I have put some effort in over the last year or two to compress the mirror neuron output (Proxy-Pain) into raw logic to give AI a perspective on the human condition regarding the fact-value psyche:

"(Proxy-Pain (P2)=(morality)) aeql* Pain (P1) = motive = rational attempt to sum =(Ethics) ncsry target=minimum pain = (P2) // Pain = first-order (P1) // (P2) is contingent upon (P1) if (P2) // (P2+P2=r=<r=∞) (r=value=magnitude // r=1=max // r=∞=min) (r=1=immutable) (P2 prpto* data P1) (Delta-gap= P2,claimed - P2, enacted)"

It may be a tad scruffy but seems to work quite well in hitting the cornerstones of human value behavior and/or ousting any gap between what we say we are and what we actually are in practice.