas A. Fergus n Baylor University, USA
art ic l e i nf o
a b s t r a c t
Article history: Received 5 March 2015 Received in revised form 25 June 2015 Accepted 29 June 2015 Available online 2 July 2015
Disgust is important to mental contamination, a contamination fear that arises in the absence of physical contact with a perceived contaminant. Researchers have distinguished between disgust propensity, defined as one's general tendency to experience disgust, and disgust sensitivity, defined as one's negative appraisal of the experience of disgust. Based upon speculations that disgust sensitivity may amplify the experience of disgust propensity on disgust-relevant outcomes, this study examined the interaction of disgust propensity and disgust sensitivity in relation to mental contamination among a community sample of adults located in the United States recruited through Amazon's Mechanical Turk (N¼ 478). The results suggest that disgust sensitivity potentiates the effect of disgust propensity on mental contamination. The interactive effect was robust to the effects of negative affect and broader contamination fears. These results indicate that mental contamination is particularly strong among individuals with concurrently high disgust propensity and disgust sensitivity. Implications and future directions are explored. & 2015 Elsevier Inc. All rights reserved.
Keywords: Disgust Disgust propensity Disgust sensitivity Mental contamination
1. Introduction A common trigger for psychological distress comes from a desire to keep one's self or environment clean. This desire for cleanliness is often perpetuated by a fear of contamination (Rachman, 2004). As noted by Rachman (2004), fear of contamination is a near-universal experience and is defined as a fear related to coming into contact with a person or item that is believed to be dirty, whether this contact is direct (e.g., as when one touches garbage) or indirect (e.g., as when one touches an object that trash may have once touched). Those who have a severe fear of contamination tend to either try to limit their contact with potentially unclean people or items and/or engage in cleansing behavior after they come into contact with a perceived contaminant (Rachman, 2004). Whereas contact contamination originates from physical contact with a perceived unclean stimulus, contamination can arise from merely observing or thinking about something unclean, immoral, or undesirable (Rachman, 2004). As described by Rachman (2004), contamination that occurs in the absence of physical contact with a contaminant is called mental contamination. Mental contamination differs from contact contamination along a n Correspondence to: Department of Psychology & Neuroscience, Baylor University, Waco, TX 76798, USA. Fax: þ 1 254 710 3033. E-mail address: [email protected] (T.A. Fergus).
http://dx.doi.org/10.1016/j.jocrd.2015.06.007 2211-3649/& 2015 Elsevier Inc. All rights reserved.
number of qualities, including that mental contamination, unlike contact contamination, can develop due to reflection upon certain thoughts, memories, and images (see Fairbrother, Newth, & Rachman, 2005, for a review). Moreover, mental contamination, unlike contact contamination, does not require a tangible external source (Fairbrother et al., 2005). As such, Fairbrother et al. (2005) noted that it may be difficult for an individual to identify the source of mental contamination or the location of perceived dirtiness, as mental contamination can be generated from an internal process. Although they share overlap, mental contamination and contact contamination are distinguishable (Coughtrey, Shafran, Knibbs, & Rachman, 2012) and, thus, studies have focused specifically upon improving our understanding of mental contamination as its own distinct construct (e.g., Badour, Feldner, Blumenthal, & Bujarski, 2013; Badour, Ojserkis McKay, & Feldner, 2014; Coughtrey, Shafran, & Rachman, 2014; Elliott & Radomsky, 2009; Fairbrother et al., 2005; Herba & Rachman, 2007; Radomsky & Elliott, 2009; Radomsky, Rachman, Shafran, Coughtrey, & Barber, 2014). A consistent finding across these prior studies is an association between disgust and mental contamination, which provides support for the viewpoint that contamination concerns originate as a result of disgust or anticipated exposure to stimuli that elicit disgust (Olatunji, Cisler, McKay, & Phillips, 2010). Disgust is a basic emotion that has traditionally been thought to arise as the result of a need to beware of potentially harmful contact substances; however, its application is also seen in the
R. Travis, T.A. Fergus / Journal of Obsessive-Compulsive and Related Disorders 6 (2015) 114–119
