|
ACKNOWLEDGEMENTS:
I should like to thank Dr Jake Empson
(my MSc Supervisor) of Hull University for providing me with
a knowledge of sleep-research technique , and my current
Supervisor, Dr Graham Wagstaff, for his valuable comments on
my work.
I should especially like to thank the
main subject in this study, Alan Worsley of Hull, for his
great co-operation over the 3 years of experimentation.
Also, all the other subjects who volunteered to spend nights
in the sleep-laboratory with no remuneration.
The Department’s workshop staff,
under Eric Britton, deserve credit for their friendly
efficiency in providing equipment, and I am grateful to
Brian Mitchell and Eddy Cookson for designing and
constructing the 'CEMOS' device to my
specifications.
Keith M.T.Hearne
Dept. of Psychology
University of Liverpool. May
1978.
ABSTRACT
:
The aim of this
research was to make original investigations into
‘lucid-dreams’ (those in which the dreamer has
insight that the experience is a dream). A new method of
ocular signalling from these dreams was discovered, so
circumventing the bodily paralysis of Stage REM sleep, and
establishing a mode of communication from the sleeping
subject.
All-night Polygraphic
recordings were obtained from 18 subjects who reported
having lucid-dreams. However, after extensive monitoring
only two of the eighteen subjects were able to produce
lucid-dreams in the laboratory. Much physiological and
psychological information on these dreams in the best
subject was made available using the new technique. All the
lucid-dreams occurred in Stage REM sleep and had a mean
duration of 4 minutes. There were no differences in the
sleep-patterns between Control and lucid-dream nights. The
temporal order of events reported on waking corresponded in
general to the signalled information. A group of simulating
Control subjects were unable to reproduce ocular signals
with REM EEG on waking from Stage REM sleep. Additional data
was analysed concerning home lucid-dreams. A 4-day cycle
accounted for 25% of the subject’s lucid-dreams and
they tended to occur more after days of above average
stimulation.
A large group of
persons who reported having lucid-dreams provided
questionnaire data. Personality and intelligence factors
were also studied in relation to these dreams, but no
significant findings resulted.
A method of induction
of lucid-dreams was tried unsuccessfully on a group of
subjects, but a later technique showed promise. A study of
2-way communication between subject and experimenter was
inconclusive.
Three inventions were
devised as a result of this research: a switching device
operated by ocular signals; a device for waking persons at
the early stage of nightmares ; a device to induce
lucid-dreams and false-awakenings.
CHAPTER I
I.1 AIMS OF THIS
RESEARCH
The purpose of this
programme of research was to investigate a remarkable type
of dream (the 'lucid' dream) in which, reportedly,
consciousness and volitional control are present i.e. the
dreamer has insight whilst dreaming that the experience is a
dream and can, to some extent, manipulate dream content and
course of action. Very little appeared to be known about
lucid-dreams, yet it seemed that, potentially, they held the
key to unravelling much about dreams generally, and also
could assist the understanding of other psychological
processes such as memory and thought.
It seemed not
unreasonable to suppose that suitable subjects could report
information from ongoing lucid-dreams in some way. This
would provide knowledge on dreams from within the dream for
the first time. Obviously, a signalling technique (from the
subject) would first have to be devised, though.
A primary aim of the
research programme therefore was to obtain subjects who
report having lucid-dreams and perform all-night polygraphic
monitoring on them. Providing a signalling method could be
established, basic electrophysiological data was to be
ascertained about lucid-dreams, together with
psychological information. One objective was to
determine whether lucid-dreams are in fact true dreams
occurring in Stage REM sleep, or whether they are a
phenomenon of imagery experienced on waking.
Electrophysiological monitoring of subjects could answer
that question. Since no previous work appeared to have been
conducted on lucid-dreams, the actual course of
experimentation in that respect would develop as findings
became available.
In addition to
polygraphic monitoring of lucid-dream subjects, it was
planned to attempt the artificial induction of lucid-dreams
in subjects in order to make research more efficient. Also,
questionnaire data from lucid-dreamers would be obtained and
analysed to seek any connections between various imagery and
sleep phenomena, in the hope of finding clues as to any
possible causes of lucid-dreams. Another aim would be to
develop any devices which might be useful regarding the
induction of lucid-dreams or as aids in
experimentation.
CHAPTER I
I.2
THE FORMAT
This thesis consists
of four main parts. The first is introductory, consisting of
information on : The methodology concerning the
electrophysiological study of sleep and dreams ; general
sleep-research findings ; the history of dreams and various
dream theories ; collated data on the waking accounts of
lucid-dreams ; philosophical aspects of dreams. These areas
are covered in five Chapters.
In the second part,
the experiments are described in detail. The programme
followed the plan outlined in I.1. One of the lucid-dreamers
was particularly co-operative and produced much valuable
sleep-lab and questionnaire data. Once a method of
signalling was perfected, one precautionary study involved
seeing whether simulating Controls could reproduce the same
type of signal when woken from Stage REM sleep. Another
study which later suggested itself on the basis of earlier
findings was that of testing personality and intelligence
factors of subjects in relation to their reported frequency
of experiencing lucid-dreams. In all, 10 Chapters catalogue
the experimentation performed in this research.
Part 3 (three
Chapters) consists of descriptions of three devices which
were developed as a direct result of this research. The
first would aid lucid-dream research, but is still in the
developmental stage. Another device is designed to wake
persons from the early stage of nightmares. The third device
is intended to induce lucid-dreams and false-awakenings.
