Search

Persecutory ideations are self-referential delusions of being the target of malevolence despite a lack of evidence. Wisner et al. (2021) found that reduced connectivity between the left frontoparietal (lFP) network and parts of the orbitofrontal cortex (OFC) correlated with increased persecutory behaviors among psychotic patients performing in an economic social decision-making task that can measure the anticipation of a partner’s spiteful behavior. If this pattern could be observed in the resting state, it would suggest a functional-structural prior predisposing individuals to persecutory ideation. Forty-four patients in the early course of a psychotic disorder provided data for resting-state functional connectivity magnetic resonance imaging across nine brain networks that included the FP network and a similar OFC region. As predicted, we found a significant and negative correlation between the lFP–OFC at rest and the level of suspicious mistrust on the decision-making task using a within-group correlational design. Additionally, self-reported persecutory ideation correlated significantly with the connectivity between the right frontoparietal (rFP) network and the OFC. We extended the previous finding of reduced connectivity between the lFP network and the OFC in psychosis patients to the resting state, and observed a possible hemispheric difference, such that greater rFP–OFC connectivity predicted elevated self-reported persecutory ideation, suggesting potential differences between the lFP and rFP roles in persecutory social interactions.

BACKGROUND: The dot pattern expectancy (DPX) task is previously validated for cognitive control deficits in psychotic disorders. Patients with schizophrenia commit more errors as well as show less activation in bilateral prefrontal cortex during proactive control in the DPX task. Recent neural recordings in rhesus macaques indicate that proactive and reactive control processes in the DPX task engaged the same neurons in the prefrontal cortex. However, it is not clear if the overlap between these two mechanisms can be observed in human blood-oxygen-level dependent signals. Further, the role of these dual mechanism regions in the cognitive control deficits of patients with psychotic disorders remains unclear. METHODS: 150 probands with a psychotic disorder, 100 first degree relatives, and 50 healthy controls in the Psychosis Human Connectome Project will complete the DPX task while undergoing 3T functional magnetic resonance imaging scan. Traditional general linear model, as well as a novel multiway method called parallel factor analysis, will be used to examine brain mechanisms involved in proactive as well as reactive control during the DPX task. RESULTS: In a preliminary analysis of 24 probands (38.3 ± 12.8 years, 9 females) and 27 healthy controls (36.2 ± 14.0 years, 11 females), probands showed behavioral deficits in proactive control as compared to healthy controls. Neuroimaging analysis in healthy controls with general linear model showed a cluster in the left inferior frontal gyrus that was involved in both proactive and reactive control. Additionally, overlap in proactive and reactive control was observed in bilateral supramarginal/angular gyrus and right inferior occipital gyrus, although these two mechanisms also involved distinct brain regions. DISCUSSION: Consistent with previous findings, probands showed a specific deficit in proactive control in the DPX task. Proactive and reactive control in healthy controls engaged the same brain regions in the prefrontal cortex, replicating findings in rhesus macaques. Further analyses will reveal the role of the prefrontal cortex as well as other brain regions in the cognitive control deficits in psychotic disorders.

BACKGROUND: Cognitive control mechanisms enable an individual to regulate, coordinate, and sequence thoughts and actions in order to obtain desirable outcomes. Cognitive control is typically conceptualized as dual processes that occur in frontoparietal regions: proactive control, which uses sustained activation to enact anticipatory planning and goal maintenance, and reactive control, which entails retrieving information as presently needed (Braver, 2012), with patients with schizophrenia being especially susceptible to proactive control impairments (Poppe et al., 2016). However, nonhuman primate research suggests most prefrontal neurons ‘switch,’ firing during both proactive and reactive control, implying overlap between neural encoding of these two processes (Blackman et al., 2016). We sought to examine the overlapping neural circuitry of proactive and reactive control in healthy and patient populations using the Dot Pattern Expectancy Task (DPX). METHODS: 47 patients with schizophrenia (SZ) and 56 matched healthy controls (HC) completed 4 blocks of the DPX through the Cognitive Neuroscience Test Reliability and Clinical applications for Serious mental illness (CNTRaCS). During a 3-Tesla fMRI scan, participants followed the ‘X-then-Y’ rule, in which they were to press one button whenever an A cue was followed by an X probe, and another button for any other non-target stimulus sequence. Dissimilarity between proactive and reactive activation was evaluated within bilateral regions implicated in cognitive control: the medial frontal gyrus, superior frontal gyrus, and anterior cingulate cortex. Neuroimaging data was processed with FMRIB Software Library (FSL) packages. Probe accuracy and reaction time data were divided into ‘first half’ and ‘second half’ groups, depending on the block during which it occurred. RESULTS: Behavioral data analysis showed HC subjects showed a greater proclivity to engaging in proactive control across the study length than SZ subjects. HC subjects were also faster than SZ subjects in trials that required successful marshalling of proactive control. However, there was no within-subject increase in proactive proclivity or speed across the study procedure, complicating recent findings that suggest proactive control increases as a function of trial set length (Janowich & Cavanagh, 2018). ROI activation analysis showed no significant difference between HC and SZ proactive – reactive dissimilarity. Interestingly, within-ROI activation levels were significantly negative for both subject groups, implying these regions may be slightly more active during reactive processes. DISCUSSION: Results point to a between-group difference of relative strengths and weaknesses in proactive control, despite shared neural substrates. The lack of distinct ROI preference for proactive control offers support for the malleable nature of regions implicated in human cognitive control. Future analysis may investigate the association between proactive—reactive ROI dissimilarity and clinical and real-world functioning measures among patients.