PhD defence E.M. (Elias) Fernandez Santoro

Below is a brief summary of the dissertation: 

The cerebellum helps us learn and refine movements, from simple actions like reaching to more complex skills such as playing the violin. Although its structure is highly organized, many details about how it supports stable and flexible learning are still unknown. This thesis investigates how different parts of the cerebellum work together to keep learning both accurate and stable over time.
A central focus is the communication between Purkinje cells (the main processing units of the cerebellar cortex) and the inferior olive, a brainstem structure that sends timing signals known as “complex spikes.” These signals help the cerebellum decide when to strengthen or weaken connections, allowing it to correct errors and adapt to new conditions.
Using computational models, this thesis explores how two types of cerebellar modules, called “upbound” and “downbound,” learn in different ways. These modules have distinct electrical properties and favor different forms of synaptic plasticity. The results show that these differences allow the cerebellum to learn multiple things at once, each at its own pace and with its own rules.
The thesis also introduces a new idea: that a single cerebellar module can combine both open-loop and closed-loop control. This means part of the circuit can learn from feedback while another part can monitor internal changes without affecting the output.
Overall, the work shows how the cerebellum balances flexibility and stability during learning. This modular organization may help explain how the cerebellum supports not only movement but also higher cognitive functions.dt tijdens het leren, een basisprincipe dat mogelijk ook bij hogere cognitieve functies een rol speelt.