Everybody agrees that at least for some subjects the experience of hypnotic age regression is a subjectively compelling one, that is, that the subject feels that he or she is really a child again. The major thrust of research on age regression to date has attempted to confirm these experiential reports by attempting to determine whether the age regressed subject's performance in some experimental situation is appropriately childlike. For example, Gidro-Frank and Bowersbuch (1948) attempted to demonstrate the return of the Babinski reflex. Reiff and Scheerer (1959) employed Piaget's "Hollow Tube Test" for assessing reversibility concepts 'in an effort to demonstrate a return to preoperational modes of thought in subjects regressed to age four. And Parrish, Lundy and Liebowitz (1969) tried to show that subjects regressed to childhood would view the Ponzo and Poggedorff illusions as children do. However, the limited success of these attempts to establish the authenticity of hypnotic age regression calls for a thorough re-analysis of the concept of hypnotic age regression and of what can be expected of the age regressed subject.
A survey of research and theory in this area indicates that the concept of hypnotic age regression typically involves three elements. First, people hold that age regression involves a functional ablation of all the person's memories acquired after the suggested ape. According to this view the college student regressed to age five can no longer write or spell difficult words or use a sophisticated vocabulary. This we will call the ablation component of hypnotic age regression. The second feature of age regression is a return to earlier modes of cognitive and emotional functioning. Thus it is not just that the age regressed subject is an adult who has forgotten memories of his adult life, but that he has become a child again and consequently functions like one. It has been claimed, for example, that the college student regressed to age five will lose formal operations and fail to conserve volume, on Piagetian type tasks. We call this the regression component of hypnotic age regression. Finally, it has been claimed that age regression leads to a reliving of earlier experiences; that is, that the age regressed person has access to memories from childhood that may only be dimly recalled in the normal waking state. This means that the age regressed college student will be able to remember the names of forgotten elementary school teachers, childhood playmates and early experiences. This is termed the revivification component of hypnotic age regression.
Research has been conducted which bears on all three of these elements. It has been repeatedly demonstrated, for instance, that the adult does not lose adult knowledge and abilities during age regression. One of the clearest illustrations of this is provided by Orne's (1951) report of the subject regressed to age 6, who wrote without any errors in spelling: "I am conducting an experiment which will assess my psychological capacities". Such evidence argues strongly against the ablation conceptualization of hypnotic age regression.
Most of the literature to date has involved investigations of the cognitive aspects of the regression component of hypnotic age regression. This is a complex issue because the extent to which a person can recapture childhood modes of functioning or schemata is not merely a function of hypnotic depth or susceptibility but is severely constrained by the fate of childhood schemata in the course of cognitive and emotional development.
In an important but little known paper, Neisser (1962), working in a Piagetian framework, outlines a theory to explain what happens to childhood schemata as they accommodate themselves to new information. Neisser describes three modes of accommodation. The first of these is absorption in which new schemata "swallow up" the old. The old schemata were important in the establishment of the new ones but they no longer have any function and so are completely absorbed as new forms of cognitive schemata develop. The second mode is displacement in which parts of the old and the new schemata coexist and can function independently. The third mode is integration in which old schema are employed by new ones. The old ones continue to function but not in an independent way. Given this analysis, an age regressed subject mayor may not be able to capture childhood schemata depending on the mode of cognitive development. According to Neisser's distinctions, schemata that have been absorbed cannot he recaptured, but displaced ones can. Integrated schemata can be recaptured too, but only by means of new cognitive structures.
Consider now the classic studies of age regression, such as those previously mentioned, in light of this analysis. For example, in the study by Reiff and Scheerer it seems possible that these investigators were attempting to recapture schemata that are usually absorbed by new ones and are thus inaccessible. Preoperational schemata are replaced by those of concrete and formal operations (Piaget, 1970), and it may therefore be unreasonable to expect an age regressed adult to genuinely lose the concept of reversibility. Childhood schemata that are displaced, however, may be more likely to be recaptured. What might these type of schemata be? The literature in clinical and personality psychology suggests that likely candidates are those cognitive structures which have to do with the person's social and emotional life. 1'his directs us to the domains of person perception rather that object perception and emotional responses rather than reflexes.
A recent, carefully conducted study by Nash, Johnson, and Tipton (1978) investigated childlike emotional responsiveness in hypnotic age regression. Regressed subjects were placed in frightening situations and their behavior was examined with respect to transitional object relationships, i.e.. the use of blankets or teddy bears to provide comfort. When regressed to age three and told, that they were alone and it was raining outside, both reals and simulators showed signs of stress--they acted lonely and scared. But the reals were significantly more likely than the simulators to spontaneously produce a blanket or a teddy bear in an attempt to comfort themselves--a behavior that is typical of three olds, but certainly not college students. The results suggest that these investigators may in fact have been able to tap emotional schemata from an earlier period.
In the realm of person perception, John Kihlstrom and I have begun to explore peer descriptions in hypnotic age regression. A recent cross sectional study by Peevers and Secord (1973) found that peer descriptions collected from students of various age groups differed systematically on a number of dimensions. For example, they found that kindergarteners describing their peers tended to do so in terms which failed to distinguish the person from his environment, which employed a relatively high proportion of egocentric referents, and which neglected to take into account situational factors. College students' descriptions, by contrast, included more items which yielded information about the person as a unique individual, were other-oriented, and showed insight into the world of the person being described. We were interested in how well peer descriptions obtained from age regressed persons would resemble those typically provided by young children.
Eight college students scoring in the range of high hypnotizability on both the Harvard Group Scale of Hypnotic Susceptibility (Shor and Orne, 1962) and the Stanford Hypnotic Susceptibility Scale: Form C (Weitzenhoffer and Hilgard, 1962) were selected to participate in this preliminary study. All of the subjects had passed the age regression item on the Stanford Scale. The peer description data was collected in a third hypnotic session involving two conditions. During the first half of the session subjects were hypnotized and regressed to kindergarten or grade one (whichever they preferred). After allowing time for the subjects to become involved in their regressed roles, the subjects were asked to describe two of their schoolmates. Suggestions were then given for the subjects to return to the present, but to remain deeply hypnotized. During the second part of the session, subjects were asked to imagine themselves on a beach with some friends. After developing this fantasy on their own, they were asked to describe two of the friends present with them on the beach. All the descriptions were tape recorded and later transcribed. They were then scored according to the coding procedure developed by Peevers and Secord.
Although the data were analyzed for four separate dimensions (descriptiveness, depth, personal involvement, and evaluative consistency), only the data for the descriptiveness dimension will be presented today, because of time limitations. In any event, these data are sufficient to characterize the pattern of results obtained in this study. The results for the descriptiveness. dimension for both the present study and the Peevers and Secord study are presented in Figure 1.
There are four levels of descriptiveness in the Peevers and Secord coding system: Level 1, or "Undifferentiating", descriptions are those which describe the person in terms of possessions or social setting. (E.g.. John lives in a big house.) Level 2, or "Simple Differentiating", descriptions involve the use of simple, superficial characteristics, global judgments, or relationships to the perceiver. (E.g.. John is nice.) In level 3, or "Differentiating", descriptions fairly specific personal characteristics such as interests, abilities and beliefs are employed. (E.g.. John is a good athlete.) And level 4, or "Dispositional", descriptions involve the use of explicit or implicit personality disposition (E.g.. John is talkative.) To control for differences in the lengths of description, the scores for each description have been converted to percentages of the total items in that description. The bars on the graphs thus represent the mean percentages for each level of descriptiveness, averaged across the two descriptions obtained in each condition and across subjects.
Concentrating on the data from the age regression condition, we see that the pattern of results closely matches that obtained by Peevers and Secord. This looks encouraging, but several other aspects of the data force us to be very cautious in drawing conclusions. For example, as Figure 1 also shows, the data from our hypnotized non-regressed college sample differs somewhat from that of Peevers and Secord, and in a way that favors obtaining the results we did for the age regression.
With the possible exception of the Nash et al. study, then, the available evidence does not strongly support the notion that there is a true regressive component in hypnotic age regression. It just does not seem to be the case that the age regressed adult gains access to displaced childhood schemata, in the sense that he or she is operating cognitive1y and/or emotionally exactly as a child would. The behavior of the regressed subject, however, does strike an observer--and indeed the subject himself--as rather childlike, and we are still faced with the task of explaining this. To this end, a more detailed analysis of the revivification component of age regression from the point-of view of integrative accommodation might be illuminating.
It may be that what happens in hypnotic age regression is that the adult subject constructs a childlike experience in accordance with the suggestion, perhaps recapturing some forgotten memories along the way. This analysis is very similar to the theory of remembering proposed by Bartlett in the early 1930s, and revived by Neisser (1967; 1976). Bartlett (1932) held that remembering involved an active process of reconstructing the past, in which a person builds a representation of an experience on the basis of only fragmentary memories. In much the same way, we argue that the hypnotized adult constructs the age regre8sion experience on the basis of two aspects of memory. First, he remembers some childhood experiences: some of which may be Quite vivid, while others may only be dimly recalled. Second, he has a great degree of knowledge and many expectations of what it is like to be a child. In age regression he combines these in an active fashion, filling the gaps to create the illusion of a continuous experience. To illustrate: a subject regressed to his fifth birthday knows that such an event involves parties, and cakes, and funny hats; he recalls an image of one of his birthday parties--he may not know which one--and he remembers the names of some of his kindergarten classmates. From this raw material he creates the suggested experience which in turn becomes subjectively compelling to him.
What are the implications of this constructive process account of hypnotic age regression? First. some aspects of the events in question may actually have happened in just the way they are reported. After all, the experience has been built in part on a foundation of residual memory fragments. Other aspects. however. may be distorted or entir1y confabulated. Moreover. the constructed experience of feeling like a child again may provide additional contextual cues that permit the person to gain access to other memories which could not be recalled at all before the age regression experience. This is consistent with Tulving's (1974) assertion that forgetting is not permanent but that forgotten memories may be recalled in the presence of appropriate retrieval cues. Finally, a person will not behave or experience the world exactly as a child would. Childhood schemata which have been subject to integrative accommodation can be accessed, but these have been modified by adult cognitive structures.
