Simon J. Sherwood

Department of Psychology
The University of Edinburgh
Edinburgh EH8 9JZ

A whole range of anomalous experiences have been reported during the borderline hypnagogic or hypnopompic states which surround periods of sleep (e.g., Gurney, Myers, & Podmore, 1886; Rose, Hogan, & Blackmore, 1997). The question is whether these states are conducive to anomalous processes or agencies, whether normal features are being misinterpreted, or both. This paper outlines the main physiological and psychological features of these hypnagogic/hypnopompic states and considers some of the evidence to address this question.

Although the hypnagogic (and probably the hypnopompic) state has unique behavioural, electrophysiological and subjective characteristics, it is also highly variable and there are large individual differences (Rechtschaffen, 1994). During these borderline states, people can experience brief and vivid imagery in one or more sensory modalities, and also temporary paralysis. People may find these kinds of experiences rather puzzling and may be keen to find an explanation for them, particularly if they have not come across them before. Evidence for the possible conduciveness of these states to anomalous experiences is discussed with reference to experimental studies (e.g., ganzfeld), spontaneous cases and surveys. Evidence for the possible misinterpretation of hypnagogic/hypnopompic experiences is discussed with reference to experiences with anomalous interpretations found in different cultures (e.g., Old Hag attacks) which have similar phenomenology to sleep paralysis plus accompanying imagery. A number of features of hypnagogic/hypnopompic experiences are very similar to features of reported anomalous experiences, such as ESP, apparitions, and OBEs (Mavromatis, 1983, 1987).

It is concluded that hypnagogic/hypnopompic features may be both conducive to anomalous experiences and misinterpreted as involving anomalous processes or agencies (e.g., deceased persons). Either way, the experiences may be interpreted correctly or incorrectly. The interpretation may depend on the specific hypnagogic/hypnopompic features experienced, on individual knowledge and beliefs, and on the context in which the phenomena occur. Further research which addresses the decision-making processes involved in interpreting these kinds of experiences would also be useful.

What is being proposed then is that, although hypnagogic/hypnopompic imagery and sleep paralysis are relatively normal experiences, occasionally they may be influenced by anomalous processes (e.g., ESP) or may facilitate anomalous experiences. More attention to the stages, features, contents, and physiology of the hypnagogic/hypnopompic states may enable us to identify, perhaps with a greater degree of accuracy, if and when anomalous processes are operating.

A whole range of anomalous experiences have been reported during the hypnagogic or hypnopompic states which surround periods of sleep. It is not uncommon for people to experience brief, vivid and often strange imagery or to find themselves temporarily unable to move or speak during these periods between wakefulness and sleep. These brief sensations and the temporary paralysis are known as ‘hypnagogic’ or ‘hypnopompic’ imagery and ‘sleep paralysis,’ respectively (American Sleep Disorders Association (ASDA), 1990).

A recent survey found that people who report more childhood experiences of hypnagogic/hypnopompic imagery or sleep paralysis also report a greater number of anomalous experiences during childhood or adulthood (Sherwood, 1997). More specifically, hypnagogic/hypnopompic imagery has been associated with reports of extrasensory perception (ESP), apparitions and communication with the dead, out-of-the-body experiences (OBEs), visions of past lives and experiences involving extraterrestrials (e.g., Glicksohn, 1989; Gurney, Myers, & Podmore, 1886; Leaning, 1925; McCreery, 1993; McKellar, 1957; Mavromatis, 1983, 1987; Spanos, Cross, Dickson, & DuBreuil, 1993). In addition to the above anomalous experiences, sleep paralysis has also been associated with reports of psychokinesis (PK), and near-death experiences (NDEs) (Baker, 1992; Green & McCreery, 1994; Rose & Blackmore, 1996; Rose, Hogan, & Blackmore, 1997; Spanos et al., 1993; Spanos, McNulty, DuBreuil, Pires, & Burgess, 1995).

The question is: are the hypnagogic/hypnopompic states conducive to anomalous processes and events or are normal hypnagogic/hypnopompic features being misinterpreted? Perhaps both statements are true? The aim of this paper is to consider evidence for each of these possibilities. Firstly, it is necessary to outline of the characteristics of the hypnagogic/ hypnopompic states and the features of some of the experiences which can occur within them.

Most research, both experimental and survey-based, seems to have focused on the hypnagogic state (the period between wakefulness and sleep, i.e., just as you are falling asleep). Comparatively little research has been carried out on the hypnopompic state (the period between sleep and wakefulness, i.e., just as you are waking from sleep). Thus, this paper will focus mainly on the hypnagogic state. The hypnagogic state, like the sleep state, is fairly complex and contains a number of steps and stages (Mavromatis, 1983; Rechtschaffen, 1994). Hori, Hayashi, and Morikawa (1994) concluded that the sleep onset period is unique and cannot be accurately categorised as either waking or sleeping. It is very difficult to determine the precise point of falling asleep, except by using arbitrary criteria, because the transition is gradual, because the changes are not always synchronised and because there are large individual differences in when the changes occur (Lavie, 1996; Rechtschaffen, 1994).

During alert wakefulness, eye movements are fairly rapid and the normal EEG trace consists of irregular waves of high frequency (Bray, Cragg, Macknight, Mills, & Taylor, 1992). As a person relaxes or becomes drowsy, there is an increase in alpha activity (8-12 Hz) and eye movements become slower and less frequent (Parker, 1975; Rechtschaffen, 1994). In fact, the presence of slow eye movements (SEMs) is considered to be an extremely accurate indicator of hypnagogic mentation (Schacter, 1976). Stickgold and Hobson (1994) found that as the period of eyelid movement quiescence lengthens, i.e., as the eyelids move less, mentation becomes more dreamlike. This contrasts with the positive association between eyelid movements and dreamlike mentation during REM sleep. As a person passes through the hypnagogic period into the early stages of NREM sleep, there is a decline in alpha activity and a concomitant increase in slower theta activity (4-7 Hz) (Baddia, Wright, Jr., & Wauquier, 1994; Bray et al., 1992; Rechtschaffen, 1994). A person is typically considered to be asleep once they reach stage 2 sleep which is characterised by theta activity and the appearance of sleep spindles (Lavie, 1996; Rechtschaffen, 1994). During the transition from wakefulness to sleep there is also a decrease in muscle tone, a slowing of the heart and respiration rates, a reduction in blood pressure, and an increase in skin temperature (Mavromatis, 1983; Mavromatis & Richardson, 1984; Schacter, 1976). Upon awakening, these changes go in the opposite direction (Mavromatis, 1983).

