go to content


Vadim S. Rotenberg, e-mail: vadir@post.tau.ac.il A. Cholostoy , P. Gurwitz, E. Shamir (Tel Aviv)

Dynamische Psychiatrie / Dynamic Psychiatry vol 41 2008(5-6) pp297-307

Complaints on sleep disorders are common in depression and usually there is no a direct relationship between these complaints and objective sleep variables. The task of the present investigation was to check whether behavioral attitudes in depression are related to their subjective sleep estimation. Subjects. 44 inpatients with major depression and 62 healthy subjects. Methods: BASE test - projective questionnaire for the estimation of the following behavioral attitudes: search activity (SA), stereotyped (St), chaotic (Ch) and passive (Pa) behavior; sleep questionnaire; 21-item HAMILTON Rating Scale for depression.

Keywords: sleep estimation, depression, behavioral attitudes, BASE


In our previous investigation of sleep estimation in depression (ROTENBERG, CHOLOSTOY, MARK 2003) we have found that in addition to the well-known general complaints on the difficulty to start sleeping, increased number of awakenings, and unpleasant dreams (GREENBERG 1977; HALL, BIJYSSE, NOWELL et al. 2000; HEALTH, EAVES, KIRK, MARTIN 1998; LEMKE, PUHL, BRODERRICK 1999; MCCALL, REBOUSSIN, COHEN 2000) depressed patients complain also of the increase in sleep latency, in the number of awakenings (more often in men) and in the number of dreams just after a negative emotional experience during the previous day. These complaints were more prominent than in healthy control subjects. At the same time, according to the most previous investigations (PASTERNAK, REYNOLDS, HOCH et al. 1992; WEISS, MEPART-LAND, KUPFER 1973; ROTENBERG, INDURSKY, KAYUMOV et al. 2000; Tsucm-YAMA, NAGAYAMA, KUDO et al. 2003) there are no direct relationships between the subjective estimation of sleep variables like sleep duration, sleep delay and the number of awakenings and the corresponding objective variables. There are some signs (TSUCHIYAMA, NAGAYAMA, KUDO et al. 2003) that the subjective sleep estimation is partly determined by the severity of depression. It can also be one of the reasons for the discrepancy between subjective and objective sleep variables by taking into consideration that this discrepancy displays itself in both directions: depressed patients not only underestimate sleep duration but often overestimate it (ROTENBERG, INDURSKY, KAYUMOV et al. 2000; TSUCHIYAMA, NAGAYAMA, KUDO et al. 2003). Thus the elucidation of the nature of sleep complaints and its relationships to the emotional reactions on events in wakefulness require search for additional (psychological or behavioral) factors that may determine such relationships.

Search activity concept (ROTENBERG 1984; 1993) differentiates four types of behavioral reactions on stressful and problematic situations usually accompanied by negative emotional experience: a) search activity (SA) which is designed to change the situation or the subject's attitude to it, with uncertainty regarding the results of this activity but with constant monitoring of the results at all stages of activity, Active estimation of the situation in order to choose a relevant strategy of behavior is also a part of search activity; b) stereotyped behavior (St) which uses habitual skills and algorithms with highly predictable results; c) chaotic or panicky behavior (Ch) which may seem to imitate SA but does not include the monitoring of the results of the activity and for this reason cannot use the results in order to improve the direction of activity. It is fraught with inadequate actions and often finally leads to renunciation of search; d) Renunciation of search (passive behavior, Pa) manifesting itself in giving up, helplessness and freezing.

Both SA and St belong to the adaptive goal-oriented forms of behavior. However while St is adaptive in the routine situations SA is adaptive in stressful and highly unpredictable situations as well as in the creative process of finding new solutions of old problems.

Our investigations performed on animals (ROTENBERG, ARSHAVSKY 1979; ROTENBERG 1984) and on humans (ROTENBERG, KOROSTELEVA 1990) have shown that SA (search activity) increases body resistance and prevents somatic disorders in stressful conditions while Ch (chaotic activity) and especially Pa (passive behavior) predisposes subjects to such disorders and St at least does not protect them from it.

Types of behavior have strong relations to the sleep stage functions and consequently to the sleep structure (ROTENBERG 1984; 1993). Prominent SA in wakefulness decreases the subject's need in the subsequent sleep, especially strong it decreases the requirement in REM sleep (usually accompanied by dreams in healthy subjects), while forms of behavior that do not contain SA, and especially Pa and Ch, increases the requirement in REM sleep. REM sleep and dreams, if they are functionally sufficient, include the particular search activity that compensates the SA deficiency in the previous wakefulness and restores SA in the subsequent wakefulness. Thus behavioral attitudes determine the alterations of sleep in front of emotionally meaningful events during wakefulness.

