The Science of the Total Environment, 4 (1975) 107-112 © Elsevier ScientificPublishing Company,Amsterdam- Printed in Belgium
Short communication
DISTURBANCES OF SLEEP BY SONIC BOOMS BARBARA GRIEFAHN and GERD JANSEN Universitiitsklinikum 43, University of Essen, Hufelandstrasse 55, Essen ( G.F.R.)
(Received January 27th, 1975)
ABSTRACT After a pilot study (2 subjects, 19 nights) we tested two different subjects during 57 nights, administering sonic booms (1 mb, 300 ms; sound level of sonic boom in the bedroom 80-85 dB (A)) and recording EEG and peripheral blood volume. After 7 nights without noise, 30 nights with either 2 or 4 sonic booms (alternately) were applied. After 10 more nights without noise, four nights with 8 and 16 bangs followed alternately. The last 6 nights were used as a comparison phase. Results showed that disturbance was obvious during all periods of noise. No adaptation could be observed during any of the experiments. On the contrary, during the night with 4 bangs there was a tendency for compensation, e.g., in the last two thirds of nights with 4 bangs, the total time of deep sleep was comparable with the nights without any noise.
EXPERIMENTAL DESIGN After a pilot study for 19 nights with two subjects we tested two different subjects during 57 further nights. In both series data was from !0:30 p.m. to 3:00 a.m. Electroencephalogram (EEG), peripheral blood volume at the fingertips (FPA) and pulse frequency (PR) were recorded through the time. During the first three nights in each test series the subjects slept without any noise disturbance. In the following l l and 30 "noise nights" sonic booms were applied altcrnatel) 2 or 4 times. Sound level of sonic booms in the bedroom was 83.5 dB (A) in the average. In the main series after 10 more nights without noise four nights with 8 and 16 booms followed altcrnatcly. The last 6 nights in the main test series as well as in the pilot series were used as comparison phase (Fig. 1). The first sonic boom in a noise night was applied if one of the two subjects had entered the deepest stage of sleep, regardless the depth of sleep of the other person. The time distance between two sonic booms was 40 rain in nights with 2 booms, 107
20 min in nights with 4 sonic booms, and in the nights with 8 and 16 sonic booms the difference was 8.6 and 4.6 rain, respectively.
.O.O-O.O.D.
[
pressure
recording EEG
]
~(fronto-occIpitol lead)
"recording finger pulse 1 OmPlitucles
j
temperature ~ 14 "C ] atmospheric h u m i d i l y . , 4 ~
P~
baim. pres. - 760 m m H g ]
g .
• #0 mn (2 sonlcbooms/night) [3 20 mn (4 sonic booms/nigAf) • •
8,6 mn (g sonic booms/rvght) 4,6 mn (16 sonic booms/night)
temperature
E
atmospheric h u m i d i t y
65 % 760 mmHG
atmospheric pressure
recording EEG (fronto-occipital lead) recording finger pulse amplitude
E
ornDient noise ( 35 dB(A)) [ pressure of sonic booms ;~.o..p;.o..~; o..o..o.o..o.o.,p..o.
..,. l
~.
14.
] ~.. o . . . . . . . . . .
19.
33.
[Pres] ,,.,, ...... 44. 47.
53.
night
Fig. 1. Experimentalarrangement. Technical details of the physical arrangement are published in a report of the Franco-German Research Institute in Saint Louis, France. An extensive report of our own results has been published by the Franco-German Research Institute in Saint Louis, too (Report 21/74). The most important results are presented here again, in addition, new results of the pulse frequency analysis are described. The evaluation of the pulse amplitude and pulse rates followed usual rules of evaluation wheras the assessment of EEG is based on a 6-stage cheme (Fig. 2):
effects
o f s o n i c b o o m s on s l e e p stage of sleep 0 - ! stage of sleep
]
stage of sleep ! - ZZ stage of sleep
IZ
.~ja%~,~,.j~..r.~j/'~
stage of sleep I I - I I I
~
stoge of sleep I I I
Fig. 2. Scheme of the evaluation of the experiments with sonic booms.
108
RESULTS
EE G-frequencies (long period) First we calculated the percentage of the E E G frequencies in the single nights by using the frequency of the initial value (10 sec before the boom) as reference phase: 100%. The two tested subjects in the pilot study showed a different slope of E E G frequencies in the p o s t b o o m phases during the eleven nights (Fig. 3). The solid lines represent the nights with 4 sonic booms, the dotted lines represent the 2 boom nights. From the main test series one can conclude that there is no adaptation to sonic booms related to E E G frequencies. These conclusions could not be drawn after the pilot series with eleven nights as the upper part of the figure shows.
