- Email: [email protected]
The daily course of the symptomatology and the impaired time estimation in endogenous depression (melancholia)
Journal of Affectioe Elsevier
Disorders,
285
17 (1989) 285-290
JAD 00658
The daily course of the symptomatology and the impaired time estimation in endogenous depression (melancholia) H. Kuhs, W. Hermann, Department
of Psychrutty,
K. Kammer
and R. Tiille
University of Miinster, D-4400 Miinster,
E R. G.
(Received 3 January 1989) (Revision received 20 March 1989) (Accepted 11 April 19X9)
Summary The importance of ‘typical’ diurnal variations in mood (depression worse in the morning) as a diagnostic criterion of endogenous depression has been challenged in previous investigations. Disturbance in time experience may contribute towards an understanding of diurnal variation in depressive symptomatology. To examine this hypothesis a series of time estimation experiments was conducted in the course of the day. Twenty-five endogenous depressive patients (according to ICD-9) and 12 healthy controls were asked to estimate prospectively a 30-s interval on two successive days at 7.30 a.m., 11.30 a.m., 3.30 p.m. and 7.30 p.m. Simultaneously the subjects assessed their state of well-being using a visual analogue mood scale. A circadian rhythm of time estimation errors could not be detected; even in daily courses with ‘typical’ diurnal variations an increasingly favourable self-assessment of well-being was not accompanied by a corresponding diurnal fluctuation of time estimation. These results cast additional doubts on the significance of ‘typical’ diurnal variations in depressive symptomatology.
Ke_v words:
Endogenous
depression;
Time estimation
Introduction ‘Typical’ diurnal variations in well-being with low spirits in the morning and an improvement in mood in the afternoon or evening have long been considered diagnostically usable symptoms of endogenous depression (melancholia). Recent inves-
Address for correspondence: Priv.-Doz. partment of Psychiatry, University of Schweitzer-Str. 11, D-4400 Miinster, F.R.G. 0165-0327/89/$03.50
Dr. H. Kuhs. DeMiinster, Albert-
Q 1989 Elsevier Science
Publishers
tigations (Graw et al., 1980; Von Knorring et al., 1977) however, have shown that previous findings need to be modified. Among other things the daily rhythm of depressive symptomatology is subject to considerable intra-individual variations from day to day (Stallone et al., 1973; Tolle and Goetze, 1987). But if it is at all possible to determine the existence of diurnal variation in depressive symptomatology, then the variation found is predominantly ‘typical’ (i.e., depression worse in the morning; see Tiille. 1989). In healthy persons diurnal variations in mood are also well-known (Engel, 1957; Hampp, 1961;
B.V. (Biomedical
Division)
286
Taub and Berger, 1974). The way in which healthy subjects and melancholies experience the course of the day does. however, reveal fundamental differences: whereas a healthy person plans his day looking ahead, the melancholic regards his day as being endlessly drawn out and insurmountable. The only source of relief for the patient is that time will pass and with it the sufferings of the day. Diurnal variations in melancholia are characterised by a special experiential quality of the low spirits as well as of the improvement in mood, which the patient passively undergoes rather than actively initiates. However. no conclusive empirical verification exists at present for either this melancholic change in experience or for its diurnal fluctuations. The theoretical background for the investigation may be found in a phenomenological aspect whereby changes in time experience are expounded as being fundamental disorders in melancholia (Minkowsky, 1923; von Gebsattel, 1928; Straus, 1928). These changes particularly concern a slowing down of the inner development period (ego time) compared to changes in the surrounding world (universal time) (Straus, 1928) and appear to be in agreement with experimental results: time estimation investigations have revealed characteristic mistakes in judgement by melancholic patients. According to our own findings melancholies prospectively underestimate an interval of 30 s by almost 6 s whereas healthy subjects overestimate the same interval by more than 10 s (Kuhs et al., 1989a). Since time estimation may serve as a quantitative measurement of the disturbance in time experience in melancholia it may be assumed that the repeated ascertainment of time estimation errors in the course of the day can provide a valuable contribution towards understanding the diurnal variation in depressive symptomatology. So far, no time estimation measurements have been carried out in melancholic patients in relation to the time of day. Prospective time estimation experiments conducted by Wever (1973) established that only one healthy person from his study group made substantial errors in time judgements before noon. These judgement errors decreased as the day progressed and reached a minimum towards midnight.
