Brain Connectivity and Mental Illness

Abstract

The remaining articles illustrate the application of specific techniques to study different disorders. Independent component analysis (ICA), a popular data-driven technique for decompos-ing fMRI data into distinct components with characteristic anat-omy and temporal dynamics (Calhoun et al., 2004), is applied by Calhoun et al. (2011) to demonstrate phenotypic continuities and discontinuities in the brain functional networks of patients with schizophrenia and bipolar disorder performing an auditory oddball task. Seed-based techniques are featured in the work of Mills et al. (2012), Mennes et al. (2011), Rabinak et al. (2011), and Davey et al. (2012). The first two articles use multiple seeds to character-ize how individual differences in functional connectivity relate to cognitive performance on tests of executive function in children with ADHD and typically developing controls. This work offers welcome new evidence supporting the behavioral relevance of resting-state measures for psychiatric disorders, which is criti-cal for validating any case-control differences observed under such experimental conditions (Fornito and Bullmore, 2010). In war veterans with combat-related PTSD, Rabinak et al. (2011) report a prominent alteration of resting-state functional con-nectivity between an amygdala seed region and posterior insu-lar cortex, extending prior evidence from task-based activation studies in this patient group. Finally, Davey et al. (2012) provide evidence that adaptive changes in the functional connectivity of the subgenual cingulate cortex may be necessary to support intact executive function early in the course of major depres-sive disorder (MDD), prior to the emergence of characteristic cognitive deficits. In a second study of MDD, Almeida et al. (2011) use dynamic causal modeling (DCM) to characterize changes in the effective connectivity (i.e., causal functional interactions) of a circuit linking ventral frontal cortical regions and the amygdala during emotional processing. Their findings shed light on how sex differences in the dynamics of this network may relate to the clinical phenomenology of female compared to male MDD. Graph analytic techniques are featured in the work of Xie and He (2011) and Collin et al. (2011). Xie and He provide a primer on the basic principles of graph theoretic analysis of neuroimaging data, and comprehensively review the application of these techniques to study disturbances of structural and functional brain network connectivity and topology in Alzheimer's disease. In a resting-state fMRI study, Collin and colleagues use graph analytic techniques to provide some of the first evidence that functional connectivity between cerebellar and cortical and subcortical regions is associated with genetic risk for schizophrenia. Despite nearly three decades of neuroimaging research in psychia-try, no imaging study to date has identified a single site of pathology in the brain that represents a causal factor in the emergence of any of the major mental illnesses described in current nosological sys-tems. Rather, the available evidence indicates that these disorders are associated with subtle abnormalities distributed throughout the brain (e.g., Fornito et al., 2009, 2012; Bora et al., 2010, 2012), implying that they arise from disordered interactions between con-nected neural systems rather than damage to any individual brain region. Accordingly, current hopes for the development of more targeted interventions in psychiatry are being placed on research attempting to map the molecular determinants and clinical cor-relates of neural circuit abnormalities in mental disorders (Insel and Scolnick, 2006; Meyer-Lindenberg, 2010). The impact of this thinking on the field is evident in the emergence of new scientific journals devoted to the study of brain connectivity 1 ; increasingly popular workshops devoted to the topic