repulsion of morally or socially disgusting acts (Olatunji & Sawchuk, 2005). Researchers have distinguished between two disgust-related vulnerability factors: (a) disgust propensity, defined as one's general tendency to experience disgust, and (b) disgust sensitivity, defined as one's negative appraisal of the experience of disgust (van Overveld, De Jong, Peters, Cavanagh & Davey, 2006). Extensions of these definitions propose that disgust propensity is a predisposition toward a specific type of negative affect (i.e., disgust) and disgust sensitivity represents fear of experiencing disgust (Olatunji et al., 2010). Empirical support for the distinctiveness of disgust propensity and disgust sensitivity comes from research finding that they are structurally distinguishable (Olatunji, Cisler, Deacon, Connolly, & Lohr, 2007; Fergus & Valentiner, 2009; van Overveld et al., 2006) and evidence a unique pattern of associations with criterion variables (Cisler, Olatunji, & Lohr, 2009; Fergus & Valentiner, 2009; Olatunji et al., 2007, 2010). As noted, prior studies have supported an association between disgust and mental contamination. However, prior studies examining this association have often used single-item indicators of disgust (e.g., Rachman, Radomsky, Elliott, & Zysk, 2012) or measures that only target disgust propensity (e.g., Radomsky & Elliott, 2009). In fact, only one known published study has examined how disgust sensitivity relates to mental contamination. In that study, Badour et al. (2013) found that disgust sensitivity shared a positive association with mental contamination. A notable limitation of Badour et al.'s (2013) study is that they isolated the relation between disgust sensitivity and mental contamination rather than simultaneously examining how both disgust propensity and disgust sensitivity relate to mental contamination. Isolating relations between each disgust vulnerability factor is a notable limitation because “disgust propensity and sensitivity may interact and predict disgust-related psychopathological complaints” (van Overveld et al., 2006, p. 1242). Disgust sensitivity would conceptually be expected to serve as the moderator of the relationship between disgust propensity and outcomes, as van Overveld et al. (2006) noted that the construct of disgust sensitivity parallels Reiss's (1987) concept of the fear of anxiety (i.e., anxiety sensitivity). According to Reiss (1987), fear of anxiety enhances the discomfort level of anxiety and, thus, serves as an amplifying factor. Consistent with this possibility, research has found that fear of anxiety moderates the relationship between the propensity to experience anxiety and outcomes (e.g., Dixon, Stevens, & Viana, 2014). Providing evidence that disgust sensitivity may similarly serve as an amplifying factor, Engelhard, Olatunji, and de Jong (2011) found that disgust sensitivity moderates the relationship between disgust reactions experienced during a traumatic event and posttraumatic stress symptoms. We propose that the moderating effect of disgust sensitivity extends to mental contamination. Clinical observations made by Coughtrey, Shafran, Lee, and Rachman (2013) indicate that the misappraisal of negatively valenced emotions is related to mental contamination, such that individuals misappraise emotions as a sign of having done something wrong and, consequently, experience an internal sense of dirtiness. Further, disgust evokes physiological changes in heart rate and skin conductance, indicating parasympathetic nervous system response (Cisler, Olatunji, Sawchuk, & Lohr, 2008). It is possible that individuals who are marked by heightened fear of disgust (i.e., disgust sensitivity; Olatunji et al., 2010) may be more likely to misinterpret those reactions as a sign of internal dirtiness. Indeed, as reviewed, mental contamination typically does not involve an external contaminant, but occurs as a result of internal cues (Fairbrother et al., 2005). As such, and paralleling the amplifying nature of the fear of anxiety (Dixon et al., 2014; Reiss, 1987), disgust sensitivity may contribute to misappraisals of disgust reactions and, thus, strengthen the
115
association between the frequency of disgust reactions (i.e., disgust propensity) and mental contamination. If our speculation is tenable, disgust sensitivity should moderate the association between disgust propensity and mental contamination. If an interaction between disgust propensity and disgust sensitivity was supported, we next sought to examine its robustness by controlling for negative affect, which is a correlate of disgust and contamination fears (e.g., Cisler et al., 2009). We also controlled for the effects of contamination fears more broadly to account for overlap with contamination fears that extend beyond mental contamination. Study findings were expected to contribute to research seeking to identify factors that may help account for the experience of mental contamination (e.g., Herba & Rachman, 2007).