During the hypnagogic/hypnopompic states, people can experience brief and vivid imagery or sensations in one or more different sensory modalities (e.g., Foulkes & Vogel, 1965; Hori et al., 1994; Mavromatis, 1987) or temporary paralysis (ASDA, 1990). Recall of hypnagogic imagery has been found to peak around the middle of standard Stage 1 sleep when the EEG mainly consists of theta activity (Hori et al., 1994).

Laboratory studies have also found that hypnagogic imagery and sleep paralysis can occur during sleep-onset REM periods (SOREMPs) and that isolated sleep paralysis is characterised by abundant alpha activity (Takeuchi, Miyasita, Sasaki, Inugami, & Fukuda, 1992; Takeuchi, Miyasita, Inugami, Sasaki, & Fukuda, 1994). SOREMPs have been associated with altered sleep schedules (Fukuda, 1994) which may predispose towards sleep paralysis (ASDA, 1990). Some consider REM sleep intrusions to be a necessary but not sufficient requirement for sleep paralysis (Takeuchi et al., 1992), though there is evidence to suggest that it is not inevitably associated with SOREMPs (see Ness, 1978). Although these SOREMP hallucinations are similar to other hypnagogic imagery, they seem to be more emotional and there is a greater awareness of the surroundings (Takeuchi et al., 1992, 1994).

Other features of the sleep onset period include a decreasing awareness of observing the contents of one’s own mind, increased absorption, a loss of volitional control over mentation, inaccurate time perception, a reduction of awareness of the environment, and a reduction in reality-testing (Foulkes & Vogel, 1965; Mavromatis & Richardson, 1984; Rechtschaffen, 1994). Further features may also include the hypnagogic/hypnopompic speech phenomenon and sleep starts. The hypnagogic/hypnopompic phenomenon occurs when a person hears him/herself uttering words, which can be nonsensical or irrelevant, just as they are falling asleep or waking from sleep (McKellar, 1989; Mavromatis, 1987). Sleep starts are sudden brief muscle contractions in one or more parts of the body which occur at sleep onset (ASDA, 1990). Sleep starts are sometimes associated with hypnagogic imagery, such as illusory sensations of movement (Nielsen, 1992; Oswald, 1959).

Studies have shown that as one moves through the sleep-onset period, the amount of visual hypnagogic imagery tends to increase (Hori et al., 1994), it becomes more dream-like (Foulkes & Vogel, 1965; Stickgold & Hobson, 1994) and the image quality, vividness, luminosity and intensity of colour also increase (Mavromatis, 1987; Nielsen, 1992). McKellar (1989) suggests that the form of hypnagogic imagery also changes from sequences of objects, faces or landscapes to more complex episodes or mini-plays.

Although the hypnagogic state (and probably the hypnopompic) has unique behavioural, electrophysiological and subjective characteristics (Hori et al., 1994), it is also highly variable and there are large individual differences (Rechtschaffen, 1994; Tart, 1969, p. 73). Before going on to consider possible relationships with anomalous experiences, it is necessary to consider the experiential features of hypnagogic/hypnopompic imagery and sleep paralysis.

The term ‘hypnagogic imagery’ was provided by Maury (1848) (cited by Mavromatis, 1987) and was defined in Warren’s Dictionary of Psychology (1934) as:

The term ‘hypnopompic imagery’ was introduced by F. W. H. Myers (1904) who defined it as:

Some writers distinguish between imagery which occurs in the hypnagogic and hypnopompic states (e.g., Glicksohn, 1989; McKellar, 1989) but others do not (e.g., Mavromatis, 1987; Mavromatis & Richardson, 1984). It is fair to say that both types are similar and so many features will apply to both. However, certain features or experiences seem to be more common in the hypnagogic than in the hypnopompic state, and vice versa. For this reason, I think it is useful to maintain the distinction.

It has also been speculated that we may enter the hypnagogic state and experience hypnagogic phenomena at times other than just prior to nocturnal sleep (Mavromatis, 1983, 1987; Tart, 1969, p.74). This has been supported by reports from a number of participants (McKellar & Simpson, 1954). Although hypnopompic imagery was originally defined as a persistence of dream imagery into wakefulness (F. W. H. Myers, 1904), it seems that they are not always continuations of dreams since they can begin after the sleeper has awoken (e.g. Leaning, 1925; Mavromatis, 1987). Thus, hypnopompic images may not necessarily be the result of REM sleep continuation.

It is not easy to distinguish hypnagogic and hypnopompic imagery from dream imagery. Any qualitative distinctions made will depend on the defining characteristics of hypnagogic/ hypnopompic imagery and dreams (Mavromatis & Richardson, 1984), both of which would benefit from stricter definitions. However, there is some evidence that hypnagogic (and also hypnopompic) imagery tend to be more vivid and realistic, shorter, more passive, have less emotion, and also tend to be more disorganised and irrelevant (Foulkes & Vogel, 1965; McKellar, 1989; McKellar & Simpson, 1954; Mavromatis & Richardson, 1984). McKellar (1989) describes how:

With hypnagogic experiences, there also seems to be a greater awareness of the true situation and more reality testing compared with dream experiences (Mavromatis & Richardson, 1984). Some people also claim to be able to generate and/or control their hypnagogic/ hypnopompic imagery to some extent (Mavromatis, 1987, p. 71). Necessary requirements seem to include a receptive attitude and passive volition.

It seems that episodes of hypnagogic/hypnopompic imagery often occur fairly sporadically although concentrated series of episodes can also occur (Mavromatis, 1987). Hypnagogic and hypnopompic images are typically very brief and dynamic and may last no more than a second or two (Nielsen, 1992).

Frequency

Generally speaking, hypnagogic imagery seems to be more common than hypnopompic imagery. Early surveys (e.g. Galton, 1883; Müller, 1848 cited by Mavromatis, 1987) estimated that about 2% of adults had experienced hypnagogic imagery of some form. More recent surveys have estimated that c. one-third (Leaning, 1925), 61-63% (McKellar & Simpson, 1954; McKellar, 1957) or c.75% (Glicksohn, 1989; Richardson, Mavromatis, Mindel & Owens, 1981; Sherwood, 1997) of people have experienced it on at least one occasion compared with 21.4% (McKellar, 1957) to 67.6% for hypnopompic imagery (Richardson et al., 1981; Sherwood, 1997). One recent survey estimated that 37% (12.5%) of the UK population had experienced some form of hypnagogic (hypnopompic) imagery at least twice a week during the preceding year (Ohayon, Priest, Caulet & Guilleminault, 1996).

Although visual and auditory seem to be the most common forms of both hypnagogic and hypnopompic imagery (Foulkes & Vogel, 1965; Hori et al., 1994; McKellar & Simpson, 1954), olfactory (smell), gustatory (taste), tactile, thermal, bodily, movement, and synesthetic sensations (where imagery in one modality triggers modality in a different modality) may also occur (e.g., Leaning, 1925; Mavromatis, 1987; Schacter, 1976).

Visual imagery typically occurs with the eyes closed though it can occur with eyes open (Gurney et al., 1886; Leaning, 1925; McKellar, 1972, 1989; McKellar & Simpson, 1954). It often begins with reports of clouds or mists of bright colours or a circle of light. Images may quickly change from one to another and may develop into progressively more complex images (Gurney et al., 1886; Leaning, 1925; Mavromatis, 1987). Occasionally the images may be in black and white rather than in colour (McKellar, 1957; McKellar & Simpson, 1954). The images may sometimes be very small (micropsias) or gigantic (megalopsias), though changes in size and shape are possible (Leaning, 1925; McKellar, 1957; McKellar & Simpson, 1954; Mavromatis, 1987). A series of continuous repetitions (polyopsias) of the same image may also be experienced. Sometimes the images appear to be strangely illuminated or may be seen from a peculiar angle (Leaning, 1925; Mavromatis, 1987; McKellar, 1957).

Visual hypnagogic/hypnopompic imagery is often pleasant, even humorous, but it can also be terrifying (McKellar & Simpson, 1954; Mavromatis, 1987). Although hypnagogic and hypnopompic imagery are characterised by variety, Mavromatis (1987) has modified Leaning’s (1925) classification scheme and identified six recurrent themes: (1) Formless, e.g. waves, clouds of colour, (2) Designs, e.g. geometric and symmetrical patterns and shapes, (3) Faces, figures, animals, objects, (4) Nature scenes, e.g. landscapes, seascapes, gardens, (5) Scenes with people, (6) Print and writing, e.g. in real or imaginary languages. Visual hypnagogic/hypnopompic imagery has often been referred to as ‘the faces in the dark phenomenon’ because the seeing of faces is so common (McKellar, 1957).

Mavromatis (1987) provided a summary of auditory hypnagogic/hypnopompic phenomena:

Other reported imagery includes music, bangs and explosions. Sometimes auditory hypnopompic imagery can take the form of a warning of impending danger or an important event; other times it may just be a feeling of foreboding (Mavromatis, 1987).

Sensations of smell (e.g. cigars, roses), taste (e.g. sweet), sensations of actively touching or being passively touched by someone or something, and hot or cold sensations, sometimes moving along the body, have also been reported during the hypnagogic and hypnopompic states (Mavromatis, 1987)..

A sensation of falling seems to be the most commonly reported sensation of movement. The experience is relatively common and is often associated with a bodily jerk and visual imagery, such as falling off a cliff (Oswald, 1959). Other sensations of movement may include floating, swinging, rocking, spinning, and being in or on a moving vehicle. Bodily sensations may include, for example, a feeling of energy flowing through the body, weightlessness, heaviness, tingling, numbness, shaking/vibrating, and elongation of the body. A sense of presence has also been reported (e.g. Ohayon, Priest, Caulet, & Guilleminault, 1996) and may coincide with both imagery and sleep paralysis (Conesa, 1995; Hufford, 1982; Rose & Blackmore, 1996; Spanos et al., 1993, 1995). Feelings of foreboding or being under threat, or more general feelings of confusion and disorientation, have also been reported.

Mavromatis (1987, p. 28) also points out that "very often hypnagogic [and hypnopompic] images are symbolic or metaphoric, and not infrequently autosymbolic, and therefore not always meaningless." The experient may sometimes become aware of the significance of the imagery (which may be known only by that person) during the experience or just afterwards. Such awareness is a characteristic of the ‘autosymbolic phenomenon’, described by Silberer (1965), which is an experience in which one’s thoughts or feelings at a given moment are translated into a symbolic form of imagery. However, Mavromatis (1987) adds that "awareness of the significance of the symbolism is not always present, and in the majority of cases imagery remains a puzzle until one begins to pay attention to it and enters into a form of ‘conversation’ with it." (p. 59).

It is clear that hypnagogic/hypnopompic imagery may be extremely vivid, is characterised by variety, may evoke both positive and negative emotions, may occur in more than one sensory modality, sometimes simultaneously, and may have some significance for the experient.

