There are about 100 billion (100 000 million) neurons in the human brain, they form about 100 trillion (100 000 billion) connections between them called synapses. We use neurons and their synapses to think with, and to store information. These synapses require electricity and neurotransmitters like acetylcholine .
There are many neurons with many connections, and they each take in chemical signals from other neurons. At some threshold this causes an electrical signal to be shot along the neuron (down its axon). This causes a chemical signal to be sent to other neurons. So neurons are like the transistors in a computer: they just switch on or stay off. This causes decisions to be made. Basically: “Yes” or “No” in answer to whatever question.
I believe that the best model we currently have of how the brain works is based on hierarchical hidden Markov models .
This is where a cluster of about 100 neurons interact together to realise something.
Then this cluster sends its signals to a higher-level cluster of 100 neurons to make more conclusions. So for example there is a cluster that recognises the cross on a ‘t’. Then this message is sent, along with others of a similar level, to a higher level. This level 2 cluster of neurons says, “That’s a letter ‘t’.”.
A higher level cluster uses that to see that the word ‘table’ is present. Then another level cluster says, “A table is a place to rest a computer, or food, or books, or whatever.” If there’s a picture of a table there, then a cluster of neurons on a higher level says, “I like the style of that table.” etc.