Dissertation Defense - Alexandra Bova
“Rodent Skilled Reaching: A Model for Investigating Neural Mechanisms of Dexterous Motor Control”
Dr. Daniel Leventhal, Chair
Dexterous motor skills are essential for normal, everyday functioning but are impaired in many neurological disorders (e.g., Parkinson Disease, dystonia) or other central nervous system pathologies (e.g., stroke, tumor). Therefore, understanding the neural mechanisms that underlie dexterous skills is essential for developing effective therapies for these conditions, as well as for advancing fundamental models of motor control. The rodent skilled reaching task is commonly used to study dexterous motor skill learning and performance, as rodent reach-to-grasp movements are strikingly similar to those of humans. Until recently, implementing the skilled reaching task required significant time and effort, and analyses were limited to simple outcome measures (e.g., success rate) or required time-intensive and subjective manual analysis of skilled reaching videos. These limitations prevented detailed analysis of how neural circuits regulate dexterous skill learning and performance.
I used a novel automated skilled reaching task combined with machine learning methods for markerless position tracking to extract detailed forelimb and digit kinematics as rats learned skilled reaching. I found that improvements in fine digit kinematics evolved over a longer timescale than improvements in gross forelimb kinematics. However, the improvements in fine digit kinematics continued even after success rate stabilized, suggesting that aspects of motor control that do not optimize task outcome are refined in skilled reaching. I also found that even after rats had achieved stable success rates, significant variability in reach-to-grasp movements still remained.
Brain dopamine is essential for normal motor function as evidenced by its role in Parkinson Disease and other movement disorders. To determine the role of dopamine in skilled reaching performance, I optogenetically stimulated or inhibited substantia nigra pars compacta dopamine neurons as rats performed reach-to-grasp movements. I found that gross forelimb kinematics and coordination between gross and fine movements gradually changed with repeated manipulations. However, once aberrant kinematics were established, rats transitioned rapidly between baseline and aberrant kinematics in a dopamine-dependent manner. Together, these results suggest history-dependent effects of precisely-timed dopamine signals on dexterous motor skill performance, distinct from simple reinforcement learning or ‘vigor’ models of dopamine.
This work establishes a paradigm for dissecting the neural circuitry that underlies dexterous motor skills, and highlights the importance of robust quantifiable measurements of movement kinematics. The characterization of how fine and gross motor control evolve as skilled reaching is learned provides a foundation for interpreting studies applying neural circuit manipulations during skilled reaching. Finally, our findings of the effects of dopamine manipulations on skilled reaching kinematics generated new hypotheses regarding dopaminergic control of coordinated movements and the pathophysiology of parkinsonism.