Psychologists have been interested in the processes that go on in the mind. Mental representation is studied as an important part of cognitive psychology. Generally, mental representation is attempted to answer what is stored in the mind by describing the form that knowledge of the world is represented in the mind (Pylyshyn, 1973). There are two basic forms of mental representations that are the symbolic representations and the analogous representations (mental imagery). Mental imagery has played a crucial role in the cognitive psychological research on the representation of knowledge in the mind. Mental imagery resembles perceptual experience but occurs when the relevant objects are not presented. There is variety of evidences show that mental representation is based on imagery. This essay will evaluate the evidences that mental representation is based o imagery. This essay begins by looking at several early studies of mental imagery followed by a brief discuss on dual-coding hypothesis and the mental rotation experiment. It will then go on to take a look at the propositional representation. The final part will focus on the depictive representation as the most influential theory of mental imagery.
Mental imagery is one of the aspects that have been studied for centuries. Over 2000 years ago, Aristotle regarded mental imagery as the most important medium of thought (Eysenck& Keane, 2005). The first systematic research on the mental imagery (visual imagery) is traced to Galton (1883), whom did his experiment by giving questionnaires to 100 people. Subjects were asked to remember their breakfast table and answer some questions about the images they had. By doing this, Galton (1883) developed a measure of imagery, which had taken the specific individual differences into consideration, such as sex, age and so on. He found that individual differences have a notable influence on the ability to form the mental images because some of the subjects reported no conscious imagery at all. Like Galton's (1883) study, many other early studies was depended on the incipient technique of introspection (Eysenck& Keane, 2005), in which subjects report their own conscious experience. This approach seems to be unfruitful and vulnerable compared to behaviouristic methods used later. By using this technique, subjects are easily disrupted and mental representation may be influenced. Therefore further researches are carried out by involving behavioral measurement such as reaction time to investigate whether imagery is a basic form of mental representation.
An important hypothesis was carried out later in 1971 by Paivio, which is the dual-coding theory. This theory evolved from Paivio's researches (1963, 1965) on the role of imagery in associative learning (Paivio, 1991). The basic principle of this theory was that the human mind creating separate representations for information processed. The incoming information may be organised into knowledge either in a verbal system or a visual system. Different kinds of information are contained in different systems. Visual and verbal information are processed differently and along distinct channels. The dual-coding theory indicates that visual imagery is one of the two basic forms in mental representation.
After the dual-coding theory, a series of experiments were done to investigate the mental imagery. One of the most outstanding experiments is the one done with showing pairs of 3-D line drawings (Shepard & Metzler, 1971). In this experiment the subjects' task was to decide whether the two images were the same or different. There were 1600 pairs generated, among which some pairs were totally different from each other while other pairs had more or less similarities. The rotation of the 3-D lines in this experiment varied from 0 to 180 degrees. By depicting the mean reaction time, Shepard and Metzler (1971) noticed that the required time for response is linear as a function of initial angular disparity. Later Cooper and Shepard (1973) extended the findings by examining whether the effects of angular disparity could be found when subjects had to recall a meaningful image rather than abstract shapes. Three letters (R, J and G) and three numbers (2, 5 and 7) were used in their experiment. The stimuli appeared in one of two ways, normal or mirrored and were presented in six orientations, 0, 60, 120, 240 or 300 degrees. Subjects had to judge whether the letter was 'normal' or presented as a 'mirrored' image. When the letters were rotated, similar results were found. The findings of Shepard and Metzler's (1971, 1973) experiments suggests subjects' response time implicating within the proportion to the spatial properties (rotation) of the figures presented. Concluding from the Shepard and Metzler's (1971, 1973) results, Kosslyn (1980) states that these experiments were done by the examination of mental representations which themselves have spatial properties (e.g. images)
However, Pylyshyn (1973) argues that no matter what the images are, it hardly can be 'a picture in our mind'. Because there are too many of the images, the brain does not have the capability to store everything being seen and there are no possible means to organize all the images. The presence of imagery is not a functional processing, it is only epiphenomenal. Pylyshyn (1973) claims that the empirical facts of Sheppard and Cooper's experiments (1971, 1973) could be explained by propositional representations. In the form of propositional representation, a person's knowledge can be represented by a limited list of propositions. More specifically, mental imagery is based on tacit knowledge which is explained as "knowledge of what things would look like to subjects in situations like the ones in which they are to imagine themselves" (Pylyshyn, 2002, 161). That means mental imagery is the result of what people know about the things in the world and what people intend the mental to represent.
At the same period, Kosslyn (1978, 1980, and 1994) carried out an extremely influential theory of mental imagery. This theory is referred to as depictive (analog) theory. Kosslyn (1994, p.5) describe this theory as "a type of picture, which specifies the location and values of configuration of points in a spaceâ€¦" The results of the experiment conducted by Kosslyn et al. (1978) showed the experimental evidence for the depictive theory. A map of fictional island with 7 specific locations (shown in appendix 1) was shown to subjects. The subjects' task is first to memorize the locations of the objects on the map by drawing their location on a blank paper. Then subjects were asked to picture mentally the map and to focus their attention on a particular location. After 5 sec focusing on a particular location, a name of another object would be presented to subjects. If this named object was on the map, they needed to examine their image of the entire map, image a little black speak zipping straight line to the second named location and to press a button when they arrived it on their image of the map. If the named object was not on the map, a second button needed to be pressed. The reaction time to press the either button was recorded. The result of this image scanning experiment showed that time to scan visual mental images increased linearly with increase of the distance to be scanned. This supports the claim that mental images have spatial properties.
In the depictive theory, images are quasi-pictorial entities that can in fact be processed and are not only epiphenomenal compared to the propositional theory (Pylyshyn, 1973). In a depictive representation, the meaning was covered via their similarities to an object -- that is part of the representation corresponding to parts of the object. According to Kosslyn (1994), all parts of an object are represented by a pattern of points. The spatial relation among these points is related to the spatial relations among the parts of the object. Further more when a depictive representation is used, not only is the shape of the represented parts available to processes, but also the size and orientation. Based on the depictive theory, visual mental image seems like has close similarities with visual perception (Eysenck& Keane, 2005). Thus further researches put forward by Kosslyn and Thompson (2003) to find out in which part of brain the depictive representation formed. The results of their study showed that the early visual cortex (consisted of primary and secondary visual cortex) is the place that the depictive representation created. The brain study of depictive representation further emphasizes that imagery is an indispensable part of mental representation.
In this essay the evidences that mental representation is based on imagery have been discussed. The first systematic research on the mental imagery is traced to Galton (1883) by recalling breakfast table followed by introducing the dual-coding hypothesis. The mental rotation studies carried out later improve the experimental methods of mental imagery studies. At the end of this easy, the propositional representation and depictive representation have been introduced in details as the main debate in the studying of mental imagery. Although the studies discussed in this essay involves in different areas of mental imagery, all of them demonstrate that mental representation is based on imagery. The scope of this essay was limited by only focusing on the visual mental imagery. It is widely known that mental imagery can occur in any form of sense, for instance auditory image, olfactory image as well as visual image.