# Somatosensory Induced Cerebellar Responses to Peripheral Nerve Stimulation: A Time and Time–Frequency EEG Study

## TL;DR

This study shows that EEG can detect cerebellar responses to nerve stimulation, with time–frequency analysis being more effective than traditional methods.

## Contribution

The study demonstrates that cerebellar activity can be non-invasively detected using time–frequency EEG analysis during peripheral nerve stimulation.

## Key findings

- Time–frequency analysis revealed sustained oscillatory power increases in cerebellar channels after nerve stimulation.
- Power increases peaked at ~20 Hz for upper-limb and ~10 Hz for lower-limb stimulation, with side specificity.
- Time-domain responses showed limited spatial specificity and subcortical contamination, unlike time–frequency results.

## Abstract

Background/Objectives: The cerebellum plays a central role in sensorimotor integration and temporal processing, yet its direct electrophysiological investigation in humans remains challenging, and cerebellar contributions to somatosensory responses remain poorly defined. This study aimed to determine whether cerebellar responses to peripheral nerve stimulation can be detected using scalp EEG and whether time–frequency analysis provides advantages over time-domain approaches. Methods: Scalp EEG was recorded during electrical stimulation of the median nerve and tibial nerve in 16 healthy participants. Electrode montages included posterior fossa placements targeting cerebellar activity, together with standard cortical and subcortical derivations. Data were analyzed in the time domain using evoked potentials and channel comparisons, including bipolar cerebellar derivations, and in the time–frequency domain using spectral power analysis. Results: Time-domain analyses revealed early and intermediate latency components following both upper- and lower-limb stimulation; however, these responses showed limited spatial specificity and were strongly influenced by reference effects and subcortical contamination. In contrast, time–frequency analysis consistently revealed sustained increases in oscillatory power in cerebellar channels. Power increases emerged approximately 50 ms after stimulation and persisted beyond 300 ms, peaking around ~20 Hz for upper-limb stimulation and ~10 Hz for lower-limb stimulation, with evidence of side specificity. Conclusions: Non-invasive EEG can detect cerebellar responses to peripheral nerve stimulation, particularly in the time–frequency domain. Oscillatory dynamics provide a more robust marker of cerebellar involvement than time-locked responses and may complement conventional somatosensory evoked potentials in studies of cerebellar physiology and spinocerebellar pathway integrity.