I think this is a fair paraphrase of Eric Kandel’s reasoning:

  1. Following Alois Riegl, Art History should focus more on the psychology of aesthetic response.
  2. Renaissance Art is “inner-directed”, whereas the Dutch Masters and later works enlist the viewer’s perspective and perceptual participation.
  3. Ernst Kris and maybe also Ernst Gombrich ground this psychological aesthetic participation in evidence of perceptual illusions.
  4. Authors of great works of art exploit these devices to create ambiguous interpretations and engimatic emotional effects.
  5. Through solving these visual puzzles, the viewer performs a diminished version of the same creative act in visual interpretation that the artist performs dramatically in visual production.
  6. This view of the creative synthesis and interpretation of ambiguous perceptual data can help us understand cognition.

This account seems to rely on a theory that visual phenomena can be encoded by the artist and reconstructed as different but to some extent equivalent mental images in the mind of the viewer, and that the artist that paints and the viewer that views are doing the same or an equivalent activity, but to different degrees. This makes it a kind of information theory of artistic communication, and subject to the same kinds of empirical criticisms.