The main questions addressed by this study are as follows: (1) does prospective time estimation reveal daytime fluctuations and (2) what relationship exists between time estimation and the state of well-being in the course of the day. Methods Patients and control group Twenty-five in-patients (10 males, 15 females) treated for endogenous depression (diagnosed according to ICD-9) were admitted to the study. Severity of depression was measured on the Hamilton Rating Scale for Depression (Hamilton, 1960). The mean score was 19.1 + 5.7, the minimum score was 10. The patients were 50.1 _t 11.8 years old. At the time of the investigation all depressive patients received drug therapy with antidepressants. Seventeen of them had undergone sleep deprivation therapy, which had stopped at least 5 days before the investigation began. The patients were compared with a healthy control group (n = 12) only slightly different from the patient group as regards age distribution and sex (five males, seven females, aged 46.8 f 18.0 years). Procedure and time of the investigation In accord with the results of our previous study (Kuhs et al., 1989) the following procedure proved to be especially suitable for determining prospective time estimation errors: the subject prospectively estimates an interval of 30 s by quietly counting up to 10 three times in succession. The number 10 is pronounced loudly each time it is arrived at. The time is registered by the investigator. The estimation error is determined as follows: time estimation error equals objective estimation interval minus subject’s estimation value. Hence, a positive differential value signifies underestimation, a negative differential value overestimation of time. The time estimation experiments were carried out on two successive days at 7.30 a.m., 11.30 a.m., 3.30 p.m. and 7.30 p.m., in each case together with a self-assessment of well-being. The patients and control subjects assessed their state of well-being using a visual analogue mood scale (VAMS) (Luria, 1975). The ends of this
287
100-mm scale read as follows: ‘I have never been as depressed as I am now’ (0) and ‘I am not depressed at the moment’ (100). Diurnal variations in well-being can be calculated from the VAMS scores as follows: daily differential value equals evening value minus morning value: 3.30 p.m. value + 7.30 p.m. value 2
The following statistical evaluation procedures were employed: Wilcoxon’s test for matched pairs. Spearman’s rank correlation coefficient (r) and Fisher’s exact test. Results
-
Daiij rhythm ofsymptomatoIogy/state of well-being On two successive days ‘typical’ daily variations in mood were ascertained in six of the 25 melancholic patients (24.0%) and in two of the healthy controls (16.7%). On only one of the two days ‘typical’ daily variations in well-being were found in eight depressive patients and in two healthy controls. No ‘typical’ diurnal variations on either of the investigation days were evident in 11 depressive patients (44.0%) and eight control subjects (66.7%).
- 7.30 a.m. value + 11.30 a.m. value 2 Diurnal variation was assumed to be present if the daily differential value was at least 10 points and/or 20% on the VAMS. Insignificant differences were not regarded as daily fluctuations. The daily differential value of time estimation errors was calculated in the same way.
VAMS
TEE Daily courses with tctypical)) diurnal va riations in mood (N=20)
804
1st
70-
14 -
60-
12 -
50-
lo-
40-
8-
30-
6-
20-
4-
lo- ’
2-
I I
I
I
7.30 am.
3.30 pm. 11.30 am.
Fig. 1. State of well-being in mood.