Part 4 of this thesis consists of two Chapters in which the
experimental results are discussed and various theoretical
speculations are proposed, and overall conclusions are
listed and suggestions for further research are
stated.
CHAPTER II
II.1 BRIEF
HISTORICAL BACKGROUND TO ELECTRO-PHYSIOLOGY
Galvani (c 1790)
discovered that the current generated by two dissimilar
metals applied to the crural nerve in the leg of a frog
caused twitching of the attached muscle. This demonstration
showed that nerves conduct electrical impulses rather than
some 'vital fluid' - a view that had held for centuries and
was most elaborately propounded by Descartes (Lindsley &
Wicke, 1974 ; Sheer, 1961.) Later, Nobili (1827) first
measured electrical activity in frog muscles.
When technical
developments in current detection permitted, Caton (1875) at
Liverpool university performed the first published
experiments in monitoring the very small electrical activity
from the exposed brains of rabbits and monkeys. Caton
observed a constantly changing background current and
changes at the sensory surface of the brain during sensory
stimulation.
At the beginning of
this century, several investigators began to study muscles
and nerves electrically, and in the 1920s electronic
amplification became available for electro-physiological
work following the development of the vacuum
tube.
The neuro-psychiatrist
Hans Berger (1929) at the university of Jena published an
account of the recording of electrical activity from the
scalps of human subjects (Gloor, 1969). He reported the
discovery of rhythmic 10Kz waves (which he termed 'alpha
waves') in subjects with eyes closed. In addition, he
observed smaller amplitude faster frequency activity which
he called 'beta waves'.
He also termed the
whole record the 'Elektrenkephalogram' (EEG). For
electrodes, Berger used two large saline pads on the
forehead and occiput.
His findings were
treated sceptically by other electro-physiologists until
Adrian & Matthews (1934) replicated his results. Many
varied investigations then began and the rapid advancements
in equipment (e.g. multiple channel recording, cathode-ray
oscilloscope monitoring) aided this work.
Apart from animal
studies, investigations were initiated to seek
physiological, psychological and pathological correlates of
the EEG in humans. Loomis, Harvey & Hobart (1935, 1936)
observed the EEG of sleep and noted vast changes during that
state.
Berger's original
observation that epilepsy and other neurological disorders
produced an abnormal EEG was taken up by others. Dawson
(1951) introduced an 'averaging' technique for teasing out
minute evoked responses from the background EEG. W.G.
Walter, Cooper, Aldridge, McCullum & Winter (1964) first
observed a slow negative potential (d.c.) shift associated
with anticipation - the Contingent Negative Variation
(CNV).
From the point of view
of sleep research, a most important discovery was that of
the different sleep-stages - including REM
(Rapid-Eye-Movement) sleep, which was shown to be associated
with subjective reports of dreaming (Aserinsky &
Kleitman, 1953) ; Aserinsky & Kleitman, 1955 ; Dement
& Kleitman, 1957b).
CHAPTER II
II.2
ELECTRO-PHYSIOLOGICAL MEASUREMENT
- a. Technical
points
AMPLIFIERS
The minuteness of
electro-physiological measures, especially the
electro-encephalogram (measured in millionths of a volt),
necessitates the use of very sensitive high-gain amplifiers
for monitoring and recording purposes. In modern research,
multiple-channel high quality instruments (polygraphs),
often linked to computers, enable the sophisticated
recording and analysis of data. A typical instrument is
equipped with variable time-constant, variable chart speed
and electronic filtering facilities.
ELECTRODES
The interface between
skin and recording instrumentation is of crucial importance
in obtaining accurate measurement. High-conductivity silver
electrodes coated with silver-chloride are commonly employed
in electro-physiological work. Their relative non-polarising
characteristic permits direct-current potentials to be
recorded without a constant signal shift. Electrodes need to
be firmly attached with collodion glue (where hair is
present) or surgical tape to the skin.
ELECTRODE
GEL
Electrolytic past or
gel - a chloride salt of a formula consistent with the
chemistry of the epidermis, is placed between the electrode
and skin, to conduct the electrical potentials. A grease
solvent such as acetone is used to cleanse the skin so
reducing skin-resistance before attachment of
electrodes.
ARTEFACTS
A number of sources of
artefact exist which can obliterate or modify measured
potentials. For instance, skin-stretching occurring when the
subject moves, can cause high-voltage transients. Electrical
interference ('mains hum') is another potential bug-bear
which may be present when electrodes are poorly attached or
the subject not grounded.
Bias potential results
from two electrodes having an imbalance in ionic transfer,
due to different metallic properties or surface
contamination.
Polarisation is a
back-electromotive force occurring as a result of
electrolysis between the electrode and electrolyte - in one
direction, so either increasing or decreasing the true
potential. (Thompson & Patterson, 1974 ; Greenfield
& Sternbach, 1977).
CHAPTER II
- b. The
electro-encephalogram (EEG), electro-oculogram (EOG) and
electro-myogram (EMG)
The
electro-encephalogram is a graph of voltage plotted over
time, measured from the most superficial layers of the
cerebral cortex (Stevens, 1974). The frequency and amplitude
of the monitored brain activity provide the basic data for
the encephalographer. Two modes of electrode placement exist
i.e. monopolar (referential) or bipolar. In the former case
there is an active recording electrode which is 'referred'
to an 'indifferent' electrode positioned on a supposedly
electrically neutral site such as ear-lobe. In bipolar
recording, the signal represents the difference electrically
between the two electrodes.
