Neuronal population dynamics during motor plan cancellation in non-human primates

Abstract

To understand the cortical neuronal dynamics behind movement generation and control most studies focused on tasks where actions were planned and then executed, using different instances of visuomotor transformations. However, to fully understand the dynamics related to movement control one must also study how movements are actively inhibited. Inhibition, indeed, represents the first level of control both when different alternatives are available and only one solution could be adopted and when is necessary to maintain the current position.We recorded neuronal activity from a multielectrode array in the dorsal premotor (PMd) cortex of monkeys performing a countermanding reaching task that requires, in a subset of trials, to cancel a potentially planned movement before its onset. In the analysis of the neuronal state-spaces of PMd we found a subspace in which activities conveying temporal information were confined during active inhibition and position holding. Movement execution required activities to escape from the plane toward an orthogonal subspace and, furthermore, surpass a threshold associated to the maturation of the motor plan.These results revealed further details in the neuronal dynamics underlying movement control extending the hypothesis that neuronal computation confined in an output-null subspace does not produce movements.

Response inhibition in premotor cortex corresponds to a complex reshuffle of the mesoscopic information network

“…A similar trend was found in all recording sessions (Table S1), confirming previous findings on the role of PMd in inhibitory motor control and the relevance of the epoch of analysis in motor decisions in Stop trials 20,[22][23][24][76][77][78][79] . This level of analysis, however, does not provide information on how the nodes in the…”

“…It also proves particularly useful when one wants to directly analyze experimental brain networks and reconstruct their generative models 44,96,97 . For all these reasons, although is still not largely used in behavioral neurophysiological studies at the small and mesoscale, a 18/30 graph theory-based conceptualization of neural interactions, possibly united with information theoretic measures, can be very profitable also compared to other common approaches relying on the analysis of covariance between neurons or mesoscopic signals 14,22,[98][99][100][101] and should be exploited more. In fact, these methods are not straightforward in distinguishing the specific contributions of single neurons (or discrete populations of neurons) to the topology of network dynamics, which is our strategy's strength.…”

“…Some studies have employed the Stop signal task 19 , that requires actively inhibiting an upcoming movement upon presentation of a Stop signal, to investigate how neuronal activity in this area supports both movement generation and cancellation. According to these studies, neuronal activity in PMd is strongly related to both the generation and the suppression of an arm movement at different scales, from single neurons 20,21 to population of neurons 22 and local mesoscopic signals 14,23,24 .…”

Movement initiation and grasp representation in premotor and primary motor cortex mirror neurons

“…Causal perturbation experiments in conjunction with state-space analyses could provide supporting evidence that action observation activity partly evolves within a ’withholding’ subspace, if for example, thresholds for inducing movement during observation were dependent on stimulation time, or observing congruent or incongruent actions differentially affect action execution. This withholding subspace could also be characterised further using, for example, a stop-signal reaction time (SSRT) task ( Pani et al, 2019 ) where failed-stop trials are frequent, although the implementation of this within an action observation paradigm is not straightforward and requires careful consideration.…”

“…At the spinal level, PTNs not only excite motoneurons via cortico-motoneuronal (CM) projections ( Porter and Lemon, 1993 ; Rathelot and Strick, 2006 ), but also exert indirect effects via segmental interneuron pathways, which in turn display their own complex activity before and during movement ( Prut and Fetz, 1999 ; Takei and Seki, 2013 ). A dynamical systems approach ( Shenoy et al, 2013 ) has recently suggested that movement-related activity unfolds in largely orthogonal dimensions to activity during action preparation, such that movement is implicitly gated during movement preparation ( Kaufman et al, 2014 ; Elsayed et al, 2016 ), and a similar mechanism has been hypothesised to operate during action observation ( Mazurek et al, 2018 ) and action suppression ( Pani et al, 2019 ). While the roles of F5 and M1 during the execution of visually-guided grasp have been studied extensively ( Umilta et al, 2007 ; Davare et al, 2008 ; Schaffelhofer and Scherberger, 2016 ), a more systematic understanding of the differences between action execution and observation activity in these two key nodes in the grasping circuitry could provide important insights into dissociations between representation of potential actions at the cortical level, and recruitment of descending pathways and muscles for actual action execution ( Schieber, 2011 ).…”