More complex dreams with emotions or aggressions are associated with longer reports Iñigo Saez-Uribarri * 5th World Congress on Sleep Medicine. Valencia (Spain), September 28 – October 2, 2013. Introduction: Report length is an indicator of dream recall, cognitive activity of the dreamer and biases due to recording data in the laboratory as opposed to at home. Report length is also related to certain personality traits (Blagrove & Pace-Schott, 2010), dreams with more characters and with more threatening content (Saez-Uribarri, 2012). On the other hand, the dreamer is the main character in dream reports. Traits of dreamers in their dreams may be different to their traits during wakefulness. The Dream Self is the representation of the traits of the dreamer in the dream. The aim of this study was to analyse how the number of characters, emotions and threats to the Dream Self are related to dream report length. Hall and Van de Castle’s normative data (Hall & Van de Castle, 1966) were used, which comprise a standard reference in dream content research. Method: Hall and Van de Castle’s normative data were taken from DreamBank.net (Schneider & Domhoff, 1999), composed of the dreams of 491 women and 499 men, all of them university students. Data were extracted on the number of characters, aggressions and emotions in the dreams. Ten records from the database did not contain a dream report. Subsequently, the relationship between these variables and report length was analysed. In order to obtain a closer fit to the Normal distribution for report length, the number of words was transformed using the value of its napierian logarithm. On the other hand, aggressions were divided into social and physical aggressions (Saez-Uribarri, 2008), which corresponded to codes A1 to A4 and A5 to A8 respectively of the Hall and Van de Castle coding system. Finally, a multiple correspondence analysis was also carried out to explore the relationship of emotions and aggressions involving the Dream Self with the number of characters and report length. Results: A significant relationship was found between report length and the number of characters, r=.47, p<.001, the number of aggressions, r=.21, p<.001, and the number of emotions, r=.26, p<.001. The data which specifically corresponded to the Dream Self were then analysed. Table 1 summarises the differences in report length produced by the presence or absence of the Dream Self as an aggressor or a victim. Table 1 also summarises the differences produced by the presence or absence of the emotions which affected it. Practically all functions and emotions produced significant differences. In the correspondence analysis (Figure 1), dimensions 1 and 2 represent trends in the relationships between the variables of the Dream Self which were studied. The most important variables are those which are furthest from the origin of the coordinate axis. Proximity between the variables indicates the existence of relationships between them. The most significant groups of variables are highlighted in the figure. Table 1 t d.f. Sig. (two way) Sex 4.0 979 .000 Dreamer as physical aggressor -3.0 979 .003 Dreamer as physical victim -0.5 979 .599 Dreamer as social aggressor -3.2 979 .002 Dreamer as social victim -4.8 979 .000 Emotion in the Dream Self (yes/no) -6.2 979 .000 Apprehension of the Dream Self -4.2 979 .000 Anger of the Dream Self -5.3 979 .000 Happiness of the Dream Self -3.9 979 .000 Sadness of the Dream Self -2.8 979 .005 Confusion of the Dream Self (1) -2.1 151.4 .037 (