Psychother. Psychosom. 27: 8-17 (1976/77)
A Longitudinal Study of Some Physiological Parameters and Autogenic Training A.K. Tebecis', Y. Ohno, H. Matsubara, H. Sugano, T. Takeya, Y. lkem iand M. Takasaki Department of Psychosomatic Medicine, Kyushu University Medical School, Fukuoka City
Abstract. Physiological parameters were studied over 4.5 months in normal high-school girls. One group frequently practised autogenic training (AT) whereas the other (control) did not. After 4.5 months the physiology of the AT, but not the control group had changed to some extent. The mean increase in finger skin temperature in the final session was significantly greater in the AT group than in the control group, especially in subjects who experienced hand warmth during AT. No significant differences between groups were appar ent for finger blood flow, but within the AT group the girls who experienced hand warmth showed a significantly greater blood flow than those who did not experience hand warmth. The electroencephalogram (EEC) of the AT group showed significantly more ‘slow’ (ft and 0) and less fast (a and (?) activity than at the beginning of the project. Microvibration (MV) of the thumb did not differ significantly in any comparisons. Both in the AT and control group means there were no significant differences between the periods before, during and after AT although significant changes did occur in some individuals of the AT group in the final session. In particular, in 3 of 14 AT subjects respiration rate decreased markedly even though weekly training had not included any breathing exercises. The results suggest that the effects of AT are predominantly long-term, resulting in gradual physiological changes that are maintained to some extent between practice periods.
Introduction
1 Visiting Professor from the Australian National University, supported by the Japan Society for the Promotion of Science.
Downloaded by: Stockholm University Library 130.237.165.40 - 2/21/2018 2:36:52 AM
Autogenic training (AT) is a very useful tool in psychosomatic medicine (9). However, the few studies done on AT are difficult to evaluate because of the different experimental designs used, the lack of controls in some cases, the use of both patients and normal subjects, the variable duration of subjects’ experi ence with AT and other reasons. In addition, few investigators have reported
Tebecis/Ohno/Matsubara/Sugano/Takeyajlkemi/Takasaki
9
group means, making it difficult to know whether the reported results of se lected individuals represent characteristic effects of AT or not (4). Some of these variables were avoided in the present study of AT and finger skin temperature, finger blood flow, electroencephalogram (EEG), microvibra tion (MV) of the thenar eminence, and respiration rate, recorded from normal high-school girls randomly divided into two groups of equal size. Every week one group was instructed in the technique of AT and urged to practise it daily whereas the other (control) group received no such instructions. Physiological recordings were made at intervals over a few months in order to determine whether the above-mentioned physiological parameters in the AT group were different (a) before, during and after AT periods within a session, and (b) between sessions, that is, with long-term experience with AT. Some of this material has been briefly reported elsewhere (12).
Subjects and training. 30 unpaid Japanese girls from a Women’s Commercial High School, 15-18 years old (mean age 16.1 years), were divided into two groups of 15 girls each. The AT group was taught AT every week and urged to practise it daily at least 2 -3 times, as it was ‘very good for sport’. The control group received no special instructions. Physiological parameters were recorded from all subjects under similar conditions prior to giving any instructions (session 1). Each subject filled out scales E (extraversion-introversion), N (neuroticism) and L (lie) of the Maudsley Personality Inventory (MPI) (1). Records were made from all subjects randomly during the afternoons of 4 -5 consecu tive days. By session 2, 1 month after session 1, most subjects of the AT group claimed they had learned to experience a ‘calm mind’ and 'heaviness of the hands’ (9; Standard Autogenic Exercise 1) during their private training. By session 3 (3 months after the beginning) these experiences were more marked, and a small proportion of subjects could also experience ‘hand warmth’ (9; Standard Autogenic Exercise 2). By session 4 (last session; 4.5 months after the beginning) most AT subjects claimed they could often experience hand warmth. No other exercises were taught as various reasons prevented continuation of the project. All subjects again filled out scales E, N and L of the MPI after the final session. Equipment and procedure. Each subject was required to lie on a couch, eyes closed, facing upwards with the arms extended parallel to the body axis, in a dim, semi-soundproof, shielded room maintained at 23 °C. The signals were displayed from an adjacent room. Skin temperature was recorded by a thermistor probe attached to the volar surface of the middle finger of the right hand and was read directly from an electronic thermometer (San-Ei Instrument Co., Model M44-4). Local pulsatile blood flow of the index finger of the right hand was recorded with a plethysmogram cuff (San-Ei, Model E-037). Respiration was recorded by a thermistor probe (San-Ei, Model E-034) fixed in the right nostril. MV of the thenar eminence of the left hand was recorded by a disc-type (23 mm diameter, about 5 g weight) crystal acceleration pick-up (Nihon Kohden Co., MT-3T Pickup). The piezoelectric pick-up was sensitive mainly (> 90 %) to vibration parallel to the surface of the disc. Monopolar EEG recordings were made from symmetrical areas on the left and right parietal scalp using conventional silver-silver chloride electrodes, the indifferent lead being attached to the right ear-lobe (13).
