Epigenetic memory is seen as the most evolutionarily plausible way of learning from experiences and gaining instinctual knowledge accross generations, as established by the answer to one of my previous questions. However, is it known how epigenetic memories are translated from the DNA into neurons? Alternatively, I may be approaching this from the wrong direction and may need to read more papers, so links helping me clarify my question would be appreciated as well.
There are (at least) two ways epigenetic traits are inherited. The important background in both cases is gene expression: there is a misconception that genes are for this or that, where the reality is that most traits come from an overlap of several genes expressing themselves in different ratios. As a simple example, consider two varieties of bird of the same species, but one has a longer beak. One might conclude that the genes involved in the morphology of the beak were expressed longer or in greater quantity in the bird with the bigger beak. Epigenetics describes inherited expression schemes.
The well known method of epigenetic transfer is DNA methylation. The DNA itself is not actually changed, but the methylation sites act like tags, altering the expression rate of those particular genes. Another way is through histones, which are part of packing and unpacking DNA between storage and use. Alterations in histones can also alter gene expression.