Surface EEG Evidence for Cerebellar Control of Distal Upper Limbs in Humans

Abstract

Background/Objectives: The cerebellum plays a crucial role in motor control, but its direct electrophysiological investigation in humans is challenging. Electrocerebellograms (ECeGs), recorded via surface electrodes below the inion, have been proposed as a non-invasive method to assess cerebellar activity. However, its interpretation is complicated by potential interference from occipital alpha rhythms and neck muscle signals. This study aimed to investigate whether ECeG signals genuinely reflect cerebellar involvement during upper limb movement and to explore possible confounding influences. Methods: We recorded electroencephalograms (EEGs) from occipital (Oz) and cerebellar electrodes (Cb1 and Cb2), alongside EMGs from forearm muscles in healthy individuals performing sinusoidal (~1 Hz) and tremor-like (~4 Hz) wrist movements. To assess occipital contamination, recordings were obtained under both eyes-open and eyes-closed conditions. Results: Occipital alpha power was present in Cb1 and Cb2 but was less affected by eye-opening than at Oz, suggesting a partially distinct neural source. During the tremor condition, movement-frequency power increased at Cb2 and C3 (corresponding to the ipsilateral cerebellar hemisphere and contralateral motor cortex), indicating authentic cerebellar activity. No significant movement-related EEG changes were observed during sinusoidal movements, likely due to weaker neuronal synchronization. Conclusions: These findings suggest that ECeGs can detect cerebellar signals linked to movement, especially during faster and rhythmic motions, and are only moderately affected by occipital contamination. This supports the feasibility of non-invasive cerebellar electrophysiology and underscores the need for further methodological refinement to enhance signal specificity.

Keywords: alpha rhythm; cerebellum; electrocerebellogram; electroencephalography; motor cortex; sinusoidal movements; tremor.

Figure 1

Experimental blocks. (A) Resting EEG with eyes closed. (B) Resting EEG with eyes open. (C) Sinusoidal movements of the wrist. (D) Simulated tremor. See text for details.

Figure 2

Placement of recording electrodes on the scalp (A) and the forearm (B). See text for details.

Figure 3

Alpha power and frequency. (A–C) depict the PSD at electrodes Oz, Cb1, and Cb2, respectively, in Rest-eo (red lines) and Rest-ec (green lines) conditions. Shaded areas indicate the standard error of the mean. The black vertical dashed lines indicate the frequency range from which values of alpha power and frequency were extracted (see text for details). (D,E) indicate pairwise comparisons for values of alpha power and frequency, respectively. Average values (bars) and standard errors of the mean (whiffs) are depicted. Brackets and asterisks indicate statistically significant differences (p < 0.05).

Figure 4

In (A), EMG traces from both muscles and recording conditions (Sinus and Tremor) in a representative subject are depicted. (B) shows EMG frequency in Sinus and Tremor conditions. Average values (bars) and standard errors of the mean (whiffs) are depicted. M1: flexor carpi radialis; M2: extensor carpi radialis. Brackets and asterisks indicate statistically significant differences (p < 0.05).

Figure 5

ECeG results from the “Sinus” condition. (A,C,E,G,I) PSD from each recording electrode (C3, C4, Oz, Cb1, and Cb2, respectively). Vertical black lines indicate the average EMG activity of the related condition (“Sinus”) while vertical dashed black lines indicate the corresponding standard error. Shaded areas indicate the standard error of the ECeG PSD. (B,D,F,H,J) Box plots of power values extracted from the PSDs corresponding to peak EMG activity. Box edges represent the standard deviation, vertical bars are the 25° and 75° percentiles, horizontal black lines are the medians, green dots are the averages, and dashed curves represent the gaussian fitting of the data.

Figure 6

ECeG results from the “Tremor” condition. (A,C,E,G,I): PSD from each recording electrode (C3, C4, Oz, Cb1, and Cb2, respectively). Vertical black lines indicate the average EMG activity of the related condition (“Tremor”), while vertical dashed black lines indicate the corresponding standard error. Shaded areas indicate the standard error of the ECeG PSD. (B,D,F,H,J): Box plots of power values extracted from the PSDs corresponding to peak EMG activity. Box edges represent the standard deviation, vertical bars are the 25° and 75° percentiles, horizontal black lines are the medians, green dots are the averages, and dashed curves represent the gaussian fitting of the data. Brackets and asterisks indicate statistically significant differences (p < 0.05).