A neurocomputational theory of action regulation predicts motor behavior in neurotypical individuals and patients with Parkinson's disease

Abstract

Surviving in an uncertain environment requires not only the ability to select the best action, but also the flexibility to withhold inappropriate actions when the environmental conditions change. Although selecting and withholding actions have been extensively studied in both human and animals, there is still lack of consensus on the mechanism underlying these action regulation functions, and more importantly, how they inter-relate. A critical gap impeding progress is the lack of a computational theory that will integrate the mechanisms of action regulation into a unified framework. The current study aims to advance our understanding by developing a neurodynamical computational theory that models the mechanism of action regulation that involves suppressing responses, and predicts how disruption of this mechanism can lead to motor deficits in Parkinson's disease (PD) patients. We tested the model predictions in neurotypical individuals and PD patients in three behavioral tasks that involve free action selection between two opposed directions, action selection in the presence of conflicting information and abandoning an ongoing action when a stop signal is presented. Our results and theory suggest an integrated mechanism of action regulation that affects both action initiation and inhibition. When this mechanism is disrupted, motor behavior is affected, leading to longer reaction times and higher error rates in action inhibition.

Fig 1. Experimental setup for action regulation tasks that require action inhibition.

A: Decision-making task, including instructed and choice trials. B: An arrow version of the Eriksen flanker task, including congruent (the flanker arrows point to the same direction as the central arrow) and incongruent (the flanker arrows point to the opposite direction from the central arrow) trials. C: A stop signal task with instructed trials. Individuals are prompted to stop the action when the arrows turn red.

Fig 2. Behavioral findings from the decision-making task, the Eriksen flanker task and the stop signal task.

A: Bar plots of the mean RT for neurotypical individuals (Neurotypical) and PD patients (PD) across all participants in each group, in the instructed and choice trials of the decision-making task. B: Bar plots of the mean RT for neurotypical individuals (Neurotypical) and PD patients (PD) across all participants in each group, in the congruent and incongruent trials of the Eriksen flanker task. C: Bar plots of the mean RT for neurotypical individuals (Neurotypical) and PD patients (PD) across all participants in each group, in the go trials of the stop signal task. Error bars correspond to standard error (SE) across all participants in each group. White dots represent mean RT for each participant.

Fig 3. Probability to successfully stop an action as a function of the SSD.

The probability to successfully stop an action as a function of the SSD for neurotypical individuals (Neurotypical, blue) and PD patients (PD, red).

Fig 4. Model architecture.

The architectural organization of the neurodynamical theory to model tasks that involve action inhibition, such as decisions between competing options, decisions in the presence of conflicting information and outright stopping of actions.

Fig 5. Simulated reach planning field neuronal activity changes in the decision making task, Eriksen flanker task and stop signal task.

A-C: Activity changes of the 181 neurons in the reach planning field during the decision making task (instructed trial and choice trial)(A), the Eriksen flanker task (incongruent trial and congruent trial)(B), and the stop signal task (go trial and stop trial)(C). D-F: Activity changes of single neurons in the reach planning field during the decision making task(D), the Eriksen flanker task(E), and the stop signal task(F).

Fig 6. Simulated pause field activity changes during the three tasks.

A: Activity changes of single neuron in the pause field during the decision making-task. Cyan trace, simulated pause field activity during an instructed trial for a neurotypical individual. Magenta trace, simulated pause field activity during an instructed trial for a PD patient. Blue trace, simulated pause field activity during a choice trial for a neurotypical individual. Red trace, simulated pause field activity during a choice trial for a PD patient. B: Activity changes of single neuron in the pause field during the Eriksen flanker task. Cyan trace, simulated pause field activity during a congruent trial for a neurotypical individual. Magenta trace, simulated pause field activity during a congruent trial for a PD patient. Blue trace, simulated pause field activity during an incongruent trial for a neurotypical individual. Red trace, simulated pause field activity during an incongruent trial for a PD patient. C: Activity changes of single neuron in the pause field during the stop signal task. Blue trace, simulated pause field activity during a stop trial for a neurotypical individual. Red trace, simulated pause field activity during a stop trial for a PD patient.

Fig 7. Simulated reaction times for the decision-making tasks, the Eriksen flanker task and the stop signal task.

A: Bar plots of the simulated RT for neurotypical individuals (Neurotypical) and PD patients (PD) in the instructed and choice trials of the decision-making task. B: Bar plots of the simulated RT for neurotypical individuals (Neurotypical) and PD patients (PD) in the congruent and incongruent trials of the Eriksen flanker task. C: Bar plots of the simulated RT for neurotypical individuals (Neurotypical) and PD patients (PD) in the go trials of the stop signal task. Error bars correspond to SE.

Fig 8. Simulated probability to successfully stop an action as a function of the SSD.

The (simulated) probability to successfully stop an action as a function of the SSD for neurotypical individuals (Neurotypical, blue) and PD patients (PD, red).