The performance improvement (increase of movement cycle) was only possible in the thenar stimulation trials but not in the wrist stimulation trials. Thus, the thenar stimulation was effective, and involvement of attentional factors in the performance improvement should be refuted, as the wrist stimulation also likely attracted the participant's attention to the hand similar to with the thenar stimulation. We also found no performance improvement in the sham stimulation trials. This indicates that either “effective” stimulation requires certain period of stimulation time or motor performance should be initiated immediately after the cessation of stimulation in order to get beneficial effects from the stimulation, or both. But, at least, this result has ruled out the possibility that the beneficial effect from the thenar stimulation is merely attributed to placebo effects from the stimulation, i.e. participants expect something good happens after the stimulation.
The behavioral changes in the thenar stimulation trials could be attributed to the net effect of electrical stimulation as somatic stimulation (see below), though the exact mechanisms of how this stimulation allowed the nervous system to improve the performance need to be clarified. This claim is supported by different evidence. First, the performance never improved even when the participants exerted their maximum effort within an experimental day () and even across the days (), but the stabilized performance was improved by the application of the stimulation with the explicit subjective experience of smooth performance (). Second, the stimulation reduced the amplitude of finger displacement while the participants did not notice this change, which was observed even during the constant rotation at the regular pace where no particular motivation to exert maximum effort was required (). Finally, even when the participants tried to switch to the motor strategy (reducing amplitude of finger displacement) that was essential for performance improvement, this was very difficult to perform () and the volitional effort never improved, but rather worsened, the performance (). All support the view that the performance improvement was unlikely attributed to other possibilities such as changes in volitional effort (motivation) and motor strategy.
The reduction in amplitude of finger displacement appeared to be a genuine behavioral outcome from the stimulation effect, as this was consistently observed irrespective of the task demands (, ). The reduction of amplitude of finger displacement accompanied reduction of its associated muscular activity (), which was generally observed in association with performance improvement in this task 
. Since reduction of amplitude of finger displacement was also associated with performance improvement in the present study (), this change must be essential for goal-directed optimal control of the task.
The present short-period stimulation most likely modulated and remodeled sensory-motor neuronal states in a non-task-specific way 
, presumably in a distinct way when using long-period stimulation that can produce long-term potentiation (LTP)-like modulation in the central motor system 
. As described, high-frequency electrical stimulation is capable of modulating the sensory-motor neuronal states so as to generate a non-task-specific effect of reducing excessive motor activity, such as muscle hypertonia (see introduction
). In this way, we found that the present stimulation could reduce redundant motor activities that are not essentially required for execution of the performance. Thus, the neuronal states likely modulated by this stimulation might be the key for the nervous system to unconsciously promote efficient control of subsequent motor performance. This indicates the potential effectiveness of short-period, high-frequency electrical stimulation. The general effectiveness of short-period somatic stimulation has been recently demonstrated: high-frequency electrical afferent stimulation in the spinal cord, even if it is short-period (less than one minute), is capable of modulating neuronal discharge in the cortico-subcortical motor circuit, which can facilitate permissive motor behavior of animals 
In the present trained participants, without stimulation, their performance never improved, even when repeating ordinary physical training across five days. In contrast, the performance improved in association with the stimulation consistently throughout the stimulation training days. This suggests the effectiveness and consistency of the thenar stimulation effect at least in the present participants and task. Even though we could not show the change of amplitude of finger displacement across days (see methods
), based on the relationship between the performance improvement and the kinematic amplitude of finger displacement (), we presume that the kinematic change was somehow associated with the performance improvement across days (). Simple repetition of physical training of the stabilized motor skill might merely reinforce habitual access to its routine neuronal circuits, thereby resulting in no substantial performance improvement across the days. In contrast, the sensory-motor remodeling by the stimulation might assist the nervous system in shifting to a further optimal state, which allows upgraded performance. Surprisingly, this shift was possible even in the very first stimulation trial (fast-acting effect; ), and its stabilization could be established through repetition of stimulation trials (retention; ) where the upgraded control processes might be consolidated in the nervous system. This type of neuronal operation must be necessary to generate the carry-over effect in the next day's performance (), which promoted motor skill learning beyond the plateau. We speculate that the daily skill improvement () was mediated by long-term plastic change in the central motor system as previously reported 
In summary, electrical somatic stimulation allowed not only for upgrading control of the plateaued motor skill but also for promoting motor skill learning at the plateau stage. Thus, this study raised the possibility that a human plateaued motor skill can be effectively improved with pre-movement somatic stimulation, which is available for anyone at any time, and in any place.