160 %
r= 0.495. dr, 9 y . 4.516x* 96~
I/.~
140
i•|. ~ ,:o r , 0.I79. ¢f=11 y,-O.931x* 127.4
~
ii
'
'
"
i
i
i
~ |
160
!
14o
n/ght
r ~-0~24. oY , 29 y ,-O#89 x * 134.3 m
mm(mm
~m
m
m
m
m
m
m
m
m
i
mmm m
m
i
12o r,-O~Ol, oY ,29 y,-OD2 x * I16~ 100
' 12.
" m
2 sonic ~ating
booms/nig~
wit~
16.
2Q
ICuratton ~
4 so~c b o o ~ / r ~ g h ¢ ( ¢ u r a ~
2~
241.
.12.
night
2 booms 40 mini
b e t w e ~ 2 booms 20 m/n)
socVc boom: inhmsify - I tabor, durot/oa 300
msec
Fig. 3. Frequency of EEG following sonic booms within 11 and 30 nights, respectively.
EE G-frequencies (single night) The same slope of curves could be seen if the EEG frequencies as reactions to sonic booms were regarded during the single nights. The analysis of each boom event in a night shows that after each boom there is always an acute reaction. The extent of the reaction and the time of the frequency changes depend on the initial value in the EEG. Moreover, the responses observed allowed the following conclusion: the more the time interval between two booms is shortened the more decreased is the extent of reaction.
Sleep stages Our next step was the calculation of the times the subjects spent in the different 109
stages of sleep (Fig. 4). We saw that only the deepest sleep stage (III) was shortened significantly by the sonic booms. These were seen with the 2 sonic boom nights as well as with the 4 sonic boom nights.
'°I
~0 r 30
t
m
I
2 sonic booms/night ] 2 s~jeet.~. 15 r~ts
II -**°
L
el' ~
,o
~ sonic booms/night interlnll 20 nVn
20 !0
00-1 I fall -wo~-sleep
I-II
1I
l l - l l l I l l stugo of Skmp Mlwl-wove-sleep
nights without sonic booms JmJ nights with sonic booms~ intensity 1 mboro durotion 300 msec p < 1 ~ ; t#~ pxf-,?.5~t
Fig. 4. Duration of sleep within differentstages of sleep in nights with and without sonic booms. The reduction of deep sleep stage in the 2 boom nights is significant at the 1% level, whereas in the 4 boom nights the reduction is significant at the 2.5% level. An analysis of the summation of these times of stage III sleep by dividing the whole test series in three parts (10 nights were calculated together) showed that in the nights with 4 booms only in the first 10-nights period was there a reduction of stage III in favour of the fast-wave-sleep. In the following 20-nights period with 4 booms we saw that the sum of the stage III sleep time was the same as in the nights with no noise (comparison phase). This result led us to the conclusion that the human organism might be able to compensate the boom disturbances caused by regularly repeated stimuli. But we keep in mind that this conclusion is only drawn by the relative rough assessment of the sleep stages which is only based on the frequency evaluation. Pulse rate
Regarding the slope of pulse rates (Fig. 5) in the pre- and postbooms we saw in all four persons a biphasic slope characterized by an initial increase of pulse frequency in all subjects and situations. A maximum was found in the 4th second, followed by a quick decrease (in three subjects below the level of the initial value). The analysis of the context between the post-stimulatory frequency slope and other II0
It
} sonic boom: intensity subject
-I mbar, duration 300 msec ,% V. 2~ sonic b~oms 110 -~
68
~oo
9O
6~ subject E. 24 sonic booms
88 f 80
% 1110
~.
'~o
IOO :
R = 103,9 %
90 %
~.
110 100 P_ 90
80
= 98,0 % ,
I1~ 66
~s9 el.
• -, .,., alp o e e • ,
v...e;_*¢'.,"."."
%
06
78 .70 62
•
subject
,,,%~
107 sonic booms
,
10
16.
R
¢..
1110 i 100 Ul ¢L
0
2,
z~
6.
8.
12.
1/,.
sec after sonic boom 2 resp~ 4 rasp 8 re3p 16 sonic booms/night duration between 2 sor~c booms 40 rain rasp 20 rain rasp ~6 rain resl3 4.6 m i n
Fig. 5. Pulse rate after sonic boom.