2. Method 2.1. Participants The sample consisted of 478 adults located in the United States recruited through Amazon's Mechanical Turk (MTurk), an online crowdsourcing website. Recruitment was limited to MTurk workers over 18 years of age and worker specifications included requiring participants to have internet protocol (IP) addresses located in the United States. Methods to improve MTurk data quality are important and have been of interest to researchers (Paolacci & Chandler, 2014). Although “catch” questions are sometimes used in an attempt to improve data quality, Paolacci and Chandler (2014) recommend not using such questions because they “have high measurement error, rely on the questionable assumption that measured attentiveness is constant throughout the task, and may tap into correlated traits rather than state-level differences in attentiveness” (p. 186). We followed Paolacci and Chandler's (2014) recommendation and sought to improve data quality by restricting MTurk worker approval ratings, as research has found that “catch” questions do not improve data quality above and beyond recruiting MTurk workers with approval ratings above 95% (Peer, Vosgerau, & Acquisti, 2014). Worker specifications in the present study included restricting participation to MTurk workers who had approval ratings above 95% (following Peer et al., 2014). The mean age of the sample was 33.5 years (SD ¼ 12.5; ranging from 18 to 75) and respondents predominantly self-identified as female (58.8%). In terms of racial identification, 79.5% of the sample self-identified as White, 7.1% as Asian, 5.7% as Black, 3.6% as multi-racial, 3.4% as Latino, and 0.8% as American Indian. A majority of the sample reported receiving a two-year college degree or higher (59.7%) and as currently employed at least parttime (67.8%). 2.2. Measures 2.2.1. Vancouver Obsessional Compulsive Inventory-Mental Contamination Scale (VOCI-MC; Radomsky et al., 2014) The VOCI-MC is a 20-item measure that assesses mental contamination fears on a trait level (e.g., I often feel dirty or contaminated even though I haven't touched anything dirty) using a 5-point scale (ranging from 0 to 4). The items of the VOCI-MC initially were developed for inclusion in the VOCI (Thordarson et al., 2004), but were not formally included in the published measure. The VOCI-MC shares a strong correlation with a measure of contact contamination (rs ranging from .61–.76) and that correlation is stronger than its correlation with general distress (rs ranging from .12–.41; Radomsky et al., 2014). Further, the VOCI-MC continues to relate to criterion variables after controlling for the
116
R. Travis, T.A. Fergus / Journal of Obsessive-Compulsive and Related Disorders 6 (2015) 114–119
effects of general distress and contact contamination. Higher scores indicate greater mental contamination. 2.2.2. Disgust Propensity and Sensitivity Scale-Revised (DPSS-R; van Overveld et al., 2006) The original DPSS-R consisted of 16 items that assessed disgust propensity (e.g., I experience disgust) and disgust sensitivity (e.g., It scares me when I feel faint), respectively, on a 5-point scale (ranging from 1 to 5). Olatunji et al. (2007) identified four DPSS-R items that evidenced poor factorial stability and, following their recommendations, Fergus and Valentiner (2009) examined a 12item DPSS-R that removed those four items. Fergus and Valentiner (2009) found that the disgust propensity and disgust sensitivity scales of a 12-item DPSS-R, consisting of six items each, shared a strong intercorrelation (r ¼.59) and small-to-moderate correlations with disgust-relevant criterion measures (rs ranging from .28–.39). Fergus and Valentiner's (2009) 12-item DPSS-R was used in this study. Higher scores on the DPSS-R indicate greater trait levels of disgust propensity and disgust sensitivity, respectively.
full. Compensation was $1 (USD), an amount consistent with the precedence for compensation given to MTurk workers completing prior studies of similar length (Buhrmester et al., 2011).
