The Perceptual Experience of Variability in Line Orientation Is Greatly Exaggerated

What is the perceptual experience of variability? Unconscious perceptual processes are well-calibrated to variability, as are unconscious motor processes, whereas cognitive processes are not well-calibrated and tend to underestimate variability. Regarding the perceptual experience of variability, perceivers are sensitive to differences in the variability of ensembles of objects, but any potential biases have not yet been explored. In the current experiments, participants viewed a set of lines at various orientations that were presented 1 at a time in a random order. Participants judged whether the orientations within each set were more similar to each other or more disperse. Although participants were sensitive to differences in spread, participants overestimated the variability of the set by 50%. The results have implications for mechanisms underlying ensemble perception, which is the extraction of summary statistics from a set of objects. In particular, there are both shared and unique processes related to perceiving similarities across objects (such as the mean orientation) and perceiving differences (such as their spread). Both visual abilities were thorough and used the full set of lines, rather than using only a subset, but the perception of spread relied more heavily on differences presented at the beginning whereas perception of the mean relied more heavily on features of the lines at the end of the animation. The results also have implications for visualizations of uncertainty, such as hurricane forecasts. A perceptual bias to overestimate variability could help counteract cognitive biases to underestimate variability. Public Significance Statement The perceptual system is sensitive to variability in a group of objects but is also highly biased to overestimate variability. This perceptual bias has implications for how people will understand visualizations and could also have implications for diagnosing cancer. If people or doctors overestimate variability, this could affect decisions such as when to evacuate in anticipation of severe weather or whether or not a radiograph contains cancer.