Are they now? Do you have some sort of breakthrough explaining ANN encodings?

Currently the state of the art is some sort of ad hoc correlation between behavior and explanations.

The topological ANN explanation makes their thinking seem vastly different from what logical heuristics would be...