50 % ~ ,=
r = 0°088. d r = 9 y • 0~350x + 35.7
40
i
~ ~
30
%
. . . .
8.
~:
r=-(2538, d f , ; l y : - 2 , 1 8 4 x + 55.8
i i 12. n i g h t
2 subjects
so
~
r = 0,098, dr=29 0 . 1 2 9 x . 28.9
4O
y =
~
~¢}
30
mm
ml
m m m m
m m m
m m m m m m m
tuna
r.
mmm
0,022, d r = 2 ?
¥ • 0,046x*
q. 20 4.
m =. m
, 8
, 12
, 16.
-20
, 24
-28
27.9
, 32. night
2 sonic b o o m s / n i g h t {duration between 2 booms 40 min) alternating with 4 sonic b o o m s / n i g h t (duration between 2 booms 20 rain) sonic boom: intensity - f mbor. duration 300 msec
Fig. 6. Reaction of finger pulse amplitudes to sonic booms within 11 and 30 nights, respectively. 111
frequency parameters and moderator variables showed the following: (1) Intensity of stimulus and change of pulse frequency rate are not correlated significantly. The constant reaction to the stimulus seems to be caused by the extreme short rise time of the sound pressure. (2) The time of the preceding, preboom time was well as the time of the whole test series has no influence on the boom-induced change of pulse rate.
Peripheral blood volume After the application of the booms, the peripheral blood volume at the fingertips showed a regular and significant decrease. A correlation between the sleep stages and the increase of neuronal activity could not be proved as it was not possible to produce constant boom pressure levels in the sleeping room. Regarding the pulse amplitudes of the 2 test series as a whole, we saw the same behavior of the sleeping subjects (Fig. 6) as was seen in the electroencephalographic recordings. The main study shows that there is no habituation to the sonic boom in the nights with 2 and 4 booms; the same peripheral reactions occur in all nights. This conclusion could not be drawn after the first test series with 11 nights as the upper part of Fig. 6 shows. CONCLUSION Considering pulse amplitudes and EEG, they seem to react in the same way (compare Figs. 3 and 6). On the other hand it must be taken into account, that the pulse amplitude showed no habituation whereas in the last periods of the sonic boom nights with 4 booms, there seemed to be a possible compensation by a more frequent achievement of the deepest sleep stage within one night.
112
Short communication
DISTURBANCES OF SLEEP BY SONIC BOOMS BARBARA GRIEFAHN and GERD JANSEN Universitiitsklinikum 43, University of Essen, Hufelandstrasse 55, Essen ( G.F.R.)
(Received January 27th, 1975)
ABSTRACT After a pilot study (2 subjects, 19 nights) we tested two different subjects during 57 nights, administering sonic booms (1 mb, 300 ms; sound level of sonic boom in the bedroom 80-85 dB (A)) and recording EEG and peripheral blood volume. After 7 nights without noise, 30 nights with either 2 or 4 sonic booms (alternately) were applied. After 10 more nights without noise, four nights with 8 and 16 bangs followed alternately. The last 6 nights were used as a comparison phase. Results showed that disturbance was obvious during all periods of noise. No adaptation could be observed during any of the experiments. On the contrary, during the night with 4 bangs there was a tendency for compensation, e.g., in the last two thirds of nights with 4 bangs, the total time of deep sleep was comparable with the nights without any noise.
EXPERIMENTAL DESIGN After a pilot study for 19 nights with two subjects we tested two different subjects during 57 further nights. In both series data was from !0:30 p.m. to 3:00 a.m. Electroencephalogram (EEG), peripheral blood volume at the fingertips (FPA) and pulse frequency (PR) were recorded through the time. During the first three nights in each test series the subjects slept without any noise disturbance. In the following l l and 30 "noise nights" sonic booms were applied altcrnatel) 2 or 4 times. Sound level of sonic booms in the bedroom was 83.5 dB (A) in the average. In the main series after 10 more nights without noise four nights with 8 and 16 booms followed altcrnatcly. The last 6 nights in the main test series as well as in the pilot series were used as comparison phase (Fig. 1). The first sonic boom in a noise night was applied if one of the two subjects had entered the deepest stage of sleep, regardless the depth of sleep of the other person. The time distance between two sonic booms was 40 rain in nights with 2 booms, 107
20 min in nights with 4 sonic booms, and in the nights with 8 and 16 sonic booms the difference was 8.6 and 4.6 rain, respectively.
.O.O-O.O.D.
