Since mental images occur in conscious states that are generally distinct from perceptual states, their ontological description depends upon how we understand consciousness and its non-perceptual objects. These mental objects do not have prior existence outside of consciousness and do not share in many of the properties of ordinary objects. Mental images are momentary "creations" of consciousness (although some are induced by sensory conditions). While some have argued that mental images really do have some of the properties of pictures, and others have argued that mental images are really verbal descriptions in disguise, I argue that they exemplify properties from both the pictorial and verbal realms. I define mental images as "visual mental particulars to which linguistic content is provisionally bound." This definition must be hedged, however, because each type of mental image has a different set of properties. Most contemporary discussions have not paid sufficient attention to imagery types). See Tables below.
Many contemporary treatments of mental images have failed to do justice to the fact that they have intentional depth. Most contemporary philosophers have insisted (correctly) that "pictures in the head" can neither literally exist nor serve (by themselves) as vehicles for meaning. As a result, contemporary philosophers have sought alternate explanations of what it means to have mental content and they been reluctant to grant mental images any status as operative components in thought. This conclusion, to put it bluntly, is simply too hasty. I call for a reexamination of the properties of mental images and a return to common sense, which holds that mental images do play a role in thought and are essential to many aspects of mental life. This reexamination, inspired both by common sense and traditional psychology, reveals that mental images are not "pictures," or pure visual data, but "objects" that have intentional depth. They are more like topics, exemplars, or designated symbols than pictures. Since they represent something absent to perception, their present "properties" are merely ways of describing our conscious experience as it is directed toward a mental particular. These mental particulars are (often, but not exclusively) created through acts of the will that change our directedness toward topics of thought, memory, or imagination. Thus, the identical visual qualia of a mental image might occur when one remembers "the lion at the zoo" (specific) and when one imagines "a lion" (non-specific), but these are two different images because they are two different types of intentional states -- or, as I prefer to understand it, two different types of visual mental images.*
The common sense view, that mental images indeed "exist" and "have meaning" turns out to be substantially correct once it is corrected by philosophic distinctions. I follow, to a certain extent, the view of H. H. Price, who held that the mind thinks in and through symbolic particulars, and that mental images are among these symbolic particulars.
The complete explanation of how mental images can do what they appear to do and how they are generated, however, remains a mystery. Part of the mystery can be alleviated by (1) reexamining our understanding of physical reality is, or (2) by admitting that mental sensa, such as mental images, have a form of independent being in a non-physical mode (this idea is borrowed from Santayana). I argue in favor of both (1) and (2).
Other accounts have sought not to alleviate, but to eliminate the mystery. That is, they have claimed to give us a complete account of what mental images are and precisely how they function. I argue that many contemporary accounts, particularly those that have sought to explain mental images in terms of linguistic analysis or computationalism, rely on questionable suppositions about physics, reductionism, and the validity of empirical measurements of mental activity. As a result, these accounts fail to do justice to the nature and function of mental images.
---------------------------------------------------------------------
* This sort of example has been used by other philosophers to prove precisely the opposite kind of point. They argue that since the two images (i.e., as pure visual qualia) are the same, it demonstrates that images themselves have no function -- or at least that they cannot be used to differentiate memory from imagination or thought. This, I argue, is simply an error. It fails to adequately describe the form of intentional state that is involved in having an image. Having an image, I argue, includes modal awareness of its type as well as cognitive awareness of its symbolic content. The argument that two such images are the same succeeds by a trick: it treats "having an image" as if it were the sudden and unpremeditated appearance of a perceptual-like object in consciousness, initially devoid of modal awareness and symbolic content, that subsequently demands an interpretation. This, I argue, is a fundamental error.
Types of Mental Imagery
The tables below represent an inventory of types of mental imagery. The division is into types is based partially on the work of Alan Richardson (Richardson, Alan, Mental Imagery, New York: Springer Publishing Company, Inc., 1969). The proposed divisions should be thought of as fluid and somewhat arbitrary. Although the exercise is useful, the classification into "types," and the resulting object/property ontology, may be somewhat misleading because the "objects," as I have suggested elsewhere, must be understood in the context of the phenomenon of what I call "intentional depth."
Each type of mental image is associated with a number of PROPERTIES. The properties are, for the most part, simply ways to characterize our experience. We know, for example, that an after-image induced by bright light is an involuntary response; it is not under conscious control. By contrast, when we call up an image in our imagination, this is a voluntary act. We can characterize our imagery experiences, therefore, as being under various degrees of conscious control. Degrees of variation also exist for the apparent location of images in space. If we consciously call up the memory of, for example, a text book, we can "see it in the mind's eye" -- but the mental image itself does not seem to have any location in the physical space around us. On the other hand, if we form a projected image of the text book, and we imagine it as "being over there," we seem to "see" the text book in a physical location. Two of the properties of mental images are therefore degree of conscious control and apparent location in space.
The introductory material on mental images concentrates on differentiating imagery types by degree of conscious control and apparent location in space. Further information on imagery types and the terminology used to describe their other properties is available in Richardson's book as well as in parts of my own work (see advanced material, Chapter 3).
Table 1. Types of Mental Imagery -- Variable Properties
CONSCIOUS PHENOMENOLOGICAL INTENTIONAL
TYPE CONTROL QUALITY CONTENT
After-images NO sense sensation itself
Eidetic NO sensation-like object
Hallucinogenic NO sensation-like unknown
Memory YES quasi-perceptual experience
Imagination YES quasi-perceptual possibilities
Projected YES quasi-perceptual shape/size
Thought YES quasi-perceptual actualities
Table 2. Types of Mental Imagery -- Presented Properties
SPATIAL
TYPE LOCATION MODE
After-images NO single
Eidetic YES single
Hallucinogenic ? multi
Memory NO amodal
Imagination NO multi
Projected YES single
Thought NO single