In the pilot investigation performed on the 20 Russian-speaking newcomers in Israel suffering from the major depression (ROTENBERG, CHOLOSTOY 2004) we have found that depressed patients differ from the normal control group in their behavioral attitudes: their SA was significantly lower and Pa significantly higher.

The task of the present investigation was to check whether the interrelationships between the behavioral attitudes and subjective sleep estimations are different in depressed subjects and health control and whether the alterations of behavioral attitudes in depression may relate to their sleep complaints. In addition we have checked whether the level of depression relates to sleep complaints.


We have investigated 44 patients with major depression (21 men and 23 women, mean age 44.8 years) and 62 healthy subjects (27 men and 35 women, mean age 35.2 years). Depressed patients have been hospitalized in Abarbanel Mental Health Center. The study was approved by the Institutional Review Board and the Israeli Ministry of Health Committee for Studies in Human Subjects. Informed consent was obtained from all participants following an explanation of the nature of the study. Diagnosis was established according to DSM-IV criteria on the basis of the structured clinical interview. There were no non-psychiatric medical conditions responsible for their depression. In a clinical interview we have excluded other diagnoses associated with altered sleep, e. g. alcoholism, sleep apnea or myoclonus. All patients were treated with SSRI partly combined with mood stabilizers and benzodiazepines. The mean duration of illness was 14.2, SD - 11.6 years. The mean number of previous hospitalizations was 2.4, SD = 2.2.

Healthy subjects had neither psychiatric nor somatic diseases and had no active complaints on sleep.


Patients were asked to complete the Behavioral Attitudes and Search Evaluation test (BASE). This test combines principles of both projective technique and a personally questionnaire (VENGER, ROTENBERG, DESIATNIKOV 1996) and was validated in our previous investigations (ROTENBERG, KUTSAY, VENGER 1998; 2001; ROTENBERG, ZUSMAN 2002). It includes descriptions of 16 open situations with four possible reactions on each of them. The tested subject has to choose two reactions (two types of behavior) in each situation: the most appropriate from his/her point of view (scored by a '+' sign ) and the least appropriate (scored by a '-' sign ). The preferences of the subject provide an opportunity for quantitative measurements of each of the above- mentioned behavioral attitudes: SA (search activity), St (stereotyped behavior), Ch (chaotic behavior), Pa (passive behavior). The total score of the four scales can range from-16 to+16.

All reactions on test situations seem equally acceptable both ethically and pragmatically. Here is an example to illustrate the general principle:

"A group of hikers went to a cavern unfamiliar to them. Just as they arrived inside the cavern, a landslide buried the exit. While discussing the situation, the following suggestions arose:

A. I suggest searching for another exit. We'll mark our way with small mounds of stones so that we might recognize the place we would once pass even by touch when our flashes are out.

B. Of course, we must search for another exit! But we cannot waste our time for excessive precautions. Remember that we do not have food.

C. Let us wait till our friends find us. If we economize our strength, so our poor provisions will be enough to maintain our life while waiting.

D. I suggest trying to dig a passage through the landscape. This can take a lot of time, but at least we know that the exit is here and don't know whether there is another exit elsewhere."

In this situation, the answer 'A' corresponds to search activity because it suggests activity with unpredictable outcome, and fixation of all intermediate results (marking the way with stones). The answer 'B' indicates chaotic tendency because it denies fixation of the intermediate results and emphasizes the urgency of the required activity. The answer 'C reflects passive attitudes because it does not suggest any types of activity. The answer 'D' shows a tendency to stereotyped behavior: high level of activity in the same direction without considering possible obstacles.

The normal configuration of BASE in healthy subjects is characterized by positive meanings of SA and St and negative meanings of Ch and Pa.

The level of depression was measured using the 21-item HAMILTON Rating Scale for depression (HAMILTON 1960). This test was applied to the depressed patients on the same day as BASE.

In the present investigation we have concentrated on the following questions of sleep questionnaire used in our previous investigation (ROTENBERG, CHOLOSTOY, MARK 2003):

1. Do you like to watch dreams during your night sleep: a. yes; b. no.

2. After the negative emotional experience during the day the duration of your night sleep became: a. longer; b. shorter; c. is the same as usually

3. After the positive emotional experience during the day the duration of your night sleep became: a. longer; b. shorter, c. is the same as usually

4. After the negative emotional experience during the day the number of your night sleep dreams: a. increases; b. decreases, c. is the same as usually

5. After the positive emotional experience during the day the number of your night sleep dreams: a. increases; b. decreases; c. is the same as usually

6. On the next night after the sleepless night you are sleeping: a. better than usual; b. worse than usual; c. sleep is the same as usually.

