Definition of perception
“The process of becoming aware of objects and relations in the world around us, in so far as that awareness depends on sensory processes. ” This is a definition of perception by George A1. Miller, yet this awareness of our environment is not always an exact representation of the reality. Our visual system can often be tricked into perceiving things incorrectly or things that don’t exist, images that can do this are known as visual illusions.
Visual illusions can tell us a lot about perceptual processes; firstly they help use learn more about the use of top down processing and bottom up processing within perception. They also illustrate the ambiguity of perception, as well as the nature of perceptual learning. We can also learn about size constancies and depth perception, as well as interesting after effects that can be perceived. There is some controversy between psychologists as to the extent which perception relies directly on sensory information. Gibson (1966) proposed the direct theory of perception, or bottom up processing.
He believed that judgements about the visual world were based solely on the information coming into the eye2. Constructivists, however, argued that there must be intermediate processes at work, which use previous knowledge and experience to influence our perception. This is known as Top down processing, which emphasises the use of cognition within perception. Illusions such as the Necker cube (a cube that can be perceived in two different ways) rely on our stored knowledge of depth and therefore uses top down processing.
Other illusions such as the Hermann Grid (see later on) occur due to physiological processes and therefore can be seen as direct processing. The study of visual illusions has led us to believe that it is a combination of top down and bottom-up processing which constructs our perceptual processes. Such is the case with the Kanizsa triangle, where physiological processes allow us to perceive lines which aren’t really there and our knowledge of depth perception also leads us to view the triangle.
Perceptions can be ambiguous, that is we can perceive some images in more than one meaningful way. A good example of this is Rubins Vase/Profile illusion, in this illusion; the image can be perceived as either two people facing each other, or as a vase. It is hard to seeing both versions simultaneously and the image alternates between the two possibilities. Figure- ground differentiation allows us to change our perception of the image reversibly and choose which object we wish to see.
How the information is processes and hence which object is recognised, is determined by which region of the image one focuses on. This illusion shows that perception is not purely based on the image projected onto the retina, the meaning of the representation can change depending on what region of the picture you choose to focus on as the figure. Another idea put forward by Biederman, suggests that whether we look at lines as concave or convex determines how we perceive the image. Curvilinity is a nonaccidental property, in that this 2d aspect that represents properties of the real world.
Sometimes the assumptions of nonaccidentalness are false, this gives rise to visual illusions such as the impossible trident. When we try and view this as a 3d object we become confused. This is because perceptual organisation, leads us to view the different components separately, however when we come to put them together, the components do not fit. If the trident was too short this illusion wouldn’t work as we would be able to view the different parts together. It is the assumption that the trident is a 3d object that leads us to get confused.
However cross cultural studies have shown that some rural tribes are not fazed by these impossible objects, this is because they perceive in 2d. These 2d perceivers show that depth perception is not innate, we learn how to perceive depth from our environmental stimuli. Once a new perception has been learnt; it becomes almost impossible to revert back to the original perception, as is the case with the Dalmatian in the snow illusion. On first impressions this picture merely looks like a random arrangements of black shapes on a white background.
Yet once you have been told what the picture actually is, it becomes very difficult to perceive the picture in its basic terms, as you did to begin with. This demonstrates how perceptual learning takes place within an instance, and once meaning has been assigned to the image, it is almost impossible not to perceive this when you look at the picture. On the other hand Gregory’s inverted face, 1974, appears to be a normal face, due to the lighting and ease of interpretation, however, once we know that the image is actually inverted, it is still difficult to perceive it that way.
The hollow face seems improbable, so the brain dismisses this hypothesis and perceives the face in a meaningful manner. This demonstrates how perceptual processes are influenced by passed experiences. Sometimes we can perceive things that are not present, this is highlighted by Hermann’s Grid. Here, a grid of black squares with white lines running between the squares appears to have grey dots in the intersections between for squares. These grey dots however are not real and disappear whenever you look at them directly. This can be explained on a cellular level.