According to ASDA (1990):

The sleep paralysis episode usually lasts from a few seconds to a few minutes although a duration as long as 70 minutes has been reported (Goode, 1962; Schneck, 1960; Spanos et al., 1995). The experience is sometimes preceded by and/or accompanied by visual imagery which can be terrifying (Blackmore, 1996; Conesa, 1995; Goode, 1962; Penn et al., 1981; Spanos et al., 1995; Takeuchi et al., 1992). The experience may end due to efforts to overcome it by the experient, may be terminated by someone else either by touch or verbally, the experience may move into a dream (possibly lucid) or the episode may simply terminate spontaneously (Firestone, 1985; Goode, 1962; Schneck, 1960; Snyder, 1983).

The hypnagogic form seems to be more frequent than the hypnopompic form (Conesa, 1995; Goode, 1962; Spanos et al., 1995) although the opposite has also been reported (Penn, Kripke & Scharff, 1981; Sherwood, 1997). The International Classification of Sleep Disorders estimates that isolated sleep paralysis (i.e. that which occurs independently of narcolepsy) occurs at least once in a lifetime in 40-50% of normal people. Surveys have found that between 4.7% to 49% of people have reported sleep paralysis (of one or other or both forms) though the most frequent estimates range between 30-45% (Goode, 1962; Everett, 1962; Penn et al., 1981; Spanos et al., 1995; Blackmore, 1996; Rose & Blackmore, 1996; Sherwood, 1997).

Accompanying features of sleep paralysis may include: acute anxiety or terror, awareness of the surroundings, hypnagogic/hypnopompic imagery, a sense of presence, difficulty breathing or a sense of suffocation, pressure on the chest, a tendency to mentally or physically struggle to overcome it, a feeling of time distortion, sexual arousal (Blackmore, 1996; Conesa, 1995, 1997; Goode, 1962; Hufford, 1982; Liddon, 1967; Penn et al., 1981; Rose et al., 1997; Schneck, 1960, 1977; Snyder, 1983; Spanos et al., 1995; Takeuchi et al., 1992):

Personal beliefs and expectations, knowledge of normal sleep-related experiences, mental set and the setting in which hypnagogic/hypnopompic experiences take place are all very important factors which can influence how these experiences are interpreted (e.g., Leaning, 1925; McKellar & Simpson, 1954; Mavromatis, 1987). People may find hypnagogic/ hypnopompic experiences rather puzzling and may be keen to find an explanation for them, particularly if they have not come across such experiences before, as this letter illustrates:

Mavromatis (1987) illustrates how, in some cases, ‘visual psi experiences are practically indistinguishable from those occurring in hypnagogia both in their content and in their nature. Also, the mental state of the subject appears to be the same.’ (p. 138). If psi does exist, then it seems that hypnagogic/hypnopompic experiences may be a vehicle for it but clearly there is room for misinterpretation. On a given occasion, a hypnagogic/hypnopompic experience may reflect genuine anomalous processes but it may or may not be interpreted as such. On another occasion, a hypnagogic/hypnopompic experience may not reflect genuine anomalous processes but it may be interpreted as if it does. Thus, false negatives and false positives are possible. Mavromatis (1987) does not really give any indication of how psi and non-psi hypnagogic/hypnopompic experiences might be differentiated. If reliable differences could be identified then this might help the experient make better judgements. This might be particularly useful in a ganzfeld context. Further investigation of this question is warranted.

So what evidence is there to suggest that hypnagogic/hypnopompic states are conducive to anomalous processes or agencies? The hypnagogic state is considered to be ‘unusually receptive’ (Schacter, 1976, p. 468) and shares features of the psi-conducive state such as physical relaxation, reduction in sensory distraction and increased internal attention (Braud & Braud, 1975; Honorton, 1977; Mavromatis, 1987). According to Mavromatis (1983), support for a relationship between psi and hypnagogia (his generic term for hypnagogic and hypnopompic imagery) comes from the practices and literature on occultism and spiritualism, the literature on controlled psi experiments, and spontaneous cases of psi during hypnagogic practices.

Some experimental studies have found that hypnagogic/hypnopompic imagery seems to be conducive to telepathy (Gertz, 1983; Schacter & Kelly, 1975), perhaps more so than dreaming (Braud, 1977; White, Krippner, Ullman & Honorton, 1971). The ganzfeld technique, which has provided some of the best evidence for ESP, is believed to induce a hypnagogic-like state (Bertini, Lewis, & Witkin, 1969). However, although the ganzfeld procedure is often assumed to induce a hypnagogic state, the extent to which it does resemble the naturally-occurring hypnagogic state is not clear (Braud, Wood, & Braud, 1975; Schacter, 1976). If the experimentally-induced state is not radically different to the naturally-occurring state at sleep onset, then this suggests that hypnagogic experiences can occur at different times of the day and in different settings. Palmer, Bogart, Jones & Tart (1977) report a ganzfeld study (and a previous unpublished study, by Palmer) which found significant correlations between ESP z performance (psi-hitting or psi-missing) and scoring on altered state of consciousness/hypnagogic imagery scales.

Bem and Honorton (1994) reviewed ganzfeld studies carried out up until the Hyman (1985) and Honorton (1985) meta-analyses, and also a group of subsequent studies carried out by Honorton which were designed to meet the more stringent standards specified by Hyman and Honorton (1986). Bem and Honorton (1994) concluded that Honorton’s studies had met the required standards and had produced results consistent with the previous ganzfeld studies, though they also acknowledged the need for further replication by a broader range of investigators. Milton and Wiseman (1997) conducted a meta-analysis of other ganzfeld studies reported since the joint Hyman-Honorton guidelines were published and concluded that these studies did not offer any evidence for ESP, though they suggested that certain psi-conducive factors may not have been exploited as effectively in these studies. Although previous ganzfeld research has provided some of the best evidence for ESP, Milton & Wiseman (1997) concluded that "what is clear is that use of the ganzfeld alone is no guarantee of a successful study." (p. 277).

