Reliability and validity of brain‐gastric phase synchronization

Recent studies have reported that various brain regions, mainly sensory, unimodal regions, display phase synchronization with the stomach's slow (0.05 Hz) myoelectrical rhythm. These gastric–brain interactions have broad implications, from feeding behavior to functional gastrointestinal disorders. However, in contrast to other interoceptive signals (e.g., heart rate) and their relation to the brain, little is known about the reliability of these gastric–brain interactions, their robustness to artifacts such as motion, and whether they can be generalized to new samples. Here we examined these aspects in 43 subjects that had undergone multiple runs of concurrent electrogastrography (EGG), brain fMRI, and pulse oximetry. We also repeated all analyses in an open dataset of a highly sampled individual. We found a set of brain regions that were coupled with the EGG signal after controlling for non‐grey matter (GM) signals, head motion, and cardiac artifacts. These regions exhibited significant overlap with previous work. However, we also showed that prior to confound regression, the spatial extent of the gastric network was largely overestimated. Finally, we found substantial test–retest reliability in both the brain and the gastric signals when estimated alone, but not for measures of gastric‐brain synchrony. Together, these results provide methodological scaffolding for future research into brain–stomach interactions and for a better understanding of the role of the gastric network.