See more speakers [icon type=”glyphicon glyphicon-chevron-right” color=”#b50024″]
[column lg=”4″ md=”12″ sm=”12″ xs=”12″ ]
[column lg=”8″ md=”12″ sm=”12″ xs=”12″ ]
Janelia Farm Research Campus, HHMI
Theoretical and Computational Neuroscience
Title of talk:
Components of the neural circuitry for an internal model
The quest of Neuroscience is to explain how cognition arises from the microscopic circuitry within the brain. Of the endless mental feats we engage in, predictions are amongst the most fundamental and the least noticed. They are so pervasive that they underlie not only highly complex decisions, but even mundane actions, like lifting a glass. Motor control in particular is riddled with prediction and planning because fine control of the body is complex. To implement predictive control requires internal models, representations from which we can infer the future state of something from its past states. Decades of work have provided evidence supporting the existence of predictions and internal models underlying behavior. These models are also a pillar of modern control systems. Anthony Leonardo’s work is focused on understanding the neuronal structure and function of such predictive internal models and how they are integrated with reactive feedback to generate complex and robust behaviors. At present his lab studies this in the context of prey capture in dragonflies. A broad range of tools are employed, ranging across outdoor recordings, feedback-controlled behavioral environments, measurements of kinematics and neurons in freely moving animals, neuroanatomy, conventional intracellular and extracellular electrophysiology, and electron microscopy. These different levels of description, which span microscopic to macroscopic phenomena, are linked together with systems level quantitative models.