Computations In Neural Systems: From Single Cell Membrane to Large Neural Networks

Psychology 506A - (This class has been archived)

- General Information.

• This course is intended to provide a basis for understanding how brains acquire, assimilate, store and retrieve information and how they compute adaptive responses to external inputs. Understanding these processes requires a basic working knowledge of both the theoretical principles and biological mechanisms underlying neural signaling, knowledge representation and data storage. The course begins with a review of the fundamentals of membrane physiology & neuronal excitability, synaptic transmission & synaptic plasticity, and the integrative properties of dendrites. This is followed by a discussion of associative (‘Hebbian’) learning in networks of neurons and on the mechanisms by which networks can be trained to produce an adaptive response. We then consider how information is ‘coded’ in the activities of single neurons and in the collective activity patterns of large networks of neurons. Finally, we explore how some specific models have been used to make sense of computation in the brain. Topics include the spatial representation in the hippocampus, consolidation and re-consolidation of memories, the transformation of neuronal representations and other examples of ‘higher’ cognitive functions.
  
• Syllabus and grading policies
• Intructor: Jean-Marc Fellous.
  Office Hours: Fridays after class, or by appointments, LSN 384.

- Readings and lecture summaries.

- Slides and other material (protected, need password).

-Class1: Introduction – Making Sense of Neural Computation through Modeling: overview of basic phenomenological computational models of single neurons, learning rules, perceptron and classifiers.

-Class2: Channels and Excitability (neuronal computations): Review of fundamental neural signaling mechanisms; membranes, channels, ionic equilibria and neuronal excitation.

-Class3: Synaptic Transmission (synaptic computations: Part I): basic synaptic physiology. Transmitter release, excitation/inhibition in post-synaptic cell. Electrical signaling.

-Class4: Dendrites and axons (dendritic computations): morphology, physiology and integrative properties.

-Class5: Synaptic Bases of Memory (synaptic computations: Part II): Modification of synaptic communication through experience; short and long-term changes; associative vs. non-associative synaptic plasticity.

-Class6: Neural Circuits (network computations): A survey of some typical and some atypical synaptic networks in the brain.

-Class7: Computing with noise: Neurophysiology of Sleep, noise, background activity, reliability and precision of spiking.

-Class8: Midterm (all material including Class7)

-Class8: Computing in the context of memory: Associative Memory in Neural Networks - Basic principles of associative memory; pattern completion; constraints on storage capacity; sequence storage and retrieval.

-Class9: Consolidation, Reconsolidation, the Hippocampus, and the Cortex: Memory consolidation in cortical hierarchies and the hippocampus. (L. Nadel)

-Class10: Computing in the context of spatial navigation: Spatial Representation in the Hippocampus - Place fields, grid fields and spatial navigation, including recent advances in understanding the origins of spatial information.

-Class11: NO CLASS (Holliday)

-Class12: Computing at the systems level: Higher brain Functions - Introduction to the fMRI technique, a discussion of some 'higher' brain functions; planning, decision making, memory, emotion, social interactions and diseases such as Alzheimer's.

-Class13: NO CLASS (Holliday, again :-(...)

-Class14: Representation in the Brain: Philosophical look at representation. Early AI and the computer metaphor, connectionism, ecological approach to perception, dynamical systems, and embodied cognition. (L. Nadel)

-Class15: Final Exam.