Downloaded by: Stockholm University Library 130.237.165.40 - 2/21/2018 2:36:52 AM
Methods
Tebecis/Ohno /Matsubara/Sugano/Takeya/Ikemi/Takasaki
10
All the signals except skin temperature were displayed on separate channels of a 14-channel electroencephalograph chart recorder (San-Ei) using a time constant of 0.3 sec and an upper frequency cut-off of 30 Hz. Skin temperature was recorded on the chart at appropriate intervals. An EEG frequency analyzer (San-Ei, Model 7P11) was used to analyse the MV and EEG (right side) into bands of 2 -4 (5), 4 - 6 (6,), 6 -8 (62), 8 -1 3 (a) and 13-20 Hz (0,) using 10-sec epochs. Three consecutive 5-min periods, before, during and after were analysed. The subject was told to relax without moving unnecessarily and not to fall asleep. For the control subjects in all sessions and the AT subjects in session 1 the instructions were repeated 5 min later and again after a further 5 min later, and 15 min after the beginning the recording was stopped. The AT subjects in the remaining sessions received these instructions at the begin ning, followed by 'Please begin AT’ 5 min later and ‘please stop AT’ a further 5 min later. During removal of the leads, the subject was asked about her subjective experiences. Analysis. For each 5-min recording, skin temperature was taken as the mean of 3 values at 30-sec intervals of the final 60 sec. Amplitude of plethysmogram pulse was the mean of the final 10 consecutive pulses. Respiration rate was the number of breaths in the final 60 sec. For the MV and EEG, the amplitudes of the frequency analyzer output pulses of three consecutive 10-sec epochs were measured after the experiment and the ‘activity’ (total voltage) of a particular frequency band was converted to a proportion of the total activity. MV-analyzed pulses were taken from the final 60 sec of each 5-min period and EEG pulses from records immediately before. Finger skin temperature and amplitude of plethysmogram pulses (mean of 10 consecutive pulses) were also taken shortly (within 1 min) after the very beginning of recording (initial values). F'or overall comparisons of means between 5-min periods of a particular group and between the two groups, or between sessions of a group, the data were subjected to analysis of variance. In comparisons of certain subgroups (see ‘Results’) the t test or MannWhitney U test (dependent or independent samples, where appropriate) was used. Probabili ty values are for two-tailed tests.
No data were used from 2 (1 AT and 1 control) of the 30 subjects who did not complete the 4.5-month program or from records if the EEG indicated that the subject was falling asleep. AT subjects’ comments during the weekly instruc tion period suggested that in most cases motivation and enthusiasm to practise AT privately was low, being of the order of several times per week rather than per day. It was not possible to reliably estimate their rate of practice more precisely. The MPI scores indicated that all subjects were within the normal range for the Japanese population (5), that there were no significant differences between the AT and control groups, and that no significant changes had occurred be tween the beginning and end of the project. Thus, for the AT and control groups, respectively, the mean scores immediately after the first session were 31.4, 34.1 (scale E), 24.8, 27.3 (scale N) and 11.8, 9.1 (scale L). After the final session, the mean scores were 31.8, 32.9 (E), 21.8,25.1 (N) and 10.6,9.8 (L).
Downloaded by: Stockholm University Library 130.237.165.40 - 2/21/2018 2:36:52 AM
Results
The Physiology of Autogenic Training
11
During AT
Before AT
After AT
Microvibration
♦t*
f ro ' »I*
“»•et»» »♦»t**
»f*
200 mV1 EEG - , ---- -
—
_ ------- - V
- --- ------------------ V , ---------------
50
mV ]
Finger blood flow
/^ W V v y V v M M V Finger skin temperature
29.9 °C
/ \ A / M M \ / S /W \ J \y \ 30.8 °C
30.A “C
Respiration
Seconds
Individual Records For most subjects, there was little or no difference in any mean physio logical parameter between the 5-min periods of the control group in all sessions, and of the AT group in sessions 1—3. In the final session, however, approxi mately one half of the AT subjects displayed significant differences between the 5-min periods in at least one parameter. Sample records from 1 subject are illustrated in figure 1. Respiration rate decreased from 19.0/min before AT to 12.0/min during AT and increased to 22.6/min after AT. Finger skin temperature increased by almost 1 °C during AT and decreased again by 0.4 °C after AT. Mean amplitude of the plethysmogram pulse increased by 14 % during AT and decreased by approxi mately 20 % after AT. These changes were statistically significant (p < 0.01). In the EEG there was an increase in total 0-activity with a concomitant decrease in a-activity during AT but the differences were not significant. There were no significant differences in MV. These records confirm that significant changes can occur during AT in an experienced individual, even though this was not the case for the whole group.
Downloaded by: Stockholm University Library 130.237.165.40 - 2/21/2018 2:36:52 AM
Fig. 1. Records taken at comparable times during the final 60 sec of periods before, during and after AT in session 4 of 1 subject (eyes closed) who experienced a ‘calm mind’ and ‘heaviness and warmth of the hands’ during AT. The records immediately below the EEG tracings (right parietal area) are the output pulses of the frequency analyser, showing the relative voltages of the different EEG frequency bands (6, 0, a, 0). Blood flow is represented by the plethysmogram pulses.
Tebecis/Ohno/Matsubara/Sugano/Takeya/lkemi/Takasaki
12
Fig. 2. Mean increases in finger skin temperature relative to the initial mean values of the control and AT groups in session 4. The AT group is further subdivided into those subjects who experienced hand warmth during AT and those who did not (50 % each).
Downloaded by: Stockholm University Library 130.237.165.40 - 2/21/2018 2:36:52 AM
Fig. 3. Mean activity (total voltage) expressed as proportions of 5 different EEG fre quency bands of the AT (a) and control (b) groups in sessions 1 and 4. Each histogram represents the mean of the sum of all the values of the periods before, during and after AT.