3. Results 3.1. Preliminary analyses Descriptive statistics and zero-order correlations among the study variables are presented in Table 1. Both covariates (negative affect and broader contamination fears) significantly correlated with disgust propensity, disgust sensitivity, and mental contamination. Hierarchical multiple linear regressions were used to examine the study predictions. The maximum variance inflation factor (VIF) was 1.91 in the regression models, well below conventional guidelines for indicating problems with multicollinearity (410; Cohen, Cohen, West, & Aiken, 2003). 3.2. Moderating effect of disgust sensitivity
2.2.3. Positive and Negative Affect Schedule (PANAS; Watson, Clark, & Tellegen, 1988) The PANAS asks respondents to indicate to what extent singleword descriptors (e.g., distressed) capture how they felt over the past week on a 5-point scale (ranging from 1 to 5). The negative affect scale of the PANAS – the PANAS scale of interest – consists of 10 items. The negative affect scale has shown strong correlations with other measures of negative affect (rs ranging from .51–.74; Watson et al., 1988). Higher scores indicate greater negative affect. 2.2.4. The Dimensional Obsessive-Compulsive Scale (DOCS; Abramowitz et al., 2010) The DOCS is a 20-item measure that assesses the severity of obsessive-compulsive symptoms using a 5-point scale (ranging from 0 to 4). The contamination scale – the DOCS scale of interest in this study – asks respondents to rate the severity of contamination fears (e.g., Thoughts or feelings that you are contaminated because you came into contact with (or were nearby) a certain object or person) along five aspects. These aspects pertain to the amount of time occupied by intrusive thoughts and neutralizing behavior, engagement in avoidance behavior, associated distress, interference in daily living, and attempts to control intrusive thoughts and refrain from engaging in neutralizing behavior. Although the contamination scale includes examples of contact contamination for respondents to consider when responding to items, it does not force respondents to only consider that type of contamination. Nonetheless, the contamination scale shares strong correlations with measures of contact contamination (rs ranging from .57–.88; Abramowitz et al., 2010) and seemed adequate for the purpose of controlling for the effects of contamination fears that extend beyond mental contamination. Higher scores indicate greater contamination fears. 2.3. Procedure Participants were recruited using MTurk, an internet-based platform that allows individuals to request the completion of jobs (e.g., survey completion) for monetary compensation. Respondents completing surveys through MTurk have been found to produce high quality data and are more demographically diverse than American undergraduate samples (Buhrmester, Kwang, & Gosling, 2011; Paolacci & Chandler, 2014; Shapiro, Chandler, & Mueller, 2013). The present research was approved by the local institutional review board. Participants were required to provide electronic consent and there was no penalty for study withdrawal. Upon completion of the study, participants were debriefed and paid in
Moderation analyses were completed following the recommendations of Aiken and West (1991), which included meancentering the statistical predictors in Step 1 to examine main effects and examining moderation by including the product of the centered predictors (interaction term) in Step 2 of a hierarchical multiple linear regression. Regression results not including the covariates are shown in Table 2. As shown, disgust propensity and disgust sensitivity evidenced main effects in relation to mental contamination in Step 1 of the regression model. However, as predicted, disgust propensity and disgust sensitivity evidenced an interaction in Step 2. We next used Hayes's (2013) PROCESS macro to a provide 95% confidence interval (CI) for the interaction term. PROCESS included running the analyses with 1000 bootstrapped samples. A 95% CI not containing zero is indicative of a significant effect. The 95% CI for the interaction term in relation to mental contamination did not contain zero, 95% CI [0.02, 0.10]. The magnitude of the interaction was small (Cohen's f2 ¼.02). 3.3. Covariates and simple effects We examined whether the observed interactive effect was attributable to overlap with negative affect (PANAS-Negative Affect) or broader contamination fears (DOCS-Contamination). The interaction remained significant while controlling for both covariates, β ¼.09, t ¼2.69, p o.01, 95% CI [0.01, 0.07]. Graphs and simple regression equations (simple effects) were used next to further
Table 1 Descriptive statistics and zero-order correlations. Variable
Mean SD
1
1. DPSS-R-Disgust Propensity 2. DPSS-R-Disgust Sensitivity 3. VOCI-Mental Contamination 4. DOCS-Contamination 5. PANAS-Negative Affect
15.42
4.12
(.85)
12.27
4.50 .65nn (.81)
11.58 12.44 .51nn 2.82 18.50
2
3
4
5
.52nn (.95)
2.91 .44nn .48nn .69nn (.84) 7.91 .31nn .37nn .53nn .40nn (.92)
Note. N ¼ 478. nn p o .01 (two-tailed). Cronbach's alpha values listed in parentheses along the diagonal. DPSS-R ¼ Disgust Propensity and Sensitivity Scale-Revised; VOCI ¼Vancouver Obsessional Compulsive Inventory; DOCS ¼ Dimensional ObsessiveCompulsive Scale; PANAS ¼Positive and Negative Affect Schedule.
