The Neuroscience of Skill

August 31, 2021

Written by: Nitsan Goldstein

As the Olympics end and the Paralympics begin, many of us are watching a lot of sports. If you’re like me, you’ve marveled at the skill required to compete at such a high level in any sport. Every athlete will tell you, though, that it took years and years of practice to make it to the Olympics. Gymnasts perform the same skill hundreds of times, basketball players take thousands of free throws, and skaters endlessly practice the same tricks until it’s time to compete. When athletes, or any of us, practice a motor skill like shooting a basketball, changes in our brain improve our performance over time and many repetitions. What happens in the brain that can explain this change? 

            One of the most important regions for motor learning is, not surprisingly, the motor cortex. The motor cortex is an area near the surface of the brain that controls movement. Since small-scale changes in human brains are extremely difficult to observe, neuroscience often turn to rodents like mice and rats to study changes in the motor cortex during learning. There are a variety of tasks that rodents can learn through repetition including walking on a rotating cylinder, grasping at an object, or moving a lever. Recent advancements in imaging technology have allowed scientists to observe changes in the motor cortex as a rodent’s performance improves in these tasks.

            The size and shape (or, morphology) of neurons changes as an animal learns a motor task. Specifically, motor learning involves a period of dendritic spine formation followed by a period of spine elimination1. Dendritic spines are protrusions from a neuron’s dendrite, or the part of the cell that receives signals from other neurons. Communication between neurons occurs at a synapse, and spines allow for the formation of more synapses, and thus more communication between sets of neurons. It makes sense then that as a skill is learned, more connections will be made among neurons that directly control the movements necessary to perform that skill. What’s less intuitive, however, is the elimination phase of learning. As an animal is reaching peak performance, the number of dendritic spines returns pre-learning levels even though performance on the task remains high2. By tracking individual spines over time, scientists discovered that spines that are formed while learning a task are maintained, while some of those that existed before learning are removed2. These constant cycles of growth and elimination ensure that only the most important and highly used neural connections survive, and those that are less frequently active are pruned away. If this process is disrupted, for example by inhibiting protein synthesis, learning is severely impaired3.  

            Mirroring the changes in individual neurons, larger-scale shifts in motor cortex function have also been observed. By imaging many neurons simultaneously, researchers can track how activity in a population of neurons in the motor cortex correlates with increased performance in a learning task. Interestingly, multiple different firing patterns can occur when an animal performs the same movement. However, the pattern of activity in motor cortex neurons more consistently predicts motor function as learning progresses4. So, in a way, the variety of different patterns of activity that are observed early in learning is like the formation of new spines. This can be thought of as the “exploration” phase where new connections and patterns of communication are established. Then, as repetition refines a motor task, spines that are not utilized are cleared away and certain patterns of activity between neurons that are less effective will no longer occur. This is the “consolidation” phase of learning, and it is equally important for learning a new skill. 

            These concepts that developed from work in rodents have been observed on a more global scale in humans with functional magnetic resonance imaging (fMRI). Using this technique, researchers have observed both increases and decreases in neural activity, potentially corresponding to areas where exploration and consolidation are occurring5. The activity in the motor cortex more easily predicts when the participants make a trained motor pattern, suggesting that, like in rodents, specialized circuits may be forming that represent the most efficient and reliable method of achieving a certain outcome. Research like this may help us all develop into better athletes, but it is also crucial for stroke survivors, people with paralysis that use prosthetic limbs, and many others. It is essential that we continue to widen our understanding of how the brain learns and refines its networks with practice, practice, practice.   

References

  1. Peters, A.J., Liu, H., & Komiyama, T. Learning in the rodent motor cortex. Ann. Rev. Neurosci. 40:77-97 (2017).
  • Xu, T., Yu, X., Perlik, A.J., Tobin, W.F., Zweig, J.A., Tennant, K., et al. Rapid formation and selective stabilization of synapses for enduring motor memories. Nature 462, 915-919 (2009).
  • Luft A.R., Buitrago M.M., Ringer T., Dichgans J., & Schulz J.B. Motor skill learning depends on protein synthesis in motor cortex after training. J. Neurosci. 24, 6515–20 (2004). 
  • Peters, A.J., Chen, S.X., & Komiyama, T. Emergence of reproducible spatiotemporal activity during motor learning. Nature 510, 263-267 (2014).
  • Weistler, T. & Diedrichson, J. Skill learning strengthens cortical representations of motor sequences. Elife 2:e00801 (2013).

Cover image by Ania Klara from Pixabay 

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Website Powered by WordPress.com.

Up ↑

%d bloggers like this: