Relative change probability affects the decision process of detecting multiple feature changes

Change in a real world setting usually consists of multiple feature changes, and one feature maychange more often than another. This study manipulated the number of changes in orientation and spatial frequency to examine the effect of the relative change probability on the comparison and decision processes in a change detection task. Results showed that when two types of features changed equally often (Experiment 1), parallel self-terminating processing was adopted. When frequency changed more often than orientation (Experiment 2), serial self-terminating processing was adopted. However, when there was a lack of cognitive resources (Experiment 3), parallel self-terminating processing was adopted even when relative saliency existed. In all three experiments, the process capacity ranged from limited to unlimited. In addition, the participants reported that they were not aware of their use of change probability information, thus, providing evidence of implicit learning of change probability. These results support the relative saliency hypothesis: relative change probability affects the relative salience between features and, consequently, influences the adoption of decision strategies. Furthermore, the results highlight that perceptual comparison and decision processes are flexible and can vary depending on the relative saliency and resources available for information processing.