Relevance ratings and salience categorizations for objects in a set of 80 pictures

Abstract

In this study, we provide normative data for objects in a set of 80 digital color pictures (e.g., nature scenes, human activities, cartoon characters, magazine covers). In Experiment 1, four objects in each picture were rated by 48 observers on a 6-point Likert scale for their relevance to the overall meaning of the scene. In Experiment 2, Salience Toolbox software (Walther & Koch, 2006) provided additional information about whether the four relevance-rated objects were located in areas that were high or low in visual salience. Brief descriptions of the four objects, their locations in the picture, their categorizations as high or low in salience, the means and standard deviations of their relevance ratings, and statistical analyses specifying which pairs of objects in a picture differed significantly on their relevance to the meaning of the scene are given in the Appendix. An example is provided of how the pictures could be used to create stimuli for a change blindness task in which detection of item onset versus offset is contrasted for low-relevance and high-relevance features. The 80 pictures are accessible as jpg files from the first author's Web site at http://marcellm.people.cofc.edu/research.htm. © 2010 The Psychonomic Society, Inc.