Tips & hints The Hermann Grid Simulation
1. For most of the following exercises, sit so that your eyes are approximately 50 - 60 cm. (20 - 24 in.) from the screen.
2. Don't stare too long at the display! Some of the images you'll generate with this simulation can be quite discomfiting, even disorienting. Negative afterimages can also be a problem. So, give your eyes and your brain a break every now and then.
3. Remember to note carefully the appearance of the illusion with each change you make in the simulation's parameters. A standard experimental technique used by psychophysical researchers is to have the experimental subjects estimate the magnitude of the illusion, i.e., the intensity of the illusory spot. This will, of course, be personal and highly subjective, but try to use a scale from 0 (no illusory spots seen) to 10 (illusory spots appear clearly defined at their maximum intensity). You might also consider making sketches of the illusion.
4. Before trying to answer some of the questions below, you might want to read about the concept of on- and off-center receptive fields in the visual system in your text, and read about the most widely accepted explanation for the Hermann grid illusion, the Simultaneous Contrast Hypothesis.
5. For assistance in interpreting your data and suggestions for further experimentation on this topic, see the paper by Spillmann (1994; Perception 23:691-708) and referenced cited therein.
Experimental Protocols
For the following exercises, work first with the basic black square-white lines illusion. After you've completed the basic exercises, try the reversed grid (white squares-black lines), then move on to other color combinations.
Basic Experiments
1. Start the simulation and observe its features. Focus your vision on one intersection near the center of the grid and note the appearance of the grid in the periphery of your visual field. What happens if you shift your gaze to another intersection? Can you see an illusory spot when you gaze directly at a particular intersection? What if you continuously scan the grid from side-to-side, top-to-bottom?
2. Vary the size of the grid squares. Does this change the appearance or strength of the illusion? If so, how? Is there an optimal size for the squares that produces the strongest illusion?
3. Use the Reset button to return the grid settings to their default values. Now, vary the width of the grid lines. Does this change the appearance or strength of the illusion? If so, how? Is there an optimal width for the grid lines that produces the strongest illusion? What is the narrowest grid line at which you can still perceive the illusion? The widest?
4. Use the Reset button to return the grid settings to their default values. Try viewing the illusion with your eyes at two or three different distances (say, 30 cm, 60 cm, and 90 cm) from the screen. Be consistent in the distances you use. Does the illusion change with viewing distance?
5. Using the default settings, view the illusion with one eye, first with the right, then with the left. Does the illusion change when viewed with monocular vision?
Advanced Experiments
6. Reduce the contrast between the intersections and the grid squares by using the radio boxes on the right side of the display to combine gray squares with black lines, and then white squares with gray lines. Does this change the illusion?
7. Tilt your head approximately 45oto the right or left and view the grid. Note any effect(s) on the appearance of the illusion.
8. Return the settings to black squares with white gridlines and set Line Width to 2 and Square Size to 3. Is it somewhat uncomfortable viewing the grid? Regardless, continue to gaze at the grid for a few seconds and note whether you observe anything unusual about the grid.
9. Return the settings to black squares with white gridlines and set line width to 5 and square size to 15. For most observers, this setting creates the Prandtl Illusion, in which dark lines pass diagonally through the squares, connecting the illusory spots in the intersections. Repeat Exercise #7 at these settings. Does the appearance of the illusion change? Experiment to identify combinations of line width and square size for which the illusion is no longer observed.
10. Return the settings to black squares with white gridlines and set line width to 16 and square size to 10. This produces what is essentially a grid of alternating black and white squares of dimension 10, with gridline width equal to 3. Many authors state that this will produce the Springer line illusion (Lindsay and Norman, 1977), a lattice of faint gray lines interspersed among the black squares. Repeat Exercise #7 at these settings. Is the illusion affected?
11. While keeping your eyes at a constant distance from the screen (say, 60 cm.), view the illusion from different horizontal angles. Accomplish this by fixing your gaze at points successively further from the center of the grid, while using your peripheral vision to continue observing the intersections as close as possible to the center of the grid. For focus points, use the center-point of the left and right edges of grid squares and progress to squares further from the center, all the way to the right edge of the grid. At each point, record the maximum line width at which you can still observe the dark spots in the intersections nearest the grid's center. Then, repeat the experiment, this time shifting your focus point to the left of the grid's center.
