There's a saying in neurophysiology that predates neural nets: "Neurons that fire together, wire together". It's not exactly the same as back-propagation, but it's quite close.
In fact, it's the exact opposite of back propagation. Long/Short Term Potentiation is all about the synaptic buton of the firing cell, not the receiving cell (or that mess of the astrocytes).
Why so you say that? Or knowledge of the brain is still very very dim and rudimentary, and ANNs are simple mathematical models (inspired by brain neural networks, to be sure) to fit parameters to data. It's a bit naive, in my view, to assert such a simplistic answer to a very hard question without conclusive evidence for anything.