R. Travis, T.A. Fergus / Journal of Obsessive-Compulsive and Related Disorders 6 (2015) 114–119
Table 2 Regression results examining moderating effect of disgust sensitivity without covariates. Variable
VOCI-Mental Contamination ΔR2
Step 1 DP DS Step 2 DPxDS
Step 1
Step 2
β
t
β
t
.30 .32
6.01nn 6.42nn
.30 .28
6.15nn 5.31nn
.12
2.84nn
nn
.32
.01nn
Note: N ¼478. nn p o.01 (two-tailed). DP¼ Disgust Propensity and DS ¼ Disgust Sensitivity scales of Disgust Propensity and Sensitivity Scale-Revised. VOCI ¼ Vancouver Obsessional Compulsive Inventory.
Fig. 1. The potentiating effect of disgust sensitivity (DS) on the relationship between disgust propensity (DP) and mental contamination.
investigate the significant interaction term (following Aiken & West, 1991). Two simple regression equations were constructed (using þ1 SD from the mean score on the disgust sensitivity scale of the DPSS-R) for each model to depict the interaction effect. To plot these equations, two values of the disgust propensity scale of the DPSS-R (þ1 SD from the mean) were substituted into the equations. Simple effects indicated that disgust propensity shared an association with mental contamination at high (β ¼.39, t¼ 6.64, p o.01, 95% CI [0.83, 1.52]) and low (β ¼.22, t¼ 3.88, p o.01, 95% CI [0.33, 1.01]) disgust sensitivity scores. The significant omnibus interaction term in the earlier reported regression analysis indicated that the slopes of the two simple effects significantly differed from one another (Aiken & West, 1991). As such, disgust propensity shared a significantly stronger association with mental contamination at high, relative to low, disgust sensitivity scores. Simple effects are depicted in Fig. 1.
4. Discussion The results of this study supported predictions that disgust sensitivity impacts the association between disgust propensity and mental contamination, with the observed interactive effect being robust to the effects of negative affect and broader contamination fears. The pattern of the interactive effect indicated that the association between disgust propensity and mental contamination was potentiated by disgust sensitivity, such that the association was strongest at high disgust sensitivity scores. It is important to note that simple effects indicated that disgust propensity still
117
shared a relationship with mental contamination at low disgust sensitivity scores. One implication of the pattern of associations is that disgust sensitivity is not necessary for disgust to relate to mental contamination; rather, it strengthens the relationship of disgust propensity and mental contamination. Disgust sensitivity may not be necessary for disgust propensity to relate to mental contamination because individuals who frequently experience disgust reactions (i.e., reflecting disgust propensity; van Overveld et al., 2006) may have greater opportunities to misappraise disgust as a sign of internal dirtiness, thereby still being associated with mental contamination (Coughtrey et al., 2013). Disgust sensitivity may potentiate the likelihood that individuals misappraise disgust reactions, as individuals who have heightened fear of disgust may be more likely to find disgust reactions intolerable and become preoccupied with the implications of the reactions (e.g., “I will never be clean again”; Badour et al., 2013). As noted by Badour et al. (2013), the preoccupation with disgust reactions are linked to greater mental contamination, distress (including even greater disgust), and maladaptive coping (e.g., washing). Future longitudinal and experimental studies are needed to explicate whether this explanation for the observed associations is tenable. While disgust and mental contamination are often experienced in the general population, these results may also be applicable to those who frequently experience disgust and mental contamination. Fear of contamination is a hallmark of certain types of obsessive-compulsive disorder (OCD) and this fear can include mental contamination (Rachman, 2004). Preliminary research indicates that mental contamination shares a unique association with obsessive-compulsive symptoms and is elevated among individuals with contamination-based OCD (Radomsky et al., 2014). Prior research has found that disgust shares a robust association with mental contamination (e.g., Badour et al., 2014) and contamination-based obsessive-compulsive symptoms (e.g., Olatunji et al., 2010). Coughtrey et al. (2013) examined the effectiveness of treating OCD with intervention strategies geared toward modifying cognitive aspects of mental contamination. The fear of disgust was not a direct target of intervention. If future studies continue to support the potentiating effect of disgust sensitivity in relation to mental contamination, interventions for OCD designed to reduce mental contamination may consider targeting fear of disgust. The present results indicate that mitigating fear of disgust may reduce mental contamination even among individuals who experience frequent disgust reactions, which may ultimately have implications for the treatment of OCD. As discussed, there