Women's dreams reported during first pregnancy


  • Yaron Dagan MD, DS ,

    1. Institute for Fatigue and Sleep Medicine, ‘Sheba’ Medical Center, affiliated to the Tel-Aviv University Medical School, Israel
    Search for more papers by this author
  • Anat Lapidot MA ,

    1. Institute for Fatigue and Sleep Medicine, ‘Sheba’ Medical Center, affiliated to the Tel-Aviv University Medical School, Israel
    Search for more papers by this author
  • Michal Eisenstein MA

    1. Institute for Fatigue and Sleep Medicine, ‘Sheba’ Medical Center, affiliated to the Tel-Aviv University Medical School, Israel
    Search for more papers by this author

Correspondence address: YaronDagan Institute for Fatigue and Sleep Medicine, The Chaim Sheba Medical Center, Tel-Hashomer 52621 Israel. Email: ydagan@post.tau.ac.il


Abstract This study analysed women's dreams reported during first pregnancy, a subject matter located at the crossroads of the psychology of dreams and the psychology of pregnancy. In the comparison of dreams reported by first-time pregnant women, to those reported by controls, we hypothesized that pregnant women's dreams would: (i) include more pregnancy-related content; (ii) display a higher degree of anxiety; and (iii) rate higher on a primary-process thinking (PPT) scale. As predicted, it was found that pregnancy-related contents significantly occupied pregnant women's dreams, a fact that might be attributed to an attempt to process and master the experience. Contrary to our expectations, it was found that anxiety and PPT were not significantly higher among pregnant women. An attempt to account for these findings raised methodological, as well as theoretical issues, consequently leading to a re-examination of the original hypotheses. Thus, it was claimed that the linkage of pregnancy to increased anxiety and PPT is grossly unbalanced.


The most salient approach currently treating the psychology of pregnancy is the developmental approach. According to Bibring,1 pregnancy is viewed as a developmental crisis: a critical stage characterized by a fast pace of change, that requires dealing with various topics, and achieving new adaptation. Previous conflicts surface again and require new solutions. Regression occurs, accompanied by a weakening of defences, an arousal of anxiety, and changes in the organization of the self.2,3 Other works stemming from this background have found that pregnant women are characterized by a relatively low level of self-acceptance, by introversion, and by an increased amount of emotional instability.3,4

It is likely that the changes discussed above express themselves in dreams as well, which are, according to Freud, a compromise between wish fulfillment and the need for camouflage.5 The disguising agent is the censor, working on behalf of the defences. In sleep, the defences grow weaker, thus the chances of the censorship cracking grow higher, and the latent contents may break through it. It is this weakness that permits a compromise between the defences and the unconscious: forbidden materials are allowed to be expressed, but only as disguised and symbolized contents.

The availability of dream material in pregnancy is also explained by the cognitive approach, focusing on a change in thought processes during pregnancy, manifested by a reduction in the logical and critical aspects of thought, together with an increase in intuitive and less-rational features.

A different cognitive explanation refers to the increasing awareness of time limitation in pregnancy. The rationed time gives a sense of urgency, and brings up more meaningful content.2

Breger et al.6 investigated the effects of a stressful experience on subsequent dreams. They found that following such an experience, the dreams either: dealt with the stressful event; showed excessive anxiety; or their contents were characterized by primitiveness, increased involvement, and implausibility.

The work of Breger et al.6 views pregnancy as a specific stress-arousing experience and has marked the three focal points of dream research in pregnancy as: pregnancy content in dreams, occurrence of anxiety in dreams, and manifestation of primary thinking in dreams.

Studies analysing the content of pregnant women's dreams found anxiety to be reflected in many of them. For example, Maybruck7 states that 40% of the dream reports of pregnant women in her study had nightmares, and that 30% included anxiety-provoking elements. Yet these results were not compared to a control group, so it is possible that they would not differ from those of a non-pregnant group, had there been one.

In Epstein's8 dissertation, results showed that women graded as ‘moderate’ on a worrying scale (in waking) reported more pregnancy-related dreams and more dreams containing threats regarding the pregnancy, in comparison to women that were graded as ‘low’ on the same scale. In other words, dreams reflected the confrontation of these threats in waking.

Breger et al.'s6 study found within-subject differences in primitiveness, implausibility and cognitive disturbance, before and after a stress-inducing experience. These differences touch the characteristics of primary-process thinking (PPT).

