Meeting Banner
Abstract #2251

Resting-State fMRI Signal Anti-Correlation Exists in Absence of Global Signal Regression

Xiao Liu1, Jeff H. Duyn1

1Advanced MRI section, LFMI, NINDS, National Institutes of Health, Bethesda, MD, United States

To understand the negative correlation between resting-state fMRI signals and how it could be affected by global signal regression (GSR) procedure, a novel technique was applied to temporally decompose the default mode network (DMN) into multiple co-activation patterns (CAPs). This decomposition of activity patterns during rest suggests that anti-correlation between brain regions is not an artifact of global signal regression but may be caused by brief periods of negatively correlated neuronal activity.

Keywords

absence according activate activation activations active activity additionally advanced almost although analyses anti apparent arousal audience averaging band brain brief caps caused certain classified clustering column commonly confirming connectivity consistent contradiction contribution correction correlated correlation correlations covering datasets decomposed decomposition demeaned deviation distinct entire even exists field filtering finding findings fluid functional furthermore global hand health highest illustration incorporate indicate indicated institutes intercept intermittent intervals introduced investigations larger largest linear local made middle mode modulate moreover motion motor much national negative network networks neuronal normalized novel nuisance occur original participants particularly pattern patterns perfectly periods positive potentials power preprocessing previous primate procedure processes processing project quadratic recorded registration regresses regression related removal replicates reported represent researchers rest resting role section seed seeded selected sensory series short similarity skipping spatial statistics step steps strong studies suggesting suggests support target task template temporal tend towards trends typical uniform various vectors visual white whole