Determinants of the Z�llner illusion

The Ponzo illusion and the perception of orientation

“…Clearly, quantitative comparisons about the magnitude of the effect of distance are difficult to make, because a small gap between the context and the test line in the fovea may have a larger effect than a large gap in the periphery. Note that increasing the distance between the context and the test portions of the stimulus reduces the magnitude of both the Zöllner (Oyama, 1975;Wallace, 1969) and the Ponzo (Fisher, 1968b) illusions.…”

“…Gibson noted that the tilt inductioneffect is greater when the context lines are near vertical than when they were near horizontal. Similarly, both the Zöllner and the Ponzo illusions are greater in the vertical than in the horizontal orientation (Fisher, 1968b(Fisher, , 1973Oyama, 1975). We do not know the cause of this difference, but it may be related to other anisotrophies in the visual field (Prinzmetal & Gettleman, 1993).…”

“…For example, Gibson (1937) showed that the tilt induction effect was strongest when the contest line was from 15º to 20º from vertical or horizontal (see also Andriessen & Bouma, 1976;Kramer, 1978;Shimamura & Prinzmetal, 2000). Similarly, the Zöllner illusion is greatest when the angle between the test and the context lines is in this same range (e.g., Oyama, 1975;Wallace & Moulden, 1973). Note that the Ponzo illusion is greatest when each of the oblique lines is 15º-20º from vertical or horizontal (Fisher, 1968b(Fisher, , 1973.…”

Determinants of the Bourdon effect

“…The purpose of Experiment 1 was to investigate the effect of the angle (0') of the figure relative to the vertical and the size of the internal apical angles (~) on the apparent collinearity of the edges AB and Be, shown in Figure 1. It was expected that, like other illusions of direction, the effect would be greater for oblique orientations of the figure than inducing and test lines of about 5 deg and, according to Oyama (1975), is almost negligible at 10 deg. However, the data from Experiments 4 and 5 below have led to a modification of this view.…”

“…The Zollner illusion and its numerous derivatives are cases in point. However, there are some notable exceptions in figures in which the angle between inducing and induced contours is about 5 deg (Adam, 1964;Oyama, 1975;Wallace & Crampin, 1969). Under this condition, the test or induced line is apparently displaced toward the inducing line.…”