Schroer9 investigated the occurrence of archetypes, which are elements of the collective unconsciousness according to Jung, in first pregnancies. Dream reports of pregnant women were compared to those of non-pregnant women on four scales: affective intensity, irrationality, non-everydayness, and parallelism to mythology. Based on the grading of the reports according to these scales, it was found that first-time pregnant women had significantly more archetypal dreams. This can clearly be taken as supporting evidence since irrationality, non-everydayness and affective intensity characterize both PPT and archetypal thinking (here operationally defined). In contrast, Davidson,10 who examined the hypothesis that pregnant women are in an enhanced state of adaptive regression, did not find supporting experimental evidence.

Blake and Rietmann11 noted that 67% of pregnant women in his study reported pregnancy-related dreams, and that 60% of these women reported having such dreams once a week. Since Blake's study relied on retrospective reports, one may assume that these figures are an underestimation, and that in fact, the incidence of dreams dealing with pregnancy is higher.

The content categories that were found to repeat themselves in studies of dreams in pregnancy were: the pregnancy itself; the theme of the woman being unattractive and fear of losing her partner; conflicts of dependency–independence (especially toward the woman's mother); environmental and aggressive threats; and past experiences. Except for pregnancy itself and past experiences, which significantly characterized women during first pregnancy in comparison to a control group, the remaining categories do not necessarily characterize pregnant women since some were found in studies not including a control group.12,13

Besides the explicit content categories, some studies describe symbols characterizing pregnant women's dreams. The statement that similar symbols are common in many people's dreams is found in many theories. Maybruck14 claims that in dreams of women in a similar stage of physical development, certain common symbols are especially salient.

Schroer,9 departing from the Jungian theory, focused on the dreams of women during first pregnancy, which he refers to as a critical life stage. He describes three main themes appearing in dream reports of women in first pregnancy: an unconscious perception of the development of the new life in the womb as something bestial and primitive that expresses itself with dream contents of animals; the theme of leaving the partner/spouse behind as the pregnant woman goes into a state or transition to which he cannot follow her; and the theme of a new feminine identity.

As mentioned before, many of the studies in the field are characterized by the following shortcomings.

  • 1Lack of a control group. Studies not including a control group do not provide comparable data, thus not allowing a conclusion of significant differences between pregnant and non-pregnant women.
  • 2Retrospection. Studies in which dreams are recollected retrospectively suffer from memory biases (affected by the passing of time and events taking place during this time) and context effects (i.e. questioning in baby care centres may encourage mothers to recall pregnancy-related dreams).
  • 3A small number of participants. In these studies the statistical significance of the results may be compromised.

Further research is necessary in order to expand knowledge concerning pregnancy and dreaming, and more specifically, the relationship between the two.

In the present study, dreams of pregnant women were investigated in regard to the following hypotheses. Dream reports of pregnant women: (i) will include more contents related to pregnancy, in accordance with the assumption concerning the role of dreams in processes involving loaded emotional contents while striving for control; (ii) will reflect more anxiety, based on the perception of pregnancy as a developmental crisis which arouses anxiety; and will be ranked higher on a PPT scale, this either as a result of defence-weakening in pregnancy and the displacement of the reality principle for the pleasure principle, or characterization of the pregnancy by primitiveness, implausibility, and cognitive disturbance, due to stress.



Subjects were recruited in a specialized centre preparing women for childbirth, were all free of medication, and none were found to have sleep disorders as measured by the Mini Sleep Questionnaire.15

Experimental group subjects were 10 married women pregnant for the first time, between the ages of 25–37 years (mean = 28.7; SD = 3.46), who had between 13 and 18 years of schooling (mean = 16.3; SD = 1.49), and had been married for 1–7 years (mean = 3.07; SD = 2.00).

Control group subjects were 11 married women who were not, nor have been pregnant, between the ages of 23–32 years (mean = 26.36; SD = 2.9). They had between 12 and 18 years of schooling (mean = 15.7; SD = 1.79), and had been married for 0.5–3.5 years (mean = 1.54; SD = 1.03).

A significant difference between the two groups was found for years of marriage (t = 2.14; P < 0.05). This difference is due to the fact that first pregnancy usually takes place after a few years of marriage.