The international
10/20 system of electrode placement (Jasper, 1958) has been
widely adopted for EEG recording. This uniform system
enables a better comparison of studies from different
laboratories. Electrodes are positioned at points on
imaginary circles 10 or 20 percent of the distance along the
axes from nasion to inion and preauricular points coronally
(Figure I.1, page 17). Gibbs & Gibbs (1964) criticised
the 10/20 system as being geometric rather than satisfying
the requirements for the best electrical placements. Remond
& Torres (1964) modified the 10/20 system for use with
infants and small children.
In the normal EEG
there are four main frequency bands. Changes in the
predominance of different bands occur during maturation
(Lindsley & Wicke, 1974). These bands are :
DELTA
W.G. Walter (1937)
introduced this term to describe certain 'high voltage'
(perhaps a few hundred microvolts) slow waves of a frequency
of 0.5 to 3Hz. Delta activity is found in the waking EEG of
infants and young children, but is abnormal in adults.
Factors which cause an increase in intra-cranial pressure,
for instance a brain tumour, are linked with the presence of
Delta waves. They are also present in Stage 4 sleep
(slow-wave sleep) and unconsciousness (Lindsley & Wicke,
1974).
THETA
This term was also
introduced by W.G. Walter (1953). Theta waves have a
frequency range of 4-7Hz. During maturation, theta
predominates in all head regions, though mainly from
posterior and temporal areas. The frequency is slightly
higher in the frontal lobes. Theta activity is abundant in
childhood and early adult life but decreases in the 20s and
is abnormal beyond the age of 30. The presence of theta from
the temporal regions of adults and teenagers is thought to
be associated with delayed cerebral maturation and is often
found in persons with severe behavioural disorders and
psychopathy (Hill, 1952). Theta waves are linked with the
hippocampus and limbic system (Green & Arduini, 1954) ;
amplitude is usually under 20 microvolts.
ALPHA
Alpha activity, of a
frequency range 8 - 13Hz, first appears in mid-childhood. It
is prominent posteriorly over the visual cortex. Typically,
it appears in bursts or 'spindles' of 20-100 microvolts.
Lindsley (1938, 1939) found the mean frequency from a large
adult population to be 10.2Hz. Its frequency may vary by
about half a cycle, however in hypothyroidism for instance,
the frequency is much reduced. In fevers, the frequency may
be elevated one or two cycles.
There is much
individual variation in the amount of alpha present in the
waking EEG. A few persons show virtually continuous alpha
('P' type of Golla, Hutton & Walter, 1943) ; a minority
have little or none ('B' type of Davis, 1941). Most people
fit between these two extremes.
The generator sites
are not yet known (Andersen & Andersson, 1968). The
activity is stronger over the sensory and associated areas
of the posterior cortex but is also present over frontal
regions. It has an underlying pacemaker mechanism in the
thalamus which is linked to the ascending reticular
activating system. Sensory input of any kind can
de-synchronise alpha - this is termed 'alpha-blocking'.
Lindsley & Wicke (1974) state that alpha is sensitive to
unexpected sensory stimuli, to factors which modify the
state of arousal and alertness or vigilance and events which
elicit or demand specific attention whether they be external
events or internal events such as thoughts, ideas, worries,
etc. A laterality effect or asymmetrical effect is observed
I about 30% of adults i.e. one hemisphere has a greater
amplitude - usually the right or 'dominant' hemisphere
(Cobb, 1963). In recent years, the volitional control of
alpha using biofeedback methods has become popular (Kamiya,
1962, 1967, 1969 ; Hart, 1967).
BETA
This common
low-voltage (usually under 20 microvolts) activity of
frequency range 14 to 30Hz is prominent from the frontal
lobes during adulthood. Their study has been much neglected
(Lindsley & Wicke, 1974). Jasper & Penfield (1949)
found that in a patient with an exposed part of the motor
cortex, beta waves at local regions could be blocked by
voluntary effort.
GAMMA
Jasper & Andrews
(1938) divided up the beta activity described by Berger
(1929) as 20-50Hz into beta waves of 14-30Hz and gamma waves
of 30-50Hz.
KAPPA
Kennedy, Gottsanker,
Armington & Gray (1948) found a frequency similar to
alpha (8-12Hz) of about 20 microvolts at the temples, which
seemed to be associated with intellectual activity. The
bursts of kappa are supposed to increase with reading,
memory and arithmetic tasks, and problem-solving. Not all
subjects evince the waves, but Chapman (1972) suggests that
where present it is a reliable effect.
MU
Gastaut (1952)
described this 9-11Hz rhythmic bursts of which appear in the
EEG of about 7% of subjects (Other names are : 'comb',
'wicket', 'rythme en arceau') : It is rare after the age of
30. It is found in the Rolandic area, usually bilaterally
asynchronous. The rhythm is apparently decreased by movement
or intention to move the contralateral limb.
LAMBDA
These are single
positive waves of 'sawtooth' appearance (over 250mS)
recorded at the occiput in some people (Gastaut, 1951 ;
Evans, 1952). They seem to be linked with visual
perception.
VERTEX
WAVES
These are single sharp
negative waves (generally under 25 microvolts) over the
vertex. They occur randomly - especially in children
(Gastaut, 1953) ; 20% of normal adults show them (Roth, Shaw
& Green, 1956).
