Principlev1
Capture every instance of surprise (prediction-expectation
Capture every instance of surprise (prediction-expectation mismatch) immediately, regardless of magnitude, because small surprises reveal systematic model blind spots that large surprises obscure.
Why This Is a Principle
This derives from Brain as Hierarchical Prediction Machine (brain as prediction machine that propagates prediction errors) and Learning occurs when outcomes differ from predictions, (learning occurs when outcomes differ from predictions). The principle prescribes capturing all prediction errors because they are the learning signal, with small errors being more diagnostic than large ones.