Daily courses without cctypicab diurnal variations in mood ( N= 30)
-
and time estimation VAMS
score;
11.30 a.m
7.30p.m. errors
- - - - - - TEE
3.30p.m
7.30a.m
in endogenous (time estimation
depressive error):
patients
with and without
underestimation
Z 30p.m ‘typical’
of a 30-s interval
diurnal
in seconds.
variations
288
Comparison of symptom frequencies (by means of Fisher’s exact test) failed to reveal significant differences between patients and healthy individuals. The calculation of the daily differential value for the depressive patients indicates a mean improvement in well-being of approximately 7.95 + 12.8 points (VAMS) in the evening as compared to the morning, whereas the corresponding value of healthy subjects ( - 0.17 i 16.9) demonstrates that their state of well-being is hardly affected by the time of the day. This difference did not, however, reach significance (P = 0.101). Time estimation
The prospective time estimation error failed to reveal circadian variations in either the depressed patients or the healthy controls. There were no statistically significant differences between any time estimation measurements on either day in the two groups. Consequently the mean daily differential value of the time estimation error is negligible and does not differ between patients and controls (0.02 + 3.9 in the depressed group: - 1.30 k 4.2 in the control group). No relationship could be found between time estimation errors and state of well-being in the course of the day. The corresponding statistical calculations reveal that there is no correlation between the mean daily differential values (= diurnal variations) of both time estimation errors and VAMS scores (r = 0.13, NS in the depressed group and r = 0.14, NS in the control group). Even in the daily courses of those patients exhibiting ‘typical’ diurnal variations in mood like in the daily courses without ‘typical’ diurnal variations - daytime fluctuations in time estimation could not be detected (Fig. 1): an increasingly favourable self-assessment of well-being in the course of the day was not accompanied by a corresponding reduction in time estimation error. Again, in daily courses with typical diurnal mood variation there is no correlation between the mean daily differential values of time estimation and VAMS measurements (r = 0.16; NS). Discussion
The findings on the daily rhythm of the symptomatology confirm that daily courses with ‘typi-
cal’ diurnal variations are by no means the rule for endogenous depressive patients. In fact, they do not even occur in the majority of these patients. It was likewise confirmed that the diurnal course of depression symptomatology varies from day to day (Stallone, 1973; Tiille and Goetze, 1987). Even healthy individuals exhibit a daily course of well-being which quantitatively hardly differs from the daily fluctuations in depressive patients. It must, however. be taken into account that a comparative investigation into the well-being of depressive patients and healthy individuals is subject to fundamental difficulties. A healthy individual is capable of accepting, for example, low spirits in the morning as a natural occurrence which can be remedied by more various and effective means than are available to a depressive patient (compare Hampp, 1961). The assessment of well-being by means of a self-rating instrument (VAMS) is thus unable to reflect the various qualities of mood. All melancholies received antidepressant drug treatment while the control persons did not. Any effect of antidepressants on diurnal variations in mood could not, however, be determined in depressive patients (Tolle and Goetze, 1987). The reproducibility of the experimental time estimation findings at various times of the day serves above all to underline the reliability of such experiments in discriminating between depressive patients and healthy controls (Kuhs et al. , 1989a; see also Melges and Fougerousse, 1966). Time estimation as a quantitative measurement of disturbances in experiencing time is nevertheless barely subject to circadian fluctuations. The data do not therefore allow a chronopathological relationship to be established between time experience/estimation and daily variations in the state of well-being. The observed circadian fluctuations in well-being are not accompanied by diurnal changes in time estimation errors. Thus, according to our findings, time estimation investigations do not contribute towards an understanding of diurnal variations in melancholic symptomatology. This negative result can possibly be accounted for by the following explanations. (a) Time estimation is an insufficiently sensitive tool to understand subjective disorders of time experience. It has to be kept in mind that quanti-
289
fied time estimation experiments are only an indirect approach to the qualitative modification of time exhibited in melancholia. Nevertheless it is worth mentioning that in melancholia the amount of time estimation error parallels the intensity of depression as measured on the VAMS (Kuhs et al., 1989a). Thus time estimation can serve as an indicator of the feeling of ill-being in endogenous depressed patients. Lehmann (1967) makes a distinction between two qualitative modes of time experience, namely time awareness (whether time passes quickly or slowly) and time perspective, which concerns the fundamental relationship of man towards past, present and future. It is by no means inevitable that the different modes in which time can be experienced change in the same way. Indeed, Beth (1975) as well as Kitamura and Kumar (1982, 1984) in depressive patients observed a slowing down of time awareness but no peculiarities in time estimation. We therefore need to find other instruments suitable for illustrating the described disorders in time experience in melancholia. Within the framework of the inquiry in hand, simply to repeatedly question the patients about their awareness of time appears to hold little promise. (b) The second consideration arises from the finding that time estimation errors in melancholic patients show no recognisable diurnal variation and accordingly no parallelism to possible daily fluctuations in depressive symptomatology. It can therefore be concluded that the altered time experience in melancholia - possibly indicated by time estimation errors - does not contribute to an interpretation of diurnal variations in mood. If these findings could be confirmed and no further evidence of a specific diurnal variation in wellbeing in melancholia (endogenous depression) brought forward, the phenomenon of diurnal variations in mood would lose further significance (synopsis Tolle, 1989). The time estimation findings reveal that the attempt to establish a relationship at a phenomenological and clinical-experimental level of inquiry has only been partially successful (evidence of a constant underestimation of prospective time intervals in melancholies as opposed to healthy
individuals; Kuhs et al., 1989a). In all other spects, however, it remains unsatisfactory.
