Background 

The ability to cancel or stop planned movements is a key aspect of self-control. The process is thought to involve a prefrontal-basal ganglia-thalamocortical network, which exerts a broad suppression over primary motor cortex (M1) to prevent planned movements from being expressed^[1]. For example, when stopping a hand movement, suppression is observed in unrelated motor cortical representations, such as leg motor areas^[2]. However, this broad suppression could be counterproductive when attempting to stop an ongoing movement. In such cases, stopping requires the activation of antagonist muscles to actively 'brake' the motion^[3]. 

Aims 
In this study, we set out to examine the hypothesis that the activation of antagonist muscles recruited to stop an ongoing movement is independent of the M1 and, consequently, remains unaffected by cortical suppression. 

Methods 

In two ongoing experiments, participants engaged in a stop signal task using a computer mouse. They responded to 'go' cues by performing reaching movements to maneuver the cursor towards targets on a screen. In ~25% of trials, they received a stop signal instructing them to halt their planned movement. The timing of the stop signal was adjusted so that participants were able to successfully prevent their response in 50% of the trials. In the other 50%, they initiated a movement but then interrupted it before reaching the target. We recorded electromyogram (EMG) activity during the movements from the right anterior and posterior deltoid muscles (AD, PD), which serve as agonists and antagonists during reaching. Transcranial magnetic stimulation (TMS) was applied to M1 to elicit motor evoked potentials (MEPs) in the task-irrelevant FDI of the left hand (Exp.1) and task-relevant AD and PD (Exp.2) at 50 ms intervals following the presentation of the stop signal.  

Results 

Preliminary results from Exp.1 (n=7, age 24±7 years, 3 males) indicated an expected suppression^[3,4] of MEPs in the task-irrelevant FDI, which emerged 150 ms after the stop signal and continued for ≥100 ms. It was observed both when movement was entirely prevented and when it was initiated but then halted by the activation of the antagonist muscle, whose activity started during the suppression phase, ~200 ms after the stop signal. Initial results from Exp.2 (n=8, age 24±7 years, 4 males) focused on examining the temporal relationship between changes in muscle activity and MEP amplitudes in agonist and antagonist muscles. If M1 is responsible for driving EMG activity, we would expect an increase in MEP amplitudes to precede the rise in EMG activity, with a time delay equivalent to the cortico-muscular conduction time^[5]. This pattern held true for the agonist muscles when initiating movements (time delay ~9 ms), but not when the antagonist was recruited to stop an ongoing movement. 

Conclusions 

The present data offer preliminary evidence of substantial antagonist "braking" activity when stopping an ongoing movement, even in the context of extensive motor system suppression, and without a preceding increase in M1 excitability. This suggests that the initiation of antagonist activity may be governed by a distinct, potentially subcortical, mechanism^[5].