Custom Essays and Free Coursework

Psychology essays | Constructivism and Perception

“Constructivists” such as Helmholtz, Rock and Gregory claim that incoming visual data are insufficient for perception to be achieved, and that the brain must supplement these data with top down processing. Outline the evidence for these claims and critically evaluate it.

This essay will begin by defining the terms perception, top down processing, and bottom up processing. Constructivism will then be introduced alongside the theoretical stance of Helmholtz, Rock and Gregory and the value of top down processing will be appraised by examining the relative utility of an opposing paradigm, the nativist approach to perception. The essay then reaches a natural conclusion summarizing the value of top-down processing in light of the evidence presented.
Perception is the way in which individuals attempt to “gain an accurate fix on the world, to recognize predators and prey and to aid survival” (Zimbardo, 1988:189). Perception occurs via two mechanisms known as ‘top-down’ and ‘bottom-up’ processing. Top down processing is where higher-level knowledge based information predominates (Bernstein, 1997:144) and bottom up is where the information that senses gather about the world by detecting various forms of energy such as sound, light, heat and physical pressure are deemed as sufficient to enable us to perceive (Bernstein, 1997:88).
Constructivists are traditionally affiliated with top down processes, an affiliation that began over 100 years ago, instigated by Herman Helmholtz (1821-1894). Helmholtz was the first to argue “human perception has few, if any, innate components, being instead the product of experience” (Warren & Warren; 1999:3). Figure 1 provides a clear example. The incoming information from the dots on this page is insufficient for perception via our bottom up processing. Instead top-down processing is required to supplement the data to aid perception.

Figure 1: Just dots?

(Taken from: Perceptual Ambiguity, 2003)

If asked: “can you see a dog?” our previous experience of what a dog looks like i.e. four legs, tail etc. aids our perception in this example. Helmholtz named this phenomenon ‘unbewusster schluss’ (unconscious interpretations), which roughly means that basic perceptual processes, (like that of previous experience, is a fundamentally unconscious process (Warren & Warren, 1993). Helmholtz was also the first theorist to introduce the concept ‘size constancy’. Size constancy is the ability to perceive the true size of an object despite variations in the size of its retinal image. Helmholtz claims several elements of knowledge facilitate this ability, including prior experience, an example of which will be provided shortly (Zimbardo, 1988:206). However, Helmholtz’ work has since been criticised because his work lacked physiological recognition.

Moving on, the contribution of Rock stems from his interest into the reasons why objects look so different when they are upside down or tilted (Rock, 1974; as cited in Rock, 1990: 113).
Rock proposed in his 1974 paper “The perception of Disorientated Figures” that when we look at a square for example, and then tilt out head 45 degrees, although the image on the retina has changed to represent what we label ‘a diamond’ “we automatically compensate for the tilt in much the same way as we compensate for the size of distal objects” (Rock, 1974; as cited in Rock, 1990:121). We do this by ‘egocentrically’ applying a top, bottom and sides to the image.

Rocks viewpoint differed from the dominant paradigm of the day in that many psychologists believed that perception could be reduced to the perception of contours and light intensity on the retina i.e. bottom up processing. Rock, with Helmholtz before him and Gregory today, proposes that even a very different retinal image will still allow recognition. To demonstrate the clear validity of this proposition and the use of top-down processing…

Figure 2: Do you recognize this figure?

Although the retinal image is not of the ‘A’ as we are used to seeing, we still recognize it as being so due to the egotistical top, bottom and sides we assign to it.
Gregory proposes that perception is a form of hypothesis testing. In short, signals are received by the sensory receptors e.g. eyes, ears, etc. These then trigger neural events where appropriate knowledge, such as previous experience interacts with the sensory input to create psychological data. Gregory’s theory of hypothesis testing is based upon a number of arguments which include: We can interpret a table with four legs for example even if we can only see one leg and the corner sticking out from behind something else, therefore the information presented to the senses is supplemented. Gregory also argues that perception can be ambiguous and that “…if a single pattern can induce two different percepts, perception cannot be tied to stimulation in a 1:1 manner” (Zimbardo, 1988).

Figure 3: Girlfriend or mother-in-law?

Can you see both an old woman and a young lady in this picture despite the same retinal image being projected onto the eye?

Robert Leeper (1935) conducted research based on the famous ambiguous picture of the old/young woman. When subjects were exposed to either an accentuated version of the ‘old woman’ / ‘young woman’, for both groups, once presented with the ambiguous picture, the version previously exposed to was more readily perceived by the subjects strongly suggesting that previous experience influenced what they saw in the ambiguous picture (Bernstein, 1997; 146).
Moreover, perception can also be influenced by the context the stimuli are surrounded by, the individual’s motivation, and even long-term deprivation. For example:

Figure 4: Is the figure in the middle a letter or a number?


