Resting state fMRI (functional magnetic resonance imaging) measures spontaneous fluctuations in brain activity when a person is not performing any specific task. Participants typically lie still with eyes closed or fixating on a cross, while the scanner records slow changes in blood oxygenation (the BOLD signal) across the brain.

These signals are temporally correlated between distant regions, revealing large scale functional networks that remain active at rest. A key example is the default mode network, involving medial prefrontal cortex, posterior cingulate cortex, and angular gyrus, which is more active during internal mentation such as mind wandering than during demanding external tasks. Other networks consistently identified include visual, auditory, somatomotor, attention, and control networks, as well as limbic and subcortical circuits.

Research uses methods such as seed based correlation, independent component analysis, and graph theory to quantify connectivity patterns and network organization. Resting state networks show robust reproducibility across individuals, species, and scanning sites, and are present even in infants, suggesting they reflect fundamental aspects of brain architecture.

Alterations in resting connectivity have been reported in many conditions including Alzheimer’s disease, schizophrenia, depression, autism, ADHD, epilepsy, and disorders of consciousness. These findings raise hopes that resting state fMRI could serve as a biomarker for diagnosis, prognosis, and treatment monitoring, and could guide individualized interventions such as brain stimulation.

However, the technique faces challenges: sensitivity to motion and physiological noise, variability across individuals, and difficulties in interpreting causality or direction of information flow. Ongoing work focuses on improving preprocessing, modeling dynamic changes in connectivity over time, and integrating resting data with structural and task based measures to better understand brain function and dysfunction.