Another example of how perspective can be influenced by the expectations of size is the Ponzo illusion. In the Ponzo illusion, the depth cues of linear perspective (the tracks converging) and height in the horizontal plane provide distance cues that make the upper bar appear farther away than the lower bar. Because it seems farther away, the perceptual system concludes that the bar in the background must be larger than the bar in the foreground, despite the fact that the two bars cast retinal images of the same size (Passer & Smith, 2001). This, known as the misapplied constancy theory, therefore provides some form of evidence that the illusion is not occurring via bottom-up processing otherwise the two bars would be perceived as being the same size.
The misapplied size constancy theory can also explain the Muller-Lyer illusion. This illusion consists of two lines besides each other of the same length. One line however has arrowheads on the ends and the other has reversed arrowheads on each end. The line contained within reversed arrowheads appears appreciably longer than the other (Rock, 1995). According to Gregory (1970), the arrowheads on the ends of the lines can be seen as angles formed by two intersecting surfaces. When they arrows point outwards the surface appears to be slanted inwards and when they point inwards, the surfaces appear to recede away. The same explanation applied to the Ponzo illusions can be applied for this illusion. Because of perspective cues, the receding corner appears farther away, and at the same time, the retinal images of the two corners are identical in size. Now there is only one way objects at different distances can cast images of equal size: the farther object must actually be larger. This is therefore, how it is perceived.
However there is another explanation of this illusion, which brings us on to perceiving via bottom-up processing. According to Gregory, the Muller-Lyer illusion occurs because our SCS system is misled by inappropriate perspect cues of the figures. However, it has been found (Morgan, Hole, & Glennister, 1990) that the illusion occurs no matter what is put on the ends of the lines.
According to the Morgan et al. (1990) explanation, the reasons we experience this illusion is due to the way we are interpreting the data signals received from the actual figures. When looking at the line with reversed arrows heads on each end it looks longer because we actually look for the centre of the mass above and below the line. So with regards to the other line, the difference between the centres of the mass attached to the line with arrowheads is smaller than the difference between the centres of the mass of the line with inverted arrowheads. This idea can be applied to all versions of the ‘Muller-Lyer’ illusion. Morgan et al. (1990) explained this, using spatial interval judgements, by placing white circles at exact distances apart on two different occasions. One of the distances however had darker circles surrounding the white circles. When subjects were asked to calculate which of the two white circles were further apart, more people believed it was the one with the darker circles. This explanation is known as the incorrect-comparison theory. It holds that we cannot succeed fully in isolating parts from wholes. Despite a clear understanding of what we need to compare (the distances between dots), we cannot avoid including other components in our judgements (the arrowheads, circles etc) (Rock, 1995). This idea therefore suggests that when perceiving these types of illusions, cognitive thinking is not involved and it is only a matter of what the visual system sees (and hence is processed and viewed from the bottom, up).
Research on visual, motion perception provides an example of the accuracy of bottom-up processing. Psychologists trained monkeys to report the direction in which a display of dots moved while the researchers observed the response of individual neurones previously identified as feature detectors for movement of a particular speed and direction (Newsome, Britten, & Moushon, 1989). They discovered that the ‘decisions’ made by individual neurones about the directions the dots moved were as accurate – and sometimes more so – than the decisions of the monkeys. Perceptual decisions on simple tasks of this sort may require little involvement of higher mental (cognitive) processes (Westen, 1996).
J.J. Gibson also believes that perception is bottom-up, that it is direct and needs little prior knowledge and that sensory information intrinsically carries meaning (Gibson 1966,1979). A leading factor in Gibson’s theory is the concept of the “Ambient Optic Array”: the pattern of light reaching the eye. This optic array provides unambiguous or invariant information about the layout of objects in space, which comes in many forms. Perception hence, involves ‘picking up’ the rich information provided by the optic array in a direct fashion via resonance with little or no information processing involved (Eysenck & Keane, 1996). Gibson argues that all the information we need to have “a perception” is already present in the ambient optic array. This direct approach still, however, cannot account for visual illusions. Gibson attempts to get around this by arguing that visual illusions are just tricks invented by psychologists.
Indeed Morgan et al (1990), Newsome et al. (1989) and Gibson’s (1966, 1979) ideas and were not wrong with regards to what they were all trying to explain, however they still do not fully describe how an why visual illusions are perceived. Similarly top-down explanations alone cannot fully account for visual illusions and perceptions as input from visual stimuli in visual illusions still need to be processed in a bottom-up manner before being interpreted. Visual illusions consequently, are maybe due to both top-down and bottom-up processing. In many cases, according to Hole (2004), the distinction between ‘top-down’ and ‘bottom-up’ maybe somewhat arbitrary. For example, Helmholtz Hollow-face illusion, which consists of a hollow mask that looks like a normal convex face when viewed from different angles. This illusion is due to the ideas of top-down processing but has also been regarded as due to bottom-up processes of the brain and visual system which infer shape from shading information. When looking at the hollow mask people assume it as being a convex face because it has all features of a normal face. The shading of the mask is also taken into account (and therefore depth cues, texture gradients etc). For many years we have come to incorporate ‘knowledge’ that light usually comes from above and thus if objects have shadows at the bottom they are convex and have shadows at the top they are concave (Hole, 2004). This idea, thus suggesting, that both top-down and bottom-up processing play a role in the perceptions of illusions.
