Though this is the prevailing view in Neuroscience, in truth there is very little information on the actual computation the brain in doing biologically.
The majority of Neuroscience literature in the CNS focuses on sensory systems signal processing and on the low levels of those pipelines( vision, sound, and taste). There has been recent work in olfaction, but there exist inherent problems with state space and encoding (also olfactory neurons seem to project all over- causing even more experimental problems) to consider (one of the reasons for the attractiveness of vision, sound and taste encoding).
All higher order processing really is a black box. This is one of the primary arguments of pursuing a connectomics based approach. Some cool work in doing targeted connectomics is being done at HHMI and the Allen Institute (both with fluorescent and em microscopy). Where the latter focuses more on fly because of the resolution
The majority of Neuroscience literature in the CNS focuses on sensory systems signal processing and on the low levels of those pipelines( vision, sound, and taste). There has been recent work in olfaction, but there exist inherent problems with state space and encoding (also olfactory neurons seem to project all over- causing even more experimental problems) to consider (one of the reasons for the attractiveness of vision, sound and taste encoding).
All higher order processing really is a black box. This is one of the primary arguments of pursuing a connectomics based approach. Some cool work in doing targeted connectomics is being done at HHMI and the Allen Institute (both with fluorescent and em microscopy). Where the latter focuses more on fly because of the resolution
https://www.janelia.org/project-team/mouselight https://www.janelia.org/project-team/flylight https://www.janelia.org/project-team/flyem https://www.alleninstitute.org/
So in short, we don't know if NN are like the brain or not.