I usually go to the park every morning for a jog and enjoy the scenery after by sitting on a bench for at least half an hour. As I went about with this usual routine, I noticed a curious thing. Across the park from the bench I usually sit in is a sidewalk, with railings at the edge consisting of two strands of chain strung between the posts. I attached a picture of my view at the park so that you may understand clearly before I describe to you what I noticed.
As you can see, the chain linked fence runs across the whole sidewalk. What I noticed was, when viewed from this distance, I seem to see that the bicycles running on the sidewalk beside this fence does not only move forward, but also bobs up and down in phase with the contour of the chains. When I decided to go to the sidewalk to take a closer look, I saw that the pavement was flat and smooth, and therefore does not serve as the reason for my perception of the bicycle's up and down motion.
What could possible account for this? I am really trying to figure this out. It has bothered me for months.
Hoping you can reply the soonest.
Senserely yours,
Confused jogger 23
Dear Confused Jogger 23,
Thank you for writing to us! I am an avid jogger myself and, believe it or not, I've noticed the exact same thing you have just described! Apart from what you have just explained to me, I also noticed another curious thing - when seen at a nearer distance at a seemingly larger point of view, the motion of the bicycle does not seem to move along similar to the contours of the chains but in the opposite direction. Between the posts, where chains have the tendency to loop downward, the motion of the bicycle seem to move upward; while in areas near the post where the tendency of the loop is to move upward, motion of the bicycle seem to move downward.
Luckily for us, a study done by Masson, Dodd, & Enns (2009) demonstrated and explained the phenomenon we both have just described. They called this phenomenon the Bicycle Illusion. Their explanation of this illusion provides an opportunity for understanding the way human vision integrates information from multiple sources - a stationary shape (the fence) and the position in space of an object in motion (the cyclist) (Masson, Dodd, & Enns, 2009).
According to their research, motion integration governs the process of the perception of the Bicycle Illusion. How does this happen? Signals are assimilated (averaged) and are derived from two sources often considered to be processed in distinct systems in the brain and their interaction - the processing of stationary shape (ventral visual stream) versus the processing of motion in the dorsal visual stream (Livingstone & Hubel, 1987; Ungerleider & Haxby, 1994; Van Essen & DeYoe, 1995 as cited in Masson, Dodd, & Enns, 2009).
One explanation of the process in the perception of the bicycle illusion is that some of the spatial properties of the stationary rails are used incorrectly by our visual system in determining the position of the bicycle in motion. Masson, Dodd, and Enns (2009) proposed that the position of the moving bicycle in space is represented with greater uncertainty than that of the stationary rails. An explanation for this disparity in certainty is that the object in motion has very little contour in comparison with the rails, that the contours of an object in motion can be represented less than those of stationary objects. The large viewing distance and the limits of acuity contribute to the differences in representation between the bicycle and rails. In other words, the higher certainty of the contours of the rails are being used to determine the position in space of the lower certainty contours of the moving bicycle (Masson, Dodd, & Enns, 2009).
As I have already described, the bicycle illusion can have two effects: an illusion of bobbing up and down along the contours of the railings, as you have described, or the illusion of its movement opposite to the contours of the railings, as I have added. These effects are determined largely by a difference in the size of the display by which motion is seen - with a small display, the bicycle appears to bob up and down, which creates and illusion of assimilation (also called motion capture). Meanwhile, a large version of the display produces an illusion of opposition (also called induced motion). What factors contribute to these effects?
In their study, Masson, Dodd, & Enns (2009) explained that motion capture (illusion of assimilation) is more likely for smaller central stimuli and that motion opposition increases in likelihood as size is increased (Murakami & Shimojo, 1993; Nawrot & Sekuler, 1990 as cited in Masson, Dodd, & Enns, 2009). Additionally, spatial frequency and luminance contrast tend to differentiate these two illusions, with motion capture more likely to occur with surrounding stimuli of a lower frequency and central stimuli of a lower contrast (Ido et al., 1997, 2000 as cited in Masson, Dodd, & Enns, 2009).Taken in the light of motion integration just described, the bicycle illusion is also size and luminance contrast dependent. The difference, however, can be noted in how the stimulus in the bicycle illusion is entirely stationary (aka railings). This fact implies that an analysis of motion may be influenced at a relatively early stage by an analysis of the stationary background in which the motion of the bicycle is occurring (Masson, Dodd, & Enns, 2009).
To further illustrate the bicycle illusion, a combination of mechanisms, including explanations for different motion illusions, are involved. As already mentioned, motion integration - combining local motion signals through perceptual grouping - helps explain the bicycle illusion. Additionally, the aperture problem helps in explaining this illusion through determining the direction of motion by the orientation of local edges. For example, the wheels of the bicycle can be thought of as being viewed through the "aperture" of the sagging chain rails. As such, the leading edge of the wheel may appear to be oriented with a negative slant in the upward portion of the rails and then oriented with a positive slant in the downward direction of the rails (Masson, Dodd, & Enns, 2009). Moreover, the stepping feet illusion provides additional explanation to the bicycle illusion problem.This illusion explains that the apparent speed of a moving object depends on the difference in the relative luminance contrast between the object's surfaces and those of the background on which it is moving (Thompson, 1982 as cited in Masson, Dodd, & Enns, 2009). Specifically, an object moving at a constant velocity will appear to move faster when moving against a background that differs sharply in contrast and to slow down when the contrast of the background is more similar.
The bicycle illusion can be explained through a difference in contrast in that edges of the moving object are weaker due to its edge contrast being more similar to that of the rails, instead of being compared to the reduced edge contrast of the background. As a result, the strongest local motion signals are coming from between the rails, and so the edge will appear to move along the direction of the sagging rails, rather than horizontally as they are really moving (Masson, Dodd, & Enns, 2009).
As a final point, opposition is experienced because at larger viewing angles, the contours of the moving object and that of the rails are more readily differentiated. This is due to increased acuity that comes with larger size. Viewing small displays has the effect of minimizing subtle differences between the edges of the moving object and the edges of rails. As the elements of the scene are enlarged and the individual shapes are represented with greater accuracy, the confusion arising from the neighboring edges of the moving objects and rails is also reduced, allowing global grouping mechanisms to link the signals from the object above, below, and between the rails (Masson, Dodd, & Enns, 2009). Perceptual grouping and segregation as a result serve to exaggerate differences between objects and as such, leads to the illusion of motion opposition described earlier.
I know it may seem a handful but I hope that I was able to shed some light on the matter! As you can see, there are several factors that are integrated for our perception of the bicycle illusion. Do continue to be observant as you might discover other illusions of motion around you!
Senserely yours,
Yana
Reference
Masson, M.J., Dodd, M.D., & Enns, J.T. (2009). The
bicycle illusion: Sidewalk science informs the integration of motion and shape
perception. Journal Of Experimental
Psychology: Human Perception And Performance, 35(1), 133-145. doi:1-.1037/0096-1523.35.1.133
No comments:
Post a Comment