To this point, what we have described is not much different from merely recalling the event in question, or enacting the role of a child; we have not said anything about what makes hypnotic age regression special. There are several elements which hypnosis may add to the normal memorial and role enactment processes. Other research has shown that hypnosis is characterized by a high degree of imaginative involvement (Hilgard. 1970). and this may greatly facilitate the reconstructive process. In addition, as Orne (1951) has pointed out, highly hypnotizable subjects may be able to structure an hallucinated environment which reinforces the suggested experiences. It will help the sub.1ect to feel like a child again if the laboratory room becomes transformed into a nursery. and the reclining chair into a sandbox.
Finally, the deeply hypnotized sub.1ect may be able to dissociate from awareness his image of himself as an adult (cf. Hilgard, 1977). This reduces the tendency for the adult self-perception to interfere with the construction of a new perception, vis that the subject is a child once more. In subjects who are not capable of this kind of dissociation, the age regression experience may be composed of both child participant and adult observer experiences.
This conceptualization of age regression leads to a different research strategy than that which has previously characterized this field. Instead of looking for the "right task" to demonstrate the authenticity of age regression, we will want to examine the cognitive concomitants of the suggested experience. For example, is the subject's capacity for age regression related to individual differences in the accessibility of childhood memories, or to the amount of contact he or she has had with children? To what extent does age regression permit the accurate retrieval of memories which cannot be recalled in the normal waking state? Are there features of the age regression experience--such as whether the subject experiences himself as a child only, or as both a child and an adult f' observer--that are related to the depth of hypnosis achieved? And what about the relationship between hypnotic age regression and individual differences in particular hypnotic abilities? Perhaps this new approach will be more fruitful than the old one.
Bartlett, F. C. Remembering. Cambridge: Cambridge University Press, 1932.
Hilgard, E. R. Divided Consciousness: Multiple Controls in human thought and action. New York: John Wiley & Sons, 1977.
Hilgard, J. R. Personality and Hypnosis: A study in Imaginative Involvement. Chicago: University of Chicago Press, 1970.
Gidro-Frank , L. & Bowersbuch, M. K. A study of the plantar response in hypnotic age regression. Journal of Nervous and Mental Disease, 1948, 107, 443-453.
Nash, M. R., Johnson, L. S. and Tipton, R. D. Hypnotic age regression and the occurrence of transitional object relationships. 1978, in press.
Neisser, U. Cultural and cognitive discontinuity. In T. E. Gladwin and W. Sturtevant (Eds.), Anthropology and Human Behavior. Washington, D. C.: Anthropological Society of Washington, 1962.
Neisser, U. Cognitive Psychology. New York: Appelton Century Crofts, 1967.
Neisser, U. Cognition and Reality: Principles and Implications of Cognitive Psychology. San Francisco: W. H. Freeman and Company, 1976.
O'Connell, D. N., Shor, R. E. and Orne, M. T. Hypnotic age regression: An empirical and methodological analysis. Journal of Abnormal Psychology Monograph, 1970, 76, 1-32.
Orne, M. T. The mechanisms of hypnotic age regression: An experimental study. Journal of Abnormal & Social Psychology, 1951, 46, 213-225.
Parrish. M.. Lundy. R. M. and Leibowitz. H. W. Effect of hypnotic age regression on the magnitude of the Ponzo and Poggendorff illusions. Journal of Abnormal Psychology, 1969, 74, 693-698.
Peevers. B. H. and Secord. P. F. Developmental changes in attribution of descriptive concepts to persons. Journal of Personality & Social Psychology, 1973, 27, 120-128.
Perry. C. and Chisholm. W. Hypnotic age regression and the Ponzo and Poggendorff illusions. International Journal of Clinical & Experimental Hypnosis, 1973, 21, 192-204.
Piaget. J. Piaget's theory. In P. Mussen (Ed.). Carmichael's Manual of Child Psychology. 3rd Ed. Vol. 1. New York: Wiley. 1970.
Reiff. R. and Scheerer. M. Memory and Hvpnotic Age Regression: Developmental Aspects of Cognitive Function Explored Through Hypnosis. New York: International Universities Press, 1959.
Shor R. E. and Orne. E. C. Harvard Group Scale of Hypnotic Susceptibility: Form A. Palo Alto. California: Consulting Psychologists Press. 1962.
Tulving. E. Cue-dependent forgetting. American Scientist, 1974, 62, 74-82.
Weitzenhoffer, A. M. and Hilgard. E. R. Stanford
Hypnotic Susceptibility Scale: Form C. Palo Alto. California:
Consulting Psychologists Press. 1962.
Presented at the 30th annual meeting of the Society for Clinical and Experimental Hypnosis, Ashville, N. C., October, 1978. Preparation of this paper was supported in part by Grant MH 29951 from the National Institute of Mental Health, United State Public Health Service. The author holds a Doctoral Fellowship from the Social Sciences and Research Council of Canada. I would like to thank John Kihlstrom and Beverly Chew for their helpful comments on earlier draft of this report.