Perceptrial and Motor Skills, 1978, 46, 559-562. @ Perceptual and Motor Skills 1978
FIELD DEPENDENCE AND THE EFFECT OF REM DEPRIVATION O N THIRST1 DAVID KOULACK9 University of Manitoba
JOSEPH DE KONINCK University of Ottazua
GENE OCZKOWSKI University of Manitoba
Summary.-Recently a number of studies have concerned the possible function of rapid eye movement (REM) sleep and the mastery of stress. The present study was designed to explore the possibility that REM sleep might play a function in reducing the potency of a stressful physiological stimulus, thirst, as well as the possibility that such a function might be specific to individuals falling at different points along the field-dependence dimension. While there was no difference between REM deprivation and noa-REM awakening nights in subsequent morning thirst, there was a significant interaction between field dependence and night on morning thirst measures for 10 college students. These results are discussed in light of previous work on stylistic differences in dreaming and their possible role in adaptation to stress.
Since Dement (1960) observed that rapid-eye-movement (REM) sleep deprivation produced detrimental effects, several attempts have been made to clarify this phenomenon. For example, Grieser, Greenberg, and Harrison (1972) found that subjects high in ego strength, who were permitted to have REM sleep after encountering both threatening and nonthreatening material, were better able to remember the threatening material than subjects who were REM-deprived. These findings suggest that REM sleep might serve the function of reducing the potency of threatening presleep stimuli and pave the way for f u m e adaptation to stress. The present study was designed to explore the possibility that REM sleep might serve the same function for an ongoing, internal, physiological stimulus, such as thirst which has been found to alter the sleep cyde (Koulack, 1970). Specifically it was hypothesized that water-deprived subjects would exhibit more thirst in the morning after REM deprivation than after a night with an equal number of awakenings during non-REM (NREM) sleep. In addition, since field-dependent and field-independent subjects have been found to differ in REM dream-content pattern and in reaction to REM deprivation (Witkin, 1970a), it was decided to examine the differences in morning thirst between these two types of subjects. No specific prediction could be made as to direction of this difference.
METHOD Subjects The subjects were 10 paid, male volunteer students at the University of Manitoba, T h i s research was supported in pan by the Medical Research Council of Canada, Grant
MA-3929 to David Koulack.
'Send reprint reauests to David Kouladc. De~artmentof Psychologv. - - . University of Manitoba, w;nnipeg, ' ~ a n i t o b a ,Canada R ~ T ' Z N ~ .
5 60
D. KOULACK, ET AL.
who ranged in age from 18 to 27 yr. and had served in previous experiments in our laboratorg. The MMPI and a questionnaire on sleep habits and past and current illness permitted the selection of subjects free from physiological and psychological difficulties. During the screening session the subjects took the Embedded Figures Test and the Rodand-frame Test.
Procedure The subjects were told that the purpose of the experiment was to examine the effects of thirst on their sleep and dreams. The night before they were to begin deprivation (Night 1) subjects came to the laboratory, were wired with electrodes, and then permitted to go to sleep. N o recordings were made on Night 1. In the morning they were awakened and were met by an experimenter who escorted them through their daily routine making sure that they did not eat, drink, or take a nap during the entire day (there were no recorded instances of violations of the ground rules of the experiment). In the evening (Night 2 ) , they returned to the laboratory and indicated by drawing a line through a 15-un. scale designated as ranging from "not thirsty at all" to "extremely thirsty" how thirsty they were at the time. Then the electrodes were attached and they were permitted to go to sleep. Throughout the night continuous electroencephalograms (EEGs), electrooculargrams ( E m s ) and electromyograms (EMGs) were recorded. At the beginning of each REM ~ e r i o d .the subject was awakened bv an 85-db. buzzer and keot awake for approximately 5 min. by asking him to tell if anything was going through his mind and by having him fill out a modified version of the Nowlis Mood Adjective Check List (Koulack, 1969). In the morning the subject was asked to indicate how thirsty he felt on the thirst scale. Then he was shown 40 out-of-focus slides of everyday scenes and asked to say the first thing that came to mind on seeing each slide. The slides were shown on a screen 8 ft. from the bed. (This measure was later dropped from the analysis because of its poor correlation with the other two measures of morning thirst.) Finally the subject was allowed to drink water and the amount consumed was surrevtitiouslv measured. A combination of the thirst scale score and the amount of water consumed in the morning constituted a measure of morning thirst. The procedure on Night 3 which took place approximately a week after Night 2 was identical to that of Night 1. Again the subject was met in the morning by an exprimenter who monitored his behavior throughout the day. In the evening the subject returned to the laboratory for Night 4. O n this night the procedure was identical to that of Night 2 with an equal number of awakenings being made; however, the awakenings were made during NREM sleep at approximarely the same time of night as the REM awakenings had taken place during Night 2, enabling the subject to have his full complement of REM sleep.' The thirst measure was obtained by combining the scores from the morning-thirst scale and the amount of water consumed. The Pearson correlation coefficient between these measures was .52 ( d f = 1 8 ) . They were combined by calculating z scores and averaging each measure for a given subject on a given night.
