| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Department of Psychology (R.B., M.A.L.), Wayne State University, Detroit; and the Sleep Disorders and Research Center (T.R.), Department of Psychiatry, Henry Ford Medical Centers, Detroit, Michigan.
Address reprint requests to: Mark A. Lumley, PhD, Department of Psychology, Wayne State University, 71 West Warren Ave., Detroit, MI 48202. Email: mlumley{at}sun.science.wayne.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: Fifty healthy, normally sleeping adults from the community completed the 20-item Toronto Alexithymia Scale and slept uninterrupted for one night in the laboratory while polysomnography was conducted. Various measures of sleep latency, sleep stages, and REM sleep–related variables were obtained, and analyses correlated these sleep measures with alexithymia, controlling for age, sex, and level of depressed affect.
RESULTS: Higher alexithymia scores were significantly related to increased stage 1 (light) sleep and decreased stage 3/4 (deep) sleep. Alexithymia was unrelated to overall sleep efficiency or percentage of stage 2 sleep. Alexithymia was related to more frequent REM episodes and more stage 1 sleep during and immediately after REM episodes but was unrelated to the absolute amount of REM sleep. Alexithymia was also related to an earlier onset of the first REM episode.
CONCLUSIONS: Alexithymia is associated with more light sleep and less deep sleep, which may contribute to subjective reports of poor sleep and increased sleepiness, fatigue, and somatic symptoms. Although alexithymia is not associated with an overall reduction of REM sleep, the increased frequency of episodes of REM that are interrupted and followed by light sleep rather than complete awakenings may contribute to limited dream recall.
Key Words: alexithymia • emotional awareness • sleep • dreams • polysomnography.
Abbreviations: CES-D = Center for Epidemiologic Studies Depression Scale; DDF = difficulty describing feelings; DIF = difficulty identifying feelings; EOT = externally oriented thinking; REM = rapid eye movement; TAS-20 = 20-item Toronto Alexithymia Scale.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
Alexithymia seems to be related to two aspects of sleep, poor sleep quality and limited dream recall. Sleep quality includes various aspects of the efficiency, continuity, and restorative properties of sleep. Poor sleep quality is suggested when people report insomnia, restless sleep, light sleep, daytime sleepiness, or feelings of fatigue or not being rested. Very little research has been conducted on the relationship between alexithymia and sleep quality. Our review of the literature found occasional case reports describing alexithymic patients with poor sleep (1); a report that among alexithymic men, nonrestorative sleep and daytime sleepiness were associated with depression and mental confusion (2); and a report that among fibromyalgia patients, less emotional awareness was associated with reports of a shorter duration of uninterrupted sleep (3). In addition, there are well-replicated relationships between alexithymia and conditions known to be associated with poor sleep, such as depression, posttraumatic stress disorder, chronic pain, and general symptom complaints (1, 4), which increase the likelihood that people with alexithymia will complain of poor sleep.
Alexithymia also has been linked with alterations in dream recall. Alexithymic patients have been described as either rarely or never recalling dreams; when they do remember dreams, they describe the dreams as short and lacking elaborate detail (5–8). One study found that alexithymia was associated with a reduced frequency of dream recall assessed retrospectively among asthmatic adults (9), and another study suggested that alexithymic patients often sense that they have dreamed but cannot recall the details (10). Our survey and dream diary studies found that alexithymia, particularly the externally oriented cognitive style, is associated with reduced frequency of dreaming, having shorter dreams, and having dreams rated as boring (11). A recent study in which people were awakened during REM sleep, the stage most closely linked with dreaming, found that alexithymia was associated with dreams that were rated as less fantastic (12).
Although alexithymia may be associated with subjective reports of poor sleep quality and limited dream recall, there has been no published study of the association between alexithymia and objective measures of sleep. Objective sleep properties are typically measured by polysomnography during uninterrupted sleep, which provides a wealth of measures of both sleep quality and REM sleep. For example, complaints of insomnia may be substantiated by polysomnographic measures of poor sleep efficiency, including long latencies to sleep onset or periods of wakefulness after sleep onset. Also, increased stage 1 sleep (light sleep) is generally experienced as having insomnia or having slept lightly, and it contributes to increased sleepiness the next day (13, 14). A reduction in stages 3 and 4 sleep (deep sleep) has been associated with feelings of fatigue and being poorly rested, and the pattern of alpha-delta sleep (deep sleep with intrusions of fast brain activity) has been linked with physical symptoms such as increased pain sensitivity (15, 16).
Alterations in REM sleep could impair the recall of dreams. For example, a reduction in overall REM sleep or shorter REM episodes can lead to reduced dreaming. In addition, dreams are typically recalled most vividly when the sleeper is abruptly awakened either during or soon after an REM period and remains awake for some time. Thus, either having fewer awakenings or experiencing increased light sleep (stage 1) during and immediately after REM episodes would likely interfere with dream recall. This is because the short-lived opportunity for attending to and recalling one’s dreams passes without the full awakening needed to create conscious, detailed memories. Finally, although little studied, the patterning of REM episodes, such as the latency to REM or the number of REM episodes, may influence the dream experience.
