H-Reflex Suppression and Autonomic Activation During Lucid REM Sleep: A Case Study

Abstract

A single subject, a proficient lucid dreamer experienced with signaling the onset of lucidity (reflective consciousness of dreaming) by means of voluntary eye movements, spent 4 nonconsecutive nights in the sleep laboratory. The subject reported becoming lucid and signaling in 8 of the 18 rapid-eye movement (REM) periods recorded. Ten lucid dream reports were verified by polygraphic examination of signals, providing a total of 12.5 min of signal-verified lucid REM. H-Reflex amplitude was recorded every 5 s, along with continuous recording of electroencephalogram, electrooculogram, electromyogram, electrocardiogram, finger pulse, and respiration. Significant findings included greater mean H-reflex suppression during lucid REM sleep than during nonlucid REM and correlations of H-reflex suppression with increased eye movement density, heart rate, and respiration rate. These results support previous studies reporting that lucid REM is not, as might be supposed, a state closer to awakening than ordinary, or nonlucid, REM; rather, lucid dreaming occurs during unequivocal REM sleep and is characteristically associated with phasic REM activation.

Full text

Sleep
12(4):374-378, Raven Press, Ltd., New York
©
1989
Association
of
Professional Sleep Societies
Short Communication
H-Reflex Suppression and Autonomic
Activation During Lucid REM Sleep:
A Case Study
* Andrew Brylowski, Lynne Levitan, and Stephen LaBerge
*University
of
Texas Medical School
at
Houston, Houston, Texas,
and
Sleep Research
Center, Stanford University School
of
Medicine, Stanford, California,
U.S.A.
Summary: A single subject, a proficient lucid dreamer experienced with sig-
naling the onset
of
lucidity (reflective consciousness
of
dreaming) by means
of
voluntary eye movements, spent 4 nonconsecutive nights
in
the sleep labora-
tory. The subject reported becoming lucid and signaling in 8
of
the
18
rapid-eye
movement (REM) periods recorded. Ten lucid dream reports were verified by
polygraphic examination
of
signals, providing a total
of
12.5 min
of
signal-
verified lucid REM. H-Reflex amplitude was recorded every 5 s, along with
continuous recording
of
electroencephalogram, electrooculogram, electromyo-
gram, electrocardiogram, finger pulse, and respiration. Significant findings in-
cluded greater mean H-reflex suppression during lucid REM sleep than during
nonlucid REM and correlations
of
H-reflex suppression with increased eye
movement density, heart rate, and respiration rate. These results support pre-
vious studies reporting that lucid REM is not, as might be supposed, a state
closer to awakening than ordinary, or nonlucid, REM; rather, lucid dreaming
occurs during unequivocal REM sleep and
is
characteristically associated with
phasic REM activation. Key Words:
H-Reflex-Lucid
REM sleep.
Electrophysiological studies have shown that
"lucid"
dreams, in which persons are
explicitly aware that they are dreaming while continuing to dream, occur almost ex-
clusively in rapid-eye-movement (REM) sleep (1). Further investigations (2,3) have
indicated that lucid REM sleep
is
characterized by higher levels
of
phasic activity and
autonomic activation than ordinary REM sleep, and therefore that it is likely that
lucidity occurs during periods
of
relatively great cortical activation. Contrary to some
earlier suggestions that lucid dreaming might be the result
of
microawakenings, brief
intrusions
of
waking consciousness, in REM periods (4), the findings that lucidity
is
Accepted for publication September
1988.
Address correspondence and reprint requests to Dr.
A.
Brylowski at Department
of
Psychiatry, South-
western Medical School at Dallas,
5323
Harry Hines Blvd., Dallas, TX 75235-9070, U.S.A.
374

