Since dreams reflect waking life experiences (the so-called continuity hypothesis of dreaming; Schredl Reference Schredl2003), dream studies can elucidate sex differences reported for waking behaviour. For example, it is a stable finding that men report more sexual dreams than do women (Schredl et al. Reference Schredl, Sahin and Schäfer1998), which reflects the meta-analytic findings of higher frequency of masturbation and sexual fantasies in males compared to females (Oliver & Hyde Reference Oliver and Hyde1993). Regarding aggression in dreams, the findings are in line with the meta-analysis reported by Archer: Men's dreams included more physical aggression than women's dreams did (Hall & Van de Castle Reference Hall and Van de Castle1966), whereas the amount of verbal aggression did not differ between the sexes (Schredl et al. Reference Schredl, Sahin and Schäfer1998). The gender difference regarding the percentage of physical aggression (50% in men's dreams vs. 34% in women's dreams; Hall & Van de Castle Reference Hall and Van de Castle1966) is quite stable over time. The data collection period of Hall and Van de Castle (Reference Hall and Van de Castle1966) ranged from the late 1940s to the early '50s. Subsequent studies in the '70s (Hall et al. Reference Hall, Domhoff, Blick and Weesner1982) and '90s (Schredl et al. Reference Schredl, Ciric, Bishop, Gölitz and Buschtöns2003), and very recent studies (Schredl & Keller Reference Schredl and Keller2008–2009) replicated the higher prevalence of physical aggression in men's dreams compared with total aggression. This means that cultural developments such as the women's movement did not affect this sex difference in dreams. And this favours the sexual selection theory over the social role theory.
Another aspect of dream aggression fits the theory put forward by Archer: the higher difference in the aggression per male character compared to the aggression per female character in men (0.23 vs. 0.13 for men, 0.13 vs. 0.10 for women). That is, women experience aggressive interactions with both men and women in almost equal frequency in their dreams, whereas men's dreams include same-sex aggression more often compared to opposite-sex aggression (Hall & Domhoff Reference Hall and Domhoff1963).
Also very interesting is the shift in the percentage of physical aggression over the life-span. Whereas children (younger than 11 years old) showed the typical gender difference of more physical aggression in boy's dreams compared with girl's dreams, the ratio of physical aggression in dreams is the same for both sexes in the age range from 12 to 17 years (Hall & Domhoff Reference Hall and Domhoff1963; Oberst et al. Reference Oberst, Charles and Chamarro2005). This fits with Archer's argument that males avoid risky encounters with older males prior to adulthood. The adults – as reported above – again showed the preponderance of physical aggression in men's dreams.
Domhoff (Reference Domhoff1996) reviews cross-cultural dreams studies. Whereas many Western countries showed higher prevalence rates of dream aggression in men compared to women, there where several exceptions. The Hopi Indians, for example, showed no gender differences in overall aggression and in the percentage of physical aggression (Domhoff Reference Domhoff1996). The term Hopi can be translated into “peaceful ones” reflecting the life-style of these Pueblo Native Americans. But for some industrial countries, such as Switzerland and Japan, the ratios of physical dream aggression were not different between the sexes. This indicates that cultural factors modulate the amount of aggression in dream. In females, dream aggression was more often found in dreams of non-traditional women, indicating again the cultural effect on aggression pointed out by Archer. Unfortunately, these studies did not differentiate between same-sex aggression and opposite-sex aggression to enable us to test Archer's claim that cultural factors might be more important in explaining the amount of opposite-sex aggression.
Two studies, by Waterman et al. (Reference Waterman, de Jong and Magdeliyns1988) and Cohen (Reference Cohen1973), investigated whether biological sex or feminine versus masculine sex role orientation explains differences in dream aggression. Whereas the finding of the first study was unambiguous (only biological sex was of importance), the second study showed effects of both variables on the amount of dream aggression. Again, it would have been fruitful to differentiate between same-sex and opposite-sex aggression.
Another interesting gender difference can be found in the bad dreams and nightmares of children (Schredl & Pallmer Reference Schredl and Pallmer1998). In Table 1, the percentages of male and female aggressors are presented.
Table 1. Human aggressors in children's dreams
Male characters threaten the dreamer most often whereas women are quite rarely aggressors in dreams. The ratio of male and female aggressors in dreams is similar for boys and girls, thus reflecting the preponderance of male aggression in mass media (news, films, etc.). It would be very interesting to study the gender of the aggressor in cross-cultural dream studies in more detail, for example, in societies without predominance of male aggression. If dream aggressors are still mostly male, one might argue that gender differences in aggression have hereditary aspects.
To summarize, dream studies seem to be a valuable tool for studying sex differences in aggression, because the findings fit very well with the findings for waking-life aggressive behaviour. One might argue that studying dream aggression might be even more promising because aggression in dreams is not labelled socially the way being aggressive in waking life is. This is supported by the fact that aggression, even physical aggression, is quite common in dreams (Hall & Van de Castle Reference Hall and Van de Castle1966). For future studies, it would be very interesting to study the effect of mediator variables mentioned by Archer on the amount of dream aggression – for instance, the effects of empathy, risk-taking behaviour, and assertiveness, as well as physiological measures such as testosterone levels, size, and strength.