Although several studies of alexithymia have involved awakening subjects during REM sleep to obtain dream reports (10, 12, 17), only Tantam et al. (17) reported REM parameters from noninterrupted sleep. They reported preliminary evidence for less REM sleep among six patients with skin disorders, four of whom were judged clinically to be alexithymic. This finding is unreliable, however, because the sample was very small, the patients had an unusual condition, only the first 4 hours of uninterrupted sleep were used, the patient with the least REM sleep took REM-suppressing medication, and another patient drank alcohol heavily (which fragments REM sleep).
Of course, subjective reports may differ greatly from objective measures of a phenomenon. For example, symptom complaints, including reports of poor sleep, may be due not to the objective nature of underlying somatic dysfunction (such as abnormal sleep latency and stages) but to negative affectivity, which is the disposition to report aversive states (18). Impaired dream recall may not be a function of REM variables but rather may depend on cognitive processes, such as a deficit in imagination or a failure to attend to intrapsychic phenomena. A literature review (4) and subsequent report (19) of the linkages between alexithymia and illness found limited evidence that alexithymia is directly related to objective biological measures of health problems but substantial evidence that alexithymia is linked with subjective processes. Thus, in this study, we evaluated whether alexithymia is related to objective alterations in sleep quality and REM sleep as assessed by polysomnography that might underlie subjective reports of poor sleep quality and limited dream recall.
We studied a relatively large sample of healthy adults from the community to provide the maximum generalizability and to avoid biases introduced by the study of narrowly defined clinical samples. Because polysomnographic findings can be influenced by a host of medical and behavioral variables (eg, drug use, caffeine intake, obesity, altered sleep timing, psychiatric disorders, and various medical problems), we excluded people manifesting such confounds. We measured alexithymia with the best validated measure available, the TAS-20 (20), and explored how its subscales (DIF, DDF, and EOT) related to sleep quality and REM measures because the subscales often have different correlates (21, 22). Finally, we statistically controlled for several demographic variables (age and sex) that have been related to either alexithymia or sleep parameters and for depressed affect, because some authors have argued that depression and negative affect may be confounded with alexithymia (22, 23) and because depression has been linked with altered sleep and REM phenomena (24).
We tested the hypotheses that alexithymia is related to reduced sleep efficiency (ie, less sleep for a given time in bed), classic sleep-onset insomnia (ie, a longer sleep onset latency), and more light sleep (stage 1). We made no hypotheses about the relation of alexithymia to stage 2 or 3/4 sleep. Based on the preliminary observations of Tantam et al.(17), we hypothesized that alexithymia is associated with reduced REM sleep and that alexithymia is related to having fewer awakenings during REM sleep and more light sleep during and after REM episodes. We made no hypotheses about the relation of alexithymia to REM patterning, such as the latency to REM sleep and the number and duration of REM episodes.
|
METHOD |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
250 mg daily) and alcohol (14 1-ounce drinks weekly), no acute or chronic medical problems, no history of substance abuse or dependence, no history of psychiatric diagnosis or use of prescription psychoactive medications, and no current medication use. Respondents meeting these criteria came to the laboratory for a physical examination and were included in the study only if they had negative results on drug and pregnancy screens, normal blood and urine chemistry values, a body mass index within 20% of the recommended Metropolitan Life Insurance Tables values, and no significantly elevated scores on the clinical scales of the Minnesota Multiphasic Personality Inventory.
Procedures
All participants presented to the laboratory in early evening and adapted to being in private bedrooms. Electrodes for recording sleep were attached at standard sites according to Rechtschaffen and Kales (25). Participants went to bed at their normal bedtimes and were allowed to remain in bed uninterrupted for 8 hours, at which point they were awakened. Polysomnographic tracings were recording using Grass Instruments model 48-D polygraphs, and sleep stages were later scored in 30-second epochs for periods of wakefulness and stages 1, 2, 3/4, or REM sleep according to standard criteria (25). Scoring was done by technicians who had been trained on a series of standardized records (90% agreement). Technicians were blind to subjects’ alexithymia and depressed affect scores. Almost all participants (N = 43) had not slept in a laboratory before; 7 participants had prior sleep laboratory experience.
Measures
Alexithymia was assessed with the TAS-20 (20), which presents 20 items rated by participants on a scale of 1 (strongly disagree) to 5 (strongly agree). We calculated the total score as well as scores on the three subscales. We obtained the following internal consistencies (Cronbach’s 
): TAS-20 total = 0.79; DIF = 0.78; DDF = 0.70; and EOT = 0.73. The three subscale scores had the following intercorrelations: DIF with DDF (r = 0.37), DIF with EOT (r = 0.15), and DDF with EOT (r = 0.36). In this sample the TAS-20 mean score was 41.3 (SD = 10.0, range = 23–64).
