The Complex Interaction Between Psychopathy, Empathy and Interoception

The project described here encompasses two major studies. The goals, hypothesis, and analysis plan for each study will be separately described, although data collection procedures were shared by both studies. More information about this project can be found at https://osf.io/hmtz9/. Study 1. Untangling Self-Reported Interoceptive Accuracy and Attention: Evidence from the European Portuguese Validation of the Interoceptive Accuracy Scale and the Body Perception Questionnaire Interoception, broadly defined as the perception of the body’s internal states (Craig, 2002) has been widely associated with several domains of cognitive and emotional processing (e.g., Barrett, 2017; Critchley & Harrison, 2013; Damasio, 1994; Marshall et al., 2018; Pereira et al., 2019; Seth & Friston, 2016) and implicated in a wide range of psychiatric disorders (e.g., Khalsa et al., 2017; Murphy et al., 2017). These findings have led researchers in the field to propose different frameworks to address the challenges of measuring individual differences in interoception (e.g., Ceunen et al., 2013, Garfinkel et al., 2015, Khalsa et al., 2017), although proposed taxonomies only focused on the different measurement methods (i.e., questionnaires vs. experimental tasks). More recently, Murphy et al. (2019) proposed a 2 x 2 factorial framework to categorize individual differences in interoception: Factor 1, differentiates what is measured (interoceptive accuracy, the ability to accurately perceive interoceptive signals vs. interoceptive attention, the degree to which interoceptive signals are the object of attention); Factor 2, distinguishes how interoception is measured (via self-report measures assessing an individual’s beliefs regarding their interoceptive ability vs. objective measures, that assess performance on objective tests of interoception). Subsequently, Murphy et al. (2020) provided further evidence for this framework, finding that self-report measures of interoceptive accuracy such as the Interoceptive Confusion Questionnaire (Brewer et al., 2016) and the newly developed Interoceptive Accuracy Scale (IAS) were uncorrelated with the Body Awareness subscale of the Body Perception Questionnaire (BPQ; Porges, 1993), a widely used self-report measure of interoceptive attention. These results support a distinction between self-report interoceptive accuracy and attention, although this should be ideally replicated in other samples to further strengthen the validity of the 2 x 2 factorial model of interoception. Finally, it is also important to highlight that studies exploring the psychometric proprieties of self-report questionnaires of interoception are still scare. For instance, although the BPQ was developed over 25 years ago and has been widely used since then, there is only one recently published study exploring its psychometric properties (Cabrera et al., 2018). Moreover, the IAS was validated quite recently and also requires further psychometric testing (Murphy et al., 2020). Thus, Study 1 has the following main goals: 1. Translate, validate, and assess the psychometric properties of the European Portuguese versions of the IAS and the BPQ. 2. Investigate the association between self-reported interoceptive accuracy and attention (measured by the IAS and the BPQ Body Awareness subscale, respectively). Study 2. Dissociating Cognitive and Affective Empathy in Psychopathic Phenotypes: The Role of Interoceptive Attention and Accuracy According to the Triarchic Model (Patrick et al., 2009), psychopathy reflects the convergence of three etiological phenotypes: boldness, characterized by fearlessness, invulnerability to stress, social influence, and dominance; meanness, defined by callous self-interested pursuit of resources, reduced empathy, and social closeness; disinhibition, characterized by lack of behavioral restraint, poor planning, and impaired emotional regulation. Boldness and meanness share low fear as their etiological pathway, while both meanness and disinhibition are associated with externalizing vulnerability. As this model is fairly recent and widely debated, further insights about the mechanisms underlying each phenotype are needed. Empathy can be a key construct to disentangle these phenotypes. Recent frameworks suggest that empathy encompasses two main domains (Stietz et al., 2019): cognitive empathy, defined as the ability to make inferences about or represent the mental states of others, and affective empathy, defined as the isomorphic sharing of the affective state of another person. There is evidence suggesting that these empathy domains are differentially associated with psychopathic phenotypes (e.g. Almeida et al., 2015; Stanley et al., 2013; Sellbom & Phillips, 2013; Paiva et al., 2020). Meanness has been associated with empathy deficits including both affective and cognitive domains. Boldness has been positively associated with cognitive empathy, while results regarding affective empathy are inconsistent. Although disinhibition seems positively associated with affective empathy, it is not clear whether cognitive empathy is negatively associated or unrelated to this phenotype. Despite these findings, little is known about the underlying mechanisms that drive the dissociation of empathy domains across psychopathic phenotypes. Interoception (perception of the body’s internal states) may be a putative candidate to explain this complex interaction. It has been suggested that atypical affective experience plays a critical role in empathy deficits in psychopathy (Blair et al., 2005; Brook et al., 2013). Interoception has been widely linked to affective experience, contributing not only to the processing of our own emotional information but also to our ability to perceive and share the emotional states of others (e.g., Barrett, 2017; Seth & Friston, 2016; Fotopoulous & Tsakiris, 2017; Ondobaka et al., 2017). This suggests that empathy deficits in psychopathy could be explained by impaired interoceptive processing, as low interoceptive sensitivity may preclude subjects with high psychopathic traits from perceiving somatic sensations that signal the emotional valence of situations (Gao et al., 2012; Hoppenbrouwers et al., 2016). Until now there are only a handful of studies exploring the link between psychopathy and interoception (Lamoureux, 2019; Lyons & Hughes, 2015; Nentjes et al., 2013; Zwets et al., 2014) and most of them were conducted in criminal samples. Also, there are not any studies exploring the association of interoception with the triarchic psychopathic phenotypes. Finally, neither of the previous studies applied a formal theoretical framework to measure individual differences in interoception. There is recent evidence supporting a bifactorial model to evaluate interoception (Murphy et al., 2019; Murphy et al., 2020). This model distinguishes what is being measured (interoceptive accuracy, defined as the ability to accurately perceive interoceptive signals vs. interoceptive attention, the degree to which interoceptive signals are the object of attention) and how is it being measured (self-report vs. objective measures). Thus, Study 2 has the following main goals: 1. Investigate the association of psychopathic traits with empathy domains (cognitive and affective) and interoception (self-reported interoceptive attention and accuracy). 2. Examine the association of psychopathy with empathy and interoception using the psychopathic phenotypes suggested by the triarchic model (boldness, meanness, and disinhibition) as well as the classical factor structure of psychopathy (primary vs. secondary). 3. Investigate the association between empathy domains (cognitive and affective) and interoception (self-reported interoceptive attention and accuracy). 4. Test whether the associations between psychopathic traits, empathy and interoception are retained after controlling for alexithymia. 5. Explore whether interoceptive attention and/or accuracy mediate the association between psychopathic traits and cognitive/affective empathy.