re-
References Beth. P. (1975) Depression: influence on time estimation and time experience. Acta Psychiatr. Stand. 51. 42-50. Engel, S.W. (1957) Die depressiven Wellen des Tages. Fortschr. Neural. Psychiatric 25, 342-354. Gebsattel. V.E. van (1928) Zeitbezogenes Zwangsdenken in der Melancholic Nervenarzt 1, 275-287. Graw. P., Hole, G. and Gastpar, M. (1980) Tagesschwankungen bei hospitalisierten Patienten wghrend der Depression and in gesunden Zeiten. Arch. Psychlatrie Nervenkr. 228. 329-339. Hamilton. M. (1960) A rating scale for depression. J. Neural. Neurosurg. Psychiatry 23, 56-62. Hampp, H. (1961) Die tagesrhythmischen Schwankungen der Stimmung und des Antriebs beim gesunden Menschen. Arch. Psychiatric Nervenkr. 201, 355-377. Kitamura. T. and Kumar, R. (1982) Time passes slowly for patients with depressive state. Acta Psychiatr. Stand. 62. 415-420. Kitamura, T. and Kumar, R. (1984) Controlled study on time reproduction of depressive patients. Psychopathology 17, 24-27. Kitanishi. K.. Graw, P. and Hole. G. (1982) Klinisch-experimentelle Untersuchungen zur Tagesschwankung bei depressiven Patienten. Schw. Arch. Neurol. Neurochir. Psychiatric 130, 245-258. Knorring. L. van. Perris, C. and Strandman. E. (1977) Diurnal variations in intensity of symptoms in patients of different diagnostic groups. Arch. Psychiatric Nervenkr. 224, 295-312. Kuhs, H., Hermann, W., Kammer, K. and Tiille. R. (1989) Time estimation and the experience of time in endogenous depression (melancholia). An experimental investigation. (In preparation.) Lehmann, H.E. (1967) Time and psychopathology. Ann. NY Acad. Sci. 138, 798-821. Luria, R.E. (1975) The validity and reliability of the visual analogue mood scale. J. Psychiatr. Res. 12. 51-57. Melges, F.T. and Fougerousse, C.E. (1966) Time sense, emotions and acute mental illness. J. Psychiatr. Res. 4. 127-140. Minkowski, E. (1923) Bleulers Schizoidie und Syntonie und das Zeiterleben. Z. Neurol. 82, 212-230. Stallone. F., Huba. G.J., Lawlor, W.G. and Fieve, R.R. (1973) Longitudinal studies of diurnal variations in depression: a sample of 643 patient days. Br. J. Psychiatry 123. 311-318. Straus. E. (1928) Das Zeiterleben in der endogenen Depression und in der psychopath&hen Verstimmung. Mschr. Neural. Psychiatric 68, 640-655. Taub, J.M. and Berger, R.J. (1974) Diurnal variations in mood as asserted by self-report and verbal content analysis. J. Psychiatr. Res. 10. 83-88.