The electro-oculogram
(EOG) is a recording of eye-movements obtained from
electrodes placed usually above and under the outer canthus
of each eye. The electrode arrangement can be varied
according to the type of ocular activity being studied e.g.
vertical, horizontal or oblique movements. The electrodes
pick up potentials caused by movements of the dipole moment
of the electrical charge on the retina and cornea of the
eye. The cornea is positive (by 1 millivolt) relative to the
retina because of the higher metabolic activity of the
latter (Greenfield & Sternback, 1972).
The electromyogram
(EMG) is a recording of muscle potentials. Electrodes placed
over a muscle indicate the general level of tonus as well as
monitoring discrete contractions (Greenfield &
Sternbach, 1972).
CHAPTER II
II.3 HUMAN
SLEEP-STAGES AND SCORING CRITERIA
Oswald (1962) defined
sleep as a healthy recurrent condition of inertia and
unresponsiveness. Its study was somewhat limited until
all-night polygraphic monitoring of subjects was performed
and the various sleep-stages discovered (Aserinsky &
Kleitman, 1953 ; Dement & Kleitman 1957b). In general
terms there are two sleep states : Rapid eye movement sleep
(REM) and non-REM (NREM). The terminology of sleep states
has varied remarkably over the years so that even totally
contradictory terms refer to the same state. Freemon (1972)
found 25 different nomenclatures for REM and NREM sleep
states in the literature (Table II.1, page 18).
Four NREM stages have
been distinguished by their different appearance in the
polygraphic record. In human sleep there is a roughly 90
minute cycle during sleep in which the different stages
appear sequentially.
Typically, the subject
enters NREM stages 1 through to 4, then reverses back to
stage 2, after which stage REM occurs. This pattern is
repeated several times throughout the night, but the amount
of stage 4 decreases each time and the duration of the REM
state increases (Figure II.2, page 19).
Rechtschaffen &
Kales (1958) published a manual for scoring sleep stages so
as to standardise scoring criteria. The authors suggested,
among other things, a minimum chart speed of 10mm/sec for
clear identification of EEG frequencies, a minimum
time-constant of 0.3 secs and a minimum pen deflection of
7.5 - 10mm for 50 microvolts. EEG monitoring from positions
C4/A1 or C3/A2 (according to the 10/20 system) was proposed.
In EMG recording, high amplification is suggested (20
microvolts or higher) with a fast time-constant to eliminate
slow potentials from other sources which could cause
amplifier blocking at high gain. Records are scored by
judging which sleep stage is present on each page (epoch) -
usually of about 20-30 seconds ; this judgement sometimes
depends on preceding or following epochs. The total
percentage of the different stages can then be computed.
The sleep stages are
:
STAGE 1 :
This stage occurs
first when falling asleep, or after gross body movements in
sleep. The EEG is low-voltage mixed-frequency activity, with
many 2-7Hz waves. In its latter part, vertex sharp waves may
occur. There are often large slow rolling eye-movements in
the EOG. The EMG level is usually lower than that of relaxed
wakefulness (Roth, 1961).
STAGE 2 :
This stage has
'k-complexes' (Loomis et al, 1938) and/or sleep-spindles
present, but the EEG amplitude is still generally low (under
75 microvolts). A k-complex is an EEG wave having a sharp
negative front followed by a positive component : for
scoring purposes it should exceed 0.5 secs. They occur in
response to sudden external stimuli - but may also occur
spontaneously (Johnson & Karpam, 1968).
Sleep-spindles are
bursts of 12-14Hz activity occurring often with a
k-complex.
STAGE 3 :
This stage has been
arbitrarily defined as one in which the EEG shows a minimum
of 20% and maximum of 50% f 2Hz or slower waves (delta)
having an amplitude of at least 75 microvolts peak-to-peak.
K-complexes and spindles may be present in stage
three.
STAGE 4 :
Here, the EEG record
shows 50% or more of 2Hz or slower waves with a minimum
amplitude of 75 microvolts ; sleep-spindles may or may not
occur.
STAGE REM :
The EEG here is of
low-voltage mixed frequency, like that of Stage 1, with -
very often - distinctive 'saw-tooth' waves (Schwartz &
Fischgold, 1960 ; Berger, Olley & Oswald, 1962). Alpha
is usually a little more prominent than in stage
1
But the frequency is
slower by 1-2Hz than during wakefulness (Johnson, Nute,
Austin & Lubin, 1967). No k-complexes or spindles are
present in stage REM. A main characteristic is the presence
of episodic REMs. Stage REM sleep is not so scored if
mental-submental muscle tonus is high in the EMG (Berger
1961 ; Jacobson, Kales, Lehman & Hoedemaker, 1964).
Complicated and specific rules for scoring stage REM under
all conceivable conditions are stated in the sleep-manual of
Rechtschaffen & Kales (1958).
The basic
electro-physiological criteria of sleep having been stated,
in the next chapter an overall view of general
sleep-research findings will be reviewed to illustrate the
nature of sleep and the various experimental
approaches.
(Keith Hearne's PhD
thesis, pages 17 - 21)
Page 17 : Figure of
the 'ten-twenty' electrode system of electrode
placement
Page 18 : List of
nomencalture of sleep stages
Page 19 : Figure of
typical night of sleep in a young adult (sleep
stages)
Pages 20 - 21 : Figure
of EEGs of different sleep stages.
