Role of the primary motor cortex in the early boost in performance following mental imagery training

Abstract

Recently, it has been suggested that the primary motor cortex (M1) plays a critical role in implementing the fast and transient post-training phase of motor skill consolidation, known to yield an early boost in performance. Whether a comparable early boost in performance occurs following motor imagery (MIM) training is still unknown. To address this issue, two groups of subjects learned a finger tapping sequence either by MIM or physical practice (PP). In both groups, performance increased significantly in the post-training phase when compared with the pre-training phase and further increased after a 30 min resting period, indicating that both MIM and PP trainings were equally efficient and induced an early boost in motor performance. This conclusion was corroborated by the results of an additional control group. In a second experiment, we then investigated the causal role of M1 in implementing the early boost process resulting from MIM training. To do so, we inhibited M1 by applying a continuous theta-burst stimulation (cTBS) in healthy volunteers just after they learnt, by MIM, the same finger-tapping task as in Experiment #1. As a control, cTBS was applied over the vertex of subjects who underwent the same experiment. We found that cTBS applied over M1 selectively abolished the early boost process subsequent to MIM training. Altogether, the present study provides evidence that MIM practice induces an early boost in performance and demonstrates that M1 is causally involved in this process. These findings further divulge some degree of behavioral and neuronal similitude between MIM and PP.

Figure 1. Schematic view of Experiment #1 protocol.

The task consisted of performing a sequential finger tapping with the left hand on a computer keyboard. The experiment was divided into six distinct phases: 1) a familiarization phase, 2) a first pre-training, practice session to evaluate subject performance baseline before learning, 3) a training phase during which 3 different group of subjects (n = 8) either physically (PP) or mentally (MIM) performed the sequence (PP) during 10 blocks of 30 seconds, each separated by a 20 second resting period or were engaged in a control (Ctrl) reading task, 4) a first post-training session, 5) a 30 minute delay period, 6) a retest session in order to evaluate the “early boost”.

Figure 2. Schematic view of Experiment #2 protocol.

The experiment was the same as that undergone by subjects from the MIM group in Experiment #1 except that, after the post-training session, cTBS was applied either over the contralateral (right) primary motor cortex (M1) in a group of 8 subjects or over the vertex in another group of 8 controls subjects.

Figure 3. Effect of training on the total number of correct sequences.

The total number of correct sequences performed during the two-block practice of the pre-training, the post-training and the re-test sessions, have been computed for the three different groups of subjects participating in Experiment #1. Both MIM and PP groups increased their performance significantly in the post-training session, while no significant difference was found in the Ctrl group. Similarly, during re-test, only in the MIM and PP groups, we found an increase in the total number of sequences, demonstrating the occurrence of an early boost effect. Error bars indicate one SD.

Figure 4. Early boost following MIM and PP training.

The increase in the total number of sequences between the post-training and re-test sessions is expressed in percent with respect to data gathered in the pre-training session. This shows that MIM and PP training led to the same increase in performance in the post-training and re-tests sessions and confirms the occurrence of an early boost in both groups. The inset shows the amplitude of the early boost, estimated as the difference between the total number of sequences in the re-test and post-training sessions - in both groups and expressed in percent; no difference in early boost amplitude was found between the two groups. Error bars indicate one SD.

Figure 5. Lack of early boost following M1 cTBS.

The increase in the total number of sequences in the post-training and re-test sessions is expressed in percent with respect to data gathered in the pre-training session. This shows that, whereas the performance in the Vertex and M1 groups was comparable in the post-training session, no early boost was found in the M1 group. This difference between groups was confirmed by data shown in the inset (same conventions as in Fig. 4). Error bars indicate one SD.