13
Group Data There were no significant differences in any mean physiological parameter between periods before, during and after AT in any session of either group, although some trends towards change during AT were apparent for the AT group in session 4. In this session, all the AT subjects reported that they had experi enced a ‘calm mind’ during AT; 78.6 and 50 % said they had also felt heaviness and warmth, respectively, of the hands. The control subjects reported no special experiences other than ‘feeling relaxed’. Finger skin temperature. Finger skin temperature ranged from 27.4 to 32.4 °C, being 29-31 °C in most cases. For almost all subjects, the initial value shortly after the beginning of recording was the lowest of the session. The mean increases during subsequent recording in the same session were significant for both groups in each session (p < 0.001). Figure 2 shows the mean increments in skin temperature relative to the initial value in session 4. The control group showed mean increments of 0.27, 0.19 and 0.32 °C before, during and after AT periods, respectively, the overall mean being a 0.26 °C increase from baseline. The AT group showed mean increments of 0.34, 0.48 and 0.28 °C before, during and after AT periods, the overall mean being 0.37 °C. The difference between these overall means was significant (p < 0.05). Thus, the AT group showed a greater overall mean in crease in skin temperature than the control group. Moreover skin temperature of the AT group, but not the control group, increased during the 5-min AT period. In those subjects who experienced warmth of the hands during AT, skin tem perature increased significantly during AT (p < 0.05) whereas there were no significant changes between the same three periods in the subjects who did not experience hand warmth. Peripheral blood flow. Mean amplitude of the plethysmogram pulse was divided by the initial mean value, the ratios reflecting peripheral blood flow relative to the beginning of the recording. Blood flow was much more labile than skin temperature in both groups. Whereas skin temperature usually remained constant during a 60-sec sampling period, blood flow often fluctuated within a few seconds. For this reason, changes in skin temperature did not always parallel those of blood flow although large increases in temperature were accompanied by increases in flow and vice versa. Mean amplitude of the plethysmogram pulse generally decreased with time in a recording session. Mean pulse amplitude was similar for both the AT and control groups (table I), there being no significant differences between corresponding means, nor between the means of the 5-min periods within each group. However, within the AT group, the means for the subjects experiencing hand warmth were signifi cantly higher than corresponding means for those who did not (p < 0.05). Electroencephalogram. The greatest differences were between sessions 1 and 4 of the AT group (fig. 3 a). Mean 5-activity in session 4 was almost 6 % higher
Downloaded by: Stockholm University Library 130.237.165.40 - 2/21/2018 2:36:52 AM
The Physiology of Autogenic Training
Tebecis/Ohno/Matsubara/Sugano/Takeya/Ikemi/Takasaki
14
Table /. Mean plethysmogram pulse amplitudes relative to the initial values (ratios) and mean respiration rates per minute of the control and AT groups before, during and after 5-min AT periods in session 4 Groups
Control group AT group Subjects experiencing hand warmth Subjects not experiencing hand warmth
Plethysmogram pulse ratio
Respiration rate
before
during
after
before
during
after
0.91 0.92
0.82 0.88
0.66 0.74
18.7 18.3
19.1 17.3
18.2 18.9
1.07
1.08
0.91
18.5
16.9
19.8
0.69
0.S9
0.48
18.0
17.8
17.5
Downloaded by: Stockholm University Library 130.237.165.40 - 2/21/2018 2:36:52 AM
than in session 1, a significant difference (p < 0.01). 0 ,- and 02-activity were each approximately 2 % greater in session 4, both increases being significant (p < 0.01). a- and /3-activities were both significantly lower (by almost 6 and 3.7 %, respectively) in session 4 (p < 0.01). That is, the EEG of the AT group displayed more slow (< 8 Hz) activity and less fast (> 8 Hz) activity in session 4 compared with session 1. Similar comparisons between other sessions showed no significant differences. In contrast, the control group showed no significant differences between corresponding means, or even trends in the data, in any comparisons of sessions. For example, for sessions 1 and 4 any difference in a particular frequency band was less than 2 % (fig. 3 b). There were no significant differences in corresponding means between the periods before, during and after AT for any session in either group, even when the AT group was subdivided into subjects who experienced hand warmth and subjects who did not. In the final session, however, the AT group did display a tendency towards increased 6- and 9- activity during AT. Microvibration. There were no significant differences in any mean frequency band between the two groups, between sessions of a group or between the 5-min periods of a session, even in a comparison between subjects who experienced hand warmth and subjects who did not. Changes in the AT group were never theless greater than those in the control group. Thus, the AT group showed 2.2% more total 9- (4 -8 Hz) and 2.6% less /3r activity in session 4 than in session 1, whereas the same frequency bands varied by less than 1 % in the control group, a-activity was remarkably constant in both groups. Respiration rate. There were no significant differences in mean respiration rate between the two groups, nor between the 5-min periods of a group. However, the AT group showed a slight mean decrease during AT as compared
The Physiology of Autogenic Training
15
with the rate before and after AT periods, and this was a little more pronounced in those subjects who experienced hand warmth (table I). The decrease was mainly due to 3 subjects whose respiration rate decreased by 5—7/min during AT (fig. 1). No control subject showed a decrease during AT in the same session.
Discussion Several reasons may explain why there were no significant differences in means between the period before, during and after AT. First, group means were compared, unlike most previous studies in which data of individuals have been reported (4). In addition, our subjects may not have become sufficiently profi cient in the AT technique or a 5-min period may have been too short for most subjects to achieve the intended conditions in a laboratory setting. However, we have also observed that some subjects with several years’ AT experience some times do not either show any physiological changes during AT (unpublished observations). The main reason for this may be that frequent experience of a mental state such as occurs during AT (as well as hypnosis or meditation) pre disposes a subject to enter the state just by adopting a relaxed attitude or posture with eyes closed, either automatically as a result of previous condi tioning or in anticipation of the state to be achieved. The greatest differences in this study were between the AT and control groups in the final session, and between the first and last sessions of the AT group, suggesting predominantly long-term effects of AT.