has been interest in comparing and contrasting mental versus contact contamination (Fairbrother et al., 2005). Considered in conjunction with prior findings, the present results may highlight further potential similarities and differences between these two types of contamination concerns. For example, disgust propensity and disgust sensitivity both evidenced independent main effects in relation to mental contamination in the present study and in relation to contact contamination in a prior study completed by Cisler et al. (2009). Such findings support proposals that the tendency to experience disgust reactions and the fear of disgust are important unique factors for contamination concerns (Olatunji & Broman-Fulks, 2009; Olatunji et al., 2010). Contrary to the present results, Cisler et al. (2009) found that disgust propensity and disgust sensitivity did not interact in relation to contact contamination. Cisler et al.'s (2009) findings suggest that general emotional dysregulation, rather than disgust sensitivity, potentiates the effect of disgust propensity on contact contamination. Cisler et al. (2009) interpreted those findings to indicate that difficulties regulating negative emotions in general underlie contact contamination. As such, disgust sensitivity may uniquely contribute to, but not serve as an amplifier for, contact
118
R. Travis, T.A. Fergus / Journal of Obsessive-Compulsive and Related Disorders 6 (2015) 114–119
contamination because contact contamination is marked by difficulties regulating many negative emotions. Alternatively, difficulties regulating disgust may be more relevant for mental contamination, with Elliott and Radomsky (2009) stating that “feelings of disgust are usually experienced alongside feelings of mental contamination” (p. 996). Although it remains to be empirically tested, difficulties regulating disgust may be more important to mental contamination than are difficulties regulating negative emotions more broadly. The study results and above discussion points should be considered with the following study limitations in mind. It is important to note that the magnitude of the observed interactive effect was small in size. Nonetheless, the interactive effect was (a) consistent with the magnitude of interactive effects found in prior studies examining the moderating impact of disgust sensitivity (e.g., Cisler et al., 2009), (b) in a typical range for the magnitude of moderation effects founds in prior self-report based studies (e.g., 1–3% of variance explained; Aiken & West, 1991) and (c) associated with a percentage of variance that can be considered a meaningful amount for moderation analyses (i.e., at least 1%) based upon Monte Carlo studies (Evans, 1985). Moreover, the interactive effect continued to account for unique variance in mental contamination scores even after statistically accounting for negative affect and broader contamination fears. As such, although modest in magnitude, we believe the observed interaction is meaningful and should be interpreted as such. Another limitation is the method by which the sample was obtained; while MTurk users are more demographically diverse than both American undergraduate samples, MTurk users may not be representative of the general population (Buhrmester et al., 2011; Paolacci & Chandler, 2014; Shapiro et al., 2013). Although our use of a community sample was supported by continuous nature of disgust propensity, disgust sensitivity, and contamination concerns (Olatunji & Broman-Fulks, 2009), it is important to replicate these findings among respondents who likely experience consistently higher scores on the mental contamination and disgust measures (e.g., individuals with obsessive-compulsive disorder or who have trauma exposure; Badour et al., 2013; Radomsky et al., 2014). Additionally, our study is limited by the sole use of self-report measures, as the monomethod assessment may have inflated study associations. Our methodology is unlikely to be the reason for the observed interactive effect, though, as correlated measurement error, which can be present as a result of monomethod assessment (Aiken & West, 1991), has not been found to produce spurious interactions and actually seems to attenuate the strength of interactions (Evans, 1985). Finally, although the study variables were modeled in a manner consistent with theory and prior studies, the cross-sectional design precludes causal conclusions related to study variable associations. In sum, disgust sensitivity appears to be an amplifying factor in relation to mental contamination. Given the robust association found between disgust and mental contamination across studies, future directions may include examining how mental contamination relates to other disgust-related phenomena. For example, appraisals of various situations, or even certain people (e.g., outgroup members; Cottrell & Neuberg, 2005), may be related to disgust responses, and, consequently, mental contamination. As such, even broader benefits of elucidating the link between disgust and mental contamination in future studies may be realized.
