Altered brain connectivity in patients with schizophrenia is consistent across cognitive contexts

Abstract

IntroductionBeyond abnormal activity of focal brain areas, 1 aberrant functional interactions between brain regions are thought to be a core feature of schizophrenia. [2][3][4] Among the current issues related to this influential idea is whether abnormal functional connectivity in patients with schizophrenia is contextdependent. 5 Resting-state fMRI characterizes intrinsic functional brain connectivity, which is guided by endogenous processes and spontaneous thoughts in the absence of any explicit task. 6 Resting-state functional brain imaging in patients with schizophrenia has revealed a widespread functional dysconnectivity between brain regions that are also known to exhibit abnormal levels of activation in relation to impairments in task performance and/or symptomatology. 7-11However, a critical issue with the study of resting-state functional connectivity in patients with schizophrenia is the poor control over the cognitive state of these individuals, which may confound the interpretation of the results. Patients with schizophrenia experience pronounced disturbances of thoughts and differ from controls with regards to their internal mentations and cognitive states during rest. 12 The question of the context-independence of abnormal functional connectivity in patients with schizophrenia may thus be best adressed by characterizing such alterations using well-defined behavioural paradigms. To date, research has focused on determining whether alterations in brain connectivity were further modulated by task complexity in single cognitive domains. 13-15However, to our knowledge, there is no sufficiently powered study that specifically examined commonalities and differences in dysconnectivity across distinct cognitive tasks. 16In the present work, we sought to assess whether the degree of cognitive similarity would impact the results by looking at 2 cognitive domains that strongly differ from one another, emotional versus visuomotor processing. We further distinguished the emotional processing task based on an additional embedded component, either incidental learning or recognition memory. These 3 paradigms were studied with a connectome-wide general linear model analysis. We characterized context-independent effects through commonalities in the main effects of abnormal functional connectivity across the 3 tasks. We evaluated context-dependent effects of abnormal connectivity using interaction analyses that looked for changes in functional connectivity between tasks in patients Correspondence to: P. Orban, CRIUGM, Université de Montréal, 4545 Queen Mary, Montreal, QC H3W 1W5; pierre.orban@criugm.qc.ca Submitted July 13, 2015; Revised Nov. 19, 2015; Accepted Jan. 5, 2016; Early-released Apr. 19, 2016 DOI: 10.1503 Background: Schizophrenia has been defined as a dysconnection syndrome characterized by aberrant functional brain connectivity. Using task-based fMRI, we assessed to what extent the nature of the cognitive context may further modulate abnormal functional brain connectivity. Methods: We analyzed data matc...

Multisite generalizability of schizophrenia diagnosis classification based on functional brain connectivity

“…Mass univariate analyses evidenced brain dysconnectivity across the entire brain, with significant effects in over a third of brain connections distributed in various large-scale brain networks, from cognitive to primary sensory networks. Abnormally decreased rather than increased functional connectivity in schizophrenia is largely consistent with previous reports in the literature (Pettersson-Yeo et al, 2011). With close to 200 schizophrenia patients and 200 controls, our fMRI connectome-wide association analysis is one of the largest to be reported to date.…”

“…Brain imaging data from 6 independent studies were obtained through either the SchizConnect and OpenfMRI data sharing platforms (http://schizconnect.org; https://openfmri.org) or local scanning (Çetin et al, 2014;Gollub et al, 2013;Kogan et al, 2016;Orban et al, 2016;Poldrack et al, 2016;Wang et al, 2016). The 6 datasets differed in terms of both scanning site and cognitive context during fMRI data acquisition (resting-state, emotional memory, Sternberg item recognition paradigm, N-back, task-switching and oddball tasks).…”

“…Since it is impossible to completely control the variations in mental states in realistic clinical situations, we further promoted the complexity of the classification problem by including data obtained in distinct cognitive task conditions across sites. Mass univariate findings have indicated that cognitive state does not further impact on the nature of functional brain connectivity alterations in schizophrenia (Kaufmann et al, 2016;Orban et al, 2016). However, the potential influence of cognitive context on classification performance in a multivariate analysis should not be rejected.…”

Neuropsychiatric mutations delineate functional brain connectivity dimensions contributing to autism and schizophrenia

“…Studies applying this strategy in ASD have shown reproducible patterns of both over-connectivity (prefrontal and parietal cortices) and under-connectivity (sensory-motor regions) 19 . In contrast, SZ patients tend to exhibit a general under-connectivity profile, mainly involving the medial prefrontal cortex, the cingulate and the temporal lobe 20 , with the exception of over-connectivity of the thalamus 21 . These altered networks do not appear to be disorder-specific and have been reported across several disorders, including ASD, ADHD, and SZ 22 .…”

Atypical Neocortical Development in theCited2Conditional Knockout Leads to Behavioral Deficits Associated with Neurodevelopmental Disorders

“…Decreased connectivity with the PFC is further correlated with restrictive and repetitive behaviors, a keystone phenotype in the diagnosis of ASD (Delbruck et al, 2019). PFC functionality deficits are prevalent in schizophrenia as well, with both hypoactivation phenotypes (Huang et al, 2010;Denardo et al, 2015) and consistent reduced connectivity reported (Woodward et al, 2012;Orban et al, 2017;Giraldo-Chica et al, 2018;Chen et al, 2019;Bulletin et al, 2020). Although additional experiments are necessary to investigate intrahemispheric connectivity disruptions in the Cited2 cKO, including those to the PFC, it is likely that disruptions in both long-distance intrahemispheric and interhemispheric connectivity contribute to the behavioral disruptions in the Cited2 cKO.…”