Finger Blood Flow The subjects of the AT group who experienced hand warmth showed a significantly greater blood flow than those who did not experience hand warmth but the AT group as a whole did not differ from the control group. Although blood flow changes were generally small, some individuals (fig. 1) showed
Downloaded by: Stockholm University Library 130.237.165.40 - 2/21/2018 2:36:52 AM
Skin Temperature Mean finger skin temperature increased significantly during recording in both the AT and control groups in all sessions. This point has not been appreci ated before in AT studies which have not used control groups (4). Skin tempera ture increases with time can therefore occur without any special mental training, and probably reflect increased relaxation. However, in the final session, skin temperature of the AT group (particularly in those subjects who experienced hand warmth) increased significantly more than in the control group, indicating that AT practice enhances rise in skin temperature. Only the subjects who experienced hand warmth showed a significant increase in skin temperature during AT as compared with the periods before and after AT.
Tebecis/Ohno /Matsubara /Sugano/Takeya/Ikemi/Takasaki
16
marked changes during AT in the final session. It is possible, therefore, that blood flow differences between the AT and control groups were only beginning to appear in session 4, and that 1 or 2 months more AT practice would have resulted in more conspicuous differences. Electroencephalogram Although the EEG has been the most commonly studied physiological pa rameter in investigations of AT (4), the reports are confusing. Even the few investigators who used frequency analysis of the EEG did not report group means. Despite this, a conclusion often made or implied is that during AT a-activity sometimes initially increases and later decreases, with a concomitant increase in 0 -(sometimes also 5)-activity (2, 3, 6, 7, 10). In the present study, mean 6- and 0-activity occurred more whereas a-and 0-activity occurred less in session 4 than in session 1 in the AT group. There were no such changes in the control group. That is, by the end of 4.5 months AT practice, the EEG had ‘slowed down’ in the AT group, whether the subjects were actually doing AT or not. Similar conclusions were made from studies of sub jects practising self-hypnosis (13) and transcendental meditation (11). In the hypnosis-study increased 0 -activity was evident when the subjects’ eyes were open as well as closed, indicating that drowsiness cannot explain the increase in 0-activity (13). There is some basis, therefore, to postulate similar physiological mechanisms underlying these altered states of consciousness, at least for the earlier stages.
Respiration Rate The greatest effect of AT on respiration in experienced subjects has been reported to be a decrease in frequency (4). In our study, no significant differ ences were apparent between group means, but again this might be due to subjects’ insufficient experience with AT, as a small proportion did show a marked decrease in respiration rate during AT in session 4. In conclusion, this longitudinal study indicates that in a group of normal subjects of variable motivation to practice AT, some physiological changes (but not others) become apparent 4.5 months after the beginning of training. These
Downloaded by: Stockholm University Library 130.237.165.40 - 2/21/2018 2:36:52 AM
Microvibration MV, or minor tremor (8) of the body surface during AT has been little studied and the few reports available are conflicting (4). In the present study, there were no significant differences in mean MV frequency bands between groups, sessions or periods within a session. However, a longer training program might have revealed greater differences, as reported by others using more experi enced subjects (4, 6). In any case, the results show that the EEG is more sensitive than the MV to changes in consciousness.
The Physiology of Autogenic Training
17
changes are not necessarily more pronounced during AT practiced over a few minutes. In other words, the physiological effects of AT are predominantly long-term.
References
Dr. A.K. Tebecis, Department of Psychosomatic Medicine, Kyushu University, Medical School, Fukuoka City 812 (Japan)
Downloaded by: Stockholm University Library 130.237.165.40 - 2/21/2018 2:36:52 AM
1 Eysenck, H.J.: The manual of the Maudsley Personality Inventory (University of London Press, London 1959). 2 Israel, I,.; Geissmann, P. et Noël, C.: Modifications des rythmes électroencéphalographiques au cours de la relaxation, observées à l’analyse de fréquence. Revue Méd. Psychosom. 3: 133-136(1960). 3 Jus, A. et Jus, C.: Etude polygraphique du “ training autogène”. Revue Méd. Psycho som. 3: 136-138 (1960). 4 Luthe, VJ.: ln Luthe Autogenic therapy, vol. 4: Research and theory (Grüne & Strat ton, New York 1970). 5 MPI Kenkyu Kai Ichido: Shin Sheikaku Kensa Ho - Maudsley Seikaku Kensa. In Japanese (Seishin Shobo, Tokyo 1969). 6 Ohno, Y.: Studies on physiological effects of autosuggestion - centered around auto genic training. In Japanese. Fukuoka Acta med. 56: 1102-1119 (1965). 7 Rohmer, F. et Israel, L. : L’clectroencéphalogrammc dans le training autogène. Revue neurol. 96: 559-563 (1957). 8 Rohracher, H. and Inanaga, K. : Die Mikrovibration (Hans Huber, Bern 1969). 9 Schultz, J.H. and Luthe, W.: In Luthe Autogenic therapy, vol. 1: Autogenic methods (Grune& Stratton, New York 1969). 10 Stojanow, W. und Heidrich, R. : Das EEG während des autogenen Trainings. Psychiatric Neurol, med. Psychol. 14: 13-18 (1962). 11 Tebëcis, A.K.: A controlled study of the EEG during transcendental meditation: com parison with hypnosis. Folia psychiat. neurol. jap. 29: 305-313 (1975). 12 Tebëcis, A.K.; Ohno, Y.; Matsubara, H.; Sugano, H.; Takeya, T , and Ikemi, Y.: A longitudinal physiological study of autogenic training. Proc. 17th A. Meet. Jap. Soc. Psychosom. Med., Tokyo 1976. 13 Tebëcis, A.K.; Provins, K.A.; Farnbach, R. W., and Pentony, P.: Hypnosis and the EEG. A quantitative investigation. J. nerv. ment. Dis. 161: 1-17 (1975).
