Levari et al. (2018) found that participants shift their perceived category of the color ”blue” when the frequency of ”blue” stimuli they were presented with changed. However, they neither took individual differences in color categories into account, nor did they relate their findings to existing theories of perceptual category shifting.
We performed several experiments to understand whether and how an individual’s category boundary shifts as a result of previously seen stimuli, and whether the shifts are also present in a similarity rating task. Additionally, different models were evaluated by comparing how well they predict data and how systematic their errors are. All stimuli were carefully selected and confounders closely monitored.
In two experiments we replicated that changing the frequency of ”blue” stimuli shifts color categorization. Time-course and effect size can be predicted using both the Range-Frequency (Parducci, 1965) and the Criterion Setting Theory Model (Treisman & Williams, 1984). In a similarity rating task without category judgments we used a triplet embedding method in two conditions: One had an equal number of stimuli from both categories, while the other had 90% from one category. By comparing the distances in the aligned embeddings, we can show that the perceived similarity also changes based on stimulus context.


Levari, D.E., Gilbert, D.T., Wilson, T.D., Sievers, B., Amodio, D.M., Wheatley, T.(2018). Prevalence-induced concept change in human judgment. Science,360(6396),1465–1467.

Parducci, A.(1965). Category judgment: a range-frequency model. Psychological review,72(6),407.

Treisman, M., Williams, T.C.(1984). A theory of criterion setting with an application to sequential dependencies. Psychological review,91(1),68.