Transcript Document
Cognition, Brain and Consciousness: An Introduction to Cognitive Neuroscience Edited by Bernard J. Baars and Nicole M. Gage 2007 Academic Press Chapter 3 Neurons and Their Connections “There is no more important quest in the whole of science probably than the attempt to understand those very particular events in evolution by which brains worked out that special trick that enabled them to add to the scheme of things: color, sound, pain, pleasure, and all the facets of mental experience.” Roger Sperry, 1976 Cognition, Brain and Consciousness: An Introduction to Cognitive Neuroscience Edited by Bernard J. Baars and Nicole M. Gage 2007 Academic Press Chapter Outline 1.0 Introduction 2.0 Working assumptions 3.0 Arrays and maps 4.0 How neural arrays adapt and learn 5.0 Coordinating neural nets 6.0 Summary Cognition, Brain and Consciousness: An Introduction to Cognitive Neuroscience Edited by Bernard J. Baars and Nicole M. Gage 2007 Academic Press 1.0 Introduction Real and idealized neurons A single bipolar neuron Cortical neurons may have 10,000 dendrites (input fibers) and 1 or more axons (output fibers) An idealized neuron A simplified neuron with dendrites shown on top and axon at the bottom Cognition, Brain and Consciousness: An Introduction to Cognitive Neuroscience Edited by Bernard J. Baars and Nicole M. Gage 2007 Academic Press 1.0 Introduction Excitation and inhibition Neurons are connected through synapses which can be excitatory or inhibitory The probability that the next neuron will fire a spike will be increased if it an excitatory connection or decreased if it is an inhibitory connection Cognition, Brain and Consciousness: An Introduction to Cognitive Neuroscience Edited by Bernard J. Baars and Nicole M. Gage 2007 Academic Press 1.0 Introduction Neural computation Simplified -- idealized -- neurons are used in artificial neural nets (ANN) to model many brain functions. ANNs are artificial but they have provided understanding of ways neural computation might work in the brain. Cognition, Brain and Consciousness: An Introduction to Cognitive Neuroscience Edited by Bernard J. Baars and Nicole M. Gage 2007 Academic Press 2.0 Working Assumptions Starting simple: receptors, pathways, and circuits Six working assumptions: 1. 2. 3. 4. 5. 6. Neurons work using an integrate-and-fire action Connections are either excitatory or inhibitory Idealized neurons are used in artificial neural nets to model brain function Neurons typically form two-way pathways, providing the basis for reentrant connectivity The nervous system is formed into arrays or maps of neurons Hebbian cell assemblies underlie the change from transient to stable, lasting connections Cognition, Brain and Consciousness: An Introduction to Cognitive Neuroscience Edited by Bernard J. Baars and Nicole M. Gage 2007 Academic Press 3.0 Arrays and Maps Maps flow into other maps: The nervous system often uses layers of neurons in giant arrays. Cognition, Brain and Consciousness: An Introduction to Cognitive Neuroscience Edited by Bernard J. Baars and Nicole M. Gage 2007 Academic Press 3.0 Arrays and Maps Neuronal arrays usually have two-way connections Cognition, Brain and Consciousness: An Introduction to Cognitive Neuroscience Edited by Bernard J. Baars and Nicole M. Gage 2007 Academic Press 3.0 Arrays and Maps Sensory and motor systems work together Cognition, Brain and Consciousness: An Introduction to Cognitive Neuroscience Edited by Bernard J. Baars and Nicole M. Gage 2007 Academic Press 3.0 Arrays and Maps Temporal codes: spiking patterns and brain rhythms Neurons have different spiking codes. These two electrode traces show the voltage of simulated neurons with differing spiking codes Cognition, Brain and Consciousness: An Introduction to Cognitive Neuroscience Edited by Bernard J. Baars and Nicole M. Gage 2007 Academic Press 3.0 Arrays and Maps Choice-points in the flow of information Ambiguous figures such as the face-vase illusion (a) and Necker cube (b) pose points at which the brain must make a decision or choice about how to perceive and interpret sensory input. Cognition, Brain and Consciousness: An Introduction to Cognitive Neuroscience Edited by Bernard J. Baars and Nicole M. Gage 2007 Academic Press 4.0 How Neural Arrays Adapt and Learn Hebbian learning: ‘Neurons that fire together, wire together’ Donald Hebb was one of the most influential theorists for cognitive science and neuroscience. He clarified the notion of the cell assembly and proposed the bestknown learning rule for neural networks Cognition, Brain and Consciousness: An Introduction to Cognitive Neuroscience Edited by Bernard J. Baars and Nicole M. Gage 2007 Academic Press 4.0 How Neural Arrays Adapt and Learn Neural Darwinism: survival of the fittest cells and synapses An example of Neural Darwinism in learning: stages of encoding a neural activation pattern until dynamic synaptic activity allows permanent connections to be strengthened, allow memories to be stored Cognition, Brain and Consciousness: An Introduction to Cognitive Neuroscience Edited by Bernard J. Baars and Nicole M. Gage 2007 Academic Press 4.0 How Neural Arrays Adapt and Learn Symbolic processing and neural nets A network which represents a large set of propositions such as ‘a robin is a bird’ and ‘a rose has petals’. Cognition, Brain and Consciousness: An Introduction to Cognitive Neuroscience Edited by Bernard J. Baars and Nicole M. Gage 2007 Academic Press 5.0 Coordinating Neural Nets An activation map of visual areas active while a subject is watching a movie. Note the correlation of neural activity in the left hemisphere (top of figure,marked with ‘l’) and the right hemisphere, (bottom of figure, marked with ‘r’) across differing visual areas such as V3 and V4. Cognition, Brain and Consciousness: An Introduction to Cognitive Neuroscience Edited by Bernard J. Baars and Nicole M. Gage 2007 Academic Press 6.0 Summary A basic question in cognitive neuroscience is how nerve cells combine to perform complex cognitive functions such as perception, memory, and action. While neurons form the basic building block of cognition, we are still unfolding how they work both as individual cells and in synchrony in large scale arrays. Some working assumptions about how neurons work -- such as the integrate-andfire neuron, two-way pathways, cell assemblies and artificial neural nets -- have allowed scientists to begin to model the complex and dynamic activity in the brain that underlies human cognition.