A Longitudinal Study of Some Physiological Parameters and Autogenic Training A.K. Tebecis', Y. Ohno, H. Matsubara, H. Sugano, T. Takeya, Y. lkem iand M. Takasaki Department of Psychosomatic Medicine, Kyushu University Medical School, Fukuoka City
Abstract. Physiological parameters were studied over 4.5 months in normal high-school girls. One group frequently practised autogenic training (AT) whereas the other (control) did not. After 4.5 months the physiology of the AT, but not the control group had changed to some extent. The mean increase in finger skin temperature in the final session was significantly greater in the AT group than in the control group, especially in subjects who experienced hand warmth during AT. No significant differences between groups were appar ent for finger blood flow, but within the AT group the girls who experienced hand warmth showed a significantly greater blood flow than those who did not experience hand warmth. The electroencephalogram (EEC) of the AT group showed significantly more ‘slow’ (ft and 0) and less fast (a and (?) activity than at the beginning of the project. Microvibration (MV) of the thumb did not differ significantly in any comparisons. Both in the AT and control group means there were no significant differences between the periods before, during and after AT although significant changes did occur in some individuals of the AT group in the final session. In particular, in 3 of 14 AT subjects respiration rate decreased markedly even though weekly training had not included any breathing exercises. The results suggest that the effects of AT are predominantly long-term, resulting in gradual physiological changes that are maintained to some extent between practice periods.
Introduction
1 Visiting Professor from the Australian National University, supported by the Japan Society for the Promotion of Science.
Downloaded by: Stockholm University Library 130.237.165.40 - 2/21/2018 2:36:52 AM
Autogenic training (AT) is a very useful tool in psychosomatic medicine (9). However, the few studies done on AT are difficult to evaluate because of the different experimental designs used, the lack of controls in some cases, the use of both patients and normal subjects, the variable duration of subjects’ experi ence with AT and other reasons. In addition, few investigators have reported
Tebecis/Ohno/Matsubara/Sugano/Takeyajlkemi/Takasaki
9
group means, making it difficult to know whether the reported results of se lected individuals represent characteristic effects of AT or not (4). Some of these variables were avoided in the present study of AT and finger skin temperature, finger blood flow, electroencephalogram (EEG), microvibra tion (MV) of the thenar eminence, and respiration rate, recorded from normal high-school girls randomly divided into two groups of equal size. Every week one group was instructed in the technique of AT and urged to practise it daily whereas the other (control) group received no such instructions. Physiological recordings were made at intervals over a few months in order to determine whether the above-mentioned physiological parameters in the AT group were different (a) before, during and after AT periods within a session, and (b) between sessions, that is, with long-term experience with AT. Some of this material has been briefly reported elsewhere (12).
Subjects and training. 30 unpaid Japanese girls from a Women’s Commercial High School, 15-18 years old (mean age 16.1 years), were divided into two groups of 15 girls each. The AT group was taught AT every week and urged to practise it daily at least 2 -3 times, as it was ‘very good for sport’. The control group received no special instructions. Physiological parameters were recorded from all subjects under similar conditions prior to giving any instructions (session 1). Each subject filled out scales E (extraversion-introversion), N (neuroticism) and L (lie) of the Maudsley Personality Inventory (MPI) (1). Records were made from all subjects randomly during the afternoons of 4 -5 consecu tive days. By session 2, 1 month after session 1, most subjects of the AT group claimed they had learned to experience a ‘calm mind’ and 'heaviness of the hands’ (9; Standard Autogenic Exercise 1) during their private training. By session 3 (3 months after the beginning) these experiences were more marked, and a small proportion of subjects could also experience ‘hand warmth’ (9; Standard Autogenic Exercise 2). By session 4 (last session; 4.5 months after the beginning) most AT subjects claimed they could often experience hand warmth. No other exercises were taught as various reasons prevented continuation of the project. All subjects again filled out scales E, N and L of the MPI after the final session. Equipment and procedure. Each subject was required to lie on a couch, eyes closed, facing upwards with the arms extended parallel to the body axis, in a dim, semi-soundproof, shielded room maintained at 23 °C. The signals were displayed from an adjacent room. Skin temperature was recorded by a thermistor probe attached to the volar surface of the middle finger of the right hand and was read directly from an electronic thermometer (San-Ei Instrument Co., Model M44-4). Local pulsatile blood flow of the index finger of the right hand was recorded with a plethysmogram cuff (San-Ei, Model E-037). Respiration was recorded by a thermistor probe (San-Ei, Model E-034) fixed in the right nostril. MV of the thenar eminence of the left hand was recorded by a disc-type (23 mm diameter, about 5 g weight) crystal acceleration pick-up (Nihon Kohden Co., MT-3T Pickup). The piezoelectric pick-up was sensitive mainly (> 90 %) to vibration parallel to the surface of the disc. Monopolar EEG recordings were made from symmetrical areas on the left and right parietal scalp using conventional silver-silver chloride electrodes, the indifferent lead being attached to the right ear-lobe (13).
