‘Look not at what is contrary to propriety’: A meta-analytic exploration of the association between religiosity and sensitivity to disgust. Zhaoliang Yu, Persefoni Bali, Myron Tsikandilakis, Eddie M. W. Tong. British Journal of Social Psychology, July 1 2021. https://doi.org/10.1111/bjso.12479
Abstract: Previous research has suggested that disgust sensitivity contributes to moral self-regulation. The relationship between religiosity and disgust sensitivity is frequently explored as a moderator of moral-regulating ideologies, such as conservative and traditional ideologies. However, religiosity is suggested to differ from these in moral attitudes against social dominance and racial prejudice. Psychological theories, such as the societal moral intuition and the evolved hazard-perception models, have proposed that there could be reasons to support a distinct relationship between religiosity and disgust sensitivity. These reasons relate to the intuitive pursuit of spiritual purity and the non-secular transcendental emotional-reward value of moral behaviour for religious individuals. In the present manuscript, we conducted the first dedicated meta-analytic review between religiosity and disgust sensitivity. We analysed a summary of forty-seven experimental outcomes, including 48,971 participants. Our analysis revealed a significant positive association (r = .25) between religiosity and disgust sensitivity. This outcome suggests that sensitivity to disgust could have distinct spiritual purity and moral self-regulatory response value for religious individuals.
Discussion
Summary of findings
The present study aimed to provide an evaluation of the relationship between religiosity and disgust sensitivity. The current meta-analytic summary showed that there is a positive correlation between religiosity and disgust sensitivity. The overall association between religiosity and sensitivity to disgust was positive (r = .253) and significant (p < .001) after controlling for heterogeneity, publication bias, parametricity, and adjusting for the effect of moderators, such as gender, sample type (student vs. general), and age (Borenstein et al., 2017; see Figure 2). The present analytic summary also revealed that gender strongly moderated positive associations between religiosity and trait-level sensitivity to disgust subtypes, including pathogen disgust (r = .127; p < .001) and sexual disgust (r = .262; p < .001; see, Schumm et al., 2013).
General discussion
Religiosity has been defined as an affiliation with a system of morals and often ritualistic practices that include the belief in the existence and a moral code associated with a transcendental entity, or entities (Fincher & Thornhill, 2012). Sensitivity to disgust has been approached as the belief or self-report that an event, cue, or elicitor will be experienced as aversive to a subject using questionnaire and self-report measures (Rozin & Haidt, 2013; Tybur et al., 2013). As mentioned in the introduction, theoretical models, such as the societal moral intuition and the EH-PS models, suggest that disgust sensitivity could relate to religiosity as part of a system of beliefs that contribute to moral cognition, emotions, and behaviours. This relationship is suggested to be distinct from the association of secular forms of morality and sensitivity to disgust in the sense that religiosity involves key conceptual differences to secular morality, such as the notions of supernatural invigilation and proportionality (Shariff, 2015).
In the current study, we showed that religiosity positively correlates with sensitivity to disgust. This finding cannot be interpreted to imply causality, such as whether sensitivity to disgust is a predictive marker for religious belief, or vice versa (Beit-Hallahmi, 2014). Instead, it offers the first meta-analytic dedicated findings that religiosity and sensitivity to disgust are, indeed, correlated (Terrizzi et al., 2013). As regards the proposed differences in the evolutionary trajectories between religious and secular factors, that could underlie the reported association, it is worth inquiring whether the relationship between sensitivity to disgust and religiosity can translate to high, or higher, compared to non-religious individuals, physiological reactions in response to disgust-related elicitors (Kapogiannis et al., 2009). The major consideration in this instance is that the result of such an empirical exploration will help to clarify whether the association between religiosity and sensitivity to disgust reflects a theoretical belief framework dissociated from behavioural (see, for example, Argyle, 2006; Beit-Hallahmi & Argyle, 1997; Nelkin, 2000) and physiological responsivity to stimulus exposure, or an actual physiological module for the avoidance of inferred/indirect threat (Henry, 2016).
This is a very critical distinction. The psychological models that have been associated with religiosity and sensitivity to disgust can be interpreted to suggest that physiological responses will occur when religious individuals are presented with emotional elicitors. The EH-PS model places sensitivity to disgust in relation to religiosity in continuity to an avoidance system related to indirect threat to fitness. Therefore, both disgust-related emotional elicitors and immoral emotional elicitors should stimulate physiological responses for religious individuals that reflect these avoidance mechanisms (Baumard & Boyer, 2013). Conversely, the societal intuitional model has been used to suggest that physiological responses to elicitors related to disgust sensitivity in religious individuals will not only be automatic and involuntary (Boyer & Liénard, 2006) but additionally possibly subject to pre-conscious or subliminal processing, given that their evolutionary origins are linked with intuitional processes (Shariff, 2015).