[
pressure
recording EEG
]
~(fronto-occIpitol lead)
"recording finger pulse 1 OmPlitucles
j
temperature ~ 14 "C ] atmospheric h u m i d i l y . , 4 ~
P~
baim. pres. - 760 m m H g ]
g .
• #0 mn (2 sonlcbooms/night) [3 20 mn (4 sonic booms/nigAf) • •
8,6 mn (g sonic booms/rvght) 4,6 mn (16 sonic booms/night)
temperature
E
atmospheric h u m i d i t y
65 % 760 mmHG
atmospheric pressure
recording EEG (fronto-occipital lead) recording finger pulse amplitude
E
ornDient noise ( 35 dB(A)) [ pressure of sonic booms ;~.o..p;.o..~; o..o..o.o..o.o.,p..o.
..,. l
~.
14.
] ~.. o . . . . . . . . . .
19.
33.
[Pres] ,,.,, ...... 44. 47.
53.
night
Fig. 1. Experimentalarrangement. Technical details of the physical arrangement are published in a report of the Franco-German Research Institute in Saint Louis, France. An extensive report of our own results has been published by the Franco-German Research Institute in Saint Louis, too (Report 21/74). The most important results are presented here again, in addition, new results of the pulse frequency analysis are described. The evaluation of the pulse amplitude and pulse rates followed usual rules of evaluation wheras the assessment of EEG is based on a 6-stage cheme (Fig. 2):
effects
o f s o n i c b o o m s on s l e e p stage of sleep 0 - ! stage of sleep
]
stage of sleep ! - ZZ stage of sleep
IZ
.~ja%~,~,.j~..r.~j/'~
stage of sleep I I - I I I
~
stoge of sleep I I I
Fig. 2. Scheme of the evaluation of the experiments with sonic booms.
108
RESULTS
EE G-frequencies (long period) First we calculated the percentage of the E E G frequencies in the single nights by using the frequency of the initial value (10 sec before the boom) as reference phase: 100%. The two tested subjects in the pilot study showed a different slope of E E G frequencies in the p o s t b o o m phases during the eleven nights (Fig. 3). The solid lines represent the nights with 4 sonic booms, the dotted lines represent the 2 boom nights. From the main test series one can conclude that there is no adaptation to sonic booms related to E E G frequencies. These conclusions could not be drawn after the pilot series with eleven nights as the upper part of the figure shows.
160 %
r= 0.495. dr, 9 y . 4.516x* 96~
I/.~
140
i•|. ~ ,:o r , 0.I79. ¢f=11 y,-O.931x* 127.4
~
ii
'
'
"
i
i
i
~ |
160
!
14o
n/ght
r ~-0~24. oY , 29 y ,-O#89 x * 134.3 m
mm(mm
~m
m
m
m
m
m
m
m
m
i
mmm m
m
i
12o r,-O~Ol, oY ,29 y,-OD2 x * I16~ 100
' 12.
" m
2 sonic ~ating
booms/nig~
wit~
16.
2Q
ICuratton ~
4 so~c b o o ~ / r ~ g h ¢ ( ¢ u r a ~
2~
241.
.12.
night
2 booms 40 mini
b e t w e ~ 2 booms 20 m/n)
socVc boom: inhmsify - I tabor, durot/oa 300
msec
Fig. 3. Frequency of EEG following sonic booms within 11 and 30 nights, respectively.
EE G-frequencies (single night) The same slope of curves could be seen if the EEG frequencies as reactions to sonic booms were regarded during the single nights. The analysis of each boom event in a night shows that after each boom there is always an acute reaction. The extent of the reaction and the time of the frequency changes depend on the initial value in the EEG. Moreover, the responses observed allowed the following conclusion: the more the time interval between two booms is shortened the more decreased is the extent of reaction.
Sleep stages Our next step was the calculation of the times the subjects spent in the different 109
stages of sleep (Fig. 4). We saw that only the deepest sleep stage (III) was shortened significantly by the sonic booms. These were seen with the 2 sonic boom nights as well as with the 4 sonic boom nights.
