Before being able to analyze the data in these spike trains however, we need to figure out how we plan to represent them.įor this example, we will consider data collected by Robert Cat from a neuron located in an alien’s posterior inferior temporal cortex responding to a rapidly-changing color stimulus. Although we’ve reduced the response of our neuron down to a train of 1s and 0s, we have the information we need to analyze what caused the original neuron to fire. This means that when neurons communicate with each other, the peak voltage never changes - all the information must be conveyed in the rate or timing of the action potentials. This time-dependent sequence of 1s and 0s is referred to as a spike train, depicted in graph Figure 6.3b.Īnother characteristic of action potentials is that they are stereotyped: they are consistent with each other in voltage. With this in mind, a voltage response curve can be reduced to time-bins that either contain a spike (represented in a binary fashion by a 1) or do not (represented by a 0). This small increment of time is called a time bin.
The length of time over which the neuron is recorded is called the integration window: we will first break up this window into lengths of time small enough that they can only contain one spike. When the voltage surpasses the threshold, a spike occurs. To do this, we need to remember that the axes of both graph A and B represent voltage over time. This characteristic allows us to represent spikes in a binary fashion. How should we analyze the information encoded in these action potentials?Īs we’ve previously discussed, action potentials are an all-or-none event. Graph A shows the recorded stimulus and graph B shows the recorded actions potentials during the stimulus.Īssume that we measured a neuron firing in response to a random sensory stimulus (shown in Figure 6.3a), and we recorded its voltage changes and displayed the signal in an oscilloscope. 12.6 Chapter 6: Reverse Correlation and Receptive Field Mappingįigure 6.3: Example of a spike train.12.3 Chapter 3: Passive Membrane Models.12.2 Chapter 2: Introduction to Computational Neuroscience.11.7 Chapter 7: Reverse Correlation and Receptive Field Mapping.
11.4 Chapter 4: Passive Membrane Models.11.3 Chapter 3: What is Computational Neuroscience?.7 Reverse Correlation and Receptive Field Mapping.4.7.4 The flow of ions and equilibrium potentials.4.7.3 Main features of an action potential:.4.7.2 An action potential has six phases:.4.4.2 Negative feedback and repolarization.4.4.1 Positive feedback and depolarization.4.3 Membrane potentials and electrochemical gradients.3.6 The future of computational neuroscience.3.5 Applications of computational neuroscience.3.3.1 Can we make models that understand?.3.3 What is computational neuroscience?.2.13 Coding Exercises for Learning Python.2.12 Conceptual Exercises for Learning Python.1.4 This book creates a public record of learning that exists after the semester ends.
0 kommentar(er)