Downloaded by: Stockholm University Library 130.237.165.40 - 2/21/2018 2:36:52 AM
Methods
Tebecis/Ohno /Matsubara/Sugano/Takeya/Ikemi/Takasaki
10
All the signals except skin temperature were displayed on separate channels of a 14-channel electroencephalograph chart recorder (San-Ei) using a time constant of 0.3 sec and an upper frequency cut-off of 30 Hz. Skin temperature was recorded on the chart at appropriate intervals. An EEG frequency analyzer (San-Ei, Model 7P11) was used to analyse the MV and EEG (right side) into bands of 2 -4 (5), 4 - 6 (6,), 6 -8 (62), 8 -1 3 (a) and 13-20 Hz (0,) using 10-sec epochs. Three consecutive 5-min periods, before, during and after were analysed. The subject was told to relax without moving unnecessarily and not to fall asleep. For the control subjects in all sessions and the AT subjects in session 1 the instructions were repeated 5 min later and again after a further 5 min later, and 15 min after the beginning the recording was stopped. The AT subjects in the remaining sessions received these instructions at the begin ning, followed by 'Please begin AT’ 5 min later and ‘please stop AT’ a further 5 min later. During removal of the leads, the subject was asked about her subjective experiences. Analysis. For each 5-min recording, skin temperature was taken as the mean of 3 values at 30-sec intervals of the final 60 sec. Amplitude of plethysmogram pulse was the mean of the final 10 consecutive pulses. Respiration rate was the number of breaths in the final 60 sec. For the MV and EEG, the amplitudes of the frequency analyzer output pulses of three consecutive 10-sec epochs were measured after the experiment and the ‘activity’ (total voltage) of a particular frequency band was converted to a proportion of the total activity. MV-analyzed pulses were taken from the final 60 sec of each 5-min period and EEG pulses from records immediately before. Finger skin temperature and amplitude of plethysmogram pulses (mean of 10 consecutive pulses) were also taken shortly (within 1 min) after the very beginning of recording (initial values). F'or overall comparisons of means between 5-min periods of a particular group and between the two groups, or between sessions of a group, the data were subjected to analysis of variance. In comparisons of certain subgroups (see ‘Results’) the t test or MannWhitney U test (dependent or independent samples, where appropriate) was used. Probabili ty values are for two-tailed tests.
No data were used from 2 (1 AT and 1 control) of the 30 subjects who did not complete the 4.5-month program or from records if the EEG indicated that the subject was falling asleep. AT subjects’ comments during the weekly instruc tion period suggested that in most cases motivation and enthusiasm to practise AT privately was low, being of the order of several times per week rather than per day. It was not possible to reliably estimate their rate of practice more precisely. The MPI scores indicated that all subjects were within the normal range for the Japanese population (5), that there were no significant differences between the AT and control groups, and that no significant changes had occurred be tween the beginning and end of the project. Thus, for the AT and control groups, respectively, the mean scores immediately after the first session were 31.4, 34.1 (scale E), 24.8, 27.3 (scale N) and 11.8, 9.1 (scale L). After the final session, the mean scores were 31.8, 32.9 (E), 21.8,25.1 (N) and 10.6,9.8 (L).
Downloaded by: Stockholm University Library 130.237.165.40 - 2/21/2018 2:36:52 AM
Results
The Physiology of Autogenic Training
11
During AT
Before AT
After AT
Microvibration
♦t*
f ro ' »I*
“»•et»» »♦»t**
»f*
200 mV1 EEG - , ---- -
—
_ ------- - V
- --- ------------------ V , ---------------
50
mV ]
Finger blood flow
/^ W V v y V v M M V Finger skin temperature
29.9 °C
/ \ A / M M \ / S /W \ J \y \ 30.8 °C
30.A “C
Respiration
Seconds
Individual Records For most subjects, there was little or no difference in any mean physio logical parameter between the 5-min periods of the control group in all sessions, and of the AT group in sessions 1—3. In the final session, however, approxi mately one half of the AT subjects displayed significant differences between the 5-min periods in at least one parameter. Sample records from 1 subject are illustrated in figure 1. Respiration rate decreased from 19.0/min before AT to 12.0/min during AT and increased to 22.6/min after AT. Finger skin temperature increased by almost 1 °C during AT and decreased again by 0.4 °C after AT. Mean amplitude of the plethysmogram pulse increased by 14 % during AT and decreased by approxi mately 20 % after AT. These changes were statistically significant (p < 0.01). In the EEG there was an increase in total 0-activity with a concomitant decrease in a-activity during AT but the differences were not significant. There were no significant differences in MV. These records confirm that significant changes can occur during AT in an experienced individual, even though this was not the case for the whole group.
Downloaded by: Stockholm University Library 130.237.165.40 - 2/21/2018 2:36:52 AM
Fig. 1. Records taken at comparable times during the final 60 sec of periods before, during and after AT in session 4 of 1 subject (eyes closed) who experienced a ‘calm mind’ and ‘heaviness and warmth of the hands’ during AT. The records immediately below the EEG tracings (right parietal area) are the output pulses of the frequency analyser, showing the relative voltages of the different EEG frequency bands (6, 0, a, 0). Blood flow is represented by the plethysmogram pulses.
Tebecis/Ohno/Matsubara/Sugano/Takeya/lkemi/Takasaki
12
Fig. 2. Mean increases in finger skin temperature relative to the initial mean values of the control and AT groups in session 4. The AT group is further subdivided into those subjects who experienced hand warmth during AT and those who did not (50 % each).
Downloaded by: Stockholm University Library 130.237.165.40 - 2/21/2018 2:36:52 AM
Fig. 3. Mean activity (total voltage) expressed as proportions of 5 different EEG fre quency bands of the AT (a) and control (b) groups in sessions 1 and 4. Each histogram represents the mean of the sum of all the values of the periods before, during and after AT.
