Large-scale functional network overlap is a general property of brain functional organization: Reconciling inconsistent fMRI findings from general-linear-model-based analyses

Abstract

Functional magnetic resonance imaging (fMRI) studies regularly use univariate general-linear-model-based analyses (GLM). Their findings are often inconsistent across different studies, perhaps because of several fundamental brain properties including functional heterogeneity, balanced excitation and inhibition (E/I), and sparseness of neuronal activities. These properties stipulate heterogeneous neuronal activities in the same voxels and likely limit the sensitivity and specificity of GLM. This paper selectively reviews findings of histological and electrophysiological studies and fMRI spatial independent component analysis (sICA) and reports new findings by applying sICA to two existing datasets. The extant and new findings consistently demonstrate several novel features of brain functional organization not revealed by GLM. They include overlap of large-scale functional networks (FNs) and their concurrent opposite modulations, and no significant modulations in activity of most FNs across the whole brain during any task conditions. These novel features of brain functional organization are highly consistent with the brain’s properties of functional heterogeneity, balanced E/I, and sparseness of neuronal activity, and may help reconcile inconsistent GLM findings.

Constructing fine-granularity functional brain network atlases via deep convolutional autoencoder

“…Even though some pairs have largely overlapped regions, e.g. ICN 3 and ICN 12, ICN 4 and ICN 5, ICN 10 and ICN 11, ICN 14 and ICN 16, they tend to exhibit different activation patterns, which further confirms that spatial overlap of different functional networks is an intrinsic functional organization property of the human brain (Xu et al, 2016). …”

“…However, the metric of spatial overlap similarity has limited spatial description power and limited robustness to noises and image registration errors. Also, spatial overlap of functional networks is an intrinsic property of functional brain organization (Xu et al, 2016), which is supported by the neuroscientific point of view that cortical microcircuits are rather interdigitated with each other instead of being independent, and it complicates the application of spatial overlap similarity for overlapping networks. Thus, in literature studies, inter-expert visual validation after the spatial overlap similarity measurement is needed to confirm and establish spatial brain networks’ correspondences across individual subjects, which renders that the overlap similarity metric is both prone to inter-expert variability and time-consuming.…”

Developmental lead and/or prenatal stress exposures followed by different types of behavioral experience result in the divergence of brain epigenetic profiles in a sex, brain region, and time-dependent manner: Implications for neurotoxicology

“…Albeit potentially limited by current costs, it will ultimately be important to assess epigenetic profiles more broadly and correlatively across brain regions, recognizing that modularity and site specific changes are only a component of integrated brain function [67, 75]. Looking with both breadth and depth at integration as well as modularity may assist in finding more generalized markers important to specific environmental events or behavioral domains.…”

“…How such widespread and yet highly localized epigenetic impacts within the CNS ultimately relate to brain function and behavior is a critical question, as the brain is a highly-networked interactive system of systems [61, 62], with an extensive body of fMRI studies demonstrating existence of overlapping functional (behavioral) networks [63]. Indeed, CNS responses represent the integrated output of a network of systems of the brain, and not the modular response of a specific brain region or site or cell type [64–66].…”

“…As current studies have tended to focus on a single or perhaps two brain regions in assessment of epigenetic profiles, a more holistic networked-based epigenetic characterization has not yet been approached and may indefinitely remain cost prohibitive, even though a sizeable literature documents the fact that brain function is characterized by overlapping large scale functional networks (the connectome, [61, 62]), and that network connectivity can actually be predicted by local clustered gene expression, corresponding to the interactive nature of genetic underpinnings of brain function [67, 75]. …”