-
Notifications
You must be signed in to change notification settings - Fork 40
ConceptNeuronCompleteness
Neuroscientists have made great strides in modeling the complexity of neural function and several detailed neural modeling packages are available which simulate neurons near the molecular level. There are two problems inherent in this approach: First, a detailed computer model of a neuron contains numerous parameters and is necessarily slower, limiting the number of neurons which can be simulated at any given time. Second, with a focus on the detail of individual neurons, the overall functionality of the brain might be ignored. Similarly, a focus on an individual heart muscle cell sheds very little light on the overall function of the heart.
The Brain Simulator takes more of a “top-down” approach. Let’s start with the simplest possible neuron models and see what can be produced with them. It’s this approach that leads to the ability to simulate a billion neurons on a desktop computer and demonstrate how even larger networks can be simulated across multiple servers on a LAN.
- Brain Simulator II: The Guide, Chapter 2, p.31