A basic difficulty for the nervous system is integrating locally ambiguous sensory information to form accurate perceptions about the outside world1–4. This local-to-global problem is also fundamental to motor control of the arm since complex mechanical interactions between the shoulder and elbow allow a particular amount of motion at one joint to arise from an infinite combination of shoulder and elbow torques5 (Fig. 1a). Here we show that a transcortical pathway through primary motor cortex (M1) resolves this ambiguity during fast feedback control. We demonstrate that single M1 neurons of behaving monkeys can integrate shoulder and elbow motion information into motor commands which appropriately counter the underlying torque within ~50 ms of a mechanical perturbation. Moreover, we reveal a causal link between M1 processing and multi-joint integration in humans by showing that shoulder muscle responses occurring ~50 ms after pure elbow displacement can be potentiated by transcranial magnetic stimulation. Our results show that M1 underlies multi-joint integration during fast feedback control, demonstrating that transcortical processing permits feedback responses to express a level of sophistication previously reserved for voluntary control and providing neurophysiological support for influential theories positing that voluntary movement is generated by the intelligent manipulation of sensory feedback6,7.

The motor system must consider a variety of environmental factors when executing voluntary motor actions, such as the shape of the goal or the possible presence of intervening obstacles. It remains unknown whether rapid feedback responses to mechanical perturbations also consider these factors. Our first experiment quantified how feedback corrections were altered by target shape, which was either a circular dot or a bar. Unperturbed movements to each target were qualitatively similar on average but with greater dispersion of end point positions when reaching to the bar. On random trials, multijoint torque perturbations deviated the hand left or right. When reaching to a circular target, perturbations elicited corrective movements that were directed straight to the location of the target. In contrast, corrective movements when reaching to a bar were redirected to other locations along the bar axis. Our second experiment quantified whether the presence of obstacles could interfere with feedback corrections. We found that hand trajectories after the perturbations were altered to avoid obstacles in the environment. Importantly, changes in muscle activity reflecting the different target shapes (bar vs. dot) or the presence of obstacles were observed in as little as 70 ms. Such changes in motor responses were qualitatively consistent with simulations based on optimal feedback control. Taken together, these results highlight that long-latency motor responses consider spatial properties of the goal and environment.

Recent theories of voluntary control predict that multiple motor strategies can be precomputed and expressed throughout movement. We examined online decisional processing in humans by asking them to make reaching movements with obstacles located just to the sides of a direct path between start and end targets. On random trials, the limb was perturbed with one of four mechanical loads that varied in direction and amplitude. Notably, we observed two different strategies when we applied a perturbation (left medium-sized) that deviated the participants' hand directly toward an obstacle. In some trials, subjects directed their hand between the obstacles and in other trials to the left of the obstacles. Importantly, changes in the muscle stretch response between these two strategies were observed in Ͻ60 ms after perturbation, during the R2 long-latency epoch (ϳ45-75 ms). As predicted, the selected strategy depended on the estimated position of the limb when it was perturbed. In our second experiment, we presented either one or three potential goal targets. Movements initially directed to the closest target could be quickly redirected to other potential targets after a perturbation. Differences in muscle stretch responses for redirected movements were observed ϳ75 ms after perturbation during the R3 long-latency epoch (ϳ75-105 ms). The results show that decisional processes are rapidly implemented during movement execution. In addition, our data suggest a hierarchical process with corrective responses on "how" to attain a behavioral goal expressed during the R2 epoch and responses on "what" goal to attain during the R3 epoch.

No abstract

An important aspect of motor control is the ability to perform tasks with the upper limbs while maintaining whole body balance. However, little is known about the coordination of upper limb voluntary and whole body postural control after mechanical disturbances that require both upper limb motor corrections to attain a behavioral goal and lower limb motor responses to maintain whole body balance. The present study identified the temporal organization of muscle responses and center of pressure (COP) changes following mechanical perturbations during reaching. Our results demonstrate that muscle responses in the upper limb are evoked first (∼50 ms), with lower limb muscle activity occurring immediately after, in as little as ∼60 ms after perturbation. Hand motion was immediately altered by the load, while COP changes occurred after ∼100 ms, when lower limb muscle activity was already present. Our secondary findings showed that both muscle activity and COP changes were influenced by behavioral context (by altering target shape, circle vs. rectangle). Voluntary and postural actions initially directed the hand toward the center of both target types, but after the perturbation upper limb and postural responses redirected the hand toward different spatial locations along the rectangle. Muscle activity was increased for both upper and lower limbs when correcting to the circle vs. the rectangle, and these differences emerged as early as the long-latency epoch (∼75-120 ms). Our results demonstrate that postural responses are rapidly and flexibly altered to consider the behavioral goal of the upper limb. The present work establishes that, when reaching to a target while standing, perturbations applied to the upper limb elicit a rapid response in lower limb muscles. Unlike voluntary movements, postural responses do not occur before corrections of the upper limb. We show the first evidence that corrective postural adjustments are modulated by upper limb behavioral context (target shape). Importantly, this indicates that postural responses take into account upper limb feedback for online control.

Humans vary greatly in their motor learning abilities, yet little is known about the neural processes that underlie this variability. We identified distinct profiles of human sensorimotor adaptation that emerged across 2 days of learning, linking these profiles to the dynamics of whole-brain functional networks early on the first day when cognitive strategies toward sensorimotor adaptation are believed to be most prominent. During early learning, greater recruitment of a network of higher-order brain regions, involving prefrontal and anterior temporal cortex, was associated with faster learning. At the same time, greater integration of this “cognitive network” with a sensorimotor network was associated with slower learning, consistent with the notion that cognitive strategies toward adaptation operate in parallel with implicit learning processes of the sensorimotor system. On the second day, greater recruitment of a network that included the hippocampus was associated with faster learning, consistent with the notion that declarative memory systems are involved with fast relearning of sensorimotor mappings. Together, these findings provide novel evidence for the role of higher-order brain systems in driving variability in adaptation.

When presented with competing potential reach targets and required to launch a movement before knowing which one will be cued as the target, people initially reach in the average target direction. Although this spatial averaging could arise from executing a weighted average of motor plans for the potential targets, it could also arise from planning a single, optimal movement. To test between these alternatives we used a task in which participants were required to reach to either a single target or towards two potential targets while grasping an object. A robotic device applied a lateral elastic load to the object requiring large grip forces for reaches to targets either side of midline and a minimal grip force for midline movements. As expected, in trials with two targets located either side of midline, participants initially reached straight ahead. Critically, on these trials the initial grip force was minimal, appropriate for the midline movement, and not the average of the large grip forces required for movements to the individual targets. These results indicate that under conditions of target uncertainty, people do not execute an average of planned actions but rather a single movement that optimizes motor costs.

Humans vary greatly in their motor learning abilities, yet little is known about the neural mechanisms that underlie this variability. Recent neuroimaging and electrophysiological studies demonstrate that large-scale neural dynamics inhabit a low-dimensional subspace or manifold, and that learning is constrained by this intrinsic manifold architecture. Here, we asked, using functional MRI, whether subject-level differences in neural excursion from manifold structure can explain differences in learning across participants. We had subjects perform a sensorimotor adaptation task in the MRI scanner on 2 consecutive days, allowing us to assess their learning performance across days, as well as continuously measure brain activity. We find that the overall neural excursion from manifold activity in both cognitive and sensorimotor brain networks is associated with differences in subjects’ patterns of learning and relearning across days. These findings suggest that off-manifold activity provides an index of the relative engagement of different neural systems during learning, and that subject differences in patterns of learning and relearning are related to reconfiguration processes occurring in cognitive and sensorimotor networks.