Visual Imagery

The Nature of Visual Imagery Explained

Visual Imagery is the natural ability to invent or re-create an experience that resembles the experience of actually perceiving an object or an event, with no retinal input.

These mental representations resemble actual depictions both subjectively and functionally and play an important role in remembering and problem solving. More broadly, they can, by virtue of their content, powerfully guide the flow of thought.

The Subjective Qualities of Visual Images

Visual images can be (and usually are) created in the absence of an actual visual stimulus—and so one can create an image of (or “visualize”) an elephant even if none is nearby. One can also create images that alter things that are in view (and so one could, for example, imagine this page with all the words printed in green ink).

More ambitiously, one can create images of things that do not exist at all (e.g., an image of a unicorn). In some cases, images are created deliberately (and so someone can, if they choose, call up an image of a beautiful sunset); in other cases, the images arise spontaneously (perhaps triggered by someone else’s mention of a sunset).

Visual images are not hallucinations—the person experiencing the image can tell that the image is “in their head,” and not a real sight. Nonetheless, there is a strong subjective resemblance between visual images and actual sights. This is reflected in the way people commonly talk about their visual images, and references to mental pictures or the mind’s eye have been common at least since Shakespeare’s time (e.g., Hamlet, Act 1, Scene ii).

It is noteworthy that people feel these terms are apt; this is a strong indication that the conscious experience of having an image does resemble the experience of seeing. More specifically, the imaged object or scene seems to be “viewed” from some particular vantage point and is typically “seen” against some background; objects in the image have colors, shapes, and surface textures that are immediately “visible” and so on.

Experimental Studies of Visual Images

Here as elsewhere, though, researchers are cautious in how they interpret these self-reports on conscious experience. However, the self-reports can be corroborated via the appropriate experiments, and the data make it plain that visual images do share many functional properties with actual sights as one would expect based on the self-report.

For example, participants in one study were first asked to memorize a map of an island, including the location of several landmarks on the island. With this done, participants were asked to create a mental image of the map and then to “scan” their image from one landmark named by the experimenter to another. When these scanning times were carefully recorded, they showed a strong linear relationship between the time needed for each scan and the distance between the relevant pair of landmarks on the original map. This result confirms that the image accurately represents all the spatial relations on the map—and so pints close together on the map are functionally close on the image; points further apart on the map are far apart on the image. In this fashion, the image seems truly to depict the layout (and thus all the shapes and patterns) in the scene that is being represented.

Experiments also indicate that images function as though they have an identifiable “view point,” just as an actual visual scene would. This is evident, for example, in the fact that aspects of the image that depict larger objects or objects that are at the front of the imaged scene are more rapidly accessible. Likewise, participants need time to “zoom in” on an image to inspect small details or to “zoom out” to survey the larger scene, suggesting that again, just like actual scenes, images are inspected from a particular “viewing distance.”

Visual images also respect spatial layout in another regard—in the pattern of transformations in the image. In many studies, for example, participants have been shown two forms and asked if the forms are the same shape but viewed from different vantage points, or actually different shapes.

In these studies, participants seem to imagine one of the forms “rotating” into alignment with the other and then make their decision only after this imagined rotation. The imagined rotation itself seems to take place at a constant velocity, and so the time needed for these judgments is a linear function of the angular distance between the initial orientations of the two forms being compared.

Neuroscience Studies of Imagery

What brain mechanisms might lie behind the behavior data described so far?

In the view of many researchers, the various parallels between visual imagery and actual vision invite the hypothesis that these two activities rely on similar brain circuits, and several lines of evidence confirm this claim. Neuroimaging results indicate an enormous overlap between the brain sites activated during visual imagery and sites activated during ordinary vision. Likewise, brain damage often has parallel effects on imagery and vision. Thus, patients who (because of stroke) have lost the ability to perceive color often seem to lose the ability to imagine scenes in color.

Similarly, if as a result of occipital damage patients have a blind spot in a particular region of visual space or some restriction of the extent of visual space, they are likely to have a corresponding limit on their visual imagery. Further confirmation comes from studies that use transcranial magnetic stimulation to produce temporary “lesions” in visual cortex. Not surprisingly, this procedure causes a disruption of vision, but crucially, it also causes a parallel disruption in visual imagery.

