Effects of Huddling on Sleep Dynamics and Neural Activity in Mice

Abstract

Huddling, or sleeping together in physical contact, is a common social thermoregulatory behavior observed in nature. Mice also exhibit this group sleeping behavior. While individual sleep patterns have traditionally been the focus, there is increasing interest in understanding how social interactions influence sleep dynamics at the group level. Recent studies have highlighted synchronization in sleep initiation and waking times during group sleep; however, the concurrent effects of group sleep on neural activities remain less understood. This study aimed to investigate changes in sleep patterns when mice huddle in groups compared to when they sleep alone. To do this, we conducted 12-hour recordings using wireless neurophysiological devices and video monitoring. Each experimental group consisted of four mice, with recordings of each mouse individually occupying a space on one day, and all four sharing the same space on the subsequent day. We collected neural signals from the frontal and parietal regions, along with electromyography (EMG) signals from neck muscles. Our findings revealed that group sleep reduced the duration of both non-rapid eye movement (NREM) and rapid eye movement (REM) sleep phases. As overall wakefulness increased, the REM to NREM ratio also decreased. Despite the stability of power across different frequencies when comparing group and individual sleep conditions in each sleep state, we noted a peak frequency shift in the theta band (5 ？ 10 Hz) during REM sleep for huddled mice compared to those sleeping alone. We observed that there was increased phase synchrony in the theta band during REM sleep among the mice in the huddling condition, suggesting enhanced inter-brain synchrony. Together, our findings underscore the significant influence of huddling on neural activity during sleep in mice as showing theta coherence during REM sleep, pointing to the potential mutual modulation of neural activity in social sleep dynamics.