How learning can guide evolution

Hinton and Nowlan are cited alongside Baldwin in dalla2026between to delimit how lifetime learning interacts with our static optimality results: learned encodings shape evolutionary trajectories on the same objective landscape rather than displacing the convergence analysis.

Hinton and Nowlan's simulation showing learning can guide evolution gives us the computational template for the two-ecology mechanism in riva2026task, where developmental optimisation under one ecology is steered by population selection under another.