A

I2 I3 I4

C

(Adapted from Gordon 1993:124)

The figure in the center is either a ‘B’ or the number ‘I3’ depending on the context. Therefore the sensory information is sufficient for the figure to be perceived, the context has to be taken into account before the figure becomes meaningful. Furthermore Sanford in 1968 demonstrated the impact of motivation on perception. Ambiguous pictures were shown to groups of children at different times of the day. The children were asked to write down what they had seen. Twice as many food related responses were given before lunch in comparison to after meal times. And finally, Bruber & Goodman, 1947 asked two cohorts of children to adjust the size of a light to match a variety of coin sizes. The wealthy cohort of children adjusted the light to a size smaller than that of the coins and the poorer children adjusted the light to be larger than the coins. It was suggested that the poor children reflected how much they valued the monetary element of the coins in the light sizes chosen (Zimbardo, 1988).
The evidence for the role of top down processes in perception is not without it’s critics. The greatest challenge for constructionists is to answer the chicken and egg question; “If perception is essentially constructive then how does it get started?” (Warren & Warren, 1968).
Also, many of the studies supporting the constructivist approach involve presenting visual stimuli very briefly thus reducing the impact of bottom up processes. Furthermore, the artificial environment in the lab may confound results or reduces the generalisability of the findings. Perhaps the greatest critique regarding the utility of top down processing comes from the theory of Ecological Optics by J.J Gibson.
Gibson’s perspective places emphasis upon the natural environment and the richness of stimulation available to active perceivers. In particular, we pay attention to patterns of light Gibson calls the ‘optic array’. The optic array describing the way in which light provides us with “unambiguous or invariant information about the layout of objects in space” (Eysenck & Keane, 1988:47) via the “physical structure of light impinging on the observers eye” (Gregory & Coleman, 1995:13). Characteristics of the optic array include for example texture gradients and natural gradients e.g. the texture of the road close to you is very vivid in comparison to when you look at the road at a point further in the distance. According to Gibson we do not stop and think or assess aspects of texture, we just perceive distance automatically based upon the deterioration of clarity.
Gibson also proposed a number of principles as being largely innate or ‘bottom up’ processes, which aid us in our perception of depth as opposed to previous experience, or ‘top down’ processes, For example; Relative size: if two objects are assumed to be the same size e.g. 2 adults then the one producing the larger image on the retina will be assumed to be closer; height in the visual field: On the ground, more distant objects are usually higher than those objects nearby. And linear Perspective: The closer together two converging lines are, the greater is the perceived distance.

Figure 5: The Ponzo Illusion


Most people perceive line A to be longer than B as a result of the depth cue ‘linear perspective’ which nativist’s such as Gibson argue to be instigated by innate mechanisms (Zimbardo, 1988:194).
If this were to be the case and there was no need for brain to supplement data with top-down processing then why do some images produce multiple interpretations (see: figure 3). Gibson’s approach has little to say about the nature of such illusions and clearly ignores the perceivers mind. How would he explain figure 3 or hungry children’s change in perception?
To depict the value of top down processing and previous experience more clearly in the role of perception the essay will return to the element of Helmholtz’ and ‘size constancy’. Helmholtz claims size constancy occurs as a result of previous experience and Gibson via our innate ability to perceive depth cues via the environment. Which viewpoint is most accurate? Is incoming data sufficient for perception to be achieved? Colin Turnbull in 1961 conducted research with pygmies and the innateness of our ability to perceive depth cues. The pygmies studied by Turnbull had lived in the forrest their whole lives and had not experienced long distance viewing before. For the pygmies it was a new experience to see across planes for miles. Turnbull selected one pygmy and took him to where he could see for miles and in the distance there was a boat. Having never experienced or used size constancy before the pygmy refused to believe that the boat was miles away and believed it to be witchcraft as the boat became larger before his very eyes (Hayes & Orrell, 1993). This evidence suggests that there is certainly a need for previous experience via top down processing to aid out perception because in this example the depth cues or ‘bottom-up’ processing was insufficient for perception to be achieved.
In conclusion, ‘top-down’ and ‘bottom-up’ processes represent two overlapping constructs rather than two separate elements of perception. Both top down and bottom up processes interact to optimize performance” (Sarter et al, 2002). If perception were bottom up only the world would be perceived as the same by everyone and not to our own personal circumstances.
Alternately, if top down processing were used alone, we would live in our own fantasy worlds filled with only those things that we want to perceive (Zimbardo, 1988:217). Therefore it can be concluded that given the evidence presented in this essay, in some cases incoming visual data is insufficient for perception to be achieved and that top down processing is required to aid perception, especially when the stimulus is ambiguous.