The Simulation Model
- This model has been used for a range of categorisation experiments.
- It is based on neurons that are fatiguing leaky integrators.
- There are discrete cycles (synchronous update).
- Connectivity is distance biased, and there are inhibitory
and excitatory neurons.
- Learning is Hebbian and occurs after each cycle.
- A compensatory learning algorithm is used so that neurons have a target
total synaptic strength.
- Neurons can spontaneously activate; there is biological evidence
for this.
