(First posted on February 5, 2007) Last week I asked if you would be interested in my take on this paper, since it is in Serbian (and one commenter said Yes, so here it is - I am easy to persuade):
Stankovic Miodrag, Zdravkovic Jezdimir A., and Trajanovic Ljiljana,
Comparative analysis of sexual dreams of male and female students (PDF). Psihijatrija danas 2000, Vol. 32, No. 4, pp. 227-242
Here is the English-language Abstract:
The subject of research is analysis of connection between sexuality as instinctive function and dreams with sexual content as cognitive function. The sample consisted of 656 students, 245 males and 411 females. Research showed significant difference between genders concerning sexual dreams their appearance, frequency, image of sexual partner, and content subjective emotional experience during dreams and talk about sexual fantasies. Based on the obtained data, the authors believe that dreams with sexual content are not learned behavior, but biologically determined sexual behavior, and that cognitive elaboration of contents and objects of sexual fantasies is secondarily environmentally conditioned.
Reading this paper was quite an interesting experience. It's been a while since I last read a paper in Serbian so the language itself (with so many Serbianized English terms) appears strange - in my opinion English is for science, Serbian is for poetry!
The first thing that hit me was the very beginning of the Introduction. I have forgotten over the years how many papers there start with pointing out the deepest historical origins of their topic, usually starting with Ancient Greece (we used to joke back when I was still there in the 1980s that every paper there starts with the same phrase: "Even the ancient Greeks thought...."). The very first mentions are of the writings of Ovid, Galen and Aristotle and what they wrote about sexual dreams.
Next comes the obligatory smackdown of Freud establishing the hard-nosed, hard-science attitudes of the authors, followed by a brief review of the old theries on dreaming (a very short version of something like this), followed by a brief review of the classic in the field of sex research - the Kinsey review, focusing on what it has to say about erotic dreams coupled with orgasms/nocturnal emissions.
I have mixed feelings about this approach. On one hand, starting every paper with Ancient Greeks is silly. On the other hand, I wish more papers in English would spend a little bit of time at the very beginning providing the historical, philosophical and theoretical background and context instead of immediatelly jumping into last month's literature on the topic (at least press releases and blog posts about the papers should cover this ground).
The main argument of the Introduction, I think, is that the world has changed (emancipation of women, etc.) yet the erotic dreams in both sexes have not changed since the Kinsey report, indicating a biological basis for them (and for the sex differences in them) instead of cultural effects.
So, what did they do?
They picked 656 students (245 men and 441 women) and gave them a quesitonnaire with some multiple choice questions and some open-ended short-essay questions. The students were given the explanation of the study and guaranteed anonymity. They were given 60 minutes to complete them (I am wondering if this was done in one room at one time or on one-by-one basis). This paper only covers the analysis of one quarter of the overall study (Part IV of their questionnaire).
The average age of men was 21.5 years and for women 22 years. At the time of the study, 89% of men and 77.1% of women reported having sexual experience. Of those, men reported losing virginity at the age of 17.15 years and women at 18.44 years.
The first question was "Do you have dreams with sexual content?" The available answers were "Never of extremely rarely", "Rarely" and "Often". The results for men: 47, 157, 41 (respectively) and for women: 143, 235, 33.
What is troublesome is that after this question, people who answered "never or extremely rarely" were excluded from further analysis. This means that the actual sample sizes for the rest of the study were not the stated 245 men and 441 women, but 198 men and 268 women. I am also wondering what happened: if this was all done in one big classroom (e.g., during Psych 101 class), were the people who answered in the negative to the first question asked to leave the classroom? Those are 20-something students. You can imagine women saying No and men saying Yes (then inventing answers for the rest of the questions) for social-pressure reasons. I wish the paper stated more clearly how was the study done.
The question #2 tried to get a more fine-grained view of the frequency of erotic dreams, dividing the answers over five possibilities ranging from "almost every night" to "once a month or less". Half of the men and three quarters of the women chose "once a month or less".