Another illusion that can be explained via top-down and bottom-up theories is the ‘moon illusion’. When looking at the moon along the horizon or above mountains it looks larger than it would if it were high up in the sky with nothing surrounding it. The surroundings affect the moons apparent size and create the illusion of it being closer to us. The moon illusion is initiated bottom-up (as for the Ponzo illusion), but this raises a curious question. Why does the moon always appear the same size when high up in the sky? It could be any size and distance, from this retinal image free of context. It seems that the visual system had default assumptions, which are accepted in the absence of scaling information. Therefore suggesting that because we know that the moon is larger than objects in its surroundings, when seen alongside these surroundings the moon is perceived to be very much bigger than when it is not. (Gregory, 2003)
Experiments conducted by Long and Olszweski (1999) also provided evidence of the combined roles of both sensory (bottom-up) and cognitive (top-down) processes in the perception of visual illusionary ambiguities. The purpose of their study was to examine the influence of bottom-up processing on reversible figures, however found that the observers’ prior experience with the figures, experience in the setting, attention and other top-down variables may also have influenced the reported reversals of the figures. This study was also further reinforced when analysed by Peterson (1999). He agreed that both bottom-up and top-down processing influence perception. The evidence above, subsequently suggests that the perceptions of visual illusions occur via both top-down and bottom-up processing.
Illusions such as the Muller-Lyer, Hollow Mask, Moon and many others not discussed all make use of both types of processing. For over a century, psychologists have been fascinated by these illusions and have focused a great deal of research on testing theories concerning their origins (Sekuler & Blake, 2002).
During the 1970’s theorists such as Neisser (1976), argued that nearly all cognitive activity consists of interactive bottom-up and top-down processes occurring together. Research has been carried out since to try and separate the two processes with regards to visual illusions however such research has not been extremely successful in providing evidence for the entire process of perceiving these illusions. It is undeniable that experience and expectations affect what we perceive. However, "unintelligent" low-level processes can account for a surprisingly large amount of our perceptual experiences. Without bottom-up processing, external stimuli would have no effect on perception; we would hallucinate rather than perceive. Without top-down processing, experience would have no effect on perception (Westen, 1996). It may be then safe to say that both "top-down" and "bottom-up" processes are important in visual perception of illusions.
REFERENCES:
Eysenck, M. W., & Keane, M. T. (1996). Cognitive Psychology: A Students Handbook (Third Edition). Psychology Press, East Sussex, UK.
Frisby (1980), cited in Passer, M. W., & Smith, R. E. (2001). Psychology: Frontiers and Applications. McGraw-Hill, New York.
Gibson, J. J. (1966). The senses considered as perceptual systems. Boston: Houghton-Mifflin.
Gibson, J. J. (1979). The ecological approach to visual perception. Boston: Houghton-Mifflin.
Gregory (1970), cited in Eysenck, M. W., & Keane, M. T. (1996). Cognitive Psychology: A Students Handbook (Third Edition). Psychology Press, East Sussex, UK.
Gregory (1980), cited in Eysenck, M. W., & Keane, M. T. (1996). Cognitive Psychology: A Students Handbook (Third Edition). Psychology Press, East Sussex, UK.
Gregory, R. (2003). Eye and Brain: The psychology of seeing (Fifth edition). Oxford university press, Oxford.
Long, G. & Olszweski, D (1999). To reverse or not to reverse: When is an ambiguous figure not ambiguous? American Journal of Psychology, 112, 41-56.
Morgan, Hole, & Glennister, (1990), cited in Hole, G. (2004). Lecture Notes. City University.
Neisser (1976), cited in Eysenck, M. W., & Keane, M. T. (1996). Cognitive Psychology: A Students Handbook (Third Edition). Psychology Press, East Sussex, UK.
Newsome, Britten, & Moushon, (1989), cited in Weston, D. (1996). Psychology: Mind, Brain, and Culture. John Wiley & Sons Inc, New York.
Passer, M. W., & Smith, R. E. (2001). Psychology: Frontiers and Applications. McGraw-Hill, New York.
Peterson, M. A. (1999). What’s it a stage name? Comment on Vecera and O’Reiley. Journal of Experimental Psychology: Human Perception and Performance, 25, 276.
Rock, I. (1995). Perception. Freeman & Co, New York.
Sekuler, B., & Blake, R. (2002). Perception (Fourth Edition). McGraw-Hill, New York.
Weston, D. (1996). Psychology: Mind, Brain, and Culture. John Wiley & Sons Inc, New York.
WORD COUNT: 2286 Words.