Fig. 1: Ponzo Control Illusion
Fig. 2: Muller Lyer Control Illusion
Fig. 3: Muller Lyer Cooperating Illusion
Fig. 4: Muller Lyer Conflicting Illusion
The framework of this experiment includes the Ponzo and Muller Lyer illusions. The Ponzo illusion consists of a bottomless triangle that contains two horizontal lines of the same length. Research on the Ponzo illusion indicates that the further line appears to be larger, and the closer line appears to be smaller. Fisher (1967). This is illustrated in Fig. 1. The Muller Lyer illusion works by putting arrows on the ends of two separate lines. These arrowheads can either be tails in or out. Fisher (1967). Depending on the direction of the arrowheads, the line appears to be either longer or shorter than its actual length. This is illustrated in Fig. 2. The other two illusions used in the experiment were both a combination of the Ponzo and Muller Lyer illusions. Muller Lyer arrows were added to each of the lines inside of the standard Ponzo illusion. One was a combination in which the two illusions cooperated with each other to make the bottom line smaller than the top. This combination, nicknamed the Muller Lyer cooperating illusion, is illustrated in Fig. 3. The other Ponzo and Muller Lyer combination was an illusion that conflicted with itself. The Ponzo-like attributes of the illusion worked to make the bottom line appear smaller than the top, where as the Muller Lyer-like attributes of the illusion worked to make the top line appear smaller than the bottom. This combination of the Ponzo and Muller Lyer illusion, nicknamed the Muller Lyer conflicting illusion, is illustrated in Fig. 4. It is hypothesized that the additive qualities of the Muller lyer cooperating illusion would make a stronger illusion. Altogether there are four conditions to this experiment. These conditions consist of the Ponzo Illusion, Muller Lyer Illusion, Muller Lyer cooperating, and Muller Lyer conflicting Illusions.
Method:
The Subjects used in this experiment were seventeen male and female undergraduate students. Their ages range from eighteen to twenty-three. It was requested that the subjects read the paper instructions that were provided in front of the computer. Subjects were told to sit upright with their chairs pushed in and their bodies near the table. Leaning in toward the computer screen was prohibited. The experiment was presented with four different stimuli. This stimuli consisted of the Ponzo illusion, Muller lyer illusion, Muller Lyer cooperating, and Muller Lyer conflicting. The inside lines of the Ponzo were 3.6 cm. The outer left angle was 245 degrees. The outer right angle was 295 degrees. Each line of the Muller Lyer were 3.6 cm long with arrowheads measuring at .8 cm. Both the Muller Lyer cooperating and conflicting illusions consisted of Muller Lyer lines that were 3.6 cm long. The tails for each line measured at .8 cm. These two illusions also had a Ponzo triangle with a left angle of 245 degrees and a right angle of 295 degrees.
Fig. 1: Ponzo Control Illusion
Fig. 2: Muller Lyer Control Illusion
Fig. 3: Muller Lyer Cooperating Illusion
Fig. 4: Muller Lyer Conflicting Illusion
Results:
As shown in fig. 5, the Muller Lyer cooperating illusion showed little difference in average error over that of the Muller Lyer control illusion. It did, however, appear significantly longer then the average error of the Ponzo control illusion. On the other hand, the Muller Lyer conflicting illusion showed a significant difference with perceived length in the opposite direction then any other illusion tried: the Muller Lyer cooperating, the Muller Lyer control and the Ponzo control.
Fig. 5: Results of Combining the Muller Lyer and Ponzo Illusions
Discussion:
On average, subjects tended to underestimate the length of the bottom line in the Muller Lyer conflicting illusion. Subjects tended to over-estimate the length of the bottom line on the Muller Lyer cooperating illusion. Subjects stayed consistent with their conclusions in the Muller Lyer cooperating illusion in comparison to the Muller Lyer control. The low difference in results between the Muller Lyer cooperating illusion and the Muller Lyer control implies that the added Ponzo element did not influence subjects' conclusions. Therefore, the hypothesis is proved incorrect. Combining the Muller Lyer and Ponzo illusions did not strengthen the effect of the individual illusions. The illusions were not additive; it did not make the bottom line appear even shorter than in the individual illusions. However, the fact that the Muller Lyer conflicting illusion showed such a significant difference in the average error is interesting. The Muller Lyer conflicting illusion's average error shows that subjects perceived that the bottom line was originally longer than the top and needed to be shorter. This may mean that, in the Muller Lyer conflicting illusion, the Ponzo-like elements overrode the Muller Lyer-like elements. Even though the original hypothesis was proven incorrect, the results do provide an interesting base for another study. Still, the results may have been influenced by the amount of time and subjects available. Further testing may be needed to support the claim founded by this study.
References:
Fisher, Gerald H. "A Common Principle Relating to the Muller-Lyer and Ponzo Illusions." The American Journal of Psychology. 80.4 (1967): 626-631. JSTOR. Alma College Library, Alma, MI. 3 March 2009 [http://www.jstor.org].