In a demographic questionnaire the participants were asked to state their age, years of education, and years of marriage. For a dream log the participants were instructed to write down their dreams every morning upon awakening, using paper placed next to their beds.

Dream analysis scales

The Incorporation Scale analyses the degree of appearance of pregnancy-related contents in the dream reports, such as: pregnancy, birth, baby/child, family, partner/spouse, the woman's body (including fears involved), and the baby's body (also including the fears involved with it). Based on Breger et al.,6 the appearance of a content category was scored while taking into account the degree of its presence in the dream.

The Gottschalk-Gleser Anxiety Scale16 includes six subscales evaluating different kinds of anxiety, scored according to the object to which they are referred (the subject herself, others, objects, and denial, in this descending order) as well as a general anxiety score calculated as the sum of the subscores.

The Auld, Goldberg and Weiss Primary-Process Thinking Scale17 situates the dream on the continuum (divided into seven segments) between, at one end, a logical story lacking odd happenings (scored 1), and, at the other, a bizarre tale with no logical links (scored 7).


The study was presented to women who fulfilled the recruitment criteria, as a study about women's dreams. The issue of pregnancy was not mentioned so as not to bias the dream reports. Women who consented to participate in the study were given a debriefing page, and a questionnaire designed to rule out sleep disorders. In the debriefing page participants were instructed to write down their dreams every morning upon awakening, for a period of 2 weeks. They were asked to write the dreams as they came to them without organizing or censoring them. At the end of the 2 weeks, participants were required to return the dream reports, sleep questionnaire and biographical data in a sealed envelope so that anonymity would be assured.

The dream reports of the two groups were typed, mixed, and passed on to two judges, who were blind to the experimental conditions. Both had practice in using the three scales, and scored the reports separately. The reports' final scores, on all three measures, were taken as the mean between the scores of the two judges.


Reliability between judges

In order to examine the reliability between judges, Pearson correlation coefficients were computed between the scores given to each dream report, on each of the nominated scales (Table 1). The highest reliability scores were found for the inclusion of pregnancy-related contents, and these varied between r = 0.895 (baby's body), and r = 0.993 (birth). High reliability was also found for judgements of the different types of anxiety, which varied between r = 0.756 (anxiety of injury), and r = 0.885 (diffuse anxiety). The correlation for the general anxiety score, computed as the total of all of the anxiety scores, was r = 0.925. The PPT scale gave rise to a reliability of r = 0.89 between judgements.

Table 1.  Reliability between judges
  1. * For all correlations, P < 0.05.

Anxiety type
Inclusion of pregnancy-related contents
PregnancyBirthBaby/childFamilyPartnerWoman's bodyBaby's body
primary-process thinking 0.890

Dream reports: analyses and examples

Fifty-three dream reports were gathered from the experimental group, and 68 from the control group, comprising a total of 121 dreams.

For each of the measures, in order to explore the differences between the pregnant and control groups, two t-tests were computed (i.e. one using the mean score of each participant, the second employing her maximal score). The results are shown in Fig. 1. Figure 1 describes the inclusion of pregnancy-related contents in the dreams of the pregnant group and the control group.

Figure 1.

Mean pregnancy-related contents in dreams of the two experimental groups. (▪) Pregnant; (bsl00004) non-pregnant.

The following results emanated from the analyses (Table 2). In accordance with our hypothesis, contents of pregnancy, baby/child, woman's body, and baby's body significantly characterized the dreams of the pregnant women. The maximal weight of the contents in the pregnant group's dreams was significantly higher than their maximal weight in the control group's dreams. The same is true for their mean appearance (except for the mean appearance of baby's body content, for which no significant difference was found, perhaps because it appeared in only three of the control group's dreams).

Table 2.  Results of analyses
Content categoryMean of first pregnancy groupMean of control groupt valueSignificance
  • * 

    Significance level of P < 0.05;

  • ** 

    significance level of P < 0.01.

Woman's body
Baby's body

In contrast to our hypothesis, it seems that contents of birth, family and partner did not significantly characterize the dreams of pregnant women. The results suggest that both groups' dreams reflect a quantitatively similar occupation with contents such as families of origin and relationship with the partner. This could be due to the common points between the two groups of women, all in their first years of marriage. If there were a difference, perhaps it could be found in the type of occupation, with the help of a finer content analysis. The lack of a difference for the content of birth may be due to its relatively low appearance in dreams: it appeared in only six of the pregnant group's dreams, and in two of the control group's.

