Binary Cell Assemblies

Cell assemblies (CAs) are the neural basis of human (and other mammalian) concepts.
That is your concept for *dog* is represented by a bunch of neurons.
When you think of dog, those neurons fire at an elevated rate, and that is a short term, and a working memory item.
It's hard to represent these in simulated neurons, but it's quite simple to represent a binary version of CAs.
We do it with a range of spiking models including in NEST and SpiNNaker with a couple of versions of LIF neurons, and in our own Fatiguing LIF model in java.
In these cases, what happens is the neurons are caused to fire.
As the neurons in the CA have sufficient internal connectivity, they continue to fire.
The firing last indefinitely, which is a problem for models of CAs, but
it's great for writing programs in neurons.
It's a solid basis for software engineering.