CHAPTER
III
GENERAL
SLEEP-RESEARCH FINDINGS
III.1 THE
PHYSIOLOGY OF SLEEP
Numerous physiological
changes are correlated with sleep, reflecting the alteration
in level of metabolism associated with the rest / activity
cycle. Body temperature is affected by metabolic rate
(measured by oxygen consumption or rate of heat-loss).
In sleep, oxygen consumption falls off gradually reaching a
nadir after some 6 hours : at that point the curve shows a
small inflection (Brebbia & Altshuler, 1965) ; rectal
temperature shows a similar decline curve (Kreider, Busirk
& Bass, 1958). Pulse rate begins to decline before sleep
when the body is fairly inactive and falls sharply at first
(Schaff, Marbach & Vogt, 1962). Respiratory depression
is another characteristic of sleep and the expired air
contains increased levels of carbon dioxide (Kleitman, 1963
). These metabolic measures are usually quite stable in NREM
sleep, but fluctuations are apparent in Stage REM
(see p 24 ). Basal skin resistance appears to alter too
throughout the night ; workers have reported that resistance
increases i.e. conductivity is decreased (Farmer &
Chambers, 1925 ; Batini, Fressy & Coquery, 1965). Landis
(1927) attributed this to drying of electrodes and
polarisation. Other experimenters have reported
different curves depending on whether a continuous or
intermittent current was used (Wenger, 1962 ; Tart, 1967).
This measure therefore remains controversial ; studies of
blood-pressure in sleep have been inconclusive for the
technical reason of accompanying sleep disturbance.
Generally though, there is evidence that systolic pressure
is positively correlated with depth of sleep (Snyder &
Scott, 1972 ). Plethysmographical studies have shown that
vascular dilation of the hands and feet occurs during sleep
(Howell, 1897; Johnson & Lubin, 1967).
Body movement is
limited during NREM sleep although motility is higher in
Stage REM. Overall, the number of movements increases slowly
after the first hour or so (Snyder & Scott, 1972 ).
Kleitman, Cooperman & Mullin (1933) reported that a
person may make 20-60 postural re-adjustments during the
night, but these total a mere 3-5 minutes. Brazier &
Beech (1952) found that 6 minutes before a movement, cardiac
acceleration occurs. During movement the EEG becomes less
synchronised. Auditory thresholds are lowest after a
movement and highest some 16-20 minutes later (Mullin,
Kleitman & Cooperman, 1937). Motility decreases with
'depth' of sleep although much individual variation is found
(Cathala & Guillard, 1961 ; Rohmer, Schaff, Collard
& Kurtz, 1965). Lienert & Othmer (1965) stated that
emotionally stable persons have more body movements than
unstable subjects.
The physiological and
psychological phenomena of REM sleep are so distinct that
the Stage is now considered by many to constitute a
separate third State, along with NREM sleep and wakefulness
(Oswald, 1962 ; Dement, 1974). Aserinsky and Kleitman (1953)
observed that pulse and respiration are generally
higher in REM than NREM sleep. Further, much variability
occurs in REM (Batini et al. 1965 ; Snyder, Hobson, Morrison
& Goldfrank, 1964). Blood pressure behaves in a similar
manner (Khatri & Fries, 1967 ; Snyder, Hobson &
Goldfrank, 1963). Mean increase of these measures in REM
sleep from the mean NREM level, was 50% (Snyder & Scott,
1972). Fluctuations also are seen in plethysmographic pulse
amplitude and finger skin-temperature (Snyder, 1967),
however the Galvanic Skin Response (GSR) and basal
skin resistance remain relatively more stable in REM than
NREM sleep (Asahina, 1962). The pupil, an index of autonomic
activity when awake, remains constricted during sleep and
REM (Rechtschaffen & Foulkes, 1965). Brain temperature,
which stays fairly constant in NREM sleep, increases
significantly in Stage REM sleep (Kawamura & Sawyer,
1965). In males penile erections are associated with Stage
REM (Ohlmeyer, Brilmayer & Huellstrung,1944) ; Fisher,
Gross & Zulch (1965a) found evidence that the phenomenon
is not affected by sexual gratification. Karacan,
Goodenough, Shapiro & Starker (1966) found, though, that
if Stage REM is prevented by wakening, the erection cycle
appears in other Stages at the expected times i.e. in the 90
minute cycle. A phenomenon associated with the phasic REM
bursts is activity of the stapedius muscle of the middle ear
(Baust & Rohrwasser, 1964). In REM sleep (but not NREM)
bodily paralysis is present, as indicated by EMG
suppression. Actively induced tonic non-reciprocal motor
inhibition occurs which blocks the frenzied activity of the
brain during REM (Dement & Mitler,1974). Only small
twitches are observed occasionally. Electrically induced
reflexes are suppressed in REM indicating active motor
inhibition (Hodes & Dement, 1964 ; Pompeiano, 1965,
1970), Tendon reflexes are abolished and voluntary movement
is impossible. Sometimes, a person may wake from Stage REM
to find the body paralysed (Sleep-paralysis, page 47).
Bremer (1974) remarks that the state of paralysis resembles
the 'apparent death' of lower vertebrates and that perhaps
nature uses this archaic inhibitory apparatus for protection
of the dreamer.
