Long Term Potentiation and Depression

Another form of modification is long term depression and potentiation.
This is often modelled as a permanent change of synaptic weight.
Spike Timed Dependent Plasticity (STDP) has some biological support and there is a lot of modelling of it.
This is Hebbian learning, as proposed in 1949.
If neuron A tends to cause neuron B to fire, the weight between them is increased.
I use a compensatory model so that the weight leaving (or entering) a neuron is forced toward a target weight.
There is also work with neurotransmitters here (e.g. GABA).
Another type of learning is Structural Plasticity.
Here new neurons or synapses grow or die.
Note the real problem here. If all psychological memory is based on these types of plasticity, they need to account for all of that learning. The models are just starting to scratch the surface of what happens because it is really quite complex.