Müller-Lyer illusion and perception of numerosity.

Illusion of numerosity can be observed in many of the classical illusions of linear extent by replacing the uninterrupted lines with rows of dots. Using the method of constant stimuli both length and numerosity illusions move in the same direction, whereas using a magnitude-estimation method the two illusions move in opposite directions. Two experiments show that this inversion occurs also in the Müller-Lyer illusion.

Organizational versus geometric factors in mental rotation and folding tasks.

The criteria used in performing mental rotation or mental folding tasks were studied with a paradigm that did not involve reaction times. The hypothesis was that, when perceptual-organizational factors come into conflict with the geometric features required for the correct execution of such tasks, it is the former that prevail. To verify this hypothesis two experiments were carried out. In experiment 1, subjects were asked to imagine quadrilaterals rotating round a rotation axis at different inclinations. Their responses were dependent both on the degree of tilt of the rotation axis and on the degree of tilt of the quadrilateral with respect to the rotation axis. Experiment 2 consisted of the mental execution of a folding task. In this case too, the responses depended on the degree of tilt of the folding axis and also on the complexity of the stimulus outline. In both experiments responses were divided into two groups: (i) geometrically correct responses and (ii) responses which, although incorrect, were based on perceptual-organizational criteria. In the light of the results, some theoretical implications regarding transformation operations executed by means of mental images are discussed.

Regularity, exposure time and perception of numerosity.

Two experiments are described. In Exp. 1, subjects compared the apparent numerosity of two kinds of dot patterns, regular vs irregular, with two different exposures (160 vs 2000 msec.). In Exp. 2, the subjects had to estimate the numerosity of the same patterns, presented one at a time. Analysis showed a relative overestimation of the regular patterns in Exp. 1 but not in Exp. 2. In general an overestimation occurred with an increase in exposure. Such results support our hypothesis of two separate processes as the basis of the two kinds of performance (estimating vs comparison).