CHAPTER
III
GENERAL
SLEEP-RESEARCH FINDINGS
III.2 THE CHANGING
CONCEPT OF SLEEP
Early ideas of sleep
inclined to a 'passive' theory that sleep occurs to prevent
fatigue or is caused by a lack of sensory stimulation
(Claparède, 1908 ; Coriat, 1912). 'Active' theories
also appeared i.e. that the brain actively inhibited
consciousness. Pavlov (1923) thought that sleep was the
result of cortical inhibition spreading from certain areas,
and Hess (1931) discovered that cats could be put to sleep
by electrical stimulation of the diencephalon. Bremer (1935)
invoked the passive notion to explain his finding that the
cerveau isolé cat (having a cut through the upper mid
brain) remained in virtually continuous sleep. He thought
the animal was not receiving enough sensory stimulation to
keep awake. In encéphale isolé animals (where
the cut is in the lower mid-brain) the sleep-wake cycle
persists (Bremer, 1935). Thus, the sleep mechanism seems to
be located between these brain areas. Moruzzi & Magoun
(1949) discovered that electrical stimulation of the
reticular formation roused a sleeping or anaesthetised cat.
'Reverberating loops' were supposed to keep the animal awake
in the absence of stimulation (Magoun, 1952).
It became generally
accepted that the reticular formation stimulates the cortex
to consciousness. Sensory information from the sense organs
is routed to the cortex whilst collateral afferents from
these nerves link with the reticular formation. Lesions of
the pathways to the cortex do not cause sleep, whereas
lesions between the reticular formation and the cortex do
(Lindsley, Schreine, Knowles & Magoun, 1950).
Apparently, impulses from the collateral afferents excite
the reticular formation to send diffuse 'activating'
impulses to the cortex, so maintaining
wakefulness.
There seems to be an
inherent rhythmic sleep-wake cycle in the upper reticular
formation but wakefulness is aided by external sensory
stimulation (Oswald, 1962). Animals without sense organs
tend to sleep excessively (Hagamen, 1959). Several factors
assist in maintaining wakefulness by stimulation of the
reticular formation, the 'gating' function of which controls
consciousness. For instance, a decrease in blood oxygen
content stimulates chemoreceptors in the carotid body which
in turn stimulate the reticular formation. An excess of
carbon dioxide in the blood also causes mid-brain
stimulation (Bonvallet et al, 1955). Hypothalamic
thermodetectors can affect the reticular formation too
(Hagamen, 1959), and various influences may also diminish
mid-brain activity, so promoting sleep. Baroreceptors in the
carotid sinus and aortic arch dampen the reticular formation
(Bonvallet, 1955). Heating of the hypothalamus encourages
sleep unless excessive (Euler & Söderburg, 1957.)
The cerebral cortex itself is capable of influencing the
organism’s own state of wakefulness (Hugelin &
Bonvallet, 1957a,b ; 1958). Worries can keep a person awake
and Cannon (1942) stated that in primitive cultures (eg
Aborigines) sudden death can occur in persons on the
receiving end of meaningful symbolic acts (eg pointing
a bone). Obviously, networks of feedback loops operate
between the activating reticular formation and the cerebral
cortex.
Not everyone
subscribes to the concept though ; Freemon (1972) states
that stimulation of the brain stem near the reticular
formation can lead to slow waves ; this is the opposite of
the Moruzzi and Magoun finding.
Freemon also says that
the reticular formation does not project diffusely to the
neocortex, but to the limbic areas and orbito frontal
cortex, returning then to the reticular formation (Scheibel
& Scheibel,1967.) Hippocampal arousal (shown by
de-synchronisation of the EEG) occurs several seconds before
neocortical arousal on external stimulation in NREM sleep
(Freemon & Walter, 1970). Some argument exists therefore
over the notion of the reticular formation’s direct
involvement in causing sleep.
CHAPTER
III
GENERAL
SLEEP-RESEARCH FINDINGS
III.3 DEVELOPMENTAL
ASPECTS OF SLEEP
Differences have been
discussed between sleep EEG waveforms for different ages
(II.2.(b)). Studies of premature babies show that a
virtually constant EEG pattern exists before full-term
(Parmalee & Wenner, 1967).
Slow waves do not
become evident in the sleeping EEG, along with spindles and
K-complexes, until 3 months of age, although Stage REM is
present at birth and may constitute 50% of the 16 hours or
so daily sleep for the first few weeks (Gibbs & Gibbs,
1950b).
The total amount of
sleep and the relative amount of REM decrease steadily until
approximately 4 years of age after which it varies within
some 2-3% over the years, averaging about 2-3% (Roffwarg,
Dement & Fisher, 1966). Kales, Kales, Jackson, Po &
Green (1967) found 30% Stage 4 and 29% Stage 3 in children
compared to 11% and 10% respectively for young
adults.
Significant changes in
the distribution of sleep also occur in the early
years of life. The new-born baby has 5 or 6 periods of
wakefulness which reduces to 3 or 4 by 6 months (elimination
of night feeding is probably responsible Kleitman 1939). At
1 year most infants have a solid 12-14 hour sleeping period
with some day sleep (Gessel & Ametruda, 1945.) Thus,
early polyphasic sleep is altered by socialisation and
maturational factors to a monophasic form. No sex
differences appear to exist between the various sleep Stages
in young adults (Williams, Agnew & Webb, 1964,
1966).
The main change in EEG
of the aged is gradual loss of Delta activity (Agnew, Webb
& Williams, 1967 ; Kales, Jacobson, Kales, Kun &
Weissbuch, 1967) although this could reflect a reduced need
for deep sleep. The percentage of Stage REM in the sleep of
aged persons has varied in different studies. Feinberg,
Koresko & Heller (1967) found over 20% Stage REM,
whereas Lairy, Cor-Mordret, Faure & Ridjanovic (1962)
give a figure of 14%. However, old persons often take
'cat-naps' during the day which may affect the natural sleep
pattern - thus a polyphasic distribution of sleep may
recur.
