More on the Geometricity of Visual Form

Any attempt to assess the effects of the configurational properties of visual targets on various types of perceptual tasks presupposes che ability to isolate the relevant form parameters and co measure them. For a long time so-called geometric forms were the stimuli most used in form perception experiments. Even after demonstrations (e.g., Attneave & Arnoult, 1956) that findings of form perception experiments based on heterogeneous stimulus samples (and the usual samples of geometric forms are certainly heterogeneous) cannot be very well generalized, geometric forms continue to be used without either defining or measuring them. In a series of experiments by Zusne and Michels (1962) in which foursided figures were used, a tentative definition of "geometric form" was formulated and the relative roles of three form parameters in the judgment of geometricity of these figures were established. Since five-sided figures not only contain more information (i.e., have more degrees of freedom to vary) but also exhibit only one type of bilateral symmetry in contrast to the three types of such symmetry shown by four-sided polygons, further study seemed indicated. Therefore, using a similar methodology, this experiment was conducted to establish how closely the results obtained with five-sided figures would agree with those obtained with four-sided figures. Forty-three five-sided polygons, representing the entire population of such polygons, were constructed by systematically distorting the regular pencagon until all possible discorrions were represented by at least one polygon. The stimuli were cut out from black construction paper and pasted on white cardboard discs 5 in. in diameter. Measurements were taken on the parameters of compactness (C) , elongation ( E ) , and areal symmetry (S), as previously defined (Zusne & Michels, 1962), using the longest axis as base line in computing E and S. Forty-nine students enrolled in an introductory psychology course served as Ss. Ss were asked to rate the figures on their geometricity by sorting the stimulus cards into five compartments of a sorting bin. They were told to use whatever concept of geometricity they had at the time. The results were as follows. ( a ) When the 43 figures were arranged from most geometric to least geometric according to their rated geometricity (Fig. 1 ), a pattern emerged which was similar to that obtained with four-sided polygons.