Does childhood trauma predict poorer metacognitive abilities in people with first-episode psychosis?

Highlights

• Childhood traumas predicted little variability in the metacognitive domains.

• Childhood traumas were associated with higher metacognitive scores.

• Especially sexual and emotional abuse were associated with metacognition.

Abstract

Research suggests that people with first-episode psychosis (FEP) report more childhood traumas and have lower metacognitive abilities than non-clinical controls. Childhood trauma negatively affects metacognitive development in population studies, while the association remains largely unexplored in FEP populations. Metacognition refers to the identification of thoughts and feelings and the formation of complex ideas about oneself and others. This study hypothesized that childhood trauma would be associated with lower metacognitive abilities in people with FEP. In a representative sample of 92 persons with non-affective FEP, we assessed childhood trauma, metacognitive abilities and symptoms of psychosis. We used the Childhood Trauma Questionnaire (CTQ) and the Metacognitive Assessment Scale––Abbreviated which includes Self-reflectivity, Awareness of the Mind of the Other, Decentration and Mastery. Hierarchical regression analyses were performed with metacognitive domains as outcome variables and childhood traumas as independent variables, while controlling for age, gender, first-degree psychiatric illness and negative symptoms. We found few significant associations between the different types of childhood trauma and metacognitive domains, and they suggested childhood trauma are associated with better metacognitive abilities. Study limitations included the cross-sectional design and use of self-report measures. Future studies could preferably be prospective and include different measures of psychopathology and neuropsychology.

Introduction

People with psychosis are highly likely to have experienced childhood trauma (Trauelsen et al., 2015, Varese et al., 2012) and they exhibit lower metacognitive abilities than non-clinical controls (Healey et al., 2016, Trauelsen et al., 2016b). As childhood trauma has been found to affect metacognition in population studies (Cicchetti et al., 2003, Edwards et al., 2005, Ensink et al., 2015, Pears and Fisher, 2005, Shipman and Zeman, 1999), it seems important to address the possible association in people with psychosis.

Metacognition refers to both simple and more complex mental activities. The simple activities include the ability to identify thoughts and feelings, both in oneself and others; the complex activities include the ability to use this knowledge to form complex ideas about oneself and others (Lysaker et al., 2018a, Lysaker et al., 2018b). Metacognition is associated with mentalization, social cognition and theory of mind (Fonagy et al., 2011, Lysaker et al., 2018a, Lysaker et al., 2018b). Their associations are sparsely examined, but it seems that social cognition is related to more distinct kinds of abilities whereas metacognition is related to more integrative kinds of metacognitive awareness (Lysaker et al., 2013). At the same time it has been shown that good social cognitive abilities are associated with better metacognitive abilities (James et al., 2017). One way of measuring metacognitive skills, is with the Metacognition Assessment Scale – Abbreviated (MAS-A) which includes the four domains Self-reflectivity, Awareness of the mind of the other (AoM), Decentration and Mastery. Self-reflectivity and AoM refers to the ability to recognize feelings and cognitions and use this information to form complex ideas about oneself and others, respectively. Decentration refers to the acknowledgement that other people are individual beings living their own lives relatively independently of oneself; and Mastery is the ability to review all this information to better master one's mental health and interpersonal problems.

Metacognitive abilities are thought to develop throughout childhood and adolescence with some continuance into adult life (Kuhn, 2000). For example, mothers’ appropriate mind-related comments at 12 months predicted superior metacognitive abilities at ages two and four (Laranjo et al., 2014, Meins et al., 2003). Metacognition is both trait and state dependent and as so may decrease in interpersonally threatening situations and increase in safe environments (Dimaggio and Lysaker, 2015, Liotti and Gilbert, 2011). It has been found that both childhood abuse (Cicchetti et al., 2003, Ensink et al., 2015, Pears and Fisher, 2005) and neglect (Edwards et al., 2005, Shipman and Zeman, 1999) is associated with lowered metacognitive abilities. One possible explanation for these associations is that trauma might decrease one's awareness of both one's own and others’ mental states to protect oneself from the painful feelings associated with the awareness (Lysaker et al., 2011, p. 1189). With such associations found in children and with findings of lower metacognitive abilities in people with psychosis, it seems reasonable to hypothesize an association between the two domains in psychosis populations.

Research has begun to address this gap; so far counting six studies with quite diverse findings. Aydin et al. (2016) found that a total score composed of four types of reported childhood abuse was associated with mentalizing impairments in a sample of 87 patients with a non-affective psychotic disorder. Mentalization was represented by the Hinting Task where the patient must recognize the beliefs or feelings of a character in order to apprehend an implicit message (Corcoran et al., 1995). Weijers et al. also found that mentalization mediated around 40% of the association between reported childhood abuse and negative symptoms. A case control study of people with early psychosis found that childhood abuse (sexual, physical and emotional) and neglect (physical and emotional), as measured by total Childhood Trauma Questionnaire (CTQ), was related to poorer social cognition measured by the MATRICS Consensus Cognitive Battery (MCCB) comprising the ability to understand others based on the selection and interpretation of social information (Garcia et al., 2016). Also, using the CTQ and the MCCB, a case-control study of people with non-affective FEP and mentally healthy traumatized participants found that childhood neglect - but not abuse - was a significant predictor for poorer social cognition in both groups (Kilian et al., 2017). Aydin et al. found that those with childhood emotional abuse had worse Decentration skills (Aydin et al., 2016). All other trauma types and metacognitive subdomains were not found to be associated. In another study of people either at ultra high risk or with a non-affective psychosis found that childhood trauma was not associated with theory of mind (Palmier-Claus et al., 2016). Trauma were measured by the CTQ total and theory of mind by the Hinting Task and the Mind in the Eyes task where participants are presented with pictures of others’ eyes and choose between four words to best describe them. The last study only examined child sexual abuse before age 13 and found it to be associated with good Self-reflectivity but low AoM (Lysaker et al., 2011) Thus, three studies found associations between childhood abuse and metacognition, while three studies found no or only scarce associations. One of the latter studies included ultra high risk participants which may have diminished possible associations in the FEP group (Palmier-Claus et al., 2016). Assessment instruments couldn't explain the mixed results as studies using the same metacognitive instruments, the MCCB and the Hinting Task, and even the same trauma instruments, the CTQ, also found differing results. Most studies had sufficient sample sizes and appeared robust, and it remains unclear whether there is an association between childhood trauma and metacognitive abilities in people with psychosis.

Therefore, we aimed to test associations between childhood trauma and metacognition in a clinical epidemiological sample of people with non-affective FEP, while controlling for other important factors - age, gender and 1st degree psychiatric illness. We furthermore had models with and without negative symptoms as these repeatedly have been found to be associated with metacognitive abilities (Hamm et al., 2012, McLeod et al., 2014, Trauelsen et al., 2016b) and therefore might affect the association.

We hypothesized that

• All 7 trauma types would predict lower Self-reflectivity, lower Awareness of the Mind of the Other, lower Decentration and Mastery abilities.

• An increasing number of childhood adversities would be associated with lower metacognitive abilities in adulthood.