12. Resize (use your mouse to drag the lower right corner of the window up and to the left) the display to reduce the number of intersections to two. (when you do this, all of the controls will become difficult or impossible to identify, but that doesn't matter for this exercise) Record the strength of the illusion. Now enlarge the window so that there are four intersections visible and record the illusion's strength. Continue resizing the display so that you increase the number of intersections to 16, then 25, then 36, and so on.
13. Try different color combinations and repeat as many of Exercises 1-8 as you wish.
Questions
1. Are all of your results consistent with the simultaneous contrast hypothesis? If not, can you suggest an explanation?
2. How do your results compare with those of your lab partner(s) and with the other students in your lab section?
3. Is the illusion the same for black lines with white squares as it was for white lines with black squares? What about for other color combinations?
4. Horemis (1970) claims that the 'strength' of the illusion weakens and that the illusory spots decrease in size when the contrast between the grid squares and the intersections of the grid lines decreases. Does this make sense in the context of the standard simultaneous contrast model? You can test this assertion by comparing the effect of black squares and gray squares contrasted with white grid lines, and by using black squares and gray lines. Do your results agree with Horemis' claim?
5. Does the basic illusion change when you observe it tilted at an angle of 45o? (for most observers, it does)? If so, does that suggest anything about what part of the visual pathway (i.e., the retina, the lateral geniculate body, or the visual cortex) might be responsible for the illusion?
6. What do the results of your monocular vision experiment suggest about the source of the illusion?
7. When observing the grid as set up in Exercise #8, virtually all subjects observe poorly defined, somewhat broad intersecting diagonal bands extending from lower left to upper right and from upper left to lower right across the display. Surprisingly, many, but by no means all, subjects see colors in these bands (remember, the grid consists of nothing but black squares and white lines). Of those who do see spurious colors, some see only blue shades, some only red (or pink), but some observers see the entire white-light spectrum, from violet through red. What was your perception? What about the perception of other observers? Can you suggest an explanation for the existence of the illusory colors in the diagonal bands?
8. Does the strength of the illusion depend on the number of intersections present in the display? Can you suggest an explanation for your observation?
9. According to most published reports, the addition of diagonal lines crossing the squares (the Lingelbach illusion) abolishes the illusory spots at the intersections. Is this what you observe? Does this assertion make sense in the context of the standard simultaneous contrast model? Test the assertion by using the "Lingelbach et al., 1985" check box to generate the appropriate grid.
10. Using the default settings (click Reset), determine the effect on the illusion of having rounded corners on the grid squares (the Dombrowsky version of the Hermann grid)? Most authors state that this "intensifies" the illusion. Is this your observation?
11. Work with the Jung version of the grid (you'll probably want to use relatively wide (white) lines (10 to 15) and small (black) squares (20 to 30) for this. It's generally reported that this causes the illusory spots to 'expand' to uniformly fill the small squares enclosing the intersections. Note whether you can see an illusory spot when you gaze directly at a particular intersection (compare this with what happened in Exercise #1). Observe the grid while tilting your head 45o to the right or left, and compare your observations with those you made in Exercise #7.
12. Does the color of the squares and/or grid lines affect the results? Are the results obtained when the squares are 'lighter' than the grid lines different from those obtained when the squares are 'darker' than the grid lines?
13. Based on your experiments involving colored versions of the grid, does it appear that the visual tracts involving the cones and their output are structured and function pretty much the same as the tracts that involve rods? Do their appear to be any significant differences? Justify your responses.
14. If you are color-deficient or color blind, compare your results with those of someone who has full color vision. Or, if you know someone who is color-deficient or color blind, have them conduct some of the above exercises to see if color vision is necessary for the perception of the illusions involved in some of the above exercises (especially Advanced Exercises #8 and #13).
15. Think about why some of the perceptions produced by certain presentations of the Hermann grid (e.g. in Exercise #8) are somewhat unpleasant to look at for very long. That is, where in the visual pathway might the signal that triggers the unpleasant sensation actually be generated? Can you think of any adaptive benefit to having the visual system respond in such a way? Can you think of any examples perhaps hypothetical where other organisms might take advantage of this characteristic of the human (or other vertebrate) visual system and thereby gain a benefit for themselves?
16. In line with the preceding question, do you think the Hermann illusion has any applicability in, or consequences for, art or other visually-based media? How about in environmental engineering?