In terms of spontaneous cases, there are a number of well-documented cases of ESP and crisis apparitions which have occurred during the hypnagogic/hypnopompic states (Gurney et al., 1886, Chapter IX, "Borderland" Cases, pp. 251-285). For example, a father reported a hypnopompic image involving his son:

The father received a letter the following day informing him that his son was ill but later found out that he had died at the time of the apparition. Some writers believe that hypnagogic visions might be an early form of ESP (Leaning, 1925; Mavromatis, 1983). In support of this, developing psychics often experience an increase in hypnagogic phenomena (Mavromatis, 1987). Gifted subjects also use hypnagogic imagery (White, 1964); for example, well-known psychics, such as Garrett and Northage, have described personal examples of telepathy and clairvoyance during the hypnagogic and hypnopompic states (Mavromatis, 1987).

Moody with Perry (1993) also described a number of cases of visionary encounters with departed loved ones inside a psychomanteum chamber which share characteristics of hypnagogic imagery. OBEs also tend to occur spontaneously during the hypnagogic/ hypnopompic states (Mavromatis, 1983); McCreery (1993) found a positive relationship between number of hypnagogic imagery episodes and number of OBEs. Mavromatis (1983, 1987) also lists a number of hypnagogic phenomena which have been reported during OBEs: e.g., sensations of floating, sinking, drifting, seeing lights, images, landscapes, hearing noises, music, name being called, sensation of being touched.

Rose et al. (1997) found that sleep paralysis has been associated with reports of a number of anomalous experiences, such as ESP, PK, OBEs, NDES, apparitions, past life experiences, and extraterrestrials. It is possible that anomalous experiences and sleep paralysis episodes might be related to each other because they might both be affected by a third variable such as the earth’s geomagnetic field. A number of studies have found that subjective (e.g., Persinger, 1985) and experimental GESP experiences (Berger & Persinger, 1991) and sleep paralysis (Conesa, 1995, 1997) tend to occur when geomagnetic activity is relatively low.

If the hypnagogic state is conducive to anomalous processes perhaps this could be due to the initial increase in alpha or the later increase in theta activity (Healy, 1986) which occurs during this period (e.g., Davis, Davis, Loomis, Harvey, & Hobart, 1938; Baddia et al., 1994). Experienced meditators have also been found to show the alpha-theta progression which characterises the transition through the hypnagogic state towards sleep (Mavromatis, 1987). Meditation has been associated with reports of a variety of anomalous experiences (Eysenck & Sargent, 1993).

There is evidence that alpha activity might be conducive to anomalous processes and experiences. Tart (1968) found that Miss Z’s OBEs tended to occur during stage 1 sleep which was dominated by alphoid activity. Laboratory episodes of sleep paralysis have also been found to be characterised by abundant alpha activity (Takeuchi et al., 1992). In terms of theta, Stanford and Stevenson (1972), cited by Healy (1986), found some evidence to suggest that lower EEG activity, such as theta, might facilitate telepathy performance. Unusual theta activity has also been found in individuals who report mediumship ability or OBEs (Nelson, 1970; Tart, 1967, 1968; Palmer, 1979; all cited by Healy, 1986).

It has been suggested that it is not so much altered states per se which are psi-conducive but the degree or the rapidity of the transition from one state to another (e.g., Murphy, 1966; Honorton, 1973; Honorton, Davidson, & Bindler, 1971; all cited by Parker, 1975). Physiological monitoring of participants in the hypnagogic/hypnopompic states might be useful in that it could potentially identify the precise point, or at least the optimal physiological conditions, at which psi processes might operate. However, it is recognised that this might be difficult to investigate experimentally given that, for many people, hypnagogic/hypnopompic imagery and sleep paralysis may occur only sporadically and tend to be rather involuntary when they do occur. But, if the ganzfeld conditions are indeed very similar to the naturally-occurring hypnagogic state then this might be a suitable compromise.

In summary, evidence for the conduciveness of the hypnagogic/hypnopompic states to anomalous processes comes from the fact that these states have physiological and psychological features believed to be psi-conducive in other contexts, from experimental studies using both naturally-occurring and induced states, from spontaneous case reports of a variety of different phenomena, and from biographical accounts of gifted subjects and psychics.

Hypnagogic/hypnopompic experiences may also have been misinterpreted as ESP, apparitions, visions of previous lives or other worlds, alien abductions, witchcraft or attacks by evil spirits or demons etc. (Baker, 1992; Blackmore, 1996; Dahlitz & Parkes, 1993; Hufford, 1982; Leaning, 1925; Liddon, 1967; McKellar, 1957, 1989; Spanos et al., 1993, 1995; Wilson & Barber, 1983; Wing et al., 1994; Zusne & Jones, 1989). Such experiences may initiate or sustain beliefs in the paranormal and the supernatural and may have contributed to mythology and folk-lore (Fukuda et al., 1987; Liddon, 1967; McKellar & Simpson, 1954; Mavromatis, 1983; Ness, 1978). Blackmore and Rose (1996) found that many people were scared by sleep paralysis and some were worried that they were going mad or being visited by supernatural entities. It is also possible that knowledge and beliefs may influence the content of hypnagogic/hypnopompic experiences (Hufford, 1982; Spanos et al., 1993).