7. Recently you awoke from your night sleep: a. once during the night; b. twice during the night; c. three time during the nigh; d. more; e. have no awakenings.

8. You are in general quite satisfied with your sleep during the recent month; a. yes; b. no

Statistic: we have used t-Test: Two Sample Assuming Unequal Variances. Only significant data are presented and discussed.


Healthy subjects:

1. In those who report the increase of the number of dreams after the negative emotional experience during the day (N=20) in comparison to those who report the decrease of the number of dreams (N=3) SA is higher (2.9, SD=3.04 vs. 0.33 SD=1.52, p=0.03); St is lower (0.45, SD=2.94 vs. 3.33, SD=1.14, p=0.008) and Ch is also lower (-1.2, SD=3.18 vs. 0.66, SD=0.33, p=0.01).

2. In those who report the increase of sleep duration after the positive emotional experience during the day (N=9) in comparison to those who complain on sleep reduction (N=12) Ch is higher (0.66, SD=2.44 vs. -1.75, SD=3.78, p=0.04).

Depressed patients:

1. In those patients (N=6) who have been satisfied with their night sleep the mean level of St was significantly higher than in patients (n=38) who were not satisfied with their sleep (3.16, SD=2.63 vs. 0.6, SD=3.8, p=0.03).

2. 33 patients with major depression demonstrated the reversed configuration of behavioral attitudes (group 1) and 11 patients displayed the normal configuration (group 2). The level of depression according to the 21-items HAMILTON Rating Scale was significantly higher in group 1 (27.78, SD=4.08 vs. 24.6, SD=3.34, p=0.009). The percentage of patients totally unsatisfied with their sleep was significantly higher in group 1 (73% vs. 18%, p<0.02). However, the dissatisfaction with sleep was not directly related to the level of depression: patients who are unsatisfied with their night sleep (independently of BASE configuration) do not display the higher level of depression in comparison to the small group of patients who are satisfied with it.

3. Those patients who like to watch dreams (N=14) in comparison to those who do not like it (N=30) display a higher SA (1, SD=2.44, vs. -0.56, SD=3.31, p=0.045), a lower Pa (-0.57, SD=3.47 vs. 1.53, SD=4.06, p=0.043) and a higher level of the HAMILTON Rating Scale (28.85, SD=1.6 vs. 26.13, SD=3.67,p=0.03).

4. In those patients who mentioned the increase of dream number after the positive experience (N=8) in comparison to those who report the decrease of dreams (N=12) St is higher (2.62, SD=3.06 vs. -0.16, SD=2.16, p=0.02). The level of the HAMILTON Rating scale correlated with the increase of the number of dreams after the positive experience during the day (0.322, p=0.038).


The limitation of the present study is that it is dealing only with the subjective estimation of sleep variables. However the role of this estimation in complaints of depressed patients as well as the abovementioned absence of the correlations between the subjective and objective variables suggest that we can consider it independently of objective sleep variables.

One preliminary statement has to be taken into consideration. The alteration of sleep variables may be not only a sign of sleep deterioration caused by disease or stress but also a sign of the compensatory reaction related to the role of sleep in adaptation.

Our findings in healthy subjects are in general in agreement with the main statements of the search activity concept that suggests the adaptive and compensatory role of sleep dreams in the restoration of search activity (ROTENBERG 1984; 1993). According to this theory, the negative emotional experience during the day requires the intensification of dream activity for its neutralization and for the restoration of SA. Thus the relatively higher level of SA and SA/St ratio in the majority of healthy subjects who report the increase of dreams after the negative emotional experience, in comparison to those few who report the decrease of dreams may reflect the preserved compensatory function of sleep dreams. This interpretation is also in agreement with the discovery of HARTMANN (1973) that the so-called long sleepers are characterized by the increased emotional sensitivity and vulnerability regularly compensated by the increased REM sleep duration (accompanied in healthy subjects by dreams) and it prevents the disturbance of the emotional balance.

Chaotic activity destabilizes waking goal-directed behavior and makes it maladaptive, especially in women (ROTENBERG, ZUSMAN, CHOLOSTOY, BA-RUCH 2006). As we have shown (ROTENBERG, ZUSMAN 2002) in stressful conditions Ch negatively correlates with the tendency to overcome obstacles on the Rosenzweig Picture Frustration Test (ROSENZWEIG 1935). Ch correlated also negatively with the performance of Raven test while Raven performance correlated positively with SA and with the fixation on self-defense (ROTENBERG, ZUSMAN, CHOLOSTOY 2006). Negative emotional experience during the day may exaggerate chaotic activity in those who are predisposed to such behavior and cause heavy sleep disturbances accompanied by dream reduction. In healthy subjects with a relatively low level of Ch, negative emotional experience may increase the requirement in sleep dreams and this requirement can be realized.