If these conditions are not met, and if, indeed, sensitivity to disgust does not translate to distinct physiological reactions in religious individuals, the arguments that have been proposed to address this association could be reduced to reflect belief systems without palpable physiological correlates (see Shariff, 2015). Along the same lines, the concepts of supernatural invigilation and proportionality suggest that to some extend religious individuals could be subject to deontic, prescriptive, and inviolate moral laws (Shariff, 2015). The expected outcome of this experiential belief to the morality of a transcendental entity – or entities – should be a sense of reduced moral self-authorship (see Dijksterhuis, Preston, Wegner, & Aarts, 2008) as well as a reduced sense of moral self-righteousness and authoritarianism (Haidt, 2012). If subsequent experimental efforts to provide evidence for these effects are not successful, the sceptical scholar could readily attribute the currently reported significant association to the strict and prescriptive laws that are often part of the participation in religious-related ritualistic practices (Barrett, 2000). The association could also be attributed to other moderating factors, such as secular sociosexual attitudes and beliefs (Weeden, Cohen, & Kenrick, 2008) and addressed as a means for ingroup socialization among morally compatible individuals (Shariff, Willard, Andersen, & Norenzayan, 2016).
As regards our additional findings, we showed that age was not a significant moderator for the association between religiosity and sensitivity to disgust (but see also Dienes, 2014). Several studies have proposed that sensitivity to disgust is higher in younger religious and non-religious adults due to inexperience and inhibitory mechanisms related to experiencing unwanted loss of control and uncertainty (Quigley, Sherman, & Sherman, 1997). Several studies have also suggested that disgust sensitivity can be higher in older/senior religious and non-religious adults possibly due to increased concerns for one’s physical and emotional well-being (Oaten, Stevenson, & Case, 2009). Our findings suggest instead that the association of religiosity and sensitivity to disgust is an enduring and not a transient or age-specific effect and that it can manifest throughout an individual’s lifespan (Zelenski & Larsen, 2000).
In the most surprising and possibly one of the most important additional findings of our analyses, we showed that gender and sample type had a weak effect-size significance trend influence on the correlation between religiosity and overall sensitivity to disgust. Gender was a significant moderator for the association between religiosity and subtypes of sensitivity to disgust. This finding suggests that gender plays an important role in the association between sensitivity to disgust and religiosity, particularly for sexual and pathogen/contamination-related cues. This could mean that, although religious individuals, independently of gender, have beliefs related to disgust sensitivity, sexual and pathogen/contamination sensitivity to disgust and religiosity are not reliably associated when we remove the effect of the responses of female participants. The meta-analysis matrix adds to this a novel finding. Gender moderated these associations as an interactive function of the effect of female participation in religiosity, although intriguingly female participation did not impact sensitivity to disgust subtypes (see Figure 4). Being female did not directly influence sensitivity to sexual and pathogen/contamination cues as previous studies proposed (Tybur et al., 2013), it increased the level of religiosity of an individual and moderated by association responses to disgust sensitivity subtypes and the interaction between religiosity and sensitivity to disgust. This is a most unexpected, promising, and novel finding that should be further explored by subsequent research.
Looking at the greater picture, in the present manuscript we explored one possible emotional and response attitude, and/or correlate – namely, sensitivity to disgust – that could influence religiosity. This can be the first step for further exploring whether shame, guilt, regret, self-reproach, and – most understatedly in previous psychological research (see, Fatima, Sharif, & Khalid, 2018) – positive-valence emotional states, such as awe, kindness, generosity, and calmness, could underlie and contribute to religious emotional experience and beliefs (Sharma & Singh, 2019). An important contribution of the current outcomes is that we provided evidence that this line of research can offer insightful results, theoretical advances, and further directions for experimental research. These can include the exploration of belief-system response attitudes and emotional correlates of religiosity and their possible distinctive functions within religiosity as possibly non-secular moral and experiential phenomena. The current manuscript could and should (Shariff, 2015) set an experimental and meta-analytic precedence towards the exploration of religiosity and belief-system response attitudes and emotional sensitivity as pathways to understanding religiosity further and in relation to human attitudes and experiences.
Limitations
In the current meta-analyses, the included studies employed questionnaire assessments for assessing the relationship between religiosity and sensitivity to disgust. Future experimental research could benefit from using psychophysiological assessments for exploring this relationship. This implementation will enable us to explore whether the current significant results reflect belief-system values, such as self-report responses, and/or psychophysiological emotional experiences for sensitivity to disgust (see, Tsikandilakis et al., 2021). The current meta-analyses included several different religiosity and disgust sensitivity questionnaire assessments. We must note that an important issue in meta-analytic research is whether the achieved meta-analytic power originates from sufficiently statistically powered studies (see, Amrhein, Trafimow, & Greenland, 2019). This is an important component of meta-analytic research that has, nevertheless, decreased impact in the current meta-analytic research due to the corrected-weighted statistical analyses (see Hedges & Schauer, 2019) of forty-seven experimental outcomes, including 48,971 participants (see also, Borenstein et al., 2017). It is very critical to mention as conclusive remarks that the vast majority of the included outcomes used samples, for which participants’ socioeconomic status and political beliefs were not measured. These variables could not be, therefore, included as moderators in the analyses. Exploring their influence should be a priority for experimental replications of the current findings. In addition, the pool of existing empirical studies did not also provide sufficient data to enable the examination across specific religious backgrounds or countries of origin. Further correspondence with the authors of the included studies did not result in sufficient information to perform a per religion analysis or the inclusion of religious affiliation as a categorical moderator in the meta-analysis. Further experimental research could benefit from exploring the effect that different religious affiliations confer on the association between religiosity and sensitivity to disgust.