Bayesian inference for variability discrimination on partial sameness

The comparative judgment about the partial sameness is widespread in practical visual search tasks. We explore the variability discrimination on the partial sameness by the means of experiments and theoretical models. In experiments, a set of bars with the aim orientation and random orientations are shown on a liquid crystal display (LCD), and the subjects judge the orientations of bars and record the number of bars with the same aim orientation. In comparative experiments, the three-digit integers replace the oriented bar as the new stimuli. By the visualized data of subjects, we find that the performances of subjects depend on the total number of bars (set size), the number of bars with the identical aim orientation and the standard deviation of the orientations' distribution. Based on Bayesian theory, we put forward a model in visual search to judge how many bars with the same orientation. The results show that Bayesian discrimination model on this kind of classified judgment matches the experimental data better than the alternative model, and it provides a normative and quantitative description on the judgment of the partial sameness. (C) 2019 Elsevier B.V. All rights reserved.