[Memory psychophysics for visual size estimation of geometric patterns]

In previous studies (e.g., Kerst & Howard, 1978) it has been shown that remembered sizes were related to actual sizes by a more compressive power function than they were to perceived sizes. These findings have been accounted for by the re-perceptual hypothesis. Two experiments were conducted to test this hypothesis for the size estimation with a geometric stimulus. Thirty undergraduate students participated in each experiment. The effects of the number of dimensions of perceived geometric objects (Experiment 1) and those of stimulus range (Experiment 2) on remembering magnitude were investigated. The major findings from Experiments 1 and 2 were as follows; (1) the power exponents were significantly smaller in the memory conditions than in the perceptual conditions, and (2) the results with 1 dimensional objects (lines, Experiment 1) and with 2 dimensional objects in the larger stimulus range (squares, Experiment 2) were consistent with the re-perceptual hypothesis. Those results were discussed in the relation to the stimulus complexity.