Figure 2 Average Difference of Adjusted Parallel Lines in Cafe Wall Illusion
The Cafe Wall illusion, discovered by Richard L. Gregory in 1973, shows parallel lines that appear to not look parallel. The illusion is that the rows appear to be larger on one side and closing in on the opposite side, alternating with each row. Though the viewers perceive that the lines of mortar are leaning slightly, they are actually parallel. Gregory and Heard (19790 found that the strength of the illusion is due to the lighter color of the mortar lines between the blocks. Schwartz (2005) found that the illusion is not affected by different orientations, so you can stand it straight up on its side and the illusion will still hold strong. Takeuchi (2005) found that how much space is between the dark and light boxes plays a role in the strength of the illusion. Ranpura (2009) found that the illusion is created on the "mortar" lines between the boxes because of stripes that are produced which give the lines a tilted effect. This study examined the effects of the mortar lines crossing through the black boxes as opposed to between the rows to see if the illusion still holds strong.
The subjects were male and female Alma College students between the ages of 17 and 24. The apparatus used was a computer program called Eye-Lines that showed a visual example of the illusions for the subjects to adjust. The computer was tilted because the image was created at a 9 degree angle. This was so that the horizontal lines would not follow the grid lines and adjust smoothly. The subjects were presented with a Cafe Wall illusion and one with parallel lines that cross through the boxes instead of in between them.
A subject was presented with 10 images that were randomly shown. 5 were of a standard Cafe Wall illusion and 5 were of the altered illusion where the parallel lines run through the black and white boxes .99 pixels above and below where the tops and bottoms of the boxes meet. The subjects were to adjust the lines that appeared to be slanted but were already parallel to what they perceived as parallel.
18 students participated in this experiment, 14 of whom were female and 4 were male. The mean difference of the parallel lines that were adjusted from their original spot in the control illusion was .014 pixels. The mean difference of the parallel lines that were adjusted from their original place in the altered illusion was .012 pixels. The students adjusted the lines away from their original spots while the lines were actually already parallel to begin with.
Figure 2 Average Difference of Adjusted Parallel Lines in Cafe Wall Illusion
This experiment supports the conclusion that the Cafe Wall illusion is not as strong when the gray lines cross through the black boxes, as opposed to when the lines pass between the rows. However, the substantial finding was that the altered illusion still held strong and worked. This experiment showed that the illusion is not solely based on the lines passing through at the local intersections between the black boxes. The lines still have a tilted effect when passing through the boxes. The adjustments were so similar that this experiment strongly suggested that it does not matter where the lines are located in the illusion, only that they are parallel.
Studies in Visual Perception
Brain Connection
Alma College Psychology Department