Assuming that the hypnagogic/hypnopompic experiences do not reflect anomalous processes, are there any general characteristics of the hypnagogic/hypnopompic states which might facilitate misinterpretations, regardless of individual knowledge, beliefs and expectations? There may be reduced sensory input from the environment and some ambiguity of external stimuli, especially if the person is in bed and it is dark. This may interfere with accurate reality-testing (Spanos et al., 1993) which also tends to reduce during the sleep-onset period (Foulkes & Vogel, 1965). It is sometimes difficult, subjectively, to distinguish wakefulness from sleep (Rechtschaffen, 1994); false awakenings and lucid dreams are a good illustration of this (Green & McCreery, 1994). One of the features of hypnagogic/ hypnopompic imagery which may lead people into believing in their reality and veracity may be that they feel awake throughout. In their survey of hypnagogic experiences, McKellar & Simpson (1954) found that

Awareness of the surroundings may be reduced during the hypnagogic/hypnopompic states to some extent (Foulkes & Vogel, 1965; Rechtschaffen, 1994) but it is an important feature, particularly during sleep paralysis episodes (e.g., Conesa, 1995; Goode, 1962; Liddon, 1967; Schneck, 1960). Mavromatis (1987) points out that ‘The ‘sense of reality, of life-likeness’ pointed out by many subjects in reference to their hypnagogic imagery often expands into ‘feelings of heightened reality’.’ (p. 30). Visual imagery can sometimes contain more detail than one might observe in more usual circumstances (Leaning, 1925). Hypnagogic/ hypnopompic imagery and sleep paralysis are also spontaneous vivid, realistic, intense and often frightening (Conesa, 1995; Schacter, 1976; Zusne & Jones, 1989). The unfamiliarity and involuntary nature of the imagery might facilitate external attributions.

Evidence for possible misinterpretation of normal hypnagogic/hypnopompic experiences is perhaps best illustrated by example. The Old Hag attack, well-known in the Canadian province of Newfoundland, is believed, by some people, to be caused by a supernatural creature, by a human in spirit form (e.g. a witch) or a combination of the two (Firestone, 1985; Hufford, 1982; Ness, 1978). The main features of an Old Hag attack are an impression of wakefulness and an accurate perception of the real environment, paralysis and fear; secondary features, which may be experienced with eyes open or closed, include a sense of presence, imagined sounds, visual images of a human (e.g. an old woman) or non-human attacker, a sense of motion, pressure (e.g. on the chest), difficulty breathing, odours and other bodily sensations (Hufford, 1982). Experiences with similar phenomenology have also been reported in Japan and China. These experiences are known as ‘kanashibari’ and ‘ghost oppression attacks’, respectively, and are believed by some to be caused by evil spirits or possession by a ghost (Fukuda et al., 1987; Wing et al., 1994). However, there is evidence to suggest that such beliefs may be more common among people who have not had the experiences themselves (Wing et al., 1994).

It is also possible that hypnagogic/hypnopompic imagery and sleep paralysis may account for some intense UFO reports and abductions (e.g., Baker, 1992; Spanos et al., 1993). Abductions are often reported around the time of sleep and may feature paralysis, awareness of surroundings, a sense of presence, bright lights and figures in the room, humming and buzzing sounds, and sensations of floating (e.g., Baker, 1992; Spanos et al., 1993).

Visual hypnagogic/hypnopompic imagery might also facilitate interpretations in terms of ghosts or apparitions. Faces may be experienced which range from the beautiful and the pleasant to the hideous and the terrifying (Leaning, 1925; McKellar, 1957; Mavromatis, 1987). These faces are often characterised as being extremely life-like and often seem to be looking at the observer (Gurney et al., 1886; Leaning, 1925). These faces can also develop into figures which may move towards the observer. Such faces/figures can be singular or in groups, of known or unknown, living or dead persons and may sometimes seem to represent particular moods and emotions (Leaning, 1925; Mavromatis, 1987). The experient may also hear their name being called which might be interpreted as attempts at communication by deceased persons.

Features which are similar to OBE/NDE accounts include feelings of floating or weightlessness, sensations of movement, changes in body image, awareness of the surroundings, and experiencing visual images such as land/seascapes, faces/figures (perhaps from an unusual angle).

There are also hypnagogic/hypnopompic features which might facilitate ESP interpretations. Hypnopompic imagery, in particular, tends to anticipate forthcoming daily events, and in connection with actual later events it may be considered to be precognitive (Zusne & Jones, 1989). Hypnopompic imagery may also appear to be warning of imminent or future danger. Visual imagery involving complex scenes characterised by movement and life may also be experienced (Leaning, 1925; Mavromatis, 1987).

In summary, it seems possible that features of the hypnagogic/hypnopompic states can facilitate possible confusions between reality and imagination in some instances. There are also a number of specific features which may facilitate anomalous interpretations. This may be more likely if a person has little knowledge of normal hypnagogic/hypnopompic features and/or if a person is within a group or culture which has certain explanations for particular phenomena, or if the person already believes in anomalous phenomena.

In conclusion, it is possible that the hypnagogic/hypnopompic states may be both conducive to, and also misinterpreted as involving, anomalous processes and agencies. In the absence of more objective information, the decision as to which interpretation is taken may depend on the individual and the context in which the experiences take place. Further research which addresses the decision-making processes involved in interpreting these kinds of experiences would be useful. Ideally, more naturalistic, experimental testing of participants who regularly experience hypnagogic/hypnopompic experiences is required. More attention to the hypnopompic state and the extent to which experimentally-induced states are physiologically and psychologically equivalent to the naturally-occurring states would also be beneficial. More attention to the physiology, features and content of the hypnagogic/ hypnopompic states may enable us to identify, with a greater degree of accuracy, if and when anomalous processes are operating.

American Sleep Disorders Association. (1990). International classification of sleep

disorders: Diagnostic and coding manual (ICSD). Rochester, MN: Author.

Baddia, P., Wright, Jr., K. P., & Wauquier, A. (1994). Fluctuations in single-Hertz EEG

Baker, R. A. (1992). Alien abductions or human productions? Some not so unusual personal experiences. Unpublished manuscript.

Bem, D. J., & Honorton, C. (1994). Does psi exist? Replicable evidence for an anomalous process of information transfer. Psychological Bulletin, 115, 4-18.

Berger, R. E., & Persinger, M. A. (1991). Geophysical variables and behaviour: LXVII. Quieter annual geomagnetic activity and larger effect size for experimental psi (ESP) studies over six decades. Perceptual and Motor Skills, 73, 1219-1223.

