August 2014
Volume 14, Issue 10
Vision Sciences Society Annual Meeting Abstract  |   August 2014
Stimulus Features Contributing to Perceptual Organization of Complex Scenes
Author Affiliations
  • Beliz Hazan
    Psychology Program, Graduate Center of CUNY
  • Daniel D. Kurylo
    Psychology Program, Graduate Center of CUNY
  • Zeynel Baran
    Experimental Psychology Department, Hacettepe University, Turkey
  • Xuan Zhao
    Electrical and Computer Engineering Department, Polytechnic Institute of New York University
Journal of Vision August 2014, Vol.14, 803. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Beliz Hazan, Daniel D. Kurylo, Zeynel Baran, Xuan Zhao; Stimulus Features Contributing to Perceptual Organization of Complex Scenes. Journal of Vision 2014;14(10):803.

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

  • Supplements

For complex, natural scenes, which contain multiple sources of visual information, high-order visual cognition relies on accurate organization of stimulus components. Perceptual organization is based upon stimulus metrics as well as top-down factors, including contextual cues and familiarity. Gestalt principles, such as common luminance, color, and surface texture, or good continuation of contrast borders, allow segregation and integration of elements across broad areas and occluded regions. It was hypothesized that component visual features, specifically color, high- and low-spatial frequencies, and surface information, each contribute significantly to perceptual organization of natural scenes. To test this, 34 participants viewed 60 briefly presented scenes, selected from a standardized data set, and categorized scenes as either forest, mountain, ocean coast, houses, highway, or city skyline. Each image was presented as a series of 19 trials, beginning with a highly occluded image, and progressively providing a greater percentage of the image. Performance was indexed as the occlusion level at which correct categorization stabilized. Five image filter conditions were examined: (1) original (unfiltered), (2) color filter (grayscale), (3) high-pass and (4) low-pass spatial frequency filter, and (5) surface field filter. In addition, images were presented either upright (familiar) or inverted (unfamiliar) for each filter condition. Results indicated that for upright as well as inverted images, high-pass, low-pass, and surface field filters significantly reduced performance (ANOVA, p <.05), whereas color filter has a modest affect performance. Greatest impairment was found for the low-pass and surface field filter conditions. These results indicate that coarse information, which reduces detail, is less beneficial in perceiving scene organization. In addition, information from edges needs to contain sufficient detail to facilitate perceptual organization of complex scenes.

Meeting abstract presented at VSS 2014


This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.