Do the Blind Literally "See" in Their Dreams? A Critique of a Recent Claim That They Do

Nancy H. Kerr

Oglethorpe University

G. William Domhoff

University of California, Santa Cruz

NOTE: If you use this paper in research, please use the following citation, as this on-line version is simply a reprint of the original article:
Kerr, N., & Domhoff, G. W. (2004). Do the blind literally "see" in their dreams? A critique of a recent claim that they do. Dreaming, 14, 230-233.


This article provides a critique of a recent inaccurate claim that the congenitally blind literally "see" in their dreams, which flies in the face of findings that were established in 3 careful previous studies. It first shows how this claim arose through a blurring of the distinction between actual seeing through the visual system and imagery that preserves spatial and metric properties without specific reliance on the visual system. It then discusses the 3 mistaken reasons for this blurring. This correction is important beyond the specific issue of seeing in dreams because the original findings lend important support for a cognitive theory of dreaming by showing that the imagery necessary for dreaming develops between ages 4 and 7.

Three careful sleep laboratory studies (Amadeo & Gomez, 1966; Berger, Olley, & Oswald, 1962; Kerr, Foulkes, & Schmidt, 1982) and at least one rigorous study of home dream reports (Hurovitz, Dunn, Domhoff, & Fiss, 1999) have shown that congenitally blind dreamers and those who became blind in infancy do not have visual imagery in their dreams, whereas those blinded in adolescence or young adulthood often retain visual mental imagery in their waking life and in their dreams. These controlled experiments confirm what has been reported in a number of earlier self-report studies reviewed by Kirtley (1975), who concluded on the basis of his extensive appraisal that individuals blinded before the age of about 5 report no visual imagery in dreams as adults, whereas those blinded after about the age of 7 are likely to retain visual imagery in dreaming.

According to Foulkes (1999), these studies have theoretical implications beyond the issue of blindness because they suggest that the mental imagery necessary for dreaming develops between the ages of 4 and 7. This suggestion fits with his finding that preschool children awakened in the sleep laboratory rarely report dreams and that the reports are bland and static on the few occasions on which they do recall dreams (Foulkes, 1982, 1999). Thus, the findings on blind dreamers add to the support for a cognitive theory of dreaming (Antrobus, 1978, 1991; Foulkes, 1985).

It is rather surprising that the findings on the lack of visual mental imagery in the dreams of congenitally blind individuals have been challenged in a recent home-report, polysomnographical study in which the authors concluded that "the congenitally blind have visual content in their dreams" (Bértolo et al., 2003, p. 277). However, Bértolo et al. have used the terms visual image and visual content in a way that is not consistent with previous studies, and they have drawn inappropriate conclusions from the fact that the congenitally blind can draw images of their dream content. They also have misapplied a system of content analysis. Because their new claim has implications for a cognitive theory of dreaming, it is the purpose of this brief article to show that their results are in fact consistent with the earlier studies and do not support conclusions that congenitally blind people can "see" in dreaming or waking imagery.

Bértolo et al. (2003) cited a number of studies of waking imagery abilities in the blind as background for their study of dream imagery. The waking imagery studies that they cited illustrate the well-documented finding that the performance of congenitally blind people on a variety of imagery tasks is similar to performance by sighted individuals, although not always identical. Both blind and sighted individuals, for example, have been shown to be capable of using imagery as a mnemonic and of performing a variety of tasks that require imaging shapes or objects that may change in orientation or position in space. What Bértolo et al. failed to note, however, is that the research studies they cited did not lead the original authors to the conclusion that blind people experience visual imagery. Instead, researchers studying waking imagery in blind and sighted individuals have generally concluded that congenitally blind individuals' imagery has characteristics that are functionally equivalent in many ways to the characteristics of visual imagery reported by sighted individuals. The images of totally congenitally blind individuals, however, lack the uniquely visual characteristics such as color and brightness and result in slight differences from the performance of sighted individuals on several imagery tasks. Researchers therefore use the terms visual image and visual content to mean an image or experience whose properties are similar to those of an object or scene viewed visually (with one's eyes). Terms such as spatial image, analog image, and even visuospatial image have been used to describe imagery that preserves spatial and metric properties without relying specifically on the visual system. When Bértolo et al. (2003) used the term visual image, they often blurred the distinction between the two descriptive categories. They did so for three mistaken reasons.

First, the authors reported that blind dreamers can draw the scenes and figures in the dream as well as can sighted people who have their eyes closed while they are drawing. On the basis of this evidence, they concluded that blind people must experience visual imagery in the dream. This claim is in direct contradiction to conclusions based on a large body of research investigating how blind people create and interpret pictorial drawings of waking experiences. Researchers generally employ raised-line drawings that blind people can trace with their fingers. Kennedy (1993, 1997), for example, has conducted extensive empirical research showing not only that blind people are capable of drawing two-dimensional figures but also that their drawings are similar to the drawings of sighted individuals in the depiction of depth, motion, perspective, vantage point, surfaces, contours, edges, and other characteristics. Kennedy did not interpret his findings as evidence that blind people experience visual imagery. Rather, he attributed the accuracy of drawings by blind people to the overlap in information obtained through visual and tactile perceptual systems. According to Kennedy, even though vision and touch are two different perceptual systems, one responsive to light waves and the other to pressure, they are both processed in an area of the brain that encodes and integrates the common elements of information. Kennedy suggested that the area of the brain that is responsive to several modalities, not just to vision, could more accurately be called "multimodal" or "amodal." Kennedy's empirical findings and the interpretations of them are consistent with the research on imagery described in the previous paragraph.

Second, Bértolo et al. (2003) reported the results of a content analysis using the visual activities category of the Hall and Van de Castle (1966, p. 90) coding system, which they accurately stated to be the most comprehensive and widely used system for categorizing the elements in dream reports. However, they did not allow for the metaphoric and idiomatic uses of visual terms. Blind people sometimes do employ visual language to describe their dreaming experiences, but they also use visual language to describe waking experiences that clearly do not include "seeing" things. Blind people talk about "watching TV," "keeping an eye on things," "seeing what you mean," "taking a look at something," and "visualizing a scene," although these visual terms do not imply a visual component. These colloquial expressions are simply the most convenient phrases that blind people have for communicating their experiences. It should be noted that blind people also use the term visual to describe imagery experiences that do not fit neatly into any other category of sensory experience (Kerr & Johnson, 1991). Blind people who know, for example, that a star appears as a small spot in the night sky may include stars in an image of a nighttime scene. Although the blind person's conception of a star may derive only from descriptions, he or she understands it as a visual phenomenon and labels it accordingly. Likewise, blind people describing dream imagery favor words such as visualize to explain their awareness of the detail of the dream environment without having to move around or touch particular aspects of the dream setting.

Hurovitz et al. (1999), who made extensive use of the Hall and Van de Castle (1966) system in their study of 372 dream reports from 15 blind adults, 6 of whom were congenitally blind and without any ability to perceive light, did not use the system for coding visual references because it does not explicitly exclude the scoring of metaphoric visual terms as actual incidences of visual activity. They therefore found no indication of visual imagery in the 6 congenitally blind participants who could not perceive light. They provided examples of how the failure to exclude metaphoric uses of visual terms would have biased their results. Similar examples are provided by Kerr et al. (1982).

Third, Bértolo et al. (2003) provided empirical support for the assertion that there is correlational evidence that alpha activity is attenuated in the visual cortex when "visual" content is reported in the dreams of both blind and sighted individuals. However, as Kennedy (1993, 1997) has pointed out and as Bértolo et al. acknowledged, activity in the visual cortex is not restricted to visual processing (Bértolo et al., 2003, p. 282). The attenuation may also be related to auditory, haptic, somatosensory, or tactile processing or, in fact, to generalized imagery processing that is tied to no specific modality. Because we do not know the context in which the visual words were used by blind dreamers in Bértolo et al.'s study, it is difficult to speculate about why visual words were most closely associated with attenuation of alpha activity. If we eschew the term visual cortex and adopt Kennedy's terms instead, the correlation may reflect the fact that blind people are most comfortable using visual terms for experiences that are multimodal or amodal in character.

In the discussion section of their article, Bértolo et al. (2003) referred to the dream images of blind people as "virtual images," characterizing these images as "concepts capable of graphical representation" (p. 282). We agree with both of these conceptualizations of the waking and dreaming imagery experienced by blind individuals. However, we conclude that there is no evidence that congenitally blind people experience waking or dreaming imagery of a visual character (as though seeing something with their eyes). Virtual imagery integrated via the brain area traditionally labeled visual is one thing; visual imagery is another. This is a semantic distinction but an important one because of the potential confusion about the visual terms in the vocabularies of blind people who have no sensory experience as a basis for those terms.

The potential confusion is not present in dream reports of individuals who are visually impaired from birth but who retain some ability to see visual characteristics such as brightness and color and are able to match waking visual experience to dreaming visual experience. These individuals universally report that visual images in their dreams appear to them as they would in waking life. They can see things in dreams with no more clarity or detail than they could see in wakefulness, yet they know the details of the dream environment through the integration of information from other sensory systems. Uniquely visual imagery is dependent on uniquely visual experience. We therefore strongly believe that the term visual imagery should be reserved for imagery that is phenomenologically similar to objects seen with one's eyes, the only sensory receptors capable of receiving and encoding the information conveyed exclusively in light waves.

In addition, we reassert the importance of the earlier sleep laboratory findings concerning the lack of visual imagery in those who are blind before age 4 as evidence for the idea that dreaming is a gradual cognitive achievement that requires the development of visual and spatial skills and other forms of imagistic skills as well.


Amadeo, J., & Gomez, E. (1966). Eye movements, attention, and dreaming in subjects with life-long blindness. Canadian Psychiatric Association Journal, 11, 501-507.

Antrobus, J. (1978). Dreaming as cognition. In A. Arkin, J. Antrobus, & S. Ellman (Eds.), The Mind in Sleep: Psychology and Psychophysiology (pp. 569-581). Hillsdale, NJ: Erlbaum.

Antrobus, J. (1991). Dreaming: Cognitive processes during cortical activation and high afferent thresholds. Psychological Review, 98, 96-121.

Berger, R., Olley, P., & Oswald, I. (1962). The EEG, eye movements, and dreams of the blind. Quarterly Journal of Experimental Psychology, 14, 183-186.

Bértolo, H., Paiva, T., Pessoa, L., Mestre, T., Marques, R., & Santos, R. (2003). Visual dream content, graphical representation and EEG alpha activity in congenitally blind subjects. Cognitive Brain Research, 15, 277-284.

Foulkes, D. (1982). Children's Dreams. New York: Wiley.

Foulkes, D. (1985). Dreaming: A Cognitive-Psychological Analysis. Hillsdale, NJ: Erlbaum.

Foulkes, D. (1999). Children's Dreaming and the Development of Consciousness. Cambridge, MA: Harvard University Press.

Hall, C., & Van de Castle, R. (1966). The Content Analysis of Dreams. New York: Appleton-Century- Crofts.

Hurovitz, C., Dunn, S., Domhoff, G. W., & Fiss, H. (1999). The dreams of blind men and women: A replication and extension of previous findings. Dreaming, 9, 183-193.

Kennedy, J. (1993). Drawing and the Blind: Pictures to Touch. New Haven, CT: Yale University Press.

Kennedy, J. (1997). How the Blind Draw. Scientific American, 276, 59-65.

Kerr, N. H., Foulkes, D., & Schmidt, M. (1982). The structure of laboratory dream reports in blind and sighted subjects. Journal of Nervous and Mental Disease, 170, 286-294.

Kerr, N. H., & Johnson, T. (1991). Word norms for blind and sighted subjects: Familiarity, concreteness, meaningfulness, imageability, image modality, and word associations. Behavior Research Methods, Instruments, & Computers, 23, 461-485.

Kirtley, D. (1975). The Psychology of Blindness. Chicago: Nelson-Hall.

Go back to the Dream Library index.

dreamresearch.net home page dreamresearch.net contact info