The third question asked who was the subject of the dream. More than half the men answered "someone unknown", with "my partner" getting only 16.7 percent and other options given very rarely. With women, 102 (37.2%) answered "my partner", 98 (35.8%) chose "someone unknown" and 60 (21.9%) answered "someone known" with all other options ("group", "a relative", "celebrity", "a person of the same-sex" and "unusual partner, e.g., old people, children or animals") given very rarely. If I was designing the questionnaire I would have probably woded some of the options differently, e.g,. "partner of same sex", "partner of opposite sex", "a flame who is, unfortunately, not my partner", etc.
The question #4 asked about the content of the dreams. 77.8% of men and 34.1% of women answered "conventional sex with penetration", 15.2% of men and 60.1% of women chose "kissing and petting" while all other answers were answered quite rarely ("walking, talking or touching", "oral or anal sex", "sexual activities in which I experience pain" and "sexual experience in which I give pain"). OK, but again, I would have done it a little differently, perhaps moving the "feel pain/give pain" to a separate question altogether.
Next question inquired about one's subjective experience and behavior. 44.7% of women and 34.5% of men said "I feel tense", 29.4% of men and 19.8% of women wake up without an orgasm, 10.7% of men and 8.4% of women reported experiencing an orgasm and subsequently waking up, 4.1% of men and 13.9% of women answered "I feel relaxed". I have some problems with the questions/answers as some are not mutually exclusive, yet, judging from the table, participants were required to choose only one answer. I would have split this into two or three separate questions, one asking about waking up and having an orgasm (purely "what happened" questions), one asking about feeling relaxed or tense (the psyhophysical responses), and another for subjective opinions on the experience and emotions.
Question #6 asked "who do you tell about your erotic dreams?". 36% of men and 39% of women chose "nobody", 16% of men and 26% of women chose "to my partner", 39% of men and 32% of women chose "to close friends", and only a very small number of people chose "acquantance" or "physician". Here, I'd let them give more than one answer, perhaps with some kind of way to indicate frequency (e.g., half the time nobody, 25% of the time my partner, 20% of the time best friend, and 5% of the time acquntance, etc.)
But the real kicker comes next, putting under doubt all the data so far. Both men and women without real-world sexual experience have very rare erotic dreams (and since the numbers are small, there is no statistical difference between the sexes). People with sexual experience have sexual dreams much more often and the sex difference becomes very pronounced: 81.2% of experienced men and 66.8% of experienced women have sexual dreams [Note to the Editor: the Table is mislabelled to say "inexperienced", where it is clear from the text that it should read "experienced"]. I guess one can dream only about familiar things and no amount of watching porn can substitute for real experience.
One can think of numerous ways the difference between experienced and inexperienced people could have affected the results of previous questions (I wonder why didn't they do a separate analysis for the two categories!?). If you are inexperienced, it is likely you do not have a partner so that answer is out for this group of people and they have to choose something else.
If you are 20 years old or so, it is likely that you are head-over-heels in love with someone who does not return the love (remember the intensity of feelings when you were that young?). If a subject dreams about that person, will the answer be "my partner" ("oh, how I wish it was true...") or "someone I know"? If you are inexperienced, is it more likely you will dream about "walking, talking or touching" and "kissing and petting" than any other option? Would you be more "tense" if you don't really know how sex feels, and more "relaxed" if you do? All those things could presumably be figured out from the data they have and should have been included in the paper.
Interestingly, the Discussion starts with the lament that most of the scientific literature is concerned with 'why' we dream and not with 'what' we dream, leaving the latter topic to the Old Wives' Tales and pseudoscience peddlers. On the other hand, sleep research is more concerned with 'how' we sleep than 'why' we sleep.
