Date created: | Last Updated:

Category: Project

Description: Magnitude information from different dimensions (e.g., space and time) interact with each other in perception, but how they occur remains unclear. In five experiments, the present study investigated whether these interactions arise from cross-dimensional encoding or memory interference. In Experiment 1, participants perceived a duration and two concurrent distances, were cued which distance to later reproduce, and reproduced the duration and then the cued distance. Reproduced duration increased as a function of the cued distance (replicated in Experiment 3), suggesting the duration memory can be biased by the memory of the cued distance. Experiment 2 showed the memory interference disappeared when the to-be-reproduced distance was cued when the duration memory had been retrieved (i.e. cued at reproduction). Experiment 4 demonstrated a similar memory interference for the time-on-space effect (replicated in Experiment 5). Experiments 3 and 5 additionally showed that cross-dimensional memory interference was modulated by memory noise: noisier unfilled distance, compared to less noisy filled distance, biased duration to a greater extent but was itself biased to a lesser extent. These findings suggest that the direction and extent of cross-dimensional memory interference depend on the relative memory noise of the target interfering dimensions. We proposed a Bayesian account whereby the inference (e.g., reproduction) of a magnitude is determined by integrating our prior experience that magnitudes co-vary across dimensions (e.g., space and time) and the encoded memory of the magnitude dimensions, each with a certain level of noise. We discussed the implications for cross-dimensional magnitude interactions in general.