CHAPTER
III
GENERAL
SLEEP-RESEARCH FINDINGS
III.4 THE
PHARMACOLOGY OF SLEEP
The two states of
sleep (NREM & REM) appear to be governed by different
neurochemical systems. Injections of 5-hydroxytryptophane
(5-HTP) (a precursor of 5-HTP or serotonin) in cats causes
NREM sleep (Jouvet, 1967). Injections of reserpine in cats
suppresses both states, but subsequent injections of 5-HTP
selectively restores NREM sleep (Matsumoto & Jouvet,
1964 ). Parachlorphenylalanine (p-PCA) selectively blocks
5-HTP synthesis, and Weitzman, Rapport, Mc Gregor &
Jacobs (1968) discovered that when injected into monkeys it
decreased the amount of sleep by reducing NREM sleep : REM
sleep was unaffected. Significantly, anaesthetics increase
the amount of serotonin in the brain (Freemon, 1972). Thus,
5-HT appears to be important regarding the presence of the
NREM state.
It is possible that
the cholinergic system however is important for the
production of REM sleep. For instance, the REM state is
enhanced in cats by carbachol (a cholinomimetic) and reduced
by atropine (a cholinergic blocking agent. In addition,
injections of acetylcholine near the locus coeruleus trigger
REM sleep in cats (George, Haslett & Jenden, 1964).
Jouvet (1969) though, implicated the nor-adrenaline system
in the control of REM sleep. Thus, after depletion of
nor-adrenaline by reserpine, Dopa (a nor-adrenaline
precursor) restored REM sleep (Matsumoto & Jouvet, l964
). The paradoxical finding that persons are hard to rouse
from REM sleep (despite the high cortical arousal) could be
supported by assuming the nor-adrenaline system is involved
in behavioural arousal and that the ascending nor-adrenaline
pathways are inactive during REM sleep. Jouvet (1967)
thought a link existed between the nor-adrenaline system of
the pontine part of the brain stem (ventral and caudal to
the locus coeru1eus) and ponto-geniculo-occipital spikes
occurring in the EEG of cats. Jouvet considered that dreams
may be initiated by PGO spikes produced by the release of
mono-amines at this site. Perhaps both neurotransmitter
substances are operating in Stage REM.
Hypnotics affect the
cerebral cortex, the reticular formation or the medulla.
Anxiety, causing insomnia may be treated by tranquillisers
such as chlordiazepoxide (Librium). Depression, which often
results early morning wakening is often alleviated by
antidepressants e.g. amitriptyline (Laroxyl), or
trimipramine (Surmontil). This latter drug does not decrease
REM or result in a rebound effect (Oswald, 1974 ). Pain
which prevents sleep can be treated with morphine or
pethidine. The barbiturates are the most effective soporific
drugs in use. Unfortunately they are lethal in overdose and
can interact with other drugs : they are also addictive
(page 44). It is not known exactly how barbiturates work
except that they produce widespread inhibition in the
cortex. They are either 'long-acting' (e.g. phenylbarbitone)
or 'short-acting' (e.g. quinal-barbitone). Newer drugs have
appeared, such as the benzodiazepines (e.g. Mogadon) or
flurazepam (e.g. Dalmane). These drugs suppress the
reticular formation and overdose is not fatal since the
medulla (controlling breathing) is not affected. The famous
'Micky-Finn' consisted of alcohol and chloral. A modern
version is dichloralphenazone (Welldorm). Sleeping tablets
frequently lead to many problems. Dement & Villablanca
(1974) stated that "with one or two exceptions, all sleeping
pills will always cause or worsen insomnia".
CHAPTER
III
GENERAL
SLEEP-RESEARCH FINDINGS
III.5 SLEEP
DEPRIVATION
Some persons claim to
require little or no sleep (Jones & Oswald, 1968 ;
Meddis, Pearson & Langford, 1973), however, for most
people total sleep deprivation leads after several days to
visual illusions and hallucinations, speech slurring,
inability to concentrate and memory lapses (Ross, 1925 ;
Kollar, Namerow, Pasnau & Naitoh, 1968 ; Cappon &
Banks, 1960 ; Bliss, Clark & West, 1959 ; Morris,
Williams & Lubin, 1960). Paranoid symptoms may also
occur in some subjects (Tyler, 1947, 1955). Boring test
situations produce, not surprisingly, the lowest
performance scores in sleep deprived subjects. Thus, such
persons, when told to signal when they observed a light spot
at any one of 8 points on a screen, over 40 minutes,
performed steadily worse though watching the screen
(Wilkinson, 1960). During auditory tasks errors of omission
occurred with the loss of alpha rhythm (Williams, Lubin
& Goodnow, 1959). Oswald (1962) attributed such
phenomena to falls in cerebral vigilance. Mental
capacities can be improved temporarily to waking
levels on some tasks if subjects can take their time and
amend mistakes.
The EEG of sleep
deprivation shows a decrease in the alpha rhythm of relaxed
wakefulness (Tyler, Goodman & Rothman, 1947).
Additionally, biochemical changes occur, probably due to
lack of restoration which mostly occurs in sleep. Plasma
iron level and plasma cholesterol both decline (Kuhn,
Brodan, Brodancva, & Friedman, l967). Amphetamines
temporarily improve the performance of sleep deprived
subjects on rote tasks (Weiss & Laties,
1962).