Bertini, M., Lewis, H. B., & Witkin, H. A. (1969). Some preliminary observations with an experimental procedure for the study of hypnagogic and related phenomena. In C. T. Tart (Ed.), Altered states of consciousness (pp. 93-111). New York: John Wiley & Sons.

Blackmore, S. (1996, September). On the edge of reality. Paper presented at the British

Association Annual Festival of Science, Birmingham, England.

Braud, W. (1977). Long-distance dream and presleep telepathy. In J. D. Morris, W.G. Roll, & R.L. Morris (Eds.), Research in Parapsychology 1976 (pp. 154-155). Metuchen, NJ: Scarecrow Press.

Braud, W. G., & Braud, L. W. (1975). The psi-conducive syndrome: Free response GESP performance following evocation of ‘left-hemispheric’ vs. ‘right-hemispheric’ functioning. In J. D. Morris, W. G. Roll, & R. L. Morris (Eds.), Research in Parapsychology 1974, (pp. 17-20). Metuchen, NJ: Scarecrow Press.

Braud, W. G., Wood, R., & Braud, L. W. (1975). Free response GESP performance during

Bray, J. J., Cragg, P. A., Macknight, A. D. C., Mills, R. G., & Taylor, D. W. (Eds.) (1992).

Lecture notes on human physiology (2^nd ed.). Oxford, England: Blackwell.

Conesa, J. (1995). Relationship between isolated sleep paralysis and geomagnetic

influences: A case study. Perceptual and Motor Skills, 80, 1263-1273.

Conesa, J. (1997) . Isolated sleep paralysis, vivid dreams and geomagnetic influences: II. Perceptual and Motor Skills, 85, 579-584.

Dahlitz, M., & Parkes, J. D. (1993). Sleep paralysis. Lancet, 341, 406-407.

Davis, H., Davis, P. A., Loomis, L., Harvey, N., & Hobart, G. (1938). Human brain potentials during the onset of sleep. Journal of Neurophysiology, 1, 24-38.

Doubleday, J. (1996, August). Scares in my sleep. [Letter]. Fortean Times, 89, 53.

Everett, H. C. (1962). Sleep paralysis in medical students. Journal of Nervous and Mental Disease, 3, 283-287.

Eysenck, H. J., & Sargent, C. (1993). Explaining the unexplained: Mysteries of the paranormal. London: PRION.

Firestone, M. (1985). The "Old Hag:" Sleep paralysis in Newfoundland. The Journal of Psychoanalytic Anthropology, 8, 47-66.

Foulkes, D., & Vogel, G. (1965). Mental activity at sleep onset. Journal of Abnormal Psychology, 70, 231-246.

Fukuda, K. (1994). Sleep paralysis and sleep-onset REM period in normal individuals.

Fukuda, K., Miyasita, A., Inugami, M., & Ishihara, K. (1987). High prevalence of isolated sleep paralysis: Kanashibari phenomenon in Japan. Sleep, 10, 279-286.

Gertz, J. (1983). Hypnagogic fantasy, EEG, and psi performance in a single subject. Journal of the American Society for Psychical Research, 77, 155-170.

Glicksohn, J. (1989). The structure of subjective experience: Interdependencies along the sleep-wakefulness continuum. Journal of Mental Imagery, 13, 99-106.

Goode, G. B. (1962). Sleep paralysis. Archives of Neurology, 6, 228-234.

Green, C., & McCreery, C. (1994). Lucid dreaming: The paradox of consciousness during sleep. London: Routledge.

Gurney, E., Myers, F. W. H., & Podmore, F. (1886). Phantasms of the living. London: Trübner, 2 vols.

Healy, J. (1986). Hippocampal kindling, theta resonance, and psi. Journal of the Society for Psychical Research, 53, 352-368.

Honorton, C. (1977). Psi and internal attention states. In B. B. Wolman (Ed.), Handbook of parapsychology (pp. 435-472). New York: Van Nostrand Reinhold.

Honorton, C. (1985). Meta-analysis of psi ganzfeld research: A response to Hyman. Journal

of Parapsychology, 49, 51-91.

Hori, T., Hayashi, M., & Morikawa, T. (1994). Topographical EEG changes and the

Hufford, D. J. (1982). The terror that comes in the night: An experience-centered study of supernatural assault traditions. Philadelphia: University of Pennsylvania Press.

Hyman, R. (1985). The ganzfeld psi experiment: A critical appraisal. Journal of

Parapsychology, 49, 3-49.

Hyman, R., & Honorton, C. (1986). A joint communiqué: The psi Ganzfeld controversy. Journal of Parapsychology, 50, 351-364.

Lavie, P. (1996). The enchanted world of sleep. New Haven: Yale University Press.

Leaning, F. E. (1925). An introductory study of hypnagogic phenomena. Proceedings of the Society for Psychical Research, XXXV, 287-411.

Liddon, S. C. (1967). Sleep paralysis and hypnagogic hallucinations: Their relationship

McCreery, C. (1993). Schizotypy and out-of-the-body experiences. Unpublished

McKellar, P. (1957). Imagination and thinking: A psychological analysis. London: Cohen & West.

McKellar, P. (1989). Abnormal psychology: Its experience and behaviour. London:

McKellar, P. & Simpson, L. (1954). Between wakefulness and sleep: Hypnagogic imagery. British Journal of Psychology, 45, 266-276.

Mavromatis, A. (1983). Hypnagogia: The nature and function of the hypnagogic state.

Mavromatis, A. (1987). Hypnagogia: The unique state of consciousness between

wakefulness and sleep. London: Routledge & Kegan Paul.

Mavromatis, A., & Richardson, J. T. E. (1984). Hypnagogic imagery. International Review

of Mental Imagery, 1, 159-189.

