Exploring Deep Magnetoencephalography via Thalamo-Cortical Sleep Spindles

Subcortical brain regions like the thalamus are integral to numerous sensory and cognitive functions. Magnetoencephalography (MEG) enables the study of widespread brain networks with high temporal resolution, but the degree to which deep sources like the thalamus can be resolved remains unclear. Functional connectivity methods may enhance differentiation, yet few studies have extended them beyond the cortex. We investigated the possibility of resolving deep sources via connectivity patterns during thalamo-cortical sleep spindles to leverage their well-characterized circuitry, and during spindle-free periods of non-rapid eye movement sleep to explore neural recordings that lack such high-amplitude bursts of activity. MEG and electroencephalography (EEG) were recorded in 19 participants during a 2-h nap. Spindle and non-spindle periods were identified, and connectivity was assessed using coherence and imaginary coherence within a spindle-related network. Graph theory was also applied to identify network hubs. As expected, functional connectivity increased during spindles within a distributed thalamo-cortical-hippocampal network. Cortical connectivity patterns allowed differentiation among small thalamic nuclei, but metric choice and contrast use influenced topography and distance effects. Graph theory revealed distinct cortical, thalamic, and hippocampal contributions to fast (13–16 Hz) and slow (10–13 Hz) sigma-band connectivity. These findings demonstrate that MEG functional connectivity can resolve deep brain networks during NREM sleep and during spindles, and demonstrate how it can be used to study the functional roles of subcortical regions non-invasively in healthy humans. By clarifying methodological influences, we aim to guide future research design and interpretation.