Physiological and head motion signatures in static and time-varying functional connectivity and their subject discriminability

It is well established that head motion and physiological fluctuations have a pronounced influence on resting-state fMRI activity. Here, we capitalize on a large sample from the Human Connectome Project to provide a comprehensive investigation of the biases in functional connectivity (FC) that arise from head motion, breathing motion, cardiac pulsatility, and systemic low-frequency oscillations (SLFOs) associated with changes in heart rate and breathing patterns. In static FC, artifactual connectivity was found in the sensorimotor network due to head motion, as well as in the visual network due to head motion and SLFOs. Breathing motion was found to increase within-hemisphere connectivity, which was attributed to the phase encoding direction. Recurrent patterns in time-varying FC were found to be partly correlated to head motion and SLFOs. Several preprocessing strategies were examined to assess their capability in mitigating the effects of nuisance processes. Nuisance signatures exhibited above-chance levels of subject discriminability; however, fMRI data corrected for these confounds improved subject identifiability.