Research
Using the cerebellar circuit to probe neural learning rules
Neural circuits perform computations that enable us to perceive the world and interact with it. How do the different types of neurons and synapses within a circuit influence the computations it performs and its ability to learn from experience? To address this question, we are combining our expertise in analyzing neural circuits using electrophysiological, behavioral, and computational approaches with powerful molecular-genetic tools for precisely manipulating specific neurons or synapses in vivo.
We focus on the cerebellum, a brain region with a simple circuit architecture that makes systematic analysis of its function experimentally and analytically tractable. A major function of the cerebellum is motor learning, the process by which movements become smooth and accurate through practice. Our laboratory is analyzing the neural learning rules in the cerebellar circuit that implement this type of learning.
We have developed a battery of oculomotor learning paradigms in mice for studying the neural mechanisms supporting motor skill learning, including learned changes in the amplitude and timing of movements, the generalization of learning, and factors influencing the persistence of memory. By determining exactly which aspects of cerebellum-dependent learning are altered and which are spared by any given manipulation of the cerebellar circuit, we can dissect cerebellum-dependent learning into its component parts and map those components onto specific signaling elements in the cerebellar circuit.
Neural circuits perform computations that enable us to perceive the world and interact with it. How do the different types of neurons and synapses within a circuit influence the computations it performs and its ability to learn from experience? To address this question, we are combining our expertise in analyzing neural circuits using electrophysiological, behavioral, and computational approaches with powerful molecular-genetic tools for precisely manipulating specific neurons or synapses in vivo.
We focus on the cerebellum, a brain region with a simple circuit architecture that makes systematic analysis of its function experimentally and analytically tractable. A major function of the cerebellum is motor learning, the process by which movements become smooth and accurate through practice. Our laboratory is analyzing the neural learning rules in the cerebellar circuit that implement this type of learning.
We have developed a battery of oculomotor learning paradigms in mice for studying the neural mechanisms supporting motor skill learning, including learned changes in the amplitude and timing of movements, the generalization of learning, and factors influencing the persistence of memory. By determining exactly which aspects of cerebellum-dependent learning are altered and which are spared by any given manipulation of the cerebellar circuit, we can dissect cerebellum-dependent learning into its component parts and map those components onto specific signaling elements in the cerebellar circuit.
Neural error signals controlling learning
In the classic model of cerebellar learning, climbing fibers carry the key instructive signals controlling the induction of learning. However, there is evidence that other signals in the cerebellum can also instruct learning. Indeed, different components of cerebellar learning are mediated by distinct neural processes (see, for example, Kimpo et al, 2005 and 2007. Rhea Kimpo’s research is elucidating those components of learning controlled by instructive signals carried by climbing fibers. To do this, she uses electrophysiological recordings of single neurons in mice undergoing learning as well as optogenetic techniques to manipulate the activity of the climbing fibers during learning.
Linking cellular properties with circuit function
Our understanding of the rules that determine which synapses change during learning is still very limited. Since most of the synapses in a circuit are capable of plasticity, a given neural circuit has the potential to be modified in many different ways. What controls the recruitment of plasticity at one set of synapses versus another during a given learning experience? This overarching question is the focus of Aparna Suvrathan’s research. Using single-cell patch-clamp recordings from cerebellar slice preparations, Dr. Suvrathan is analyzing how synaptic learning rules control the distribution of plasticity within a circuit in vivo in response to different types of training.
Dual role of cerebellar Purkinje cells in motor performance and learning
The brain constantly learns on the job, carrying out familiar tasks while simultaneously implementing modifications to refine its responses to various inputs. Hannah Payne’s research is focused on how Purkinje cells, the sole output cells of the cerebellum, play dual roles instructing both immediate motor output and long-term motor learning. To do this, she has adopted a circuit-level approach, using computational modeling as well as optogenetic and pharmacological tools, to understand how Purkinje cells shape movements on different time scales.
Cellular-level model simulations related to synaptic plasticity
It is possible to build sophisticated computational models to study the brain on different levels. Tiina Manninen’s research focuses on both deterministic and stochastic simulations related to synaptic plasticity at the cellular level. She is studying how different Purkinje cell stimuli affect the activation of a cerebellar nuclei neuron.
Other Interests
Contribution of cerebellar interneurons to learning
Interneurons shape local computations in neural circuits, hence an understanding of their function is critical to understanding the neural basis of behavior. Grace Zhao’s research focuses on the role of cerebellar interneurons in learning and memory. Dr. Zhao is generating and applying genetic tools for imaging and manipulating neurons in a temporal-, spatial- and cell type-specific manner. By combining these genetic tools with in vivo physiology and behavioral assays, she is investigating the contribution of the feedback and feedforward inhibition provided by cerebellar interneurons to the induction, expression, and persistence of learning and memory.
Cellular and circuit-level factors influencing the capacity for learning
Synaptic plasticity is widely believed to be the mechanism of learning, yet it is often difficult to predict the effect of impaired or enhanced synaptic plasticity on circuit function and behavior. Barbara Nguyen’s research is bridging this gap, by analyzing motor learning in mice after a range of genetic, pharmacological, and optogenetic manipulations of cerebellar neurons. She is especially interested in how the properties of a synaptic plasticity mechanism interact with the recent history of activity in the circuit to determine the capacity for additional learning. In addition, she is analyzing the role of Purkinje cell activity in the induction of learning.
How are memories transformed over time?
How are transient signals converted into long-lasting representations in the brain? Olivia Winter’s research is investigating how the representation of a learned motor program develops over time. Using a range of techniques, she is analyzing both structural and functional changes in the cerebellum during the formation, consolidation, and retention of different types of motor skills.