After a stone is thrown, the circles on water vanish, but the liquid retains microscopic oscillations. The slightest gust, and ripples appear again. Neural networks behave similarly. When electrical signals quiet down, the memory of them remains in the contact points — synapses. Like water after a storm, these traces harbor hidden energy. Scientists showed that a single snapshot of connection states can predict with near-perfect accuracy whether a new wave of activity will arise. Most surprising: even after complete silence, without any external prodding, the network can spontaneously trigger a repeated burst. In the entropic chaos of residual connections, an ordered impulse sometimes emerges, giving rise to a new flash of activity, like a sunbeam on the crest of a wave. This logic traces back to the ideas of John von Neumann about self-reproducing automata. Thus, the mechanism of hidden memory explains how the brain retains impressions when neurons are silent.
The tiny nematode worm (1 mm long) also has hidden memory: it holds onto the smell of food for several seconds after the source disappears.
As in "Eternal Sunshine of the Spotless Mind": even erased memories leave hidden traces capable of reviving feelings.