Disorders,
285
17 (1989) 285-290
JAD 00658
The daily course of the symptomatology and the impaired time estimation in endogenous depression (melancholia) H. Kuhs, W. Hermann, Department
of Psychrutty,
K. Kammer
and R. Tiille
University of Miinster, D-4400 Miinster,
E R. G.
(Received 3 January 1989) (Revision received 20 March 1989) (Accepted 11 April 19X9)
Summary The importance of ‘typical’ diurnal variations in mood (depression worse in the morning) as a diagnostic criterion of endogenous depression has been challenged in previous investigations. Disturbance in time experience may contribute towards an understanding of diurnal variation in depressive symptomatology. To examine this hypothesis a series of time estimation experiments was conducted in the course of the day. Twenty-five endogenous depressive patients (according to ICD-9) and 12 healthy controls were asked to estimate prospectively a 30-s interval on two successive days at 7.30 a.m., 11.30 a.m., 3.30 p.m. and 7.30 p.m. Simultaneously the subjects assessed their state of well-being using a visual analogue mood scale. A circadian rhythm of time estimation errors could not be detected; even in daily courses with ‘typical’ diurnal variations an increasingly favourable self-assessment of well-being was not accompanied by a corresponding diurnal fluctuation of time estimation. These results cast additional doubts on the significance of ‘typical’ diurnal variations in depressive symptomatology.
Ke_v words:
Endogenous
depression;
Time estimation
Introduction ‘Typical’ diurnal variations in well-being with low spirits in the morning and an improvement in mood in the afternoon or evening have long been considered diagnostically usable symptoms of endogenous depression (melancholia). Recent inves-
Address for correspondence: Priv.-Doz. partment of Psychiatry, University of Schweitzer-Str. 11, D-4400 Miinster, F.R.G. 0165-0327/89/$03.50
Dr. H. Kuhs. DeMiinster, Albert-
Q 1989 Elsevier Science
Publishers
tigations (Graw et al., 1980; Von Knorring et al., 1977) however, have shown that previous findings need to be modified. Among other things the daily rhythm of depressive symptomatology is subject to considerable intra-individual variations from day to day (Stallone et al., 1973; Tolle and Goetze, 1987). But if it is at all possible to determine the existence of diurnal variation in depressive symptomatology, then the variation found is predominantly ‘typical’ (i.e., depression worse in the morning; see Tiille. 1989). In healthy persons diurnal variations in mood are also well-known (Engel, 1957; Hampp, 1961;
B.V. (Biomedical
Division)
286
Taub and Berger, 1974). The way in which healthy subjects and melancholies experience the course of the day does. however, reveal fundamental differences: whereas a healthy person plans his day looking ahead, the melancholic regards his day as being endlessly drawn out and insurmountable. The only source of relief for the patient is that time will pass and with it the sufferings of the day. Diurnal variations in melancholia are characterised by a special experiential quality of the low spirits as well as of the improvement in mood, which the patient passively undergoes rather than actively initiates. However. no conclusive empirical verification exists at present for either this melancholic change in experience or for its diurnal fluctuations. The theoretical background for the investigation may be found in a phenomenological aspect whereby changes in time experience are expounded as being fundamental disorders in melancholia (Minkowsky, 1923; von Gebsattel, 1928; Straus, 1928). These changes particularly concern a slowing down of the inner development period (ego time) compared to changes in the surrounding world (universal time) (Straus, 1928) and appear to be in agreement with experimental results: time estimation investigations have revealed characteristic mistakes in judgement by melancholic patients. According to our own findings melancholies prospectively underestimate an interval of 30 s by almost 6 s whereas healthy subjects overestimate the same interval by more than 10 s (Kuhs et al., 1989a). Since time estimation may serve as a quantitative measurement of the disturbance in time experience in melancholia it may be assumed that the repeated ascertainment of time estimation errors in the course of the day can provide a valuable contribution towards understanding the diurnal variation in depressive symptomatology. So far, no time estimation measurements have been carried out in melancholic patients in relation to the time of day. Prospective time estimation experiments conducted by Wever (1973) established that only one healthy person from his study group made substantial errors in time judgements before noon. These judgement errors decreased as the day progressed and reached a minimum towards midnight.