Differences Between Visual Imagery and Vision

Even acknowledging these important parallels, however, there are also differences between visual imagery and vision and between mental pictures and actual pictures. For example, some discoveries that are easily made from a picture (the reinterpretation of an ambiguous drawing) are enormously difficult if the participant is relying on a mental image of that picture.

According to some authors, this is because the image—as a mental representation—is accompanied by a “perceptual reference frame” that organizes the depiction, specifying the figure/ground organization, how the form is parsed, where the form’s top is located, and so on. This reference frame guides how the image is interpreted and so can place obstacles on image reinterpretation (or image-based problem solving of any sort). Pictures do not on their won have this sort of reference frame (the frame must be created by the perceiver), and so pictures (“unorganized depictions”) are more readily reinterpreted than images (“organized depictions”).

A further distinction between images and pictures is suggested by cases in which brain damage has disrupted someone’s vision but spared their ability to perform imagery tasks; the reverse pattern (disrupted imagery but intact vision) has also been observed. In addition, studies have often documented normal or near-normal performance on various imagery tasks from individuals who have been blind since birth—individuals who are unlikely to be relying on a picturelike mode of representation.

Findings like these have led several researchers to propose a difference between visual imagery and spatial imagery. The former type of imagery yields a representation that bears a closer resemblance to a picture (and so visual images, but not spatial images, depict an object’s color or surface texture), and the processes needed to create visual images rely heavily on brain sites ordinarily used for actual vision.

Spatial imagery, in contrast, relies more heavily on brain sites ordinarily used for guiding movements through space (both overt bodily movement and covert movements of attention). Presumably, individuals blind since birth rely on spatial imagery, not visual, and likewise, spatial imagery is what allows patients with disrupted vision (because of brain damage) to perform normally on many imagery tasks.

The distinction between visual and spatial imagery is also valuable in explaining a different point—namely, the ways in which (neurologically intact, sighted) individuals differ from one another in their imagery abilities.

Both self-report and behavioral testing indicate that this variation is considerable: Some people report rich, vivid, visual images; some report no visual images at all. Some people perform well on paper-and-pencil tests requiring them to imagine folding pieces of paper or spinning forms; others perform much less well. Recent studies suggest that these individual differences need to be assessed separately for visual and spatial imagery (especially since self-report measures of imagery are powerfully shaped by someone’s strengths in visual imagery, while the paper and pencil measures often reflect someone’s ability in spatial imagery).

The Role of Visual Imagery in Cognition

Finally, what is imagery’s role within the broader context of mental processing?

The answer has many elements. For some purposes, imagery is essential for remembering. (Imagine someone trying to describe a previously viewed face; that person may have no choice but to call up an image of the face and attempt to describe the contents of that image.)

For other purposes, imagery may not be essential for memory but is nonetheless enormously helpful. We know, for example, that easily imaged words are easier to remember, and that deliberate attempts to form images of the to-be-remembered material usually aid memory. (Indeed, the use of imagery is a frequent component of many deliberate mnemonic strategies Opens in new window.) Imagery also seems to play a role in autobiographical memory Opens in new window, and memories of past episodes often take the form of images of those previous episodes.

In addition, imagery can play a role in problem solving Opens in new window. This is plainly the case when the problem involves spatial arrangement. (Imagine trying to decide whether a sofa, viewed in a store, will fit well in your living room. Most people would try to solve this problem by visualizing the sofa in place to “see” how it looks.)

However, imagery also plays a role in other sorts of problems, including a variety of mathematical word problems. Moreover, casting any problem’s elements in terms of a visual image can shape the sequence of thoughts that come to mind in thinking through the problem. For example, thinking about your pet cat in terms of an image will make the cat’s appearance prominent for you, and this can call to mind other animals with a similar appearance. If you had thought about the cat without an image, the appearance might have been much less prominent so that some other set of ideas would be likely to come to mind. In this way, the mere step of casting the problem in terms of an image can guide the selection of available ideas, and this may have important consequences for the flow of thought.

It should be mentioned, though, that there are contexts in which imagery can be an impediment to problem solving. These include cases in which someone is better served by drawing an overt sketch rather than relying on an image (in part because this helps the person set aside the image’s reference frame) and also cases in which someone is better served by relying on some more abstract, perhaps algebraic, mode of representation. Visual images can powerfully shape the flow of thought, but this does not mean that images reliably enhance or improve the flow of thought.