Mcasidres o f Field Dependence Only Series 111 of the rod-and-frame task was administered (Witkin, Dyk, Faterson, T h e REM deprivation and NREM awakening nights were not counterbalanced because it was necessary to know the number and distribution of awakenings needed for REM deprivation for each subject in order to be able to have an equivalent distribution and number of NREM awakenings.
FIELD DEPENDENCE, REM DEPRIVATION, THIRST
561
Goodenough, & Karp, 1962). In chis series the subject is upright and 8 trials are presented with two presentations of each of the four possibilities of tilt (28") combinations. Embedded figures were presented in the usual manner (Wiekin, 1970b) but instead of using cards, the 12 figures were presented on a screen For each subject a z score was calculated for scores on both tests and the mean of the two scores was used as a measure of field dependence. The 10 subjects were divided into equal groups with the five subjects below the median being relatively field-independent and the five subjects above the median being relatively field-dependent. For the former subjects the mean total error on the rod and frame was 11.21" and the mean total time to find the embedded figures was 117.6 sec. For the latter subjects the mean total error on the rod and frame was 20.7" and the mean total time to find the embedded figures was 408.2 sec. The Pearson coefficient of correlation between the scores was .75.
RESULTSAND DISCUSSION To be certain that there were no differences between groups of subjects in presleep thirst for both the REM deprivation and the NREM awakening nights, a two-factor analysis of variance with repeated measures was calculated using scores on the presleep thirst scale. There are no mean differences in presleep thirst between groups (7.96 and 9.04; F = 0.18) or between the REM-deprivation and NREM-awakening nights (8.01 and 8.04; F = 0.00). No significant interaction (F = 0.01) was noted. To test the main hypothesis that following REM deprivation (Night 2 ) the subjects would be more thirsty than following NREM awakenings, a similar analysis of variance was calculated utilizing the combined thirst measures. The results are presented in Table 1. Contrary to expectations there was no over-all difference in the morning thirst scores between Nights 2 and 4 (4.96 and 4.03) and there was no difference between groups of subjects when scores for the two nights were combined (4.42 and 5.53). However, there was a significant inter.01). Simple main effects were calaction of field dependence by night ( $ culated. These showed that, while field-dependent subjects exhibited the same level of thirst on both REM deprivation and NREM awakening nights (4.77 and 4.07, respectively), field-independent subjects exhibited a significant increase in morning thirst following the non-REM awakening night (5.15 and 6.00; F = 6.00, p < .05 ) .
FIELD DEPENDENCE AND THE EFFECT OF REM DEPRIVATION O N THIRST1 DAVID KOULACK9 University of Manitoba
JOSEPH DE KONINCK University of Ottazua
GENE OCZKOWSKI University of Manitoba
Summary.-Recently a number of studies have concerned the possible function of rapid eye movement (REM) sleep and the mastery of stress. The present study was designed to explore the possibility that REM sleep might play a function in reducing the potency of a stressful physiological stimulus, thirst, as well as the possibility that such a function might be specific to individuals falling at different points along the field-dependence dimension. While there was no difference between REM deprivation and noa-REM awakening nights in subsequent morning thirst, there was a significant interaction between field dependence and night on morning thirst measures for 10 college students. These results are discussed in light of previous work on stylistic differences in dreaming and their possible role in adaptation to stress.
Since Dement (1960) observed that rapid-eye-movement (REM) sleep deprivation produced detrimental effects, several attempts have been made to clarify this phenomenon. For example, Grieser, Greenberg, and Harrison (1972) found that subjects high in ego strength, who were permitted to have REM sleep after encountering both threatening and nonthreatening material, were better able to remember the threatening material than subjects who were REM-deprived. These findings suggest that REM sleep might serve the function of reducing the potency of threatening presleep stimuli and pave the way for f u m e adaptation to stress. The present study was designed to explore the possibility that REM sleep might serve the same function for an ongoing, internal, physiological stimulus, such as thirst which has been found to alter the sleep cyde (Koulack, 1970). Specifically it was hypothesized that water-deprived subjects would exhibit more thirst in the morning after REM deprivation than after a night with an equal number of awakenings during non-REM (NREM) sleep. In addition, since field-dependent and field-independent subjects have been found to differ in REM dream-content pattern and in reaction to REM deprivation (Witkin, 1970a), it was decided to examine the differences in morning thirst between these two types of subjects. No specific prediction could be made as to direction of this difference.
METHOD Subjects The subjects were 10 paid, male volunteer students at the University of Manitoba, T h i s research was supported in pan by the Medical Research Council of Canada, Grant
MA-3929 to David Koulack.
'Send reprint reauests to David Kouladc. De~artmentof Psychologv. - - . University of Manitoba, w;nnipeg, ' ~ a n i t o b a ,Canada R ~ T ' Z N ~ .