At the same time, positive emotional experience during the day may temporarily decrease the negative outcome of chaotic activity on sleep and increase sleep duration if sleep was initially suppressed due to emotional tension related to chaotic activity. On the other hand, in subjects with the relatively lower level of chaotic activity and consequently without initial sleep disturbances the positive emotional experience during the day may temporarily reduce the need of sleep dreams and cause sleep reduction that is not accompanied by negative outcome on psychic state.

A positive emotional excitement during the day is often accompanied by sleep reduction. It was shown that expectation of success, praise and positive emotional state entail a drop in the REM sleep proportion in the night sleep in healthy subjects; at the same time REM sleep percentage increases with the predomination of negative forecast and fixation on past failures as well as with the relations of rivalry between the motive of achievement of success and the motive of the avoidance of failures that causes disorganizing emotional tension * and blocks search activity ( s. ROTENBERG, KOROSTELEVA 1990).

A tendency of SA/St ratio to be higher in those subjects who are satisfied with their sleep in comparison to those who are not satisfied corresponds also to the essence of the search activity concept. It can reflect the bi-directional relationships: on one hand subjects with the domination of the most adaptive style of waking behavior are able to preserve their sleep. On the other hand normal sleep, that subjects are satisfied with, helps to maintain search activity.

In depressed patients the general dissatisfaction with night sleep is determined by the abnormal configuration of behavioral attitudes: SA and/or St (adaptive behavioral attitudes) have negative meanings while Ch and/or Pa (maladaptive behavioral attitudes) have positive meanings. It corresponds also with data that patients dissatisfied with their night sleep display lower St than patients satisfied with their sleep. Our investigations show that the configuration of behavioral attitudes is more important for the subjective estimation of sleep than the level of depression.

In this context it is interesting that St is higher in patients who display the increase of dreams after the positive emotional experience during the day. It is possible to suggest that St as an adaptive behavior with definite probability forecast of the outcome of behavior has at least moderate stabilizing effect on night sleep. For this reason it increases the satisfaction with sleep, as it was already mentioned. It also provides an opportunity to satisfy the requirement in dreams that is high in depressed patients. It corresponds to the increase of dream experiences in patients in the process of successful psychotherapeutic treatment (BURBIEL, BOTT, FINKE 1982; MELSTROM, CARTWRIGHT 1983). The increased requirement in dreams in depression is confirmed by weak but significant positive correlation between the HAMILTON Rating Scale and increased number of dreams.

This proposition also does not contradict data of the difference between patients who like and do not like to watch dreams. Those who like to watch dreams (independently of the emotional experience during the day) display a relatively higher SA and a relatively lower Pa. It may be interpreted as a sign that dreams are at least partly performing their functional task. In our previous investigation performed on depressed patients (ROTENBERG, INDURSKY 1997) we have shown that sometimes after night sleep (in 20% of all night polysomnography investigations) mood was estimated by patients as being better in the morning than in the evening, which is very atypical data for depression. In such cases eye movements (EM) density in REM sleep increased from the first to the fourth cycle just like in healthy subjects while in all other nights EM density was higher in the first than in the subsequent cycles. Mood improvement correlated positively with EM density in the fourth cycle. We have suggested that in these relatively rare cases REM sleep performs its function sufficiently.

However, it is necessary to take into consideration that in comparison to the majority of healthy subjects SA was not high enough and Pa was not low enough in these patients. It can explain the seemingly paradoxical higher level of the HAMILTON Rating Scale in those who like to watch dreams. The combination of the relatively increased level of depression with the moderate SA may explain, on the one hand, the high requirement in dreams, and on the other hand may be a sign of the relative functional sufficiency of dreams at least in some nights that can explain why these patients like to watch dreams.

The results of the present investigation on healthy subjects and on depressed patients in general confirms the proposition of search activity concept and shows that behavioral attitudes play an important role in the subjective estimation of sleep.


Burbiel, I.; Bott, Ch.; Finke, G. (1982) Wissenschaftstheoretische Grundlagen und Forschungsmethodik der Dynamischen Psychiatric In: G. Ammon (Hrsg.): Hdb. Dynam. Psychiat; Bd. 2, S. 531-573. Miinchen, Basel: Reihardt.