13
Group Data There were no significant differences in any mean physiological parameter between periods before, during and after AT in any session of either group, although some trends towards change during AT were apparent for the AT group in session 4. In this session, all the AT subjects reported that they had experi enced a ‘calm mind’ during AT; 78.6 and 50 % said they had also felt heaviness and warmth, respectively, of the hands. The control subjects reported no special experiences other than ‘feeling relaxed’. Finger skin temperature. Finger skin temperature ranged from 27.4 to 32.4 °C, being 29-31 °C in most cases. For almost all subjects, the initial value shortly after the beginning of recording was the lowest of the session. The mean increases during subsequent recording in the same session were significant for both groups in each session (p < 0.001). Figure 2 shows the mean increments in skin temperature relative to the initial value in session 4. The control group showed mean increments of 0.27, 0.19 and 0.32 °C before, during and after AT periods, respectively, the overall mean being a 0.26 °C increase from baseline. The AT group showed mean increments of 0.34, 0.48 and 0.28 °C before, during and after AT periods, the overall mean being 0.37 °C. The difference between these overall means was significant (p < 0.05). Thus, the AT group showed a greater overall mean in crease in skin temperature than the control group. Moreover skin temperature of the AT group, but not the control group, increased during the 5-min AT period. In those subjects who experienced warmth of the hands during AT, skin tem perature increased significantly during AT (p < 0.05) whereas there were no significant changes between the same three periods in the subjects who did not experience hand warmth. Peripheral blood flow. Mean amplitude of the plethysmogram pulse was divided by the initial mean value, the ratios reflecting peripheral blood flow relative to the beginning of the recording. Blood flow was much more labile than skin temperature in both groups. Whereas skin temperature usually remained constant during a 60-sec sampling period, blood flow often fluctuated within a few seconds. For this reason, changes in skin temperature did not always parallel those of blood flow although large increases in temperature were accompanied by increases in flow and vice versa. Mean amplitude of the plethysmogram pulse generally decreased with time in a recording session. Mean pulse amplitude was similar for both the AT and control groups (table I), there being no significant differences between corresponding means, nor between the means of the 5-min periods within each group. However, within the AT group, the means for the subjects experiencing hand warmth were signifi cantly higher than corresponding means for those who did not (p < 0.05). Electroencephalogram. The greatest differences were between sessions 1 and 4 of the AT group (fig. 3 a). Mean 5-activity in session 4 was almost 6 % higher
Downloaded by: Stockholm University Library 130.237.165.40 - 2/21/2018 2:36:52 AM
The Physiology of Autogenic Training
Tebecis/Ohno/Matsubara/Sugano/Takeya/Ikemi/Takasaki
14
Table /. Mean plethysmogram pulse amplitudes relative to the initial values (ratios) and mean respiration rates per minute of the control and AT groups before, during and after 5-min AT periods in session 4 Groups
Control group AT group Subjects experiencing hand warmth Subjects not experiencing hand warmth
Plethysmogram pulse ratio
Respiration rate
before
during
after
before
during
after
0.91 0.92
0.82 0.88
0.66 0.74
18.7 18.3
19.1 17.3
18.2 18.9
1.07
1.08
0.91
18.5
16.9
19.8
0.69
0.S9
0.48
18.0
17.8
17.5
Downloaded by: Stockholm University Library 130.237.165.40 - 2/21/2018 2:36:52 AM
than in session 1, a significant difference (p < 0.01). 0 ,- and 02-activity were each approximately 2 % greater in session 4, both increases being significant (p < 0.01). a- and /3-activities were both significantly lower (by almost 6 and 3.7 %, respectively) in session 4 (p < 0.01). That is, the EEG of the AT group displayed more slow (< 8 Hz) activity and less fast (> 8 Hz) activity in session 4 compared with session 1. Similar comparisons between other sessions showed no significant differences. In contrast, the control group showed no significant differences between corresponding means, or even trends in the data, in any comparisons of sessions. For example, for sessions 1 and 4 any difference in a particular frequency band was less than 2 % (fig. 3 b). There were no significant differences in corresponding means between the periods before, during and after AT for any session in either group, even when the AT group was subdivided into subjects who experienced hand warmth and subjects who did not. In the final session, however, the AT group did display a tendency towards increased 6- and 9- activity during AT. Microvibration. There were no significant differences in any mean frequency band between the two groups, between sessions of a group or between the 5-min periods of a session, even in a comparison between subjects who experienced hand warmth and subjects who did not. Changes in the AT group were never theless greater than those in the control group. Thus, the AT group showed 2.2% more total 9- (4 -8 Hz) and 2.6% less /3r activity in session 4 than in session 1, whereas the same frequency bands varied by less than 1 % in the control group, a-activity was remarkably constant in both groups. Respiration rate. There were no significant differences in mean respiration rate between the two groups, nor between the 5-min periods of a group. However, the AT group showed a slight mean decrease during AT as compared
The Physiology of Autogenic Training
15
with the rate before and after AT periods, and this was a little more pronounced in those subjects who experienced hand warmth (table I). The decrease was mainly due to 3 subjects whose respiration rate decreased by 5—7/min during AT (fig. 1). No control subject showed a decrease during AT in the same session.
Discussion Several reasons may explain why there were no significant differences in means between the period before, during and after AT. First, group means were compared, unlike most previous studies in which data of individuals have been reported (4). In addition, our subjects may not have become sufficiently profi cient in the AT technique or a 5-min period may have been too short for most subjects to achieve the intended conditions in a laboratory setting. However, we have also observed that some subjects with several years’ AT experience some times do not either show any physiological changes during AT (unpublished observations). The main reason for this may be that frequent experience of a mental state such as occurs during AT (as well as hypnosis or meditation) pre disposes a subject to enter the state just by adopting a relaxed attitude or posture with eyes closed, either automatically as a result of previous condi tioning or in anticipation of the state to be achieved. The greatest differences in this study were between the AT and control groups in the final session, and between the first and last sessions of the AT group, suggesting predominantly long-term effects of AT.