Milton, J. & Wiseman, R. (1997). Ganzfeld at the crossroads: A meta-analysis of the new generation of studies. Proceedings of the Parapsychological Association 40th Annual Convention held in conjunction with The Society for Psychical Research, Brighton, England, 267-282.

Moody, R. with Perry, P. (1993). Reunions: Visionary encounters with departed loved ones. London: Little, Brown and Company.

Myers, F. W. H. (1904). Human personality and its survival of bodily death (Vol. I.) New York: Longmans, Green, & Co..

Ness, R. C. (1978). The Old Hag phenomenon as sleep paralysis: A biocultural

interpretation. Culture, Medicine and Psychiatry, 2, 15-39.

Nielsen, T. A. (1992). A self-observational study of spontaneous hypnagogic imagery using the upright napping procedure. Imagination, Cognition and Personality, 11, 353-366.

Ohayon, M. M., Priest, R. G., Caulet, M., & Guilleminault, C. (1996). Hypnagogic and hypnopompic hallucinations: Pathological phenomena? British Journal of Psychiatry, 169, 459-467.

Oswald, I. (1959). Sudden bodily jerks on falling asleep. Brain, 82, 92-103.

Palmer, J., Bogart, D. N., Jones, S. M., & Tart, C. T. (1977). Scoring patterns in an ESP Ganzfeld experiment. Journal of the American Society for Psychical Research, 71, 121- 145.

Parker, A. (1975). States of mind: ESP and altered states of consciousness. New York:

Taplinger.

Penn, N. E., Kripke, D. F., & Scharff, J. (1981). Sleep paralysis among medical students. Journal of Psychology, 107, 247-252.

Persinger, M. A. (1985). Geophysical variables and behavior: XXX. Intense paranormal experiences during days of quiet, global, geomagnetic activity. Perceptual and Motor Skills, 61, 320-322.

Rechtschaffen, A. (1994). Sleep onset: Conceptual issues. In R. D. Ogilvie & J. R. Harsh (Eds.), Sleep onset: Normal and abnormal processes (pp. 3-18). Washington, DC: American Psychological Association.

Richardson, J. T. E., Mavromatis, A., Mindel, T., & Owens, A. C. (1981). Individual differences in hypnagogic and hypnopompic imagery. Journal of Mental Imagery, 5, 91-96.

Rose, N. & Blackmore, S. (1996, August). Two pilot surveys of unusual personal

Rose, N., Hogan, J. & Blackmore, S. (1997). Experiences of sleep paralysis. Proceedings of the Parapsychological Association 40th Annual Convention held in conjunction with The Society for Psychical Research, Brighton, England, 358-369.

Schacter, D. L. (1976). The hypnagogic state: A critical review of the literature. Psychological Bulletin, 83, 452-481.

Schacter, D. L., & Kelly, E. F. (1975). ESP in the twilight zone. Journal of Parapsychology, 39, 27-28.

Schneck ,J. M. (1960). Sleep paralysis without narcolepsy or cataplexy. Journal of the American Medical Association, 173, 1129-1130.

Sherwood, S. J. (1997). Relationship between childhood hypnagogic/hypnopompic experiences, childhood fantasy proneness and anomalous experiences and beliefs: An exploratory WWW survey. Proceedings of the Parapsychological Association 40th Annual Convention held in conjunction with The Society for Psychical Research, Brighton, England, 370-386.

Snyder, S. (1983). Isolated sleep paralysis after rapid time zone change (‘jet lag’) syndrome. Chronobiologia, 10, 377-379.

Spanos, N. P., Cross, P. A., Dickson, K., & DuBreuil, S. C. (1993). Close encounters: An examination of UFO experiences. Journal of Abnormal Psychology, 102, 624-632.

Spanos, N. P., McNulty, S. A., DuBreuil, S. C., Pires, M. & Burgess, M. F. (1995). The frequency and correlates of sleep paralysis in a university sample. Journal of Research in Personality, 29, 285-305.

Stickgold, R. & Hobson, J. A. (1994). Home monitoring of sleep onset and sleep-onset mentation using the Nightcap©. In R. D. Ogilvie & J. R. Harsh (Eds.), Sleep onset: Normal and abnormal processes (pp. 141-160). Washington, DC: American Psychological Association.

Takeuchi, T., Miyasita, A., Sasaki, Y, Inugami, M., & Fukuda, K. (1992). Isolated sleep paralysis elicited by sleep interruption. Sleep, 15, 217-225.

Takeuchi, T., Miyasita, A., Inugami, M., Sasaki, Y, & Fukuda, K. (1994). Laboratory- documented hallucination during sleep-onset REM period in a normal subject. Perceptual and Motor Skills, 78, 979-985.

Tart, C. T. (Ed.) (1969). Altered states of consciousness. New York: Wiley & Sons.

White, R. A. (1964). A comparison of old and new methods of response to targets in ESP experiments. Journal of the American Society for Psychical Research, 58, 21-56.

White, R. A., Krippner, S., Ullman, M., & Honorton, C. (1971). Experimentally-induced telepathic dreams with EEG-REM monitoring: Some manifest content variables related to psi operation. In W. G. Roll, R. L. Morris, & J. D. Morris (Eds.), Proceedings of the Parapsychological Association, Number 5, 1968 (pp. 85-87). Durham, NC: Parapsychological Association.

Wilson, S. C. & Barber, T. X. (1983). The fantasy-prone personality: Implications for understanding imagery, hypnosis, and parapsychological phenomena. In A. A. Sheikh (Ed.), Imagery: Current theory, research, and application (pp. 340-387). New York: John Wiley & Son.

Wing, Y. K., Lee, S. T., & Chen, C. N. (1994). Sleep paralysis in Chinese: Ghost oppression phenomenon in Hong Kong. Sleep, 17, 609-613.

Zusne, L., & Jones, W. H. (1989). Anomalistic psychology: A study of magical thinking

(2nd ed.). Hillsdale, NJ: Lawrence Erlbaum.