The Muller-Lyer illusion's strength has been found to be related to the amount of exposure to the illusion. It has been found that the illusion still had an effect after subjects had practiced the task extensively, (Anii, 1997) but that extended exposure to the Muller-Lyer Illusion has lead to an extreme decrement in the magnitude of the illusion. In one experiment, after being constantly exposed to a stimuli, the results concluded that the illusion never completely disappeared but did near zero. (Dewar, 1967) This is thought to occur because of a "reorganization of cognitive components involved in visual-geometric illusions." (Girgus, 1975) The next step is to test the strength of the illusion in relation to the ability to match the lengths of two plain lines.
The purpose of this experiment was to find out if there is an increase or decrease in illusion when a subject is repeatedly exposed to the illusion. We then compared one's ability to judge the length of a plain line before and after the stimulus. It is thought that repeated exposure to the illusion will affect the perceived illusion therefore increasing the illusion decrement and affecting how a plain line is judged by the subject.
METHOD
30 college students ages 18-22 were individually tested. Each subject participated voluntarily. The computer software program, Eyelines, was used for this experiment. A tails out Muller- Lyer line was created using this program. The center line of the Muller-Lyer figure was 90 mm long. Each of the four outward tails measured 24 mm long and the acute angle measured to be 70 degrees.
The Eyelines software was programmed to run 40 trials which would take approximately 5 minutes for each subject to complete. Participants were presented two trials of control line task, followed by 40 trials of the Muller- Lyer tails out figure, and lastly the participants were presented two more control line figures. Each participant completed a total of 44 trials. The computer screen showed two lines at a time. For the control tasks, both lines were straight lines, the top line was the controlled stimulus, and the lower line served as the adjustable length.
In all trials, the task was for the participant to adjust a given line (the lower line) until he/she felt it was the same length as the controlled stimulus (the top line). Participants moved the computer mouse to increase and decrease the length of the adjustable line. At the completion of each trial, the participant hit any key on the keyboard to signify they felt the lines were equal in length. Once a key was hit, another trial was presented.
RESULTS
Initial analysis of the data showed a relatively even division of the illusionary increase versus illusionary decrease. (as shown in Fig. 1) Half saw an increase in the illusion and half saw a decrease in the illusion after extended exposure to the illusion. There was also an even distribution in the control results. (as shown in Fig. 2) About half of the subjects estimated the control line to be shorter than they had previously estimated after repeated exposure to the illusion. In further analysis, half of the subjects who saw a stronger illusion after 40 trials, saw the control line as shorter than they previously saw it, the other half of the subjects percieved the control line to be longer. (as shown in Fig. 3) This is the same with the subjects who percieved a weaker illusion after 40 trials. Half estimated the control line longer than previously and half estimated it shorter. (as shown in Fig. 4)
Figure 1. Average illusional perception of repeated exposure (trials 1-5) and repeated exposure (trials 36-40).
Figure 2. Average estimation of the length of a plain line before illusion experiment (control 1) and after illusion experiment (control 2).
Figure 3. Change in perception of the control line for subjects who percieved the illusion to be stronger.
Figure 4. Change in perception of the control line for subjects who percieved the illusion to be weaker.
DISCUSSION
Repeated exposure to the Muller-Lyer illusion with 26mm tails decreased the illusion for some subjects and increased the illusion for some subjects. These results are contrary to the findings of the studies mentioned previously. In general, half of the subjects saw an increase in the illusion, although it was sight, and half saw a decrease. When comparing the results of the perceived illusion with the change in estimation of the line length of the control, we found no direct correlation.
A final increase in the illusion may have been due to a subject's self-doubt. The longer they were exposed to the stimulus, the more they may have doubted their ability to estimate the length, guessing the purpose of the experiment was to make them estimate wrongly. A decrease in the illusion may be due to an adaptation to the illusion, allowing the subject to better tell what exactly the mind is measuring when it looks at the illusion. change in illusion does not affect change in percieved link of a plain line.
REFERENCES
Anii, Akira and Kazutoshi Kudo, (1997) "Effects Of Instruction and Practice On the Length-Reproduction Task Using the Muller-Lyer Figure," Perceptual and Motor Skills Vol.85 pg. 819-825.
Beagly, W. K. (1990) Eye Lines [Computer Program]. Alma, MI: Alma College.
Dewar, Robert E., (1967) "Effect of Length of Oblique Lines and Prominence Of the Horizontal Line On the Decrement of the Muller-Lyer Illusion With Extended Practice," Psychonomic Science Vol. 8(12), pg 509-510.
Girgus, Joan S., Stanley Coren, Mitchell Durant, and Clare Porac, (1975) "The Assessment Of Components Involved in Illusion Formation Using a Long-term Decrement Procedure," Perception and Psychophysics Vol 18 (2) pg. 144-148.
Goldstein, E. Bruce, (2002) Sensation and Perception, Pacific Grove, CA, Wadsworth Group.