On the first recovery
night after sleep deprivation a marked increase in NREM
sleep is observed (Berger & Oswald, 1962 ; Williams,
Hammack, Daly, Dement & Lubin, 1964), whilst the REM
percentage remains the same (Kales, Tan, Kollar, Naitoh,
Preston & Malmstrom, 1970). On subsequent nights REM
sleep is higher. Thus, NREM sleep has priority in the
recovery process. Studies have been conducted on the
selective suppression of REM sleep by means of waking the
subject at its onset, or pharmaceutically by drugs which
suppress the state. A remarkable finding is that a 'rebound'
effect occurs when uninterrupted sleep is once again
permitted. In the case of drugs (most suppress REM), a sharp
decrease in the percentage of REM is seen at first.
Gradually, the percentage rises to normal, due to
physiological tolerance. On cessation of the drug, a rebound
occurs (and the amount is larger than by selective
awakenings) so that it amounts to 150-200% of the loss. A
'need to dream’' has been postulated on such evidence.
Early studies suggested that REM deprivation led to profound
psychological changes such as irritability (Dement, 1960),
extreme hunger (Dement & Fisher, 1963), and oral
behaviour with oral symbolism (Fisher, Gross & Zulch,
1965c). However, Kales, Hoedemaker, Jacobson &
Lichtenstein(1968) failed to detect any psychological
alterations with long term REM deprivation. Also, depressed
patients are not adversely affected by REM deprivation
(Vogel, Traub, Ben-Horin & Meyers, 1968) neither are
schizophrenics (Vogel & Traub, 1968). Indeed,
mono-amine-oxidase-inhibitors (MAOI) which apparently
totally suppress REM do not cause abnormalities (Wyatt, Fram
& Kupfer, 1971).
CHAPTER
III
GENERAL
SLEEP-RESEARCH FINDINGS
III.6 MEMORY AND
SLEEP
Jenkins &
Dallenbach (1924) found evidence that rote-learnt material
was recalled better after 8 hours of sleep than wakefulness,
presumably because of the lack of interference by
subsequently learnt material. Empson & Clarke (1970)
discovered that REM sleep seems to be important for the
consolidation process. 20 yoked pairs of subjects listened
to tapes of nonsense phrases before bed. One subject was
later chosen to be REM deprived, by waking. At that time the
other subject was woken too. The Experimental subjects
showed less recall than the Control subjects woken at random
sleep Stages.
The idea of
'sleep-learning' whereby auditory information is supposedly
absorbed and consolidated in the absence of wakefulness is
popularly believed to be an established effect.
However, Simon & Emmons (1956, 1956a) discovered
failings in methodology in such work. They used all-night
EEG monitoring on 21 subjects. After establishing their
baseline general-knowledge, they presented each of 96
questions and, 5 seconds later, the answer. Subjects had to
call out if they heard an answer during sleep. Subjects were
tested the next day to see if they could recall the answers.
When subjects were awake (showing alpha blocking) at the
original presentations, recall was very good - when drowsy
(alpha present) recall was not so good (50%). Recall was
minimal otherwise. The authors also tried repeating stimuli
- in the form of 10 one-syllable numbers (Emmons &
Simon, 1956b). The Experimental subjects performed no better
than Controls. It therefore appears that memory traces
(engrams) are not laid down during sleep.
CHAPTER
III
GENERAL
SLEEP-RESEARCH FINDINGS
III.7 EXTERNAL
STIMULI AND SLEEP
External stimuli can
affect a sleeping person. In Stage 2 of NREM sleep a sudden
sensory stimulus causes the appearance of a 'K-complex' in
the EEG (Davis, Loomis & Harvey, 1939 ; Roth, Shaw &
Green, 1956).The size of the response appears to be related
to the meaningfulness of the stimulus (Oswald, Taylor &
Treisman (1960) found that subjects responded more to their
name being called than other names or to their name being
played backwards.
In REM sleep external
stimuli may be incorporated into dreams reported on waking.
Berger (1963) found that spoken names were included - often
in a distorted fashion. Thus, 'Naomi' became 'an aim to
ski', and 'Jenny' became 'jemmy'. Dement & Wolpert
(1958) tried stimulating subjects with a tone, light and
water-spray. They found the incorporation amounted to 9%,
23% and 42% respectively. Koulack (1969) used an electrical
stimulator positioned on the median nerve at the wrist and
obtained direct incorporation in 40% of cases and indirect
in 24% , when the stimulus was applied 3 minutes after the
start of the REM and where wakening occurred 3 minutes after
stimulation.
CHAPTER
III
GENERAL
SLEEP-RESEARCH FINDINGS
III.8 SIGNALLING
FROM SLEEP
Several studies have
claimed to show that animals and humans can make movements
or signals (motor acts or speech) from the 2 states of
sleep. The movements though are of a simple repetitive kind,
or single responses to stimulation, not requiring higher
processing. Oswald (1959c, 1960a) had subjects moving arms
and legs rhythmically to music with eyes closed or taped
open. The movements continued but less vigorously with the
cessation of alpha rhythm, but sometimes movements ceased.
Defensive and reflexive movements may occur in sleep, but
these are also found in decorticate animals (Kleitman &
Camille, 1932).
Vaughan (1963) used
macaca monkeys in an experiment to test whether they could
respond (by pressing a bar) to imagery in sensory isolation
as they had previously been tra | 