In contrast to our hypotheses, no significant differences were found for levels of anxiety or for PPT between the two experimental groups.

In addition to the pursuit of differences between the two experimental groups, we performed a few analyses that focused on dream reports of the pregnant group alone. These analyses investigated the relationship between the occupation with pregnancy in dreams, and other measures (Table 3).

Table 3.  Correlation between content category and anxiety in the dreams of women during first pregnancy
  • * 

    Significance level of P < 0.05;

  • ** 

    significance level of P < 0.01.

Mean scoreMaximal guilt anxiety− 0.733*
 Maximal shame anxiety− 0.657*
Maximal scoreMean separation anxiety0.723*
Mean scoreMean death anxiety0.696*
Mean scoreMean injury anxiety− 0.805**
 Mean diffuse anxiety− 0.795**
Maximal scoreMaximal injury anxiety0.646*
 Maximal separation
Woman's body
Mean scoreMean death anxiety0.757*
Baby's body
Mean scoreMean death anxiety0.859*

A positive correlation was found between occupation with pregnancy and separation anxiety, while a negative one was found between the same occupation and guilt and shame anxiety. The occupation with the baby/child, baby's body and woman's body was found to positively correlate with death anxiety. Occupation with the partner positively correlated with injury anxiety, separation anxiety, and generalized anxiety. In contrast, occupation with family was found to negatively correlate with injury anxiety and diffuse anxiety.

For the sake of comparison, the correlation coefficients between content categories and anxiety levels in dream reports of the control group are shown in Table 4. The data give rise to the fact that many of the correlations characterizing the dreams of pregnant women are missing in the dreams of the controls. Only the positive correlations between the occupation with pregnancy or the partner, and separation anxiety repeats itself in the dreams of the control group.

Table 4.  Correlation between content category and anxiety in the dreams of controls
  • * 

    Significance level of P < 0.05;

  • ** 

    significance level of P < 0.01.

Mean scoreMaximal diffuse anxiety− 0.794**
Maximal scoreMaximal separation anxiety0.752**
 Maximal diffuse anxiety− 0.677*
 Mean diffuse anxiety− 0.694*
Mean scoreMean shame anxiety0.604*
Mean scoreMaximal guilt anxiety− 0.722*
 Mean guilt anxiety− 0.608*
Maximal scoreMaximal separation anxiety0.636*
 Mean separation anxiety0.628*
Baby's body
Mean scoreMean injury anxiety0.628*
Maximal scoreMean guilt anxiety0.734*

When investigated only in the pregnant group, no significant differences were found for levels of anxiety between dreams dealing with pregnancy, and those not dealing with it. The following examples illustrate some of the content categories found in pregnant women's dreams: the first two deal with the pregnancy itself, the third clearly expresses anxiety for the life of the fetus, the fourth deals with birth, and the last regards the baby yet is accompanied by anxiety.

  • Dream 1: ‘I dreamt that the baby was sitting on me, leaning on me, yet she was still inside me. This meaning that she was still inside my skin, but one could clearly see her body forms.’

  • Dream 2: ‘I dreamt that I was taking the baby out from inside me, but she was still connected by the umbilical cord. I played with her and it was fun. When I wanted to put her back, I was told that it's impossible because there was a problem with the umbilical cord, and I really did see that in a certain location the cord was too narrow. And now she couldn't be put back. I don't remember how it ended.’

  • Dream 3: ‘I was in Holland with my sister-in-law. She asked me why I didn't tell her that the baby died and was buried. I didn't answer, and ran until I found a place where I could be alone, and with my stethoscope looked for the baby's heartbeats in my abdomen. For a moment I couldn't find them and I panicked. Immediately afterwards I found the baby's heartbeat in my abdomen, and then my husband came too, and I told him what happened.’

  • Dream 4: ‘My delivery at home with my friend. Then I'm home with two puppies – a male and a female – and my baby is almost forgotten, he was in distress. My husband was not present at the birth.’

  • Dream 5: ‘As the baby is born, I see that she is kidnapped while I'm given another baby instead. And I yell at the staff not to do it, since I see the kidnapping. They switched my baby to one that is black, ugly, with lots of hair, and her facial features are those of Molly (a girl from my class, underprivileged with difficult problems at home, and very neglected – externally speaking). I'm upset, crying, not believing, and I explain to them that this is not my baby because we're both very fair and it can't be that we'd have such a black baby. They don't switch her back’.


