Sensory processing reallocation from external to internal signals in REM sleep
Jacinthe Cataldi, Andria Pelentritou, Sophie Schwartz, Marzia De Lucia
bioRxiv (Cold Spring Harbor Laboratory) March 18, 2026 DOI: 10.64898/2026.03.16.712081 (opens in new tab) via OpenAlex
Summary
AI-generated from the abstractThe brain continuously processes signals from the outside world and from inside the body. How this balance shifts across different states of consciousness is not well understood. This study measured brain responses to external sounds (auditory evoked potentials) and to heartbeats (heartbeat evoked potentials) in 25 healthy participants during wakefulness and during two types of REM sleep: tonic REM and phasic REM, which have progressively lower responsiveness to the environment. Auditory responses decreased from wakefulness to tonic REM and were smallest during phasic REM. Heartbeat responses remained strong across REM sleep and were even larger than during wakefulness.
Study at a glance
| Characteristics | Observational cohort Peer reviewed |
|---|---|
| Sample size | 25 |
| Population | Healthy participants |
| Keywords | Tonic physiology Wakefulness Vigilance psychology Sensory system Electroencephalography |
| Key finding | Auditory evoked potentials decreased from wakefulness to tonic REM and were most attenuated during phasic REM, while heartbeat evoked potentials were preserved and enhanced during REM sleep, indicating a shift from external to internal processing. |
Abstract
Abstract The brain continuously integrates information from the external environment (exteroception) and the internal bodily milieu (interoception). How the balance between these two processing streams shifts across vigilance states with differing levels of environmental responsiveness, however, remains poorly understood. Here, we examined neural responses to external auditory and internal cardiac signals across wakefulness and REM sleep microstates - tonic and phasic REM - which are characterized by progressively reduced responsiveness to external stimulation. High-density EEG was recorded in healthy participants (n=25). Auditory evoked potentials (AEPs) and heartbeat evoked potentials (HEPs) served as indices of exteroception and interoception, respectively, and were compared across vigilance states. AEPs progressively decreased from wakefulness to tonic REM and were most attenuated during phasic REM. In contrast, HEPs were preserved across REM microstates and were enhanced relative to wakefulness, indicating sustained - and even amplified - processing of cardiac signals during REM sleep. To quantify the relative weighting of external and internal signals, we introduce an exteroceptive–interoceptive index, defined as the ratio of auditory to cardiac neural responses. This index decreased systematically across vigilance states, revealing a graded shift from externally oriented processing during wakefulness to internally oriented processing during phasic REM, with tonic REM occupying an intermediate position. Together, these findings demonstrate that while responsiveness to external stimuli diminishes during phasic REM, the brain continues to prioritize physiologically relevant internal signals. The exteroceptive-interoceptive balance may thus provide a novel, mechanistically grounded marker of altered consciousness, particularly informative in contexts where behavioural responsiveness cannot be assessed. Graphical abstract