Spencer Ivy, Aleksandra Mroczko-Wąsowicz (2025), Review of Philosophy and Psychology, 16(3), 969–996 — DOI-10.1007/s13164-024-00763-8
Key takeaways
- Perceptual learning can install “perceptual frames” — adaptations that work bottom-up and organize sensory input in a subject-specific way — so cognition can change which perceptual objects a perceiver can individuate without penetrating any single percept.
Seeded from the abstract and a skim, before the reading in NOTE-150. What follows is what the work says about itself.
The authors propose that object perception is shaped by “perceptual frames”: adaptations of the perceptual system, acquired through perceptual learning and domain training, that govern how perceptual objects are singled out from the sensory scene. Because frames work through bottom-up processes but differ between perceivers, two subjects given the same sensory input can perceive different objects. The paper supports the idea with empirical work on multisensory integration (frames help bind features), amodal completion (which needs frames whether or not mental imagery is involved) and predictive anticipation of motion. It concludes that framing effects are a form of perceptual diversity that theories of perception should accommodate.
Standing in the record
Filed from the 2026-09-26 sweep of the papers-feed tracker for philosophy: 0 seconds of active reading over 0 sessions. It was filed Deferred, unread. NOTE-150 is the close reading of 2026-09-26, and it placed the work: Active — worth reading as a clear, short statement of a “system-level versus episode-level” way through the cognitive-penetrability debate, provided the reader knows that its empirical sections interpret others’ results and do not test the frame concept against its rivals.
When seeded, before the reading in NOTE-150, the entry gave a priority for a deeper reading: low — never opened (t = 0). It is philosophy of perception with no direct bearing on ML practice, and the skim captures the thesis.
What the seeding asked a deeper reading to check (NOTE-150 is that reading):
- The frame idea is structurally a learned inductive bias. It is a possible bridge to how training shapes which “objects” learned perceptual models segment, though the paper does not discuss ML.
- Check the empirical studies in §3 (expertise, multisensory binding, amodal completion) to see how strongly they support a bottom-up realization rather than top-down attention.
- Check how the view relates to the cognitive penetrability debate (Pylyshyn, Firestone & Scholl) and to predictive processing, both of which the keywords name.
Access when seeded: link.springer.com returned a JS client challenge, and the PhilSci-Archive/PhilPapers preprints were behind a captcha or Cloudflare. The open-access (CC BY) publisher PDF came through rd.springer.com (28 pp). Read the abstract, keywords, §1, the heads and openings of §§2–4, and §5. Dates from the PDF: accepted 2024-11-20, online 2025-01-07.