Depressed affect was assessed with the CES-D (26). Participants rated the frequency that they had experienced each of 20 depressed mood and cognition symptoms during the past week on a scale of 0 to 3. The scale has excellent reliability and has been validated against other measures of depression. We obtained an value of 0.88. In this sample the CES-D mean score was 8.5 (SD = 7.8, range = 0–27).
Sleep Quality Measures
The following parameters of sleep quality were measured: 1) sleep stage percentages: percentage of the total sleep time that was spent in stage 1, 2, or 3/4; 2) latency to persistent sleep: duration of time in minutes from "lights out" until the first 10 minutes of uninterrupted sleep in any stage (1, 2, 3/4, or REM); and 3) sleep efficiency: the proportion of the entire time in bed that was spent in any sleep stage.
REM Sleep–Related Measures
The following parameters of REM sleep were measured: 1) REM sleep percentage: percentage of total sleep time spent in REM sleep; 2) REM episodes: number of distinct REM episodes, for which the onset of an episode was defined as the first 30 seconds of REM sleep and the end of an episode was defined by the start of at least 10 consecutive minutes of non-REM sleep or wakefulness; 3) REM episode duration: the mean duration in minutes of all REM episodes during the night; 4) REM-related stage 1: the percentage of total sleep time that was spent in stage 1 sleep and that occurred during an REM episode or within 10 minutes after a REM episode had ended; 5) REM awakenings: number of entries to wakefulness that occurred during or within 10 minutes of a REM episode divided by total hours of REM sleep; and 6) REM latency: duration of time in minutes from the start of persistent sleep to the start of the first REM episode.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
value of 0.05 (two-tailed) was considered significant.
Relationships of Alexithymia to Sleep Quality Measures
Table 1 shows the sample descriptive statistics for the sleep quality measures as well as their correlations (zero-order and partial) with the TAS-20 total score. Alexithymia was positively associated with the percentage of time spent in stage 1 sleep and inversely associated with the percentage of time spent in stage 3/4 sleep even after controlling simultaneously for age, sex, and depressed affect. Alexithymia was unrelated to the time spent in stage 2 sleep. Although alexithymia did not reach significance in its relation to the latency to persistent sleep (p = .07) and was unrelated to overall sleep efficiency, it is important to note that the valence of these correlations suggests a trend in the direction opposite to that hypothesized. Alexithymia tended to be associated with falling asleep faster and better sleep efficiency.
|
Relationships of Alexithymia to REM Sleep–Related Measures
Table 2 shows the sample descriptive statistics for the REM sleep–related measures and their relationships with TAS-20 scores. Alexithymia was significantly related to a higher frequency of distinct REM episodes; all three subscales contributed to this relationship, although DIF had the strongest relationship (r = 0.27, p = .06). Alexithymia was unrelated to the overall percentage of time spent in REM sleep. These two observations, that alexithymia was related to more frequent REM episodes but not related to the overall percentage of REM sleep, suggests that alexithymia was associated with shorter REM episodes. This trend was found in the data, although the relationship did not reach significance.
|
The 7 participants who had slept previously in the laboratory had TAS-20 scores comparable to those of the 43 participants who slept in the laboratory for the first time during this study (t(48) = 0.35, p = .73). Nonetheless, we statistically controlled for the presence or absence of prior laboratory experience and found no change in the significance levels of any of the correlations in Tables 1 and 2. In addition, we tested for the effects of extreme values of all sleep quality and REM-related measures by examining the distribution of each and recalculating correlations after recoding any extreme values to 2 SDs from the sample mean. None of the correlations changed in significance after extreme values were recoded.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
With respect to sleep stages and sleep quality, alexithymia was associated with decreased time spent in the deepest stage of sleep, stage 3/4, and a parallel increase in the amount of time spent in the lightest of stage of sleep, stage 1. There are two possible clinical implications of this tradeoff of deep sleep for light sleep among those with greater alexithymia. The decrease in deep sleep may contribute to a sense of nonrestoration or a feeling of mental and physical fatigue during the day and perhaps even muscular symptoms such as pain (15, 16). This observation suggests the provocative hypothesis that the frequently observed association of alexithymia with depression, somatic symptoms, and pain (1, 4) is mediated by decreases in amount of deep sleep that alexithymic people obtain. The increase in stage 1 sleep is likely to result in a sense of having slept poorly and perhaps even of being awake during the night. Many people report that they are "not sleeping" when they are actually in stage 1 sleep, which might account for reports of poor sleep or insomnia among alexithymic people. It is noteworthy that increased stage 1 sleep was the only objective sleep alteration found that may underlie subjective reports of insomnia among alexithymic people. Alexithymia was unrelated to sleep efficiency throughout the night, and there was a nonsignificant trend for alexithymia to be associated with falling asleep faster rather than slower, as occurs with sleep-onset insomnia.