H-REFLEX
SUPPRESSION
IN
LUCID
REM
375
associated with more intense, activated REM sleep suggest rather that lucid REM is a
deep sleep state and not nearer to waking than ordinary
REM
sleep.
The primary goal
of
this experiment was to investigate the status
of
H-reflex sup-
pression in lucid dreams. Given that both human (5,6) and feline (7,8) studies have
demonstrated tonic reflex inhibition during phasic
REM
and that lucidity
is
associated
with phasically active REM, it was predicted that the H-reflex during lucid
REM
would
be
at
least as suppressed as it is in ordinary REM.
In the H-reflex analysis, the polygraph records were divided into 5-s epochs instead
of
the standard 30-s epochs (9), enabling correspondingly finer resolution
of
the
state
changes during REM and especially lucid
REM
sleep.
It
was possible to identify mi-
croawakenings and transitions to non-REM (NREM) sleep if any occurred during the
periods
of
lucidity.
METHODS
The subject was a 25-year-old medical student,
an
experienced lucid dreamer who
was familiar with the method
of
signaling from lucid dreams by means
of
eye move-
ments verifiable
on
the polygraph record (1). His sleep was normal during an all-night
polysomnogram without continuous H-reflex monitoring. On 4 nonconsecutive nights
his H-reflex was measured as he slept with the intention to recognize when he was
dreaming and to signal the onset
of
lucidity.
The experimental protocol called for the subject to signal the onset
of
lucidity with
two pairs
of
extreme left-right eye movements and to signal awakening with four pairs
of
left-right eye movements. The subject wrote reports
of
mental activity after each
spontaneous awakening, including the times
of
awakening. Following previously de-
scribed methodology
(I),
reports were scored as signal-verified lucid dreams
if
(a)
the
subject reported that he had become lucid and made the eye movement signals and (b)
the corresponding section
of
the polygraph record showed the reported signals.
A Grass model 8-18D was used to record several channels
of
electroencephalogram,
induding
the
standard central positions used for sleep scoring, vertical and horizontal
electrooculogram, electrocardiogram, nasal airflow, finger pulse, and H-reflex ampli-
tude. Standard procedures for measuring H-reflex were used, modified for continuous
measurement (10,11). The subject's knee was immobilized in a knee immobilizer set
at
100
flexion and a O.I-ms
50-
V stimulus was delivered to the right posterior tibial nerve
every 5 s as the contraction
of
the soleus muscle was monitored. A
TECA
model TE42
electromyogram (EMG) machine provided the electrical stimulus and amplified the
EMG signal, which was sent to the J5 external input
on
the Grass polygraph.
The polygraph records were scored according to the standard criteria (9). The
REM
periods, including the
30
s pre and
post
REM, were subdivided into 5-s epochs, one
epoch for each H-reflex sample. These 5-s epochs were again scored for sleep stages
according to the standard criteria modified for 5-s epochs. An epoch was scored as
Stage 2 if no change in stage had occurred between K-complexes
or
sleep spindles
or
if
a K-complex
or
sleep spindle was present anywhere in the epoch. Also,
an
epoch was
scored as lucid REM only if it contained an eye movement signal
or
lay between epochs
containing signals corresponding with the subject's written report (see Fig.
1).
In addition to sleep staging, several other physiological measures were collected and
analyzed in 5-s epochs. These were H-reflex amplitude (millimeters
of
pen
deflection),
eye movement density (number
of
eye movements
per
epoch), heart rate (beats
per
Sleep, Vol.
12,
No.4,
1989

