Functional Role of Cerebellar Gamma Frequency in Motor Sequences Learning: a tACS Study

Abstract

Although the role of the cerebellum in motor sequences learning is widely established, the specific function of its gamma oscillatory activity still remains unclear. In the present study, gamma (50 Hz)-or delta (1 Hz)-transcranial alternating current stimulation (tACS) was applied to the right cerebellar cortex while participants performed an implicit serial reaction time task (SRTT) with their right hand. The task required the execution of motor sequences simultaneously with the presentation of a series of visual stimuli. The same sequence was repeated across multiple task blocks (from blocks 2 to 5 and from blocks 7 to 8), whereas in other blocks, new/pseudorandom sequences were reproduced (blocks 1 and 6). Task performance was examined before and during tACS. To test possible after-effects of cerebellar tACS on the contralateral primary motor cortex (M1), corticospinal excitability was assessed by examining the amplitude of motor potentials (MEP) evoked by single-pulse transcranial magnetic stimulation (TMS). Compared with delta stimulation, gamma-tACS applied during the SRTT impaired participants' performance in blocks where the same motor sequence was repeated but not in blocks where the new pseudorandom sequences were presented. Noteworthy, the later assessed corticospinal excitability was not affected. These results suggest that cerebellar gamma oscillations mediate the implicit acquisition of motor sequences but do not affect task execution itself. Overall, this study provides evidence of a specific role of cerebellar gamma oscillatory activity in implicit motor learning.

Keywords: Cerebellar stimulation; Implicit motor learning; Serial reaction time task (SRTT); Transcranial alternating current stimulation (tACS); Transcranial magnetic stimulation (TMS).

Fig. 1

The figure shows the experimental procedure (left side) and the task’s structure (right side). In the pre-tACS phase, participants first performed the serial reaction time task (SRTT), followed by the assessment of the SI-1 mV (i.e., the lowest stimulus intensity needed to elicit MEP of 1 mV peak-to-peak) and of the input-output curve. In the online-tACS phase, stimulation was delivered during the execution of the SRTT (5 min). The electrodes were placed over the right cerebellum and the ipsilateral buccinator muscle. In the post-tACS phase, SI 1mV and the input-output procedure were again assessed. The panel on the right side shows the SRTT. The asterisk appeared in one of four possible positions marked by low dashes and corresponding to a key on a keyboard (V, B, N, M). Subjects responded to each asterisk by pressing with their right hand (index, middle, ring, pinkie) the key on the keyboard. Five hundred milliseconds after each response, the asterisk appeared at a new location. The task counted eight blocks, the same sequence was repeated in each block with the exception of blocks 1 and 6 were asterisks’ locations were randomly distributed

Fig. 2

Mean response times (RTs) expressed in milliseconds (ms) across task blocks and stimulation frequency (Delta, 1 Hz vs. Gamma, 50 Hz), before (pre) or during (online) tACS. The repeating sequence was presented from block 2 to block 5 and in blocks 7 and 8, respectively. A random sequence was embedded in blocks 1 and 6. Error bars represent standard deviation of mean RT

Fig. 3

Difference between mean response times (RTs) in the pre-tACS phase and mean RTs in the online-tACS phase across task blocks and stimulation types (Delta, 1 Hz vs. Gamma, 50 Hz). Error bars represent standard error of mean RTs difference. Here, RTs are expressed in milliseconds (ms) whereas statistical analyses were performed on log-transformed data, baseline corrected to block 1

Fig. 4

Mean amplitude of motor-evoked potential (MEP) in pre- and post-tACS phases of the two stimulation frequencies (Delta, 1 Hz vs. Gamma, 50 Hz) across transcranial magnetic stimulation (TMS) pulse intensities (from 100 to 140% of the TMS intensity that elicited a 1 mV peak-to-peak MEP, SI-1mV). Error bars represent standard error of mean MEP amplitude