The study of motor cortex (dorsal premotor cortex and primary motor cortex) has been greatly aided by the development of a conceptual paradigm that has emerged over the past decade. In contrast to established frameworks, which view neural activity within motor cortex as a representation of particular movement parameters, the ‘dynamical systems paradigm’ posits that motor cortex is best understood via the low-dimensional neural processes that allow the generation of motor commands. This framework largely evolved from, and has been most successfully applied to, simple reaching tasks, where the sequential stages of movement generation are largely separated in time – motor cortex absorbs an input that specifies the identity of the upcoming reach, a second input initiates the movement, and strong, autonomous dynamics generate time-varying motor commands. However, while the dynamical systems paradigm has provided a useful scaffolding for interrogating motor cortex, our understanding of the mechanisms that generate movement is still evolving, and many questions remain unanswered.
Prior work has established that the neural processes within motor cortex that generate descending commands are initiated by a large, condition-invariant input. But are movements made under different behavioral contexts initiated via the same mechanisms? Lesion studies suggest that the generation of so-called ‘self-initiated movements’ is uniquely dependent on the supplementary motor area (SMA), a premotor region immediately upstream of motor cortex. In contrast, SMA is thought to be less critical for generating externally-cued movements. To characterize the degree to which SMA is able to impact movement initiation across behavioral contexts, we trained two monkeys to make reaches that were either internally or externally cued. On a subset of trials, we disrupted activity within SMA via microstimulation and asked how this perturbation impacted the monkeys’ behavior. Surprisingly, we found that the effect of stimulation was largely preserved across contexts; the behavioral effects of stimulation could be explained by a simple model in which a context-invariant, time-varying kernel multiplicatively altered the odds of movement initiation. These results suggest that SMA is able to impact movement initiation across behavioral contexts.
The question of how sequences of discrete actions are generated has been investigated for over one hundred years. It is commonly thought that once a given sequence (particularly a rapid sequence) becomes well-learned, individual actions that were once produced separately become ‘merged’, such that multiple actions are generated as a single, holistic unit. But what does it mean to generate multiple actions as a single unit? The dynamical systems paradigm offers the ability to translate this notion into specific predictions about the timing and structure of neural activity within motor cortex during sequence production. Importantly, it also offers predictions for the alternative hypothesis – that motor cortex generates the component actions of a sequence independently. To determine whether the production of rapid sequences requires motor cortex to merge multiple actions into a single ‘movement’, we trained monkeys to make sequences of two reaches. Surprisingly, we found that the same set of neural events are used to produce rapid sequences and isolated reaches. Rather than merging individual actions into a single unit, motor cortex generated rapid sequences by overlapping the neural activity related to reach preparation and execution. These results demonstrate that the performance of extremely fast, well-learned movement sequences does not require motor cortex to implement a sequence-specific strategy; the same neural motif that produces a simple reach can also generate movement sequences.
| Identifer | oai:union.ndltd.org:columbia.edu/oai:academiccommons.columbia.edu:10.7916/d8-p5ca-zv88 |
| Date | January 2021 |
| Creators | Zimnik, Andrew James |
| Source Sets | Columbia University |
| Language | English |
| Detected Language | English |
| Type | Theses |
Page generated in 0.0018 seconds