Neural Networks Trained on Natural Scenes Exhibit Gestalt Closure Journal Article uri icon



  • AbstractThe Gestalt laws of perceptual organization, which describe how visual elements in an image are grouped and interpreted, have traditionally been thought of as innate. Given past research showing that these laws have ecological validity, we investigate whether deep learning methods infer Gestalt laws from the statistics of natural scenes. We examine the law of closure, which asserts that human visual perception tends to “close the gap” by assembling elements that can jointly be interpreted as a complete figure or object. We demonstrate that a state-of-the-art convolutional neural network, trained to classify natural images, exhibits closure on synthetic displays of edge fragments, as assessed by similarity of internal representations. This finding provides further support for the hypothesis that the human perceptual system is even more elegant than the Gestaltists imagined: a single law—adaptation to the statistical structure of the environment—might suffice as fundamental.

publication date

  • September 1, 2021

has restriction

  • hybrid

Date in CU Experts

  • April 18, 2021 9:32 AM

Full Author List

  • Kim B; Reif E; Wattenberg M; Bengio S; Mozer MC

author count

  • 5

Other Profiles

International Standard Serial Number (ISSN)

  • 2522-0861

Electronic International Standard Serial Number (EISSN)

  • 2522-087X

Additional Document Info

start page

  • 251

end page

  • 263


  • 4


  • 3