Effects of visual feature distributions on intertrial priming

Abstract

Recent studies have revealed the importance of measuring people’s ability to encode the visual field into highly informative distributions of features, rather than solely relying on independent summary values to explain perception. These studies have mainly focused on our ability to learn and use the feature distributions of distractors to guide our vision. Straying from this path, Kristjánsson, Driver and Hansmann-Roth (2019) found a weak effect for learning the probability distribution for target features. The aim of the current study was to strengthen this effect by using the same paradigm, apart from colour blocking the target probability distributions intended to make the colours more distinct and the added inverted Gaussian distribution meant to provide further support for the target probability learning. 35 undergraduates in the University of Iceland took part in the experiment. Their task was to indicate the location of the diamond with the most dissimilar colour out of three. There was a total of three sessions, each defined by a different probability distribution for the target colours: Uniform, Gaussian and inverted Gaussian. The results gave partial support for our ability to learning the probability of targets by displaying a greater difference between the Gaussian- and uniform distributions than in previous results. The observers were, however, unable to fully learn the inverted Gaussian distribution possibly due to its bimodal qualities.