University of Sussex
Browse
underlying-infotropism.pdf (210.61 kB)

Infotropism as the underlying principle of perceptual organization

Download (210.61 kB)
journal contribution
posted on 2023-06-08, 21:29 authored by Chris ThorntonChris Thornton
Whether perceptual organization favors the simplest or most likely interpretation of a distal stimulus has long been debated. An unbridgeable gulf has seemed to separate these, the Gestalt and Helmholtzian viewpoints. But in recent decades, the proposal that likelihood and simplicity are two sides of the same coin has been gaining ground, to the extent that their equivalence is now widely assumed. What then arises is a desire to know whether the two principles can be reduced to one. Applying Occam's Razor in this way is particularly desirable given that, as things stand, an account referencing one principle alone cannot be completely satisfactory. The present paper argues that unification of the two principles is possible, and that it can be achieved in terms of an incremental notion of `information seeking' (infotropism). Perceptual processing that is infotropic can be shown to target both simplicity and likelihood. The ability to see perceptual organization as governed by either objective can then be explained in terms of it being an infotropic process. Infotropism can be identified as the principle which underlies, and thus generalizes the principles of likelihood and simplicity.

History

Publication status

  • Published

File Version

  • Accepted version

Journal

Journal of Mathematical Psychology

ISSN

0022-2496

Publisher

Elsevier

Volume

61

Page range

38-44

Department affiliated with

  • Informatics Publications

Full text available

  • Yes

Peer reviewed?

  • Yes

Legacy Posted Date

2015-07-07

First Open Access (FOA) Date

2016-03-22

First Compliant Deposit (FCD) Date

2017-03-19

Usage metrics

    University of Sussex (Publications)

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC