Published in Research in Parapsychology 1985
Ed. D.Weiner and D.Radin, Metuchen, N.J. Scarecrow, 163-164, 1986

Does anything leave the body in an out-of-body experience (OBE ) or does it just seem that way? Some theories suppose that something (though what is the problem ) leaves the body and actually travels to distant locations. Alternatives include the “psychological theories” but Rogo (JP , 1982 , 29- 45) has argued that these cannot account for the phenomenology of the OBE. I have since proposed a psychological theory to account for the nature of the experience and why it seems so real (JP, 1984, 201- 218).

I proposed that at any given time the cognitive system must choose which one of its many mental models represents external reality, and that it chooses the most stable one. Normally this is the model based on sensory input, but if this becomes destabilized (e.g. ,under stress, near death, or during sensory deprivation), other models may take over. The system may then try to regain “input control” by reconstructing a model of the scene as remembered. If this is most stable, it will take over as “reality”. If a good model is constructed everything will seem normal, but remembered scenes are quite frequently constructed in bird’s-eye view or “observer perspective”.   If such a model becomes “reality”,  then an OBE has occurred.

Everything seems real but from a different viewpoint. Indeed everything is “real” in just the same sense as anything is ever real. This state will continue until the input-driven model becomes more stable again and “normality” is restored . The person will then seem to have left the body and returned again. According to this theory, the OBE is a convincing “illusion of reality”. The only thing which has moved is the viewpoint.

There has been much argument over whether the psychological theories of OBEs predict positive correlations between OBES and imagery skills. In general the evidence suggests no such correlation (Blackmore JASPR, 1982, 301- 317, Cook and Irwin, JP, 1983, 23-25). The present theory might even predict a negative correlation. In the attempt to regain “normality”, the good imager may produce a better eye-level view and so never have an OBE. It is the person who constructs an ”observer” image who is likely to have the OBE.

People may do this for two reasons: either when the situation (such as being in pain or near death) provides an incentive to use an observer viewpoint, or because they naturally tend to use that viewpoint in imagery. We would, therefore, expect that a person’s ability to switch viewpoints when desired and their tendency to use observer viewpoints should be more important than the vividness of their imagery. This leads to the following predictions. First, OBErs should be better able to construct imagery from alternative viewpoints, especially from the typical OBE viewpoint above the head. Second, OBErs should be better able to switch viewpoints from one position to another, again especially to the viewpoint above the head. Third, OBErs should be those who tend to use an observer viewpoint in memory and imagination. The theory also implies that the viewpoint above the head should be the easiest to imagine, and that unpleasant scenes should more often be recalled in observer perspective.

I have previously found (RIP 1982, 231- 232) that OBErs were better able to switch viewpoints in remembered scenes, but they did not more often use an observer viewpoint. The two studies reported here extend these previous findings to test all five predictions.

Experiment 1 

Method. There were 135 subjects , 68% female , aged 17 to 94 with a mean of 43.1 years. They were asked to imagine the room they were sitting in from four or five different locations and to rate their image on a scale from one (no image) to seven (a perfectly clear and detailed image). The locations were: eye level in the doorway, by the ceiling above their own head, by the ceiling above someone else’s head, by their feet, and at their own eye level (this for only 63 subjects).

They were all asked where they normally perceived their “self” or “center of awareness” to be and were asked to try to switch this to above their head and back, and to their feet, rating their ability to do this on a scale from one (not at all) to seven (extremely easily).

Other questions dealt with the frequency of lucid dreams and out-of-body experiences. It was predicted that, compared to non-OBErs, OBErs would rate their images from different locations as more clear and detailed, and that they would more easily be able to switch viewpoints from one position to another.

Results. As expected, the room was easiest to imagine from the subjects’ own eye level and most difficult from the feet. However, the second easiest was at eye level in the doorway, not above the head.

As expected, it was much easier to switch viewpoints to a position above the head (mean rating = 3.7) than to one at the feet (mean rating = 2.1). During these various exercises, 22% of the subjects claimed that they saw themselves from the outside and 15% that they had another body in their imagined location.

The most commonly reported position of the normal center of awareness was behind the eyes (65%). The whole body was next (10%), with top of the head and the forehead reported by only 6% of the sample. Twenty-two percent of subjects reported having had at least one OBE, and 39% lucid dreams. There was no association between having OBEs and lucid dreams (chi-square = .04, 1 df).

For all positions combined, OBErs rated their images as clearer and more detailed (mean = 15.8) than non-OBErs (mean = 13.8, t = 1.71, 52 df , p = .047, one-tailed) . The biggest difference between OBErs and non-OBErs was in their ability to imagine the room from a position on the ceiling above their own head (means = 4.2 , 3.3; t = 2.56, 54 df , p = .007, one-tailed).

OBErs were significantly better at switching from one viewpoint to another, especially from the normal position to one above the head (means = 4.5, 3.4; t = 2.72, 60 df , p = .0045, one-tailed). In conclusion, this study confirmed that OBErs are better at using different viewpoints in imagination and at switching from one position to another. However, the common OBE viewpoint, above the head, does not seem to be particularly easy to imagine.

Experiment 2 

Method. Subjects were 187 students, 52% female, aged 17 to 75 (mostly 17 to 20). They were given a questionnaire regarding dreams and OBEs. They were then asked to remember seven events, to write brief descriptions of them, and to say whether they imagined them as though from the position of an observer (who would see himself or herself included in the scene) or from a position as they would have seen it through their own eyes. There were three ”real-life” situations (first thing that morning, the same time the previous Sunday, and the same time on Christmas Day) and four dreams (the previous night’s, one from the previous week, and the best and worst dreams they could remember).

It was predicted that more recent events would be less often recalled in observer perspective, that bad dreams would more often be recalled in observer perspective, and that OBErs would more often use the observer perspective in all tasks.

Results. Fifty percent of subjects initially reported having had an OBE. A second questionnaire was sent out to these and on this basis only 23% claimed OBEs.

As predicted the most recent event (first thing that morning) was least often recalled in observer perspective (14% compared with “last Sunday” (43%) and Christmas Day (46%) ). Also, worst dreams (34%) were far more often recalled in observer perspective than best dreams (23%) . “Last night’s” dream produced 31% and “last week’s” 33%.

Subjects were divided into those who most commonly use observer perspective versus those who most commonly use eye-level perspective, but there was no relationship between this factor and having OBEs. For real-life recall, the percentage of OBErs and non-OBErs using observer perspective was the same but for the dream-recall tasks, the percentage was higher for OBErs (mean of four tasks = 36%) than for non-OBErs (23%) .


Several of the predictions were confirmed. Unpleasant dreams were more often recalled in observer perspective, supporting the idea that this viewpoint is used as an escape from unpleasant situations and so may precipitate an OBE. OBErs were found to prefer using the observer viewpoint in dream recall but not in the recall of real-life events. Further research is needed to explore the reasons for this difference.


Several predictions have been derived from a psychological theory of the OBE, and many of them confirmed by experiment. We may conclude that OBErs are good at switching from one viewpoint to another (especially to that above the head), are proficient at producing clear and detailed images from different viewpoints, and tend to use the observer viewpoint in dream recall. These findings do not answer the implicit question of whether anything leaves the body in an OBE, but they do at least provide some support for a theory that suggests it does not.