THE NEURAL BASIS OF COGNITIVE CONTROL OF MOVEMENT INHIBITION
Embargo until
Date
2016-05-06
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Johns Hopkins University
Abstract
A crucial component of executive control of behavior is the ability to voluntarily inhibit an action when it has become inappropriate or dangerous in a given situation. A driver cruising through several green lights on a street for a while would have to respond quickly and correctly by stopping his/her car when a red traffic light turns on. Effective inhibitory control is also context-dependent. In the same example, red is only associated with a stop signal in a driving situation. Numerous studies have suggested that the right ventrolateral prefrontal cortex (rVLPFC) is responsible for executive control of movement inhibition, but several questions remain to be answered, how does the rVLPFC dynamically and flexibly exert its control over movement inhibition to accommodate a changing environment? And how does the information flow from rVLPFC to the rest of the brain during stop control? The current thesis research aims to examine these questions using human psychophysical and functional magnetic resonance neuroimaging (fMRI) methods in conjunction with eye tracking and advanced statistical pattern classification and machine learning techniques.
We first identified two sub-regions within rVLPFC that showed differential activations during a context-dependent stop signal task (SST) in Experiment 1. Activation in the dorsal part of the rVLPFC was associated with detecting and encoding the meaning of the signals that were behaviorally relevant, whereas the activation in the ventral part of the rVLPFC was associated with the requirement to withhold an eye movement, regardless of the stop outcome. Furthermore, we found that the frontal eye field (FEF), an ocuolomotor area controlling for eye movement in the frontal cortex, encoded information regarding movements but not the meaning of the signals or the contexts.
We next explored the behavioral variability on Stop and Continue trials in selective stopping, a by-product of the context-dependent SST in Experiment 2, by asking whether the Continue signal-induced delay in motor response is the same form of response inhibition that was induced by the Stop signal. We found evidence in favor of a two-stage stopping process. Specifically, we found that participants slowed down their reaction time in Continue trials despite the instruction to ignore the Continue signal when preparing for a speeded go response. There was no detectable difference in the time it took to respond to a Stop signal compared with responding to a Continue signal (first stage of stopping). However, merely delaying the planned response did not change behavior in the subsequent trial. Either successful stop trials, or negative feedback in failed stop trials (second stage of stopping), resulted in greater engagement of inhibitory processes in the subsequent trial. Although there was no direct measure of brain activity during the selective stopping experiment, we speculate that the dorsal part of the rVLPFC may be the neural basis of pausing in order to focus attention on the relevant stimulus and make a decision about the course of action (first stage), and then if the signal requires stopping, ventral part of rVLPFC and FEF may be recruited to complete stopping.
In Experiment 3, we further validated human fMRI findings by recording single neuron activity in one macaque monkey and comparing the results from rVLPFC (area 45) versus FEF while the monkey performed a context-based SST. We found that FEF neurons encode for movement-related activities, but not the meaning of the signal. On the other hand, neurons in area 45, a primate homologue of human rVLFPC, encoded context information by either selectively increasing their firing rate to a particular context or by increasing their firing rate at the time of context switching.
Finally, we examined the functional connectivity of ventral and dorsal parts of rVLPFC with FEF and the rest of the brain with additional analyses using the data collected in Experiment 1. We found that dorsal and ventral rVLPFC modulated stop-related activity for the rest of the brain differentially, with the former influencing an attention-related network and the latter influencing a motor-related network. Our results indicated that functional separation exists in the rVLPFC during executive control of movement inhibition, such that subregions within this area form distinct distributed networks with the rest of the brain to collaboratively instantiate cognitive control.
From a reverse engineering perspective, understanding the biological and computational processes required for the brain to control for complicated situations such as context-dependent movement inhibition will support advancement of artificial intelligence (e.g. the development of self-driving cars). On the other hand, understanding the contribution of rVLPFC to inhibitory control is also highly significant from a clinical point of view, because it might lead to improvements in assessing or even treating disorders such as attention deficit and hyperactivity disorder (ADHD), Parkinson’s disease, and obsessive–compulsive disorder (OCD).
Description
Keywords
Cognitive control, frontal lobe, fMRI, eye tracking, electrophysiology, primate, stop signal task, executive control, eye movement, saccade, inhibition, countermanding