Finger Blood Flow The subjects of the AT group who experienced hand warmth showed a significantly greater blood flow than those who did not experience hand warmth but the AT group as a whole did not differ from the control group. Although blood flow changes were generally small, some individuals (fig. 1) showed
Downloaded by: Stockholm University Library 130.237.165.40 - 2/21/2018 2:36:52 AM
Skin Temperature Mean finger skin temperature increased significantly during recording in both the AT and control groups in all sessions. This point has not been appreci ated before in AT studies which have not used control groups (4). Skin tempera ture increases with time can therefore occur without any special mental training, and probably reflect increased relaxation. However, in the final session, skin temperature of the AT group (particularly in those subjects who experienced hand warmth) increased significantly more than in the control group, indicating that AT practice enhances rise in skin temperature. Only the subjects who experienced hand warmth showed a significant increase in skin temperature during AT as compared with the periods before and after AT.
Tebecis/Ohno /Matsubara /Sugano/Takeya/Ikemi/Takasaki
16
marked changes during AT in the final session. It is possible, therefore, that blood flow differences between the AT and control groups were only beginning to appear in session 4, and that 1 or 2 months more AT practice would have resulted in more conspicuous differences. Electroencephalogram Although the EEG has been the most commonly studied physiological pa rameter in investigations of AT (4), the reports are confusing. Even the few investigators who used frequency analysis of the EEG did not report group means. Despite this, a conclusion often made or implied is that during AT a-activity sometimes initially increases and later decreases, with a concomitant increase in 0 -(sometimes also 5)-activity (2, 3, 6, 7, 10). In the present study, mean 6- and 0-activity occurred more whereas a-and 0-activity occurred less in session 4 than in session 1 in the AT group. There were no such changes in the control group. That is, by the end of 4.5 months AT practice, the EEG had ‘slowed down’ in the AT group, whether the subjects were actually doing AT or not. Similar conclusions were made from studies of sub jects practising self-hypnosis (13) and transcendental meditation (11). In the hypnosis-study increased 0 -activity was evident when the subjects’ eyes were open as well as closed, indicating that drowsiness cannot explain the increase in 0-activity (13). There is some basis, therefore, to postulate similar physiological mechanisms underlying these altered states of consciousness, at least for the earlier stages.
Respiration Rate The greatest effect of AT on respiration in experienced subjects has been reported to be a decrease in frequency (4). In our study, no significant differ ences were apparent between group means, but again this might be due to subjects’ insufficient experience with AT, as a small proportion did show a marked decrease in respiration rate during AT in session 4. In conclusion, this longitudinal study indicates that in a group of normal subjects of variable motivation to practice AT, some physiological changes (but not others) become apparent 4.5 months after the beginning of training. These
Downloaded by: Stockholm University Library 130.237.165.40 - 2/21/2018 2:36:52 AM
Microvibration MV, or minor tremor (8) of the body surface during AT has been little studied and the few reports available are conflicting (4). In the present study, there were no significant differences in mean MV frequency bands between groups, sessions or periods within a session. However, a longer training program might have revealed greater differences, as reported by others using more experi enced subjects (4, 6). In any case, the results show that the EEG is more sensitive than the MV to changes in consciousness.
The Physiology of Autogenic Training
17
changes are not necessarily more pronounced during AT practiced over a few minutes. In other words, the physiological effects of AT are predominantly long-term.
References
Dr. A.K. Tebecis, Department of Psychosomatic Medicine, Kyushu University, Medical School, Fukuoka City 812 (Japan)
Downloaded by: Stockholm University Library 130.237.165.40 - 2/21/2018 2:36:52 AM
1 Eysenck, H.J.: The manual of the Maudsley Personality Inventory (University of London Press, London 1959). 2 Israel, I,.; Geissmann, P. et Noël, C.: Modifications des rythmes électroencéphalographiques au cours de la relaxation, observées à l’analyse de fréquence. Revue Méd. Psychosom. 3: 133-136(1960). 3 Jus, A. et Jus, C.: Etude polygraphique du “ training autogène”. Revue Méd. Psycho som. 3: 136-138 (1960). 4 Luthe, VJ.: ln Luthe Autogenic therapy, vol. 4: Research and theory (Grüne & Strat ton, New York 1970). 5 MPI Kenkyu Kai Ichido: Shin Sheikaku Kensa Ho - Maudsley Seikaku Kensa. In Japanese (Seishin Shobo, Tokyo 1969). 6 Ohno, Y.: Studies on physiological effects of autosuggestion - centered around auto genic training. In Japanese. Fukuoka Acta med. 56: 1102-1119 (1965). 7 Rohmer, F. et Israel, L. : L’clectroencéphalogrammc dans le training autogène. Revue neurol. 96: 559-563 (1957). 8 Rohracher, H. and Inanaga, K. : Die Mikrovibration (Hans Huber, Bern 1969). 9 Schultz, J.H. and Luthe, W.: In Luthe Autogenic therapy, vol. 1: Autogenic methods (Grune& Stratton, New York 1969). 10 Stojanow, W. und Heidrich, R. : Das EEG während des autogenen Trainings. Psychiatric Neurol, med. Psychol. 14: 13-18 (1962). 11 Tebëcis, A.K.: A controlled study of the EEG during transcendental meditation: com parison with hypnosis. Folia psychiat. neurol. jap. 29: 305-313 (1975). 12 Tebëcis, A.K.; Ohno, Y.; Matsubara, H.; Sugano, H.; Takeya, T , and Ikemi, Y.: A longitudinal physiological study of autogenic training. Proc. 17th A. Meet. Jap. Soc. Psychosom. Med., Tokyo 1976. 13 Tebëcis, A.K.; Provins, K.A.; Farnbach, R. W., and Pentony, P.: Hypnosis and the EEG. A quantitative investigation. J. nerv. ment. Dis. 161: 1-17 (1975).
