Bias against disconfirmatory evidence in the ‘at-risk mental state’ and during psychosis

Highlights

• The bias against disconfirmatory evidence (BADE) in early psychosis is investigated.

• Patients with a first episode of psychosis (FEP) were most impaired in BADE.

• At-risk-mental-state-patients showed less evidence integration than healthy controls.

• BADE seemed to be independent of neurocognitive functioning to a great extent.

• Evidence integration plays a role in the pathogenesis of psychosis.

Abstract

Prior studies have confirmed a bias against disconfirmatory evidence (BADE) in schizophrenia which has been associated with delusions. However, its role in the pathogenesis of psychosis is yet unclear. The objective was to investigate BADE for the first time in subjects with an at-risk-mental-state for psychosis (ARMS), patients with a first episode of psychosis without antipsychotic treatment (FEP) and healthy controls (HC). A standard BADE test presenting written scenarios was employed. In addition, psychometric rating scales and a neuropsychological test battery were applied. A three-staged image was revealed. FEP-patients showed a significant BADE compared to the other groups. The performance of ARMS-patients lay in between HC and FEP-patients. A trend towards significance became evident for a bias against confirmatory evidence (BACE) in FEP-patients. Results were not attributable to antipsychotic or other medication or depressive symptoms. Correlations with delusions reached medium effect sizes but failed significance after Bonferroni-corrections. These results provide evidence for aberrations in evidence integration in the pathogenesis of psychosis and contribute to our knowledge of metacognitive functioning which can be used for (meta-)cognitive intervention in psychosis.

Introduction

The investigation of cognitive aberrations in schizophrenia has gained increasing attention because they represent potential cognitive mechanisms of the development and maintenance of delusions (Garety et al., 2001). So-called metacognitive biases, namely impairments in controlling and monitoring one's own cognitive processes, include for example the jumping to conclusion bias (i.e. a bias in data-gathering leading to an early decision making on the basis of little information) (Esslinger et al., 2013, Fine et al., 2007, Jolley et al., 2014, Rausch et al., 2014, Rausch et al., 2015) or a metamemory bias (i.e. a limited differentiation in confidence between correct and false memories) (Eifler et al., 2014b, Eisenacher et al., 2015). Furthermore, many recent studies have concentrated on belief inflexibility (So et al., 2012), such as a bias in evidence integration (e.g. Speechley et al., 2012) made of a bias towards both disconfirmatory evidence (BADE) and confirmatory evidence (BACE) (Eifler et al., 2014a, Woodward et al., 2008). The BADE describes a decreased ability to adjust one’s beliefs according to information which does not support one’s hypothesis. The BACE, on the other hand, describes a difficulty to integrate information which supports the true answer. The ability to integrate new information into the reasoning process is important in order to adapt one’s beliefs according to new knowledge. Not changing one's beliefs in the presence of disconfirming information and holding them with strong conviction though they are proved to be false is a hallmark of delusions as defined by Karl Jaspers (Jablensky, 2013).

Woodward et al. (2006b) were the first to introduce a paradigm to measure BADE and BACE. Pictures were disambiguated over time by additional confirmatory and disconfirmatory information. Later the paradigm was refined and written scenarios were presented (Buchy et al., 2007). The change of plausibility ratings for lure or true interpretations over the course of information growth indicated BADE or BACE, respectively. Strength of these paradigms is their usage of delusion neutral material precluding tautological reasoning. The authors reported findings of BADE in delusional- compared to non-delusional patients and in all patients compared to healthy controls (Woodward et al., 2006b). In subsequent studies, a BADE in schizophrenia patients was often replicated in comparison to healthy (e.g. Riccaboni et al., 2012, Veckenstedt et al., 2011, Woodward et al., 2006a, Woodward et al., 2008) and psychiatric control groups (Moritz and Woodward, 2006, Veckenstedt et al., 2011). Summarized, these findings propose that patients with schizophrenia are less able to integrate into their reasoning process new information that could call their current beliefs into question. In contrast, findings about a bias against confirmatory evidence (BACE) are mixed (Eifler et al., 2014a, Riccaboni et al., 2012, Veckenstedt et al., 2011). In addition, a third aberration, namely a liberal acceptance tendency, was found in several studies, demonstrating that patients with schizophrenia often rated implausible interpretations with higher plausibility than controls (Eifler et al., 2014a, Moritz and Woodward, 2004, Woodward et al., 2006b).

In a prior theoretical paper it was outlined that the integration of disconfirmatory evidence might be a risk factor for the maintenance of delusions (Garety et al., 2001). Supporting this theory, multiple experimental studies also revealed that a BADE was specifically associated with delusion severity (Riccaboni et al., 2012; Sanford et al., 2014, Speechley et al., 2012, So et al., 2012, Woodward et al., 2006b). Some other studies which did not observe associations between BADE and delusions discussed BADE as a potential trait-factor for delusions rather than a state-dependent bias (Moritz and Woodward, 2006, Veckenstedt et al., 2011). As most studies have usually included chronic patients with antipsychotic medication, the question emerges whether the findings of BADE might be confounded by this factor. So et al. (2010) proposed that belief flexibility mediates the effect of antipsychotic medication on symptoms implying that a symptom reduction by antipsychotics is reached by a reduction of belief inflexibility. Similarly, improved evidence integration ability under single-dose haloperidol compared to single-dose l-dopa in healthy controls suggests an association between both variables (Andreou et al., 2015). To exclude antipsychotic effects as confounds on evidence integration ability it is important to recruit patients prior to this medication.

Based on this set of findings, the role of evidence integration in the dynamic development of psychosis seems to be important. Prior literature has shown that healthy participants displayed a BADE in correlation with higher delusion ideation (Buchy et al., 2007, Orenes et al., 2012, Woodward et al., 2007, Zawadzki et al., 2012). However, no study up to date has explored this topic in risk constellations for psychosis, meaning those patients in an “at-risk mental state” (ARMS) for psychosis either presenting with cognitive basic symptoms and/or attenuated psychotic symptoms and/or brief limited intermittent psychotic symptoms (Fusar-Poli et al., 2012, Fusar-Poli et al., 2013, Schultze-Lutter et al., 2010, Schultze-Lutter et al., 2012). Such an investigation could enhance our knowledge about potential associations with early psychosis-prone symptoms. By differentiating between those symptoms, it is also possible to classify and examine subgroups of ARMS-patients. Some evidence regarding other metacognitive aberrations suggests a three-staged image of performance with ARMS-patients being less impaired than patients with a first episode of psychosis (FEP) but more than healthy controls (Eisenacher et al., 2015). Within this scope, it is also worth investigating possible associations between evidence integration and neurocognitive performance. Previous studies have demonstrated inconsistent results, reporting either no associations (Moritz et al., 2010, Woodward et al., 2007), associations with executive functioning, verbal learning, and theory of mind (Riccaboni et al., 2012), or with executive functioning, processing speed, working memory and vigilance (Eifler et al., 2014a) in patients with schizophrenia. Whether associations can be found in risk constellations for psychosis is open to question.

To our knowledge, the present study is the first focusing on evidence integration in ARMS-patients. Our main aim was to investigate patterns of evidence integration between ARMS-patients, antipsychotic-free FEP-patients and healthy controls. We expected three levels of performance with increasing impairment from healthy controls over ARMS-patients to FEP-patients. Furthermore, we hypothesized that BADE would be associated with measures of delusions and neurocognitive performance in the patient groups. In a tertiary analysis, we compared subgroups of ARMS-patients regarding BADE and BACE. We hypothesized to find less evidence integration in groups displaying higher delusional conviction.