Neural networks are thus a kind of analogy engine
Context:
“One of the basic principles which drives learning in neural networks is similarity. Similar inputs tend to yield similar outputs. Thus, if a network has learned to classify a pattern, say 11110000, in a certain way then it will tend to classify a novel pattern, e.g., 11110001, in a like fashion. Neural networks are thus a kind of analogy engine. The principle of similarity is what lets networks generalize their behaviors beyond the cases they have encountered during training.”
Rethinking Innateness: A Connectionist Perspective on Development
Jeffrey L. Elman, Elizabeth A. Bates, Mark H. Johnson, Annette Karmiloff-Smith, Domenico Parisi, Kim Plunkett
Page 59