'°I
~0 r 30
t
m
I
2 sonic booms/night ] 2 s~jeet.~. 15 r~ts
II -**°
L
el' ~
,o
~ sonic booms/night interlnll 20 nVn
20 !0
00-1 I fall -wo~-sleep
I-II
1I
l l - l l l I l l stugo of Skmp Mlwl-wove-sleep
nights without sonic booms JmJ nights with sonic booms~ intensity 1 mboro durotion 300 msec p < 1 ~ ; t#~ pxf-,?.5~t
Fig. 4. Duration of sleep within differentstages of sleep in nights with and without sonic booms. The reduction of deep sleep stage in the 2 boom nights is significant at the 1% level, whereas in the 4 boom nights the reduction is significant at the 2.5% level. An analysis of the summation of these times of stage III sleep by dividing the whole test series in three parts (10 nights were calculated together) showed that in the nights with 4 booms only in the first 10-nights period was there a reduction of stage III in favour of the fast-wave-sleep. In the following 20-nights period with 4 booms we saw that the sum of the stage III sleep time was the same as in the nights with no noise (comparison phase). This result led us to the conclusion that the human organism might be able to compensate the boom disturbances caused by regularly repeated stimuli. But we keep in mind that this conclusion is only drawn by the relative rough assessment of the sleep stages which is only based on the frequency evaluation. Pulse rate
Regarding the slope of pulse rates (Fig. 5) in the pre- and postbooms we saw in all four persons a biphasic slope characterized by an initial increase of pulse frequency in all subjects and situations. A maximum was found in the 4th second, followed by a quick decrease (in three subjects below the level of the initial value). The analysis of the context between the post-stimulatory frequency slope and other II0
It
} sonic boom: intensity subject
-I mbar, duration 300 msec ,% V. 2~ sonic b~oms 110 -~
68
~oo
9O
6~ subject E. 24 sonic booms
88 f 80
% 1110
~.
'~o
IOO :
R = 103,9 %
90 %
~.
110 100 P_ 90
80
= 98,0 % ,
I1~ 66
~s9 el.
• -, .,., alp o e e • ,
v...e;_*¢'.,"."."
%
06
78 .70 62
•
subject
,,,%~
107 sonic booms
,
10
16.
R
¢..
1110 i 100 Ul ¢L
0
2,
z~
6.
8.
12.
1/,.
sec after sonic boom 2 resp~ 4 rasp 8 re3p 16 sonic booms/night duration between 2 sor~c booms 40 rain rasp 20 rain rasp ~6 rain resl3 4.6 m i n
Fig. 5. Pulse rate after sonic boom.
50 % ~ ,=
r = 0°088. d r = 9 y • 0~350x + 35.7
40
i
~ ~
30
%
. . . .
8.
~:
r=-(2538, d f , ; l y : - 2 , 1 8 4 x + 55.8
i i 12. n i g h t
2 subjects
so
~
r = 0,098, dr=29 0 . 1 2 9 x . 28.9
4O
y =
~
~¢}
30
mm
ml
m m m m
m m m
m m m m m m m
tuna
r.
mmm
0,022, d r = 2 ?
¥ • 0,046x*
q. 20 4.
m =. m
, 8
, 12
, 16.
-20
, 24
-28
27.9
, 32. night
2 sonic b o o m s / n i g h t {duration between 2 booms 40 min) alternating with 4 sonic b o o m s / n i g h t (duration between 2 booms 20 rain) sonic boom: intensity - f mbor. duration 300 msec
Fig. 6. Reaction of finger pulse amplitudes to sonic booms within 11 and 30 nights, respectively. 111
frequency parameters and moderator variables showed the following: (1) Intensity of stimulus and change of pulse frequency rate are not correlated significantly. The constant reaction to the stimulus seems to be caused by the extreme short rise time of the sound pressure. (2) The time of the preceding, preboom time was well as the time of the whole test series has no influence on the boom-induced change of pulse rate.
Peripheral blood volume After the application of the booms, the peripheral blood volume at the fingertips showed a regular and significant decrease. A correlation between the sleep stages and the increase of neuronal activity could not be proved as it was not possible to produce constant boom pressure levels in the sleeping room. Regarding the pulse amplitudes of the 2 test series as a whole, we saw the same behavior of the sleeping subjects (Fig. 6) as was seen in the electroencephalographic recordings. The main study shows that there is no habituation to the sonic boom in the nights with 2 and 4 booms; the same peripheral reactions occur in all nights. This conclusion could not be drawn after the first test series with 11 nights as the upper part of Fig. 6 shows. CONCLUSION Considering pulse amplitudes and EEG, they seem to react in the same way (compare Figs. 3 and 6). On the other hand it must be taken into account, that the pulse amplitude showed no habituation whereas in the last periods of the sonic boom nights with 4 booms, there seemed to be a possible compensation by a more frequent achievement of the deepest sleep stage within one night.
112
