Cognition and Action In The Cortico-Cerebellar System

Lead Research Organisation: Royal Holloway University of London
Department Name: Psychology

Abstract

Skill learning is the process of acquiring the ability to execute a task not only more efficiently, but also more automatically so that it is not prone to disruption from other tasks being conducted at the same time. The ability to acquire skills must depend in some way on physical changes that take place during learning within the circuits of the brain, but little is understood about these processes. Previous work has suggested that during the learning of skilled movements these changes may take place within the circuits of the cerebellum. The integrity of the cerebellum is important for the learning of skilled movements because this ability is severely impaired in patients who have suffered damage to cerebellar circuits. Brain imaging methods enable researchers to find activity in the brain that is related to particular tasks executed during scanning (giving a 'snapshot' of activity related to specific tasks). These methods have been used to demonstrate that during the execution of skilled movement, activity of specific cerebellar areas is increased in healthy human volunteers. The cerebellum operates as part of a well-known circuit that it forms with areas of the cerebral cortex (the cortico-cerebellar system), and it has been suggested that an important neural process that underlies our ability to acquire skilled movements includes a transition from cortical control to cerebellar control. The purpose of this project is to use brain imaging methods to test the idea that the cortico-cerebellar system is engaged not only during the learning of skilled actions, but also during the development of mental skills (arithmetic skills in particular). In particular, it will test the idea that during motor skill learning activity in the motor cortex declines, but activity in interconnected cerebellar areas increases as these areas gradually 'take over' the role of the cortical areas. It will also test the idea that as mental skills become more efficient and increasingly immune to the disrupting influences of external events during scanning, activity in the prefrontal cortex (the cortical area most heavily implicated in higher functions) will decrease, while interconnected parts of the cerebellar cortex will increase. The detection and correction of errors is important for the process of learning. Experiments will therefore also test the idea that these cerebellar areas are sensitive to systematically introduced errors in the control of skilled movement and mental information processing. Newly developed methods have made it possible to use functional neuroimaging to study the ways in which brain areas communicate to achieve particular information processing goals. Studies in this project will examine the ways in which the communication between cortical and cerebellar areas gradually changes as normal healthy volunteers learn mental and motor skills in the scanner. Collectively, these experiments will provide a better understanding of how information processing in the human brain changes during the course of skill learning.

Technical Summary

The overarching goal of the proposed research is to use functional magnetic resonance imaging (fMRI) to investigate the temporal dynamics of activity in the human brain during the acquisition of motor and cognitive skills. Although previous work has shown learning-related activity changes in prefrontal cortex, motor cortex and cerebellum during motor learning, there is no coherent explanation of their involvement in motor skill learning that is consistent with the recently characterised anatomy of the primate cortico-cerebellar system. Recent evidence suggests that this system may also be involved in the acquisition of cognitive skills. I propose the specific involvement of two separate cortico-cerebellar networks in the acquisition of motor and cognitive skills. Experiments will test the long-standing idea that skill learning involves the transition of control from cortical areas during the early phases of learning, to cerebellar circuits in the later phases when the execution of skilled information processing becomes automatic and is less susceptible to additional demands. Specifically, I propose that during the acquisition of motor skills, the cortical motor areas that are initially engaged relinquish control to interconnected cerebellar areas (lobules HV and HVI). I also suggest that during the acquisition of cognitive skills, a similar transition occurs from prefrontal area 46 to its interconnected counterpart in the cerebellar cortex. Studies will additionally examine the responses of these areas to systematically introduced errors, and will also examine the temporal dynamics of information flow between cortical and cerebellar areas in each form of learning. Collectively, these experiments will give new insights into the way that anatomically well-defined cortico-cerebellar circuits become engaged in the acquisition of motor and cognitive skills in the human brain.

Publications

10 25 50
 
Description The aim of this work was to better understand the basis of skill learning in the human brain. The cortico-cerebellar system is known to play an important role in the acquisition of motor skills: Some parts of the cerebellum contain circuitry capable of storing information important for skilled movement. These exchange information with neocortical motor areas. However, other parts of the cerebellar cortex are known to contain similar circuits but are connected with the prefrontal cortex - an area responsible for decision-making and cognition. Do these areas play important roles in the acquisition and storage of cognitive skills? We tested for this possibility using brain imaging by scanning healthy individuals as they acquired cognitive skills under experimental control. We report three new findings:

1. During the acquisition of cognitive skills, excitability changes were present in cerebellar cortical Crus I and Crus II that connect with the prefrontal cortex (first grant objective). There were gradual activity decreases during learning, consistent with models of cerebellar plasticity. These were independent of any changes in error processing. This shows that activity in cerebellar circuitry changes in a manner that theoretical ideas would predict.

2. Rule errors activated cerebellar Crus I and Crus II (second objective) in over-trained subjects. Importantly, there were also subtle increases in reaction times in subsequent, error-free trials, evidencing error-induced instability of learned internal models. These post-error trials were accompanied by increases in activity in cerebellar Crus II, compared with pre-error trials. This shows that errors related to rules impact upon cerebellar circuitry.

3. The acquisition of cognitive skills was accompanied by increased information flow between the prefrontal cortex and cerebellar cortical CrusII (third objective). Connectivity gradually increased over the timecourse of learning. These effects were consistent across different experimental designs and statistical approaches. This shows that the communication between the prefrontal cortex and connected areas of the cerebellar cortex changes with skill learning, in ways that our theoretical ideas would predict.
Exploitation Route This work tests a set of theoretically motivated hypotheses about the ways in which the cortico-cerebellar system can contribute to the acquisition and storage of cognitive skills. We have localised areas in the cerebellar cortex which we think are important for fast, automatic execution of cognitive processes. This information may be important in the clinical domain for understanding the impact of cerebellar damage. It may also contribute to computing, engineering and robotics research because it tests ideas based on control theory.
Sectors Digital/Communication/Information Technologies (including Software),Electronics,Healthcare,Pharmaceuticals and Medical Biotechnology