There has also been research done by Blakemore, Carpenter, and Georgeson (1970) which suggests that cortical detectors geared toward vertical orientation are partially inhibited by detectors geared towards crossing lines. Their research has shown that two lines forming an acute angle should appear to be shifted away from each other in orientation and the angle between them should seem to expand. Also the two lines forming an obtuse angle should appear to be shifted towards each other in orientation and the angle between them should seem to contract.
This background research leads to an interest in the Poggendorf illusion. Pressey and Wilson (1974) have said that the Poggendorf illusion (as shown in fig. 1) leads viewers to perceive the oblique line on the left to point below the oblique line on the right so that they do not seem to be able to connect together. They have also stated that the two main arguments for the reason explaining this illusion are either because of the peculiar distribution of excitation at the highest point, or because of lateral inhibition, that the acute angle is perceptually enlarged.
It seems to be that the angles formed from the oblique line in the Poggendorf illusion most likely have an effect on the perception of that lineŐs orientation. The line connected on the left side has an obtuse angle on top and acute angle on the bottom. This leads us to perceive the line as farther away from the vertical than it really is. The line connected on the right side has an acute angle on top and an obtuse angle on the bottom. This leads us to perceive the line as farther away from the vertical than it really is as well.
It may be believed that by adapting the eyes to an orientation slightly steeper than that of the oblique line in the Poggendorf illusion before viewing the Poggendorf illusion, it will cause the oblique line to be perceived as even farther from the vertical, or less steep than it really is, therefore strengthening the illusion. Also it may be believed that by adapting the eyes to an orientation slightly less steep than that of the oblique line in the illusion, it will cause the oblique line to be perceived as closer to the vertical or more steep, which will bring the line back to itŐs true place and will weaken the illusion. The reason why this should weaken the illusion is because when looking at the Poggendorf illusion alone it already effects the way the oblique line is perceived. When we adapt our eyes to a less steep line beforehand this causes those specific orientation neurons to be fired less when we look at the Poggendorf illusion and the neurons specified for the orientation of the oblique line will be firing more making the contrast between the two greater and causing it to appear more steep than it really is. This will then bring it back to look like itŐs actual orientation.
Methods
Subjects
14 subjects took part in this experiment between the ages of 18-22. 7 of the subjects were female and 7 were male.
Apparatus & Stimuli
The after effects of line orientation adaptation on the Poggendorf illusion was tested. The experiment was created by using the Eye Lines Mac 3.2 program in the laboratory room of the Psy department at Alma College. The subjects were presented with 2 different adaptation stimuli, the Poggendorf illusion, as well as a controlled stimuli (which was the oblique separated line in the Poggendorf illusion, only without the rectangle). The 2 different adaptation stimuli looked like figures 2 and 3. The oblique line in the Poggendorf illusion had an angle of 30 degrees from the vertical. The more steep line adaptation (as shown in fig. 2) had an angle that was 25 degrees less than that of the line in the illusion, making it 5 degrees from the vertical. The less steep line adaptation (as shown in fig. 3) had an angle that was 25 degrees more than that of the line in the illusion, making it 55 degrees from the angle.
Procedures
The subjects each sat in a chair directly in front of the computer screen and typed in their initials. They then were presented with the Poggendorf illusion in which the top right portion of the oblique line was adjustable to move either directly up or down along the line of the rectangle. The subjects were told to adjust the line so that it looked as though it would connect directly with the oblique line on the left side of the rectangle. They were also told that this adjustment was only for practice before doing the real experiment. Next they were randomly presented with either the 5 degree adaptation lines, the 55 degree adaptation lines, or no lines (a blank screen) for 30 seconds. They were directed to look at the lines (or blank screen) for the entire 30 seconds. The screen then switched randomly to either the adjustable Poggendorf illusion, or the adjustable controlled stimuli, and were then directed to as quickly as possible, adjust the top right line so that it looks as though it could connect to the bottom left line. After they adjusted the line they pressed enter to switch to the next random adaptation before adjusting the next figure. The subjects continued with this same procedure until they had made 6 different adjustments total. 3 adjustments were for the Poggendorf illusion after they had adapted their lines to either the 5 degree lines, 55 degree lines, or no lines. The other 3 adjustments were for the controlled stimuli after adapting their eyes to either the 5 degrees lines, 55 degree lines, or no lines.
Fig. 1 - Poggendorf illusion
Fig. 2 - adaptation lines, 5 degrees from vertical
Fig. 3 - adaptation lines, 55 degrees from vertical
Results
The main result from my data is that the Poggendorf illusion was significantly strengthened by having the subjects adapt their eyes to the lines of different orientations before viewing and making the adjustment on the Poggendorf illusion (see figure 4). However both the adaptation lines caused a stronger effect of the illusion, not just the 5 degree adaptation lines (the steeper lines).
The results of the data show that the controlled variable is closer to 0 than the Poggendorf illusion. This shows that there was a significantly higher error rate when the subjects adjusted the Poggendorf illusion compared to when they adjusted the controlled variable, which was expected because this shows that the illusion is successful. The data also shows that the adjustments that the subjects made after no adaptation had a lower error rate than the adjustments made after both the 55 degree adaptation lines were viewed and the 5 degree adaptation lines were viewed. This demonstrates that the adaptation lines strengthened the effect of the illusion. The stronger effect on the illusion occurred when the subjects viewed the 5 degree (more steep) adaptation lines previously, however this effect was not much higher than the effect that the 55 degree lines (less steep) had on the illusion. Therefore this difference between the two effects are not significant enough to support the intended theory that the adaptation of the 5 degree lines would strengthen the illusion and the adaptation of the 55 degree lines would weaken the illusion.
Fig. 4 - Mean results of the subjects from both the control and the Poggendorf adjustments after adapting eyes to 55 degree lines (plotted on left), no lines (plotted in middle), and 5 degree lines (plotted on right)
Discussion
All of the error rates are negative errors and not positive errors, so we know from this that on average all the subjects moved the line too low and not too high. Also the data that was obtained from this experiment seems to argue against the adaptation and lateral inhibition theories because the adaptation of the two different orientation lines had basically the same effect.
It is unsure why the results did not turn out to support the adaptation and lateral inhibition theories. It could be that there simply may not have been enough subjects tested to provide sound data. Or it may be that the adaptation lines only caused the effect of confusion within the brain and therefore it didnŐt matter what orientation they were, they only contributed to strengthening the illusion.
References
Blakemore C., Carpenter R., Georgeson M. 1970 Lateral Inhibition between Orientation Detectors in the Human Visual System. Nature Land. Vol. 228, 37-39.
Hubel, D., Wiesel, T. 1959 Receptive fields of single neurons in the catŐs striate cortex. Journal of Physiology. Vol. 148, 574-591.
Magnussen S., Kurtenbach W. 1980 Adapting to Two Orientations: Disinhibition in a Visual Aftereffect. Science. Vol. 207, 908-909.
Pressey A., Wilson A. 1974 The Poggendorf illusion in imagination. Bulletin of the Psychonomic Society. Vol. 3, No. 6, 447-449.