376
A.
BR YLOWSKI
ET
AL.
C4·A2
~~I~
I
V."""V~"'Ii.rl.~
...
r'~<'f\"\A~
Cz·A2
~I~~~l~~'\i'"
YJ!},,j~NJtA,,,,\~f,MwJ(I!~
.....
~/r'"h\f\.
T3·A1
,~
......
--
....
tv~/'
~,~----~
T4·A2
~V\f''''----V~t
01·A1
02·A2
LOC·A1
ROC·A2
F.
Pulse
EMG
EKG
H·Reflex
Resp
\ \ \
~
A.
Awake
B.
REM
Sleep
---==-----,150
!l
V
5
Sec
FIG. 1.
A:
Awake, eyes closed. Two sets
of
eye movement signals are shown.
B:
REM sleep. Two sets of
eye movement signals are shown, with the onset
of
a lucid dream at the first (arrow). Note the indications of
autonomic activation [skin potential artifact in the electroencephalogram and electrocardiogram (EKG) chan·
nels and finger pulse amplitude suppression] and the profound H-reflex suppression accompanying lucidity.
epoch converted for analysis to beats per minute), and respiration interval (measured
for the breath cycle occupying >50%
of
the
5-s
epoch and converted for analysis to
breaths per minute).
RESULTS
Eighteen REM periods were recorded; eight included portions
of
lucid REM with a
total of
10
lucid dreams. A total
of
150
epochs (12.5
min)
of
lucid REM were collected
across the 4 recording nights out of
2,470 REM epochs
(206
min). The number of epochs
oflucid REM for nights 1-4 was
18
(1.5
min),
18
(1.5
min),
48
(4
min), and
66
(5.5
min),
respectively. The percentage
of
total sleep time spent in REM from nights 1 through 4
was 5.4%
(19
min/349.5 min), 11.1%
(60
min/538.5 min), 18.4%
(64
min/348 min), and
29.4%
(63
min/214.5 min), respectively. The increasing percentages
of
REM are pre-
sumably due to the subject adapting to sleeping under the difficult experimental con-
ditions. Only the epochs within REM periods actually scored as REM were analyzed.
None
of
the portions
of
lucid REM were found to contain any microawakenings or
transitions to NREM.
The mean H-reflex amplitude was determined for the nonlucid epochs and lucid
epochs
of
each REM period that included lucid REM. In all eight cases, the mean
H-reflex amplitude for lucid REM was lower than that for nonlucid REM (binomial test,
Sleep, Vol. 12,
No.4,
1989

H-REFLEX
SUPPRESSION
IN
LUCID
REM
377
p = 0.(04). A one-tailed paired t test
of
the REM period means
of
lucid REM versus
nonlucid also revealed lucid REM H-reflex amplitude
to
be significantly lower than
nonlucid
(t(7)
= 4.43, p < 0.005).
As mentioned above, previous studies have shown that the initiation
Of
lucidity in
REM sleep
is
associated with autonomic activation (2,3). To examine the relationship
between the onset
of
lucidity and H-reflex suppression, the mean values
of
H-reflex
amplitude for the 30 s preceding and the 30 s following lucidity initiation were compared
with the REM period median H-reflex amplitude.
For
both pre- and postlucidity peri-
ods, 8
of
10
mean amplitudes were lower than the all-REM period median amplitudes
(binomial test, p = 0.05). The same type
of
comparison was performed
on
eye move-
ment density in the 30 s pre- and postlucidity initiation.
For
all
10
lucid dreams, the eye
movement density means were higher
just
before and after lucidity onset than the
all-REM period medians (binomial test, p = 0.001), bearing
'Out
the earlier findings on
this measure
of
autonomic arousal.
Direct comparisons
of
the relationship between H-reflex suppression and other au-
tonomic variables were done with Spearman rank correlations
of
H-reflex amplitude
with eye movement density, respiration rate, and heart rate. These tests showed highly
significant (p < 0.001) negative correlations between H-reflex amplitude and all other
physiological variables when the values from nights 1-4 were pooled.
For
the individual
nights,
11
of
12
coefficients showed significant correlations
(p
< 0.05) (see Table
O.
DISCUSSION
The results
of
the above comparisons
of
H-reflex amplitude during lucid and nonlucid
REM sleep clearly show not only that H-reflex is suppressed during REM lucid dreams,
as it is in ordinary REM sleep, but that it tends to be more suppressed during lucid than
nonlucid REM. Increased H-reflex suppression was also strongly associated with pha-
sic REM activity and autonomic activation, two factors that previous studies (2,3) have
shown to also be associated with lucid REM sleep.
The findings presented here demonstrate that lucid dreaming occurs during unam-
biguous REM sleep, without microawakenings or other arousals, and also that lucid
REM sleep
is
associated with relatively high phasic activity, autonomic activation, and
TABLE
1.
Correlations
of
H-reflex amplitude with eye movement density, heart rate, and
respiration rate for the 4 nights individually and pooled using Spearman's
correlation coefficient
Night 1
Night 2
Night 3
Night 4
All nights pooled
Q P < 0.001.
b P <
O.OS.
Eye movement density
-0.3539"
(n
=
192)
-0.4265a
(n =
981)
-0.5475a
(n
=
1,057)
-0.4725a
(n =
970)
-0.3239"
(n
= 3,200)
H-Reflex amplitude with
Heart rate
-0.0538
(n
=
192)
-O.3508a
(n =
980)
-O.0698b
(n
=
I,OS9)
-0.2434
a
(n
=
970)
-O.3198a
(n
= 3,201)
Respiration rate
-0.IS20
b
(n =
191)
-0.184SQ
(n =
980)
-0.3134Q
(n =
I,OS6)
-O.l164Q
(n
==
962)
-0.5119"
(n = 3,189)
Sleep, Vol.
12,
No.4,
1989