This work examined the reflection of the experience of a first pregnancy on dream reports of pregnant women. From the comparison of these dream reports to those of women who were never pregnant, three main findings emerged.

  • 1Women during first pregnancy report of dreams in which there is increased occupation with contents related to the pregnancy, the baby/child, the woman's body, and the infant's body.
  • 2Dreams of women during first pregnancy are not characterized by significantly higher levels of anxiety.
  • 3Dreams of women during first pregnancy are not characterized by increased primary-process thinking (PPT).

The first finding testifies to the fact that women during first pregnancy are occupied with the pregnancy while dreaming. The increased occupation with pregnancy-related contents expresses the psychological ‘digestion’ of the experience. The pregnant woman processes what she is undergoing, emotionally and cognitively, through different channels – one of which is dreaming.

As to anxiety levels in dream reports, these were not found to be higher in dreams of pregnant women than in dreams of controls. This may very well be the case, suggesting that, contrary to common belief, pregnancy is not a traumatic event. This seems to contradict Maybruck's14 findings of high levels of anxiety in dream reports of pregnant women, but since her particular study did not include a control group, the meaning of its findings is limited.

On the other hand, it is not entirely unlikely that participants had trouble (or more trouble) remembering anxious and/or charged dreams. This forgetting may be conscious (i.e. the wish to avoid dealing with especially difficult or frightening dreams, as is clear from the reports of several participants who acknowledged having a nightmare they did not want to recall in writing), or unconscious (i.e. the dream may simply slip from the dreamer's awareness).

Besides the difficulty in remembering anxious dreams, it is conceivable that a difficulty in reporting them existed as well. Although participants were explicitly requested not to censor any content, one may presume that occasionally, anxious dreams, experienced as charged, intimate, or uncomfortable, were omitted.

The same line of reasoning holds regarding aspects of PPT. As PPT was not found to be elevated in dreams of pregnant women compared to those of controls, one may assume that pregnancy does not consist of an event encouraging this kind of thought processes, as postulated by psychoanalytic theory. However, it is possible that the mediation of the dreamer had camouflaged dreams with increased levels of primary-process thinking, experienced as strange, and therefore threatening or embarrassing. Another difficulty may have been encountered when trying to recollect and report such dreams.

An important conclusion that arises from these results concerns the indiscriminate tying of the experience of pregnancy with increased anxiety and PPT, and incites a discussion as to the question of a possible bias in regard to pregnancy in psychological theory, and in our culture. In the introduction, psychological theory referred to pregnancy as a ‘developmental crisis’. The term ‘crisis’ is meant to describe the confrontations that pregnancy invites, yet the term has negative connotations, while a term such as ‘developmental stage’ could have offered a more neutral alternative. In fact, feminist criticisms claim that the choice of words is not incidental, but rather conforms with numerous theoretical and problematic models of pregnancy.

According to Kolker,18 pregnancy is conceived as the ultimate fulfillment of the feminine role in popular culture, yet at the same time as a period in which the woman may justifiably be relieved from norms of rationality and adult behaviour. A concrete example of this is the convention of bizarre food cravings, acceptable in pregnant women.

The bias described above is not restricted to pregnancy, and appears in psychological references to additional aspects of biological femininity. There is a tendency to perceive the feminine hormonal fluctuations as ‘storms’ taking over women, impairing their rationality, and rendering them ‘abnormal’. Labels like ‘PMS’ or ‘postpartum depression’ are some of the manifestations of this view.

It seems therefore that conceiving pregnancy as a crisis period, and the pregnant woman as anxious and regressive requires careful examination. Terms such as ‘crisis’ are misleading in the sense that they intensify only one of pregnancy's many facets; beside the fears and obstacles, pregnancy is basically accompanied by excitement and positive feelings. The pregnant woman deals with various changes, yet one must keep in mind that the general context – that of the creation of a new life, and expected motherhood – is essentially a positive one.

Subsequent research may continue clarifying to what extent this bias truly embodies a cultural image. The findings on dreams of first-time pregnant women offers an important avenue in this direction. Dream analysis could elucidate the variety of shades with which pregnancy-related contents are dyed, orienting research towards personal experiences, instead of restraining it to cultural conventions.