Sleep State Influences Early Sound Encoding at Cortical But Not Subcortical Levels

In sleep, the brain balances protecting processes like memory consolidation with preserving responsiveness to significant external stimuli. Although reductions in higher-level auditory processes during deeper sleep have been described, the sleep-dependent changes across levels of auditory hierarchy, particularly as regards early sound representations, remain undefined. The frequency-following response (FFR) is an evoked auditory response that indexes neural encoding of sound periodicity. It is generated by neural populations in the brainstem, thalamus, and auditory cortex that phase-lock to periodic auditory stimuli and encode pitch information. The FFR's neural sources, which can be resolved using magnetoencephalography, allow evaluation of neural representation strength throughout the auditory neuraxis as a function of sleep state, as well as neural events like slow waves and sleep spindles that are hypothesized to attenuate acoustic processing as a means of preserving the sleep state. We recorded FFRs during a 2.5 h nap from 14 healthy male and female human adults to investigate how sleep depth and microarchitecture affect auditory encoding. We show that FFR strength is maintained across non-rapid eye movement sleep stages in subcortical nuclei, yet decreases in deeper sleep in the auditory cortex. FFR strength was not influenced by slow wave or spindle activity, but rather by reduced communication between the thalamus and cortex. This differentiation in sound representation across the auditory hierarchy suggests a means by which the brain might balance environmental monitoring with preserving critical restorative processes.