Greenberg, R. (1977) On understanding sleep disorders and their psychopathology. McLean Hosp. J. II, 3:139-146.

Hall, M.; Buysse, D. J.; Nowell, P. D.; Nofzinger, E. A.; Houck, P. et al. (2000): Symptoms of stress and depression as correlates of sleep in primary insomnia. Psychosomatic medicine 62:227-230.

Hamilton, M. (1960): A rating scale of depression. J. Neurol. Neurosurg. Psychiatry 23:56-62.

Hartmann, E. L. (1973): The Functions of Sleep. New Haven, London: Yale University Press.

Health, A. C; Eaves, L. J.; Kirk, K. M.; Martin, N. G. (1998): Effects of life-style, personality, symptoms of anxiety and depression, and genetic predisposition on subjective sleep disturbance and sleep patterns. Twin Res. 1:176-188.

Lemke, M. R.; Puhl, P.; Broderrick, A. (1999): Motor activity and perception of sleep in depressed patients. J. Psychiatr. Res. 33:215-224.

McCall, W. V.; Reboussin, B. A.; Cohen, W. (2000): Subjective measurement of insomnia and quality of life in depressed inpatients. J. Sleep Res. 9:43-48.

Melstrom, M. A.; Cartwright R. D. (1983): Effects of successful vs. unsuccessful psychotherapy outcome on some dream dimensions. Psychiatry 46:51-65.

Pasternak, R. E.; Reynolds, C. F.; Hoch, C. C. et al. (1992): Sleep in spousally bereaved elders with subsyndromal depressive symptoms. Psychiatry Res. 43:43-53.

Rosenzweig, S. (1935): A test for types of reaction to frustration. Am J of Orthopsychiatry 5:395-403.

Rotenberg, V. S. (1984): Search activity in the context of psychosomatic disturbances, of brain monoamines and REM sleep functions. Pavlovian J. Biological Sciences 19:1-15.

------(1993): REM sleep and dreams as mechanisms of search activity recovery. In: A. Kramer, R.

Hoffmann (Eds.) Functions of Dreaming; pp. 261-292. Albany: State Univ. of NY Press.

Rotenberg, V. S.; Arshavsky, V. V. (1979): Search activity and its impact on experimental and clinical pathology. Activitasa Nervosa Superior 21:105-115.

Rotenberg, V. S.; Cholostoy, A. (2004): Behavioral attitudes in major depression: A pilot investigation. Homeostasis 43:16-18.

Rotenberg, V. S.; Cholostoy, A.; Mark, M. (2003): Sleep estimation in depressed men and women. Homeostasis 42:13-16.

Rotenberg, V. S.; Indursky, P. (1997): The change of mood during sleep and REM sleep variables. Internat. J. of Psychiatry in Clinical Practice 2:47-52.

Rotenberg, V. S.; Indursky, P.; Kayumov, L.; Sirota, P.; Melamed, Y. (2000): The relationship between subjective sleep estimation and objective sleep variables in depressed patients. Intern. J. Psychophy-siolog 37:291-297.

Rotenberg, V. S.; Korosteleva, I. S. (1990): Psychological aspects of the search activity and learned helplessness in psychosomatic patients and healthy testees. Dynam. Psychiat. 23:1-13.

Rotenberg, V. S., Kutsay, S.; Venger, A. (1998) Behavioral attitudes and distress in adolescents: Relationship to age and gender. Homeostasis 39:57-64.

------(2001) Behavioural attitudes and the level of distress in the process of adaptation to the new society. Stress and Health 17:187-193.

Rotenberg, V. S., Zusman, Y. (2002): The dynamics of behavior attitudes in the process of group therapy. Dynam. Psychiat. 35:390-400.

Rotenberg, V. S.; Zusman, Y.; Cholostoy, A.; Baruch, Y. (2006): Behavioral attitudes and the duration of search for job in unemployed women. Homeostasis 44:97-103.

Tsuchiyama, K.; Nagayama, H.; Kudo, K.; Kojima, K.; Yamada, K. (2003): Discrepancy between subjective and objective sleep in patients with depression. Psychiatry and Clinical Neurosciences 57:259-264.

Venger, A.; Rotenberg V. S.; Desiatnikov, Y. (1996): Evaluation of search activity and other behavioral attitudes in indefinite situation. Dynam. Psychiat. 29:368-377.

Weiss, B. L.; Mepartland, R. J.; Kupfer, D. J. (1973): Once more: the inaccuracy of non-EEG estimation of sleep. American J. Psychiatry 130:1282-1285.