The retina is made up of ganglion cells, when light shines onto the centre of these cells it promotes activity, however, an stimulation from surrounding cells inhibits this activity- this is known as lateral inhibition. At the intersections, the ganglion cell activity is inhibited from all four directions, meaning it looks comparatively darker than the white lines. Hence we perceive the illusory black dots. However, we also perceive imaginary objects on more than a physiological level. The Kanizsa Square, a cognitive illusion, allows us to something, namely a square, to be present, when it actually isn’t.
The illusion consists of four incomplete circles arranged at 90 degrees to each other, yet we perceive the image as being four black circles with a white square on top, overlapping with the circles. Here, the illusory lines can be accounted for by the contrast of black and white (bottom-up processing) as well as higher cognitive functions which infer the presence of complete shapes. This is a very strong illusion and it is very difficult to try and imagine the figure in its basic form. Illusions such as this allow us to read a text by its shadow alone.
This gives us more insight into spatial perception and how shadowing and interposition can act as a cue for the depth. Our perceptual processes appear to have a preference to perceive objects in 3d rather than flat, this is illustrated by the Muller lyer illusion, which is composed of two lines. One of the lines has outward pointing fins (similar to arrow heads), whilst the other has fins pointing inwards. This gives the illusion of one line being longer than the other. When presented with the illusions, participants see the lines in 3 dimensions, for example, as being the inward and outward facing corners of a room.
The figure representing the inside corner is assumed to be further away than its fins, where as the outside corner is assumed to be closer to us than its fins. When we process this information we use a principle known as size constancy (ability to perceive an object as being the same size regardless of whether it is closer or further away from you). As both lines are the same size, this notion of size constancy tells us that the line with the outward facing arrows is longer as it is further away. This phenomenon is known as misapplied size constancy.
This theory relies on use of depth cues, however, it had been noted that when the fins are replaced with circles, similar effects are observed. The circles add little depth perception, but this can be explained due to our eye movements. We may not get our fixation point exact when viewing the tips of the lines, so when the circles are placed further apart we have to move our eyes further, this gives us a sense that the line is longer. Motion after effects can be produced by visual illusions; this is when one perceives an inanimate object to be moving after staring at a moving stimulus for a period of time.
Joseph Plateau developed the first version of this illusion with a spinning spiral. After looking at the spiral for one minute, participants diverted their eyes to the back of their hands and reported that their skin appeared to be moving. Further study into the physiology behind this phenomenon revealed the following explanation. Motion neurones that signal motion in the direction that the spiral is moving become fatigued due to continual stimulation. The brain represents motion in terms of the activity of neurones coding for clockwise motion relative to that of those coding for anticlockwise motion.
If the neurones coding for clockwise motion become exhausted after a period of continuous activity, a disruption between the two groups is created. This means that when you then look at a motionless object, the neurones coding for clockwise motion remain inhibited in relation to the others. The brain therefore interprets that the object is moving in the opposite direct to the original stimulus. However, if this explanation is the case, it doesn’t explain why, after travelling on the motorway for long periods of time, an abrupt halt does not lead us to perceive the world moving backwards.
Visual illusions help us to distinguish between sensations, as direct experiences of stimuli, and perception, which ultimately organises the sensory information within the brain to give it meaning. We need to be careful not to assume too quickly that what we see is reality, visual illusions demonstrate that our visual system is not merely a camera. We do not see a retinal image but a combination of sensory signals that gives rise to an integrated model of what we perceive.
There are conflicting views of top-down and bottom-up processing with regards to perceptual processes, however, it is possible for the two to complement each other, “Perception is not just a single task… but contributes in many ways to everyday life… Some of these… are obviously more difficult than others and it seems likely that some can be accomplished directly, as Gibson maintained, whilst others may require sophisticated internal knowledge and are thus better described by the indirect approach. “3 This is a view that has been supported by the study of visual illusions.