The main questions addressed by this study are as follows: (1) does prospective time estimation reveal daytime fluctuations and (2) what relationship exists between time estimation and the state of well-being in the course of the day. Methods Patients and control group Twenty-five in-patients (10 males, 15 females) treated for endogenous depression (diagnosed according to ICD-9) were admitted to the study. Severity of depression was measured on the Hamilton Rating Scale for Depression (Hamilton, 1960). The mean score was 19.1 + 5.7, the minimum score was 10. The patients were 50.1 _t 11.8 years old. At the time of the investigation all depressive patients received drug therapy with antidepressants. Seventeen of them had undergone sleep deprivation therapy, which had stopped at least 5 days before the investigation began. The patients were compared with a healthy control group (n = 12) only slightly different from the patient group as regards age distribution and sex (five males, seven females, aged 46.8 f 18.0 years). Procedure and time of the investigation In accord with the results of our previous study (Kuhs et al., 1989) the following procedure proved to be especially suitable for determining prospective time estimation errors: the subject prospectively estimates an interval of 30 s by quietly counting up to 10 three times in succession. The number 10 is pronounced loudly each time it is arrived at. The time is registered by the investigator. The estimation error is determined as follows: time estimation error equals objective estimation interval minus subject’s estimation value. Hence, a positive differential value signifies underestimation, a negative differential value overestimation of time. The time estimation experiments were carried out on two successive days at 7.30 a.m., 11.30 a.m., 3.30 p.m. and 7.30 p.m., in each case together with a self-assessment of well-being. The patients and control subjects assessed their state of well-being using a visual analogue mood scale (VAMS) (Luria, 1975). The ends of this
287
100-mm scale read as follows: ‘I have never been as depressed as I am now’ (0) and ‘I am not depressed at the moment’ (100). Diurnal variations in well-being can be calculated from the VAMS scores as follows: daily differential value equals evening value minus morning value: 3.30 p.m. value + 7.30 p.m. value 2
The following statistical evaluation procedures were employed: Wilcoxon’s test for matched pairs. Spearman’s rank correlation coefficient (r) and Fisher’s exact test. Results
-
Daiij rhythm ofsymptomatoIogy/state of well-being On two successive days ‘typical’ daily variations in mood were ascertained in six of the 25 melancholic patients (24.0%) and in two of the healthy controls (16.7%). On only one of the two days ‘typical’ daily variations in well-being were found in eight depressive patients and in two healthy controls. No ‘typical’ diurnal variations on either of the investigation days were evident in 11 depressive patients (44.0%) and eight control subjects (66.7%).
- 7.30 a.m. value + 11.30 a.m. value 2 Diurnal variation was assumed to be present if the daily differential value was at least 10 points and/or 20% on the VAMS. Insignificant differences were not regarded as daily fluctuations. The daily differential value of time estimation errors was calculated in the same way.
VAMS
TEE Daily courses with tctypical)) diurnal va riations in mood (N=20)
804
1st
70-
14 -
60-
12 -
50-
lo-
40-
8-
30-
6-
20-
4-
lo- ’
2-
I I
I
I
7.30 am.
3.30 pm. 11.30 am.
Fig. 1. State of well-being in mood.