With respect to REM sleep–related measures, we found no evidence linking alexithymia to the overall percentage of time spent in REM sleep. This finding counters an earlier preliminary observation (17), although, as noted in the Introduction, the finding of that study was probably unreliable. Yet we found that alexithymia was associated with having a greater number of REM episodes, and there was a nonsignificant trend for those episodes to be shorter in duration. Furthermore, alexithymia was associated with increased stage 1 sleep during and immediately after REM episodes but was not related to the number of full awakenings that occurred during REM sleep. These findings suggest that the REM episodes of the more alexithymic subjects were abbreviated by arousals to the lighter sleep of stage 1, although not to full awakenings. Briefer REM episodes may give rise to dreams of shorter duration, thus limiting the length or details of dream reports. An interpretation more consistent with our data, however, is that an increase in stage 1 sleep during and after REM episodes may prevent alexithymic subjects from having the abrupt, full, and extended awakenings needed for dream recall. Thus, the results of this study suggest that there may be some objective, REM sleep–related factors that affect dream recall in alexithymic people.
Alexithymia was also related to an earlier onset of the first REM episode. Research has linked a shorter REM latency with depression (24), yet the relation of alexithymia with REM latency was independent of depressed affect in this study. This suggests that the emotional regulation deficits underlying alexithymia, those that are unique to alexithymia and independent of general depressed affect, may contribute to a shorter REM latency.
There are several limitations of this study. First, the sleep data were obtained almost exclusively from subjects sleeping for the first time in the laboratory, and it is known that many people have poorer sleep and altered sleep characteristics during their first laboratory night. It is possible that alexithymia interacts with sleep laboratory experience in its relationship to sleep characteristics such that the relationships found between alexithymia and sleep characteristics from the first night will not generalize to stable sleep patterns of laboratory-adapted subjects. Although we know of no research or theory that would support such an interaction among healthy people, the current findings need to be replicated on subsequent laboratory nights.
A second limitation is that only people who were mentally and physically healthy and who viewed themselves as normal sleepers were examined. These inclusion criteria greatly increase our confidence that the observed relationships between alexithymia and sleep measures were not confounded by somatic, psychiatric, or behavioral problems that might negatively affect sleep. Also, we believe that by excluding potential confounds and controlling for depressed affect, we were able to obtain a purer assessment of the underlying affective/cognitive deficits of alexithymia. Yet by excluding people with abnormal sleep, psychiatric and substance abuse disorders, obesity, hypertension, and other maladies that have been linked to alexithymia (1), we disproportionately excluded alexithymic people, resulting in a relatively low distribution and restricted range of TAS-20 scores. This not only potentially biased our study against detecting statistical relationships but also limits the generalizability of the findings to relatively healthy people who tend to fall within the nonalexithymic range of TAS-20 scores. Future research should examine people with a broader spectrum of alexithymia scores and people who report poor sleep to test the reliability of these findings.
Future research on personality, dreams, and polysomnography should also consider several other measures. First, there is some evidence that the density of eye movements in REM episodes is related to dream content and to affective disorders such as depression (24). Thus, it would be interesting to know whether REM density is reduced in alexithymia. Second, although our technicians woke subjects after 8 hours in bed, it would be interesting to see how alexithymia is associated with the process of spontaneous morning awakening. This might include an examination of whether alexithymia is associated with preferential awakening from non-REM sleep, a tendency to drift back to non-REM sleep before fully awakening (such as we found for increased stage 1 sleep related to REM), or a tendency to engage in external cognitive or behavioral activities immediately on awakening, all of which would interfere with recall of dreams.
In summary, there are some objective differences in sleep stages associated with alexithymia that may contribute to complaints of poorer sleep quality or increased fatigue. We suspect that alterations in stage 1 and stage 3/4 sleep interact with the alexithymic disposition to report aversive states, giving rise to the final symptom complaints. In addition, poor dream recall or dream reports lacking details may stem from shorter dreams ending in stage 1 sleep rather than full awakenings during REM episodes. We suspect, however, that poor dream recall is probably also a function of constricted imaginal processes, decreased interest in introspection, and a preference for external attention allocation in alexithymic people, which have been suggested by studies of dream reports obtained from REM awakenings (10, 12, 15).
Received for publication August 2, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  L. De Gennaro, M. Ferrara, R. Cristiani, G. Curcio, V. Martiradonna, and M. Bertini Alexithymia and Dream Recall Upon Spontaneous Morning Awakening Psychosom Med, March 1, 2003; 65(2): 301 - 306. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