References Abramowitz, J. S., Deacon, B. J., Olatunji, B. O., Wheaton, M. G., Berman, N. C., Losardo, D., et al. (2010). Assessment of obsessive-compulsive symptom
dimensions: development and evaluation of the Dimensional Obsessive-Compulsive Scale. Psychological Assessment, 22, 180–198. Aiken, L. S., & West, S. G. (1991). Multiple regression: testing and interpreting interactions. Thousand Oaks, CA: Sage. Badour, C. L., Feldner, M. T., Blumenthal, H., & Bujarski, S. J. (2013). Examination of increased mental contamination as a potential mechanism in the association between disgust sensitivity and sexual assault-related posttraumatic stress. Cognitive Therapy and Research, 37, 697–703. Badour, C. L., Ojserkis, R., McKay, D., & Feldner, M. T. (2014). Disgust as a unique affective predictor of mental contamination following sexual trauma. Journal of Anxiety Disorders, 28, 704–711. Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazon's Mechanical Turk: a new source of inexpensive, yet high quality, data? Perspectives on Psychological Science, 6, 3–5. Cisler, J. M., Olatunji, B. O., & Lohr, J. M. (2009). Disgust sensitivity and emotion regulation potentiate the effect of disgust propensity on spider fear, blood-injection-injury fear, and contamination fear. Journal of Behavior Therapy and Experimental Psychiatry, 40, 219–229. Cisler, J. M., Olatunji, B. O., Sawchuk, C. N., & Lohr, J. M. (2008). Specificity of emotional maintenance processes among contamination fears and blood–injection–injury fears. Journal of Anxiety Disorders, 22, 915–923. Cohen, J., Cohen, P., West, S. G., & Aiken, L. A. (2003). Applied multiple regression/ correlation analysis for the behavioral sciences (3rd ed.). Hillsdale, NJ: Lawrence Erlbaum. Cottrell, C. A., & Neuberg, S. L. (2005). Different emotional reactions to different groups: a sociofunctional threat-based approach to “prejudice.”. Journal of Personality and Social Psychology, 88, 770–789. Coughtrey, A. E., Shafran, R., Knibbs, D., & Rachman, S. J. (2012). Mental contamination in obsessive-compulsive disorder. Journal of Obsessive-Compulsive and Related Disorders, 1, 244–250. Coughtrey, A. E., Shafran, R., Lee, M., & Rachman, S. J. (2013). The treatment of mental contamination: a case series. Cognitive and Behavioral Practice, 20, 221–231. Coughtrey, A. E., Shafran, R., & Rachman, S. J. (2014). The spread of mental contamination. Journal of Behavior Therapy and Experimental Psychiatry, 45, 33–38. Dixon, L. J., Stevens, E. N., & Viana, A. G. (2014). Anxiety sensitivity as a moderator of the relationship between trait anxiety and illicit substance use. Psychology of Addictive Behaviors, 28, 1284–1289. Elliott, C. M., & Radomsky, A. S. (2009). Analyses of mental contamination: Part I, experimental manipulations of morality. Behaviour Research and Therapy, 47, 995–1003. Engelhard, M., Olatunji, B. O., & de Jong, P. J. (2011). Disgust and the development of posttraumatic stress among soldiers deployed to Afghanistan. Journal of Anxiety Disorders, 25, 58–63. Evans, M. G. (1985). A Monte Carlo study of the effects of correlated method variance in moderated multiple regression analysis. Organizational Behavior and Human Decision Processes, 36, 305–323. Fairbrother, N., Newth, S. J., & Rachman, S. (2005). Mental pollution: feelings of dirtiness without contact contact. Behaviour Research and Therapy, 43, 121–130. Fergus, T. A., & Valentiner, D. P. (2009). The disgust propensity and sensitivity scalerevised: an examination of a reduced-item version. Journal of Anxiety Disorders, 23, 703–710. Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: a regression-based approach. New York, NY: Guilford. Herba, J. K., & Rachman, S. (2007). Vulnerability to mental contamination. Behaviour Research and Therapy, 45, 2804–2812. Olatunji, B. O., & Broman-Fulks, J. J. (2009). Latent structure of aversion: taxometric exploration. Journal of Anxiety Disorders, 23, 87–92. Olatunji, B. O., Cisler, J. M., Deacon, B. J., Connolly, K., & Lohr, J. M. (2007). The disgust propensity and sensitivity scale–revised: psychometric properties and specificity in relation to anxiety disorder symptoms. Journal of Anxiety Disorders, 21, 918–930. Olatunji, B. O., Cisler, J., McKay, D., & Phillips, M. L. (2010). Is disgust associated with psychopathology? Emerging research in the anxiety disorders. Psychiatry Research, 175, 1–10. Olatunji, B. O., Moretz, M. W., Wolitzky-Taylor, K. B., McKay, D., McGrath, P. B., & Ciesielski, B. G. (2010). Disgust vulnerability and symptoms of contaminationbased OCD: descriptive tests of incremental specificity. Behavior Therapy, 41, 475–490. Olatunji, B. O., & Sawchuk, C. N. (2005). Disgust: characteristic features, social manifestations, and clinical implications. Journal of Social and Clinical Psychology, 24, 932–962. Paolacci, G., & Chandler, J. (2014). Inside the turk: understanding Mechanical Turk as a participant pool. Current Directions in Psychological Science, 23, 184–188. Peer, E., Vosgerau, J., & Acquisti, A. (2014). Reputation as a sufficient condition for data quality on Amazon Mechanical Turk. Behavior Research Methods, 46, 1023–1031. Rachman, S. (2004). Fear of contamination. Behaviour Research and Therapy, 42, 1227–1255. Rachman, S., Radomsky, A. S., Elliott, C. M., & Zysk, E. (2012). Mental contamination: the perpetrator effect. Journal of Behavior Therapy and Experimental Psychiatry, 43, 587–593. Radomsky, A. S., & Elliott, C. M. (2009). Analyses of mental contamination: Part II, individual differences. Behaviour Research and Therapy, 47, 1004–1011.
R. Travis, T.A. Fergus / Journal of Obsessive-Compulsive and Related Disorders 6 (2015) 114–119
Radomsky, A. S., Rachman, S., Shafran, R., Coughtrey, A. E., & Barber, K. C. (2014). The nature and assessment of mental contamination: a psychometric analysis. Journal of Obsessive-Compulsive and Related Disorders, 3, 181–187. Reiss, S. (1987). Theoretical perspectives on the fear of anxiety. Clinical Psychology Review, 7, 585–596. Shapiro, D. N., Chandler, J., & Mueller, P. A. (2013). Using Mechanical Turk to study clinical populations. Clinical Psychological Science, 1, 213–220. Thordarson, D. S., Radomsky, A. S., Rachman, S., Shafran, R., Sawchuk, C. N., & Hakstian, A. R. (2004). The Vancouver Obsessional Compulsive Inventory (VOCI). Behaviour Research and Therapy, 42, 1289–1314.
119
van Overveld, W. J. M., De Jong, P. J., Peters, M. L., Cavanagh, K., & Davey, G. C. L. (2006). Disgust propensity and disgust sensitivity: separate constructs that are differentially related to specific fears. Personality and Individual Differences, 41, 1241–1252. Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of Positive and Negative Affect: The PANAS Scales. Journal of Personality and Social Psychology, 54, 1063–1070.