Purchase this article with an account.
Arni Kristjansson, Gianluca Campana, Andrey Chetverikov; Representing color ensembles: Mapping internal probability density functions with attentional priming. Journal of Vision 2017;17(10):1085. doi: https://doi.org/10.1167/17.10.1085.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
The natural environment is rich with colors. Objects contain a multitude of hues that depend on texture and shape, the positions of other objects and light sources. Yet, little is known about how human observers represent such color distributions. Previous research shows that observers can estimate means and variance of feature distributions, but whether shapes of distributions can be represented is less well known. We introduce a new method for studying representations of color ensembles based on intertrial learning in visual search. Observers looked for the oddly colored diamond among multicolored diamonds taken from either uniform or Gaussian color distributions. Color space was corrected for inequalities in average sensitivity to different colors, so that adjacent hues were separated by 1 average JND (Witzel and Gegenfurtner, 2013, 2015). Within "streaks" of 3-4 trials, distractor colors were randomly drawn from either a Gaussian or uniform distribution. On test trials the targets had various distances in color space from the mean of the preceding distractor color distribution allowing us to estimate the shape of representations of distractor sets. Targets therefore served as probes into representations of distractor colors. We analyzed response times on test trials as a function of differences between target color and the preceding distribution. Our results show that after only 3-4 exposures to a particular color distribution, observers obtained a detailed representation of that distribution revealing a striking similarity between the physical distribution of colors and their internal representations. The results demonstrate that the visual system represents color ensembles in a more detailed way than previously thought, coding not only mean and variance but most surprisingly, the actual shape of the distribution of colors in the environment.
Meeting abstract presented at VSS 2017
This PDF is available to Subscribers Only