Daily courses without cctypicab diurnal variations in mood ( N= 30)
-
and time estimation VAMS
score;
11.30 a.m
7.30p.m. errors
- - - - - - TEE
3.30p.m
7.30a.m
in endogenous (time estimation
depressive error):
patients
with and without
underestimation
Z 30p.m ‘typical’
of a 30-s interval
diurnal
in seconds.
variations
288
Comparison of symptom frequencies (by means of Fisher’s exact test) failed to reveal significant differences between patients and healthy individuals. The calculation of the daily differential value for the depressive patients indicates a mean improvement in well-being of approximately 7.95 + 12.8 points (VAMS) in the evening as compared to the morning, whereas the corresponding value of healthy subjects ( - 0.17 i 16.9) demonstrates that their state of well-being is hardly affected by the time of the day. This difference did not, however, reach significance (P = 0.101). Time estimation
The prospective time estimation error failed to reveal circadian variations in either the depressed patients or the healthy controls. There were no statistically significant differences between any time estimation measurements on either day in the two groups. Consequently the mean daily differential value of the time estimation error is negligible and does not differ between patients and controls (0.02 + 3.9 in the depressed group: - 1.30 k 4.2 in the control group). No relationship could be found between time estimation errors and state of well-being in the course of the day. The corresponding statistical calculations reveal that there is no correlation between the mean daily differential values (= diurnal variations) of both time estimation errors and VAMS scores (r = 0.13, NS in the depressed group and r = 0.14, NS in the control group). Even in the daily courses of those patients exhibiting ‘typical’ diurnal variations in mood like in the daily courses without ‘typical’ diurnal variations - daytime fluctuations in time estimation could not be detected (Fig. 1): an increasingly favourable self-assessment of well-being in the course of the day was not accompanied by a corresponding reduction in time estimation error. Again, in daily courses with typical diurnal mood variation there is no correlation between the mean daily differential values of time estimation and VAMS measurements (r = 0.16; NS). Discussion
The findings on the daily rhythm of the symptomatology confirm that daily courses with ‘typi-
cal’ diurnal variations are by no means the rule for endogenous depressive patients. In fact, they do not even occur in the majority of these patients. It was likewise confirmed that the diurnal course of depression symptomatology varies from day to day (Stallone, 1973; Tiille and Goetze, 1987). Even healthy individuals exhibit a daily course of well-being which quantitatively hardly differs from the daily fluctuations in depressive patients. It must, however. be taken into account that a comparative investigation into the well-being of depressive patients and healthy individuals is subject to fundamental difficulties. A healthy individual is capable of accepting, for example, low spirits in the morning as a natural occurrence which can be remedied by more various and effective means than are available to a depressive patient (compare Hampp, 1961). The assessment of well-being by means of a self-rating instrument (VAMS) is thus unable to reflect the various qualities of mood. All melancholies received antidepressant drug treatment while the control persons did not. Any effect of antidepressants on diurnal variations in mood could not, however, be determined in depressive patients (Tolle and Goetze, 1987). The reproducibility of the experimental time estimation findings at various times of the day serves above all to underline the reliability of such experiments in discriminating between depressive patients and healthy controls (Kuhs et al. , 1989a; see also Melges and Fougerousse, 1966). Time estimation as a quantitative measurement of disturbances in experiencing time is nevertheless barely subject to circadian fluctuations. The data do not therefore allow a chronopathological relationship to be established between time experience/estimation and daily variations in the state of well-being. The observed circadian fluctuations in well-being are not accompanied by diurnal changes in time estimation errors. Thus, according to our findings, time estimation investigations do not contribute towards an understanding of diurnal variations in melancholic symptomatology. This negative result can possibly be accounted for by the following explanations. (a) Time estimation is an insufficiently sensitive tool to understand subjective disorders of time experience. It has to be kept in mind that quanti-
289
fied time estimation experiments are only an indirect approach to the qualitative modification of time exhibited in melancholia. Nevertheless it is worth mentioning that in melancholia the amount of time estimation error parallels the intensity of depression as measured on the VAMS (Kuhs et al., 1989a). Thus time estimation can serve as an indicator of the feeling of ill-being in endogenous depressed patients. Lehmann (1967) makes a distinction between two qualitative modes of time experience, namely time awareness (whether time passes quickly or slowly) and time perspective, which concerns the fundamental relationship of man towards past, present and future. It is by no means inevitable that the different modes in which time can be experienced change in the same way. Indeed, Beth (1975) as well as Kitamura and Kumar (1982, 1984) in depressive patients observed a slowing down of time awareness but no peculiarities in time estimation. We therefore need to find other instruments suitable for illustrating the described disorders in time experience in melancholia. Within the framework of the inquiry in hand, simply to repeatedly question the patients about their awareness of time appears to hold little promise. (b) The second consideration arises from the finding that time estimation errors in melancholic patients show no recognisable diurnal variation and accordingly no parallelism to possible daily fluctuations in depressive symptomatology. It can therefore be concluded that the altered time experience in melancholia - possibly indicated by time estimation errors - does not contribute to an interpretation of diurnal variations in mood. If these findings could be confirmed and no further evidence of a specific diurnal variation in wellbeing in melancholia (endogenous depression) brought forward, the phenomenon of diurnal variations in mood would lose further significance (synopsis Tolle, 1989). The time estimation findings reveal that the attempt to establish a relationship at a phenomenological and clinical-experimental level of inquiry has only been partially successful (evidence of a constant underestimation of prospective time intervals in melancholies as opposed to healthy
individuals; Kuhs et al., 1989a). In all other spects, however, it remains unsatisfactory.
