Refer to psychophysical method of adjustment. Maybe call this topic Shape instead of Size? Tell instructors that students will hear two types of explanations: functional and physiological, and that they may not conflict.
There are many illusions in which we misjudge the size, location, or orientation of an object. Some were first described in the mid 1800s, others have been recently discovered. All are still the subject of active research. There have been many attempts to explain these illusions, but none are universally accepted. Some of the sections below refer to the most common explanations, but see Purves and Lotto (2003) for a different type of explanation.
In each of the following experiments, the point is to adjust size or position until two items are subjectively equal. You can then measure your estimate to see how far subjective equality is from objective equality. In other words, the point is not to get the “right” answer, but to investigate the workings of your visual system.
. Your task here is to adjust the length of one line until it looks equal to another line. Under some circumstances, you tend to over- or underestimate length. A variety of explanations have been proposed for this illusion. One common explanation holds that a line between inward or outward angles is interpreted as a corner jutting toward the viewer (↕) or recessed away from the viewer. Thus a 10 cm line jutting towards the viewer should seem smaller than a 10 cm line recessed away from the viewer. Is this explanation consistent with your results from the various control conditions (e.g., dumbbell and detached corners)?
. Your task here is to adjust the position of one line segment until it looks continuous with another. Try the first case (the usual illusion) and then try the controls. How much difference does the occluding shape make? Does it matter whether the shape has dark edges?
. Your task is again to adjust the length of one line until it looks equal to another line. As with the Müller-Lyer illusion, the most common explanations invoke perspective cues (e.g., Gillam, 1980). We tend to perceive converging vertical lines (/\) as converging on a distant point, so a horizontal line near the apex will seem larger than an equal line near the base. Because it is perceived as being further away, it must represent a larger object to take up the same width. Is this explanation consistent with your results from the upside-down and sideways Ponzo illusions? There are other explanations as well. Prinzmetal et al. (2001) make a convincing case that the Ponzo and Poggendorf illusions result from visual mechanisms that give an accurate perception of orientation regardless of the position of the head.
. Unlike the other line-adjustment experiments, this time you compare a vertical line with a horizontal one. The misjudgment of length in this case is often described as a tendency to overestimate the relative length of vertical items. Is this consistent with your results in the sideways T case or in the L case? Are the error amounts in those cases as great as in the upside-down T? The upside-down T illusion probably has two components. What might they be?
. Here, your task is to place a dot in the center of a line or shape. Under some conditions, other elements in the figure may bias our perception of center. Is the effect equally strong in horizontal and vertical cases? Which surrounding elements are most effective in misleading your judgment of center?
. Again, your task is to adjust one object until it is equal in size to another object. This time, however, the objects are circles. Once again, the presence of surrounding objects affects our size judgment. One interpretation is that we tend to exaggerate size contrasts, so a circle surrounded by small circles is made to look larger, while one surrounded by large circles is made to look smaller. Ninio (1998), on the other hand, suggests that we tend to normalize the total area taken up by a figure, so a circle surrounded by large circles looks smaller than one surrounded by large circles because we tend to reduce the entire large figure and enlarge the entire small figure.