Divyan Bavan
Introduction
One of the primary responsibilities of the brain is to dictate movement. This process is highly complex, involving many parts of the brain. Thus, it can be helpful to view it through the perspective of a specific example: picking up a ball. To initiate this process, the brain must first process sensory information—primarily visual—into understandable features. This is followed by the planning of movement. Finally, the motor cortex executes this plan; the result is the activation of skeletal muscle and grasping of the ball. Therefore, the entire process can be described as a sequential execution of three steps: perception, planning, and action. While this pathway is discussed in the context of picking up a ball, it can easily be abstracted to a variety of other tasks.
Integration of Sensory Information
To pick up a ball, it is necessary to process its traits: shape, size, and position. This starts in the primary visual cortex (V1). It is here where visual features—colour, orientation, and edges, among others—are extracted. More information is processed in the higher order cortices, such as V2 and V3. At this point, visual processing begins to specialize. This occurs through a split between processing for action versus identification. These are the dorsal and ventral streams, respectively. As expected, the dorsal stream is more relevant to this topic.
Several features of this stream enable it to process visual information for motor output. Besides visual primaries extracted from V1, other important information such as depth and motion are also computed in the dorsal stream. However, the most important area for action-specific processing is the posterior parietal cortex (PPC). The PPC is critical for integrating vestibular and proprioceptive input with visual information. This enables the encoding of object location with respect to the body. The PPC can also be split into specialized areas for specific tasks. For example, the anterior intraparietal area (AIP) aids in extracting information relevant to grasping. The ventral and medial intraparietal areas (VIP and MIP) are involved in the visual control of reaching.
Therefore, when the ball is seen by the subject, visual information is processed by V1, carried through the dorsal stream, and integrated with various forms of sensory input in the PPC. The result of this is information about the position of the ball with respect to the subject.
Planning Motor Output: The Premotor Cortex
Once sensory input has been integrated into the PPC, it travels to the premotor cortex (PMC). This area of the brain is critical for planning motor output. It is here where input from across the brain converges to decide what action is going to be executed. For example, the decision whether to pick up a ball or not is made by the prefrontal cortex (PFC). This area of the brain is split into two sections: the dorsolateral PFC and the ventromedial PFC. The former is involved in maintaining the goal of picking up the ball, whereas the latter decides whether you want to pick up the ball. Thus, this area of the brain is critical in determining the output of the PMC.
When these inputs converge in the PMC, plans are represented as actions as opposed to the activation of specific muscles. Furthermore, these actions are represented in parallel, even if they oppose each other. For example, the actions of reaching for the ball and kicking the ball are both represented in the PMC at the same time. This has been coined the affordance competition hypothesis. The decision for which action is selected ultimately resides in which one has stronger support. Stronger support increases lateral inhibition of other actions, further stabilizing the choice.
Once a decision is made, plans become more concrete. This leads to the recruitment of specific areas. Like the visual pathway, the PMC also splits into dorsal and ventral streams. The former is important for reaching, whereas the latter is important for grasping. Thus, it is no surprise that the VIP and MIP primarily provide input to the dorsal premotor cortex while the AIP gives input to the ventral premotor cortex. Furthermore, sequences of actions are also planned in the PMC, with help from the supplementary motor area (SMA). Finally, this input is sent to the primary motor cortex for execution.
The Primary Motor Cortex
M1 receives input from the premotor cortex in layers II and III. This causes a change in the population of active neurons, facilitating the activation of motor pathways. At a high level, the primary motor cortex (M1) is responsible for transforming abstract actions into specific muscle activation signals. Neurons within M1 are tuned to specific levels of motor features: force, direction, and posture. This enables the population to encode what movements need to occur for a specific action. These neurons map to multiple muscle-activating regions, coordinating movement.
The population coding of neurons was shown through a study done in monkeys (Georgopoulos et al., 1982). In the experiment, microelectrodes were placed near single neurons to record firing. The monkeys were then tasked with reaching for a target that was placed at different angles. When the scientists looked at the recordings, they found that neurons were maximally firing at specific angles. However, the neurons were still firing, albeit less frequently, at other angles. This meant that the direction in which the monkey was reaching could not be determined from a single neuron alone. However, once the firing rates of several neurons were factored into a single ‘vector’, the direction could be predicted. This shows that rather than single neurons determining the output for specific limbs, the entire motor cortex is involved in determining the action.
Since input is constantly changing, population coding also enables smooth transitions in motor output. This output is transmitted through layer V of M1, which is made up of pyramidal neurons. These neurons form the corticospinal tract of the spinal cord.
Motor Output and Adjustments
The corticospinal tract controls motor output to skeletal muscle. As axons extend from the pyramidal cells and form the tract, they synapse with interneurons in the spinal cord. These interneurons then synapse with α motor neurons, which innervate skeletal muscle fibres. By releasing acetylcholine at the neuromuscular junction, the motor neurons can activate contraction in the muscle fibres.
While this is happening, inputs are still being sent to the motor cortex. This provides a mechanism for feedback. For example, proprioceptive input from muscle spindles and Golgi tendon organs go through the dorsal column medial lemniscus (DCML) pathway, eventually reaching the primary somatosensory cortex (S1). S1 provides dense projections to M1, helping with error correction during an action. S1 also receives input from cutaneous receptors. This is essential for when the ball has been touched. Mechanoreceptors in the hand provide input to S1, which then provides information to M1 on adjusting hand position to better grip the ball. Finally, the cerebellum is also heavily involved in shaping motor output. This occurs through feedforward mechanisms; the cerebellum creates an internal model of the action and predicts what changes need to be made. This shapes the output alongside the other mechanisms.
Conclusion
The control of motor output is a complex pathway in the brain. For actions such as picking up a ball, movement is coordinated in three steps: sensation, planning, and execution. The first step takes inputs from the environment, through the sensory cortices, and internal decisions, through the PFC. Once an action has been decided on, it is planned out in the premotor cortex. Multiple plans are represented in parallel, with one being decided through its level of support. The SMA is also important at this step, as it plans out sequences of actions. This dense information is then sent to M1, where population coding is used to encode movement-related features. The result is signals being sent through the corticospinal tract and to skeletal muscle through α-motor neurons. Feedback and feedforward mechanisms modulate motor activity throughout this process, playing an essential role in error correction. By understanding this pathway, it is evident that picking up a ball, a simple task in the eyes of many, requires deeply complex mechanisms.
Works Cited
Edwards, Lauren L., et al. “Putting the “Sensory” into Sensorimotor Control: The Role of Sensorimotor Integration in Goal-Directed Hand Movements after Stroke.” Frontiers in Integrative Neuroscience, vol. 13, no. 16, 22 May 2019, https://doi.org/10.3389/fnint.2019.00016.
Georgopoulos, AP, et al. “On the Relations between the Direction of Two-Dimensional Arm Movements and Cell Discharge in Primate Motor Cortex.” The Journal of Neuroscience, vol. 2, no. 11, 1 Nov. 1982, pp. 1527–1537, https://doi.org/10.1523/jneurosci.02-11-01527.1982. Accessed 3 May 2021.
Lysea Haggie, et al. “Linking Cortex and Contraction—Integrating Models along the Corticomuscular Pathway.” Frontiers in Physiology, vol. 14, 10 May 2023, https://doi.org/10.3389/fphys.2023.1095260.
OpenAI. “ChatGPT.” ChatGPT, OpenAI, 2025, chatgpt.com/.