5 60
D. KOULACK, ET AL.
who ranged in age from 18 to 27 yr. and had served in previous experiments in our laboratorg. The MMPI and a questionnaire on sleep habits and past and current illness permitted the selection of subjects free from physiological and psychological difficulties. During the screening session the subjects took the Embedded Figures Test and the Rodand-frame Test.
Procedure The subjects were told that the purpose of the experiment was to examine the effects of thirst on their sleep and dreams. The night before they were to begin deprivation (Night 1) subjects came to the laboratory, were wired with electrodes, and then permitted to go to sleep. N o recordings were made on Night 1. In the morning they were awakened and were met by an experimenter who escorted them through their daily routine making sure that they did not eat, drink, or take a nap during the entire day (there were no recorded instances of violations of the ground rules of the experiment). In the evening (Night 2 ) , they returned to the laboratory and indicated by drawing a line through a 15-un. scale designated as ranging from "not thirsty at all" to "extremely thirsty" how thirsty they were at the time. Then the electrodes were attached and they were permitted to go to sleep. Throughout the night continuous electroencephalograms (EEGs), electrooculargrams ( E m s ) and electromyograms (EMGs) were recorded. At the beginning of each REM ~ e r i o d .the subject was awakened bv an 85-db. buzzer and keot awake for approximately 5 min. by asking him to tell if anything was going through his mind and by having him fill out a modified version of the Nowlis Mood Adjective Check List (Koulack, 1969). In the morning the subject was asked to indicate how thirsty he felt on the thirst scale. Then he was shown 40 out-of-focus slides of everyday scenes and asked to say the first thing that came to mind on seeing each slide. The slides were shown on a screen 8 ft. from the bed. (This measure was later dropped from the analysis because of its poor correlation with the other two measures of morning thirst.) Finally the subject was allowed to drink water and the amount consumed was surrevtitiouslv measured. A combination of the thirst scale score and the amount of water consumed in the morning constituted a measure of morning thirst. The procedure on Night 3 which took place approximately a week after Night 2 was identical to that of Night 1. Again the subject was met in the morning by an exprimenter who monitored his behavior throughout the day. In the evening the subject returned to the laboratory for Night 4. O n this night the procedure was identical to that of Night 2 with an equal number of awakenings being made; however, the awakenings were made during NREM sleep at approximarely the same time of night as the REM awakenings had taken place during Night 2, enabling the subject to have his full complement of REM sleep.' The thirst measure was obtained by combining the scores from the morning-thirst scale and the amount of water consumed. The Pearson correlation coefficient between these measures was .52 ( d f = 1 8 ) . They were combined by calculating z scores and averaging each measure for a given subject on a given night.
Mcasidres o f Field Dependence Only Series 111 of the rod-and-frame task was administered (Witkin, Dyk, Faterson, T h e REM deprivation and NREM awakening nights were not counterbalanced because it was necessary to know the number and distribution of awakenings needed for REM deprivation for each subject in order to be able to have an equivalent distribution and number of NREM awakenings.
FIELD DEPENDENCE, REM DEPRIVATION, THIRST
561
Goodenough, & Karp, 1962). In chis series the subject is upright and 8 trials are presented with two presentations of each of the four possibilities of tilt (28") combinations. Embedded figures were presented in the usual manner (Wiekin, 1970b) but instead of using cards, the 12 figures were presented on a screen For each subject a z score was calculated for scores on both tests and the mean of the two scores was used as a measure of field dependence. The 10 subjects were divided into equal groups with the five subjects below the median being relatively field-independent and the five subjects above the median being relatively field-dependent. For the former subjects the mean total error on the rod and frame was 11.21" and the mean total time to find the embedded figures was 117.6 sec. For the latter subjects the mean total error on the rod and frame was 20.7" and the mean total time to find the embedded figures was 408.2 sec. The Pearson coefficient of correlation between the scores was .75.
RESULTSAND DISCUSSION To be certain that there were no differences between groups of subjects in presleep thirst for both the REM deprivation and the NREM awakening nights, a two-factor analysis of variance with repeated measures was calculated using scores on the presleep thirst scale. There are no mean differences in presleep thirst between groups (7.96 and 9.04; F = 0.18) or between the REM-deprivation and NREM-awakening nights (8.01 and 8.04; F = 0.00). No significant interaction (F = 0.01) was noted. To test the main hypothesis that following REM deprivation (Night 2 ) the subjects would be more thirsty than following NREM awakenings, a similar analysis of variance was calculated utilizing the combined thirst measures. The results are presented in Table 1. Contrary to expectations there was no over-all difference in the morning thirst scores between Nights 2 and 4 (4.96 and 4.03) and there was no difference between groups of subjects when scores for the two nights were combined (4.42 and 5.53). However, there was a significant inter.01). Simple main effects were calaction of field dependence by night ( $ culated. These showed that, while field-dependent subjects exhibited the same level of thirst on both REM deprivation and NREM awakening nights (4.77 and 4.07, respectively), field-independent subjects exhibited a significant increase in morning thirst following the non-REM awakening night (5.15 and 6.00; F = 6.00, p < .05 ) .
