People who more frequently recall their dreams have less cortical volume in the medial portion of the right fusiform gyrus and parahippocampal gyrus, as well as lower fractional anisotropy in the white matter fibers connected to those regions. These relationships were not affected by regular sleep. The findings provide direct evidence that brain structure is linked to individual differences in dream recall frequency.
Dream recall frequency and REM sleep percentage are linked to different brain functional networks. In 43 healthy adults, resting-state fMRI and polysomnography showed that both measures negatively correlated with multiple networks. Dream recall frequency was mainly associated with connectivity in the lateral visual network and thalamus, while REM sleep percentage was mainly associated with connectivity in frontoparietal networks and cerebellum. A time-of-day effect emerged: dream recall frequency had stronger coupling with the lateral visual network at night, and REM sleep percentage had stronger coupling with the cerebellum in the morning. The findings indicate that the neural substrates for dream recall and REM sleep are distinct.