Their data (more men than women have sexual dreams and men have more frequently such dreams) is explained by the menstrual cycle (once a month) for women and circadian rhythm of testosterone (once a day) for men, which makes no sense. Why should hormonal surges have anything to do with dream content? And, if I remember correctly (literature on humans is not at my fingertips like the avian is), the testosterone peak is very low (i.e., the amplitude of the rhythm is small) and it appears in the morning, thus presumably not affecting what is happening during the night. They take this interpretation as favouring a more biological explanation for dreams as opposed to cultural.
The second argument they use to bolster their case for Nature over Nurture is the high concordance between their results and Kinsey results. Whatever discrepancies exist, they explain away by the differences in methodology (e.g., Kinsey asked men about nocturnal emission irresepective of dreams, while they asked about nocturnal emissions only as sa subset of quesitons about dreams), or differences in the age of subjects (Kinsey covered several age groups, not just 22-y-olds), or differences in geography/history/culture. What is left is that men have more dreams than women (as in their study) and they argue that this means the pattern is universal, thus biologically determined.
But wait a minute! Kinsey did his studies in 1948 and 1952, one of the peaks of patriarchy in the US history. The current study used subjects from the Universities of Nis (Southern Serbia) and Podgorica (Montenegro) at the end of 1990s where wars and sanction ushered in a local peak of patriarchy as well (those tend to come right after wars, I guess), in already highly patriarchal (Mediterranean macho-style) society. Furthermore, they did not use the students from Nis and Podgorica (mid-size towns, quite large for that part of the world) who may be more modern and less patriarchal and who tend to live with their parents while attending the University. They questioned only students who live in the dorms - those are not kids from Belgrade, but kids from surrounding small towns and rural areas that are even more traditional and patriarchal. And, as much as living in the dorm is a life-changing experience, I doubt that by the age of 22 or so they all managed to shed all of their patriarchal upbringing.
Then, they argue that the difference between inexperienced and experienced subjects is due to their rates of psychophysical development and maturation, as if this correlates with the date of losing virginity, i.e., once you reach a particular level of physical maturation (hormones and such), you immediatelly go and get laid. Really? Perhaps in other animals, but not in humans.
Finally, they argue for a biological determination of sexual dreams because there is great similarity between these data and the data of their other study not yet published at the time (it came out a year later - I have it and will review it), on the prevalence of sexually-themed daydreaming. They argue that we can exert a greater control over daydreams than dreams, so the similarity in frequences must be due to biology? How? Doesn't it say the opposite - if we have control (presumably guided by the culture) over daydreams, then having the same frequency of dreams suggests that dreams themselves are affetced by culture. And how can one make such a strong statement by using a study nobody has seen yet at the time (including peer-reviewers).
To their credit, they conclude that while frequency and sex-differences in sexual dreams is biologically determined, the content (and the rest of it, e.g,. who they tell afterwards, how they feel about it, etc)., are influenced by culture.
Now, I really trashed the paper, but actually it is not as bad as I portrayed it. I, blogger-style, focused on weaknesses because they are fun to debunk. Obviously from the reading this, though, these are serious, smart, dedicated and professional researchers. So why did they produce a weak paper? I think the answer lies in the List Of References!
There are only 19 references total! Two are to Kinsey reports (male and female). Freud's 'The interpretation of dreams". Jouvet's book on sleep and dreams. A local book on interpretation of dreams that does not sound scientific. Nikola Rot's 1966 textbook on the Psychology of Personality, from which my brother studied (in class taught by Rot himself) back in the 1980s. I would expect that psychology of personality has made some advances in the past 40 years. Then, there are several other books (some in English, some in Serbian, some in translation). A conference paper. Several papers in Serbian. Only a couple of papers in English. Everything quite ancient, too.
I would venture a guess that this study would have been much better designed, performed, analyzed and written if these researchers had access to the literature! If they could read papers in their field, online or off, as soon as papers get published, they would have had a much better scientific mindset, not just a much greater knowledge-base. It is people like them, smart and enthusiastic, but unfortunate to live outside the West, who suffer the most from the Closed Science. It is a hundred times more important to them than to us - and it is darned important to us - that science be made Open.