378
A.
BR
YLOWSKI
ET
AL.
deep reflex suppression. By showing that the H-reflex
is
even more suppressed during
lucid REM than nonlucid REM, these analyses suggest that lucid REM is not, as might
be supposed, a state closer to awakening than ordinary,
or
nonlucid REM. Addition-
ally, a direct relationship is suggested between the depth
of
H-reflex suppression and
increased phasic activation in REM sleep. Further research is indicated to confirm
these findings with more subjects and to investigate in more detail the relationship
between various modes
of
physiological activation, tonic reflex suppression, and the
mental activity
of
the dreamer.
Acknowledgment: The support
of
The Institute
of
Human Development
is
gratefully acknowl-
edged.
REFERENCES
1.
LaBerge SP, Nagel LE, Dement WC, Zarcone VP. Lucid dreaming verified by volitional communication
during REM sleep. Percept Mot Skills
1981
;52:727-32.
2.
LaBerge SP, Nagel
LE,
Taylor WB, Dement WC, Zarcone VP. Psychophysiological correlates
of
the
initiation
of
lucid dreaming. Sleep
Res
1981
;10: 149.
3. LaBerge S, Levitan L, Dement WC. Lucid dreaming: physiological correlates
of
consciousness during
REM sleep. J Mind Behav 1986;7:251-8.
4.
Schwartz BA, Lefebvre A. Contact veille/P.M.O. II. Les P.M.O. morcelles.
Rev
Electroencephalogr
Neurophysiol Clin 1973;3: 165-76.
5. Hishikawa Y, Sumitsugu N, Matsumoto K, Kaneko
Z.
H-reflex and EMG
of
the mental and hyoid
muscles during sleep, with special reference to narcolepsy. Electroencephalogr Clin Neurophysiol
1965;18:487-92.
6. Hodes R, Dement WC. Depression of electrically induced reflexes ("H-reflexes") in man during low
voltage EEG sleep. Electroencephala/?r Clin NeurophysioI1964;17:617-29.
7.
Pivik R, Metz J. Phasic EMG inhibition and spinal reflex modulation during synchronized sleep in the
cat. Exp Neural 1975;48:493-501.
8. Pivik R, Metz J. Spinal and lateral geniculate nucleus activity during sleep: quantitative relationships.
Exp NeuroI1982;77:142-62.
9.
Rechtschaffen
A,
Kales A, eds. A manual
of
standardized terminology, techniques,
and
scoring system
for sleep stages
of
human subjects. Los Angeles: University
of
California
at
Los Angeles, Brain Infor-
mation Service/Brain Research Institute, 1968.
10.
Magladery J, McDougal
G.
Electrophysiologic studies
of
nerve and reflex activity in normal man.
I.
Identification of certain reflexes
in
the electromyogram and the conduction
of
peripheral nerve fibers.
Bull Johns Hopkins Hosp 1950;86:265-90.
11. Magladery
J,
Teasdale R, Park
A.
Electrophysiologic studies of nerve and reflex activity in patients with
lesions
of
the nervous system: a comparison
of
spinal motorneuron excitability following afferent nerve
volleys
in
normal persons and patients with upper motor neuron lesions. Bull Johns Hopkins Hosp
1952;91 :219-44.
Sleep, Vol. 12,
No.4,
1989