The Effects of Color Contrast on the Zollner Illusion

The Zollner Illusion can be described as a visual illusion in which the perception of one set of parallel lines is distorted by the addition of another set of lines at a different orientation. The strength of the basic Zollner Illusion relies heavily on the angles at which the crossing lines hit the parallel lines, which causes a tilted appearance (Kitaoka & Ishihara, 2000). Some studies have been done resulting in the finding that luminance contrast between the lines and the grating of the background may also have an effect of the perceived illusion (Wallace, G. K., 1974). The aim of this study is to look into the effects of different color contrasts and their effects of the standard Zollner Illusion. A similar study using the Muller-Lyer Illusion found that color does, indeed, have an effect on illusion strength (Sadza & de Weert, 1984). Substituting colored short end lines for the original short end lines made the illusion less apparent, which supports the idea that the higher contrast makes for a stronger illusion. This study is an attempt to test the effects of using contrasting colors instead of the standard black lines in the Zollner Illusion.

Methods:
Participants in this study were 16 Alma College students, 10 female and 6 male. Each subject was exposed to a series of different stimuli, which were constructed by the Eye Lines software program and produced on a Visual Sensations monitor. The different stimuli were combinations of the following sets of lines: all black set, all orange set, all purple set, a combination orange/purple set (the slash marks and bottom line being orange and the top line beign purple), all yellow set, all blue set, a combination yellow/blue set (the slash marks and bottom line being blue and the top line being yellow), all red set, all green set, a combination red/green set (the slash marks and bottom line being red and the top line being green.) The different combinations were presented, in cycles of ten trials, twice to each subject, for a total of 20 presentations, and they adjusted the bottom line with the mouse until they made the lines appear parallel. The program will then identify the amount of error produced by the subjects for each of the different sets of stimuli. This will show whether ot not color has an effect on the contrast illusion and whether it is a positive or negative effect.

Results:
The results show that color does, in some cases, have an effect on the presence and strength of the Zollner Illusion. The following graphs were produced using the total participant's mean amount of error for each of the provided stimuli. The columns represent the different stimuli, and the units represent the average units of error.

Figure 1: Blue vs. Yellow Contrast

Figure 1 shows the differences in perceived illusion error produced by the all yellow, all blue, and combination yellow/blue stimuli, with the combination stimuli having the highest error.

Figure 2: Green vs. Red Contrast

Figure 2 shows the differences in perceived illusion error related to the all green, all red, and combination green/red stimuli, with the combination stimuli having the highest error.

Figure 3: Orange vs. Purple Contrast

Figure 3 shows the differences in perceived illusion error related to the all orange, all purple, and combination orange/purple stimuli, with the combination stimuli having the least amount of error.

Figure 4: Range of error in solid color contrast

Figure 4 shows the correlation between line color and amount of error for each stimuli, with the lightest color, yellow, on the low end and the standard black color on the high end.

Discussion:
These results show that differences in color can have an impact on the appearance and strength of the Zollner Illusion. In Figure A., it can be seen that the yellow stimuli had the least average effect, followed by the blue and then the combination stimuli. Likewise, in Figure B., the green stimuli had had the least average effect, followed by the red and then the combination stimuli. In the third figure, Figure C., the orange stimuli had the strongest average effect on error, and the purple had the same effect, whereas the combination stimuli produced the least average amount of error. This third graph lends much support to the idea that using two opposite colors in the stimuli provided the participant with a seperation of colors, making it easier to adjust the lines closer to the proper parallel form. However, the first two graphs provide little support, if not simply arguing against this point. It was found here that the bi-colored stimuli (blue/yellow and red/green) produced the strongest illusion in each comparison; quite the opposite of the results in the orange/purple test. In the final figure, Figure D., the average amount of error for each of the single color stimuli is presented. The results show that, with the exception of the orange stimuli, the average amount of error goes up as the lines get progressively darker. This helps support the idea that the higher the contrast between lines and background color, the stronger the illusion will appear, i.e. the white background will have a higher contrast with black than it will with yellow, therefore strengthening the illusion. The combination of these results lends some insight into the effects of color contrast in the Zollner Illusion, however, further testing would be necessary, beginning with more subjects and less variable stimuli.

References:
Beagley, W. (1990) Eyelines. [Computer program]. Alma, MI: Alma College.
Kitaoka, Akiyoshi & Ishihara, Masami. (2000). Three elemental illusions determine the Zoellner Illusion. Perception and Psychophysics v. 62 pp. 569-575.
Sadza, Karel J. , & de Weert, Charles M. M. (1984). Influence of color and luminance on the Muller-Lyer illusion. Perception & Psychophysics v. 35 pp. 214-220.
Wallace, G. K. (1974). The Effect of Contrast on the Zollner Illusion. Vision Research v. 15 pp. 963-966.