re-
References Beth. P. (1975) Depression: influence on time estimation and time experience. Acta Psychiatr. Stand. 51. 42-50. Engel, S.W. (1957) Die depressiven Wellen des Tages. Fortschr. Neural. Psychiatric 25, 342-354. Gebsattel. V.E. van (1928) Zeitbezogenes Zwangsdenken in der Melancholic Nervenarzt 1, 275-287. Graw. P., Hole, G. and Gastpar, M. (1980) Tagesschwankungen bei hospitalisierten Patienten wghrend der Depression and in gesunden Zeiten. Arch. Psychlatrie Nervenkr. 228. 329-339. Hamilton. M. (1960) A rating scale for depression. J. Neural. Neurosurg. Psychiatry 23, 56-62. Hampp, H. (1961) Die tagesrhythmischen Schwankungen der Stimmung und des Antriebs beim gesunden Menschen. Arch. Psychiatric Nervenkr. 201, 355-377. Kitamura. T. and Kumar, R. (1982) Time passes slowly for patients with depressive state. Acta Psychiatr. Stand. 62. 415-420. Kitamura, T. and Kumar, R. (1984) Controlled study on time reproduction of depressive patients. Psychopathology 17, 24-27. Kitanishi. K.. Graw, P. and Hole. G. (1982) Klinisch-experimentelle Untersuchungen zur Tagesschwankung bei depressiven Patienten. Schw. Arch. Neurol. Neurochir. Psychiatric 130, 245-258. Knorring. L. van. Perris, C. and Strandman. E. (1977) Diurnal variations in intensity of symptoms in patients of different diagnostic groups. Arch. Psychiatric Nervenkr. 224, 295-312. Kuhs, H., Hermann, W., Kammer, K. and Tiille. R. (1989) Time estimation and the experience of time in endogenous depression (melancholia). An experimental investigation. (In preparation.) Lehmann, H.E. (1967) Time and psychopathology. Ann. NY Acad. Sci. 138, 798-821. Luria, R.E. (1975) The validity and reliability of the visual analogue mood scale. J. Psychiatr. Res. 12. 51-57. Melges, F.T. and Fougerousse, C.E. (1966) Time sense, emotions and acute mental illness. J. Psychiatr. Res. 4. 127-140. Minkowski, E. (1923) Bleulers Schizoidie und Syntonie und das Zeiterleben. Z. Neurol. 82, 212-230. Stallone. F., Huba. G.J., Lawlor, W.G. and Fieve, R.R. (1973) Longitudinal studies of diurnal variations in depression: a sample of 643 patient days. Br. J. Psychiatry 123. 311-318. Straus. E. (1928) Das Zeiterleben in der endogenen Depression und in der psychopath&hen Verstimmung. Mschr. Neural. Psychiatric 68, 640-655. Taub, J.M. and Berger, R.J. (1974) Diurnal variations in mood as asserted by self-report and verbal content analysis. J. Psychiatr. Res. 10. 83-88.