Disturbances across whole brain networks during reward anticipation in an abstinent addiction population

Graphical abstract Network based statistics (NBS) analyses detected a graph sub-network comprising 153 edges between 59 nodes of the connectome where the ADD group demonstrated significantly less connectivity compared with the CON group. These differences in connectivity were mostly intra-hemispheric (55%), the majority (38%) being in the right hemisphere. The anatomical distribution of these connectivity differences between the two groups involved frontal (insula, inferior frontal gyrus, orbitofrontal cortex), limbic-associated (anterior cingulate gyrus, thalamus), and striatal (accumbens, caudate, pallidum) regions. The connectivity differences reported in this ADD sample indicate alterations between cognitive, striatal and limbic-associated regions during reward anticipation that persist into extended abstinence.
Highlights • Analytical methods can capture key features of whole brain networks in addiction. • We compared reward network connectivity in addiction (ADD) and control (CON) groups. • The ADD group showed disruptions in global network connectivity. • Global network measures may be more sensitive than traditional voxel-wise analyses.
The prevalent spatial distribution of abnormalities reported in cognitive fMRI studies in addiction suggests there are extensive disruptions across whole brain networks. Studies using resting state have reported disruptions in network connectivity in addiction, but these studies have not revealed characteristics of network functioning during critical psychological processes that are disrupted in addiction populations. Analytic methods that can capture key features of whole brain networks during psychological processes may be more sensitive in revealing additional and widespread neural disturbances in addiction, that are the provisions for relapse risk, and targets for medication development. The current study compared a substance addiction (ADD; n = 83) group in extended abstinence with a control (CON; n = 68) group on functional MRI (voxel-wise activation) and global network (connectivity) measures related to reward anticipation on a monetary incentive delay task. In the absence of group differences on MID performance, the ADD group showed reduced activation predominantly across temporal and visual regions, but not across the striatum. The ADD group also showed disruptions in global network connectivity (lower clustering coefficient and higher characteristic path length), and significantly less connectivity across a sub-network comprising frontal, temporal, limbic and striatal nodes. These results show that an addiction group in extended abstinence exhibit localised disruptions in brain activation, but more extensive disturbances in functional connectivity across whole brain networks. We propose that measures of global network functioning may be more sensitive in highlighting latent and more widespread neural disruptions during critical psychological processes in addiction and other psychiatric disorders.