Purchase this article with an account.
Jun Saiki; Motion severely reduces capacity and life of object visual working memory. Journal of Vision 2002;2(7):294. doi: 10.1167/2.7.294.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Visual working memory is said to hold a set of four coherent representations of objects (object files). However, their spatiotemoral characteristics are largely unknown. Using a paradigm called multiple object-permanence tracking (MOPT), which measures memory for feature-location bindings, this study revealed that ‘life’ and capacity of object files critically depends on objects' motion. Observers were asked to detect any color switch in the middle of regular rotation of a pattern of multiple colored disks behind a windmill-shaped occluder. Experiment 1 examined the effects of disks' motion by manipulating the angular velocity of the disks using relative motion of the pattern and occluder, keeping the exposure and occlusion durations fixed. The color switch detection performance dramatically declined with the rotation speed. Experiment 2 evaluated the life of object files, by examining the effect of occlusion duration. The estimated life (.75 threshold) of four object files decreased from 1158 ms with the stationary pattern, to 56 ms with 126 deg/s angular velocity (about 1/3 r.p.s.). Experiment 3 evaluated the capacity using MOPT with target cueing used in multiple object tracking. Target disks selected from a set of 6 were cued by flash, and observers were asked to track only the targets. The estimated capacity declined from 5.1 disks in the stationary condition, to 2.1 disks in the 126 deg/s condition. Velocity and the number of targets affect the performance independently, which is inconsistent with the view that processing cost takes effects only above the capacity limit. The data could be accounted for by a probability summation model, where the number of targets and velocity independently affect the threshold. The characteristics of visual working memory dramatically change with spatiotemporal factors, and cognitive predictability is of little use. The MOPT paradigm is a useful tool to quantitatively evaluate such dynamics. Supported by JMESC grants (11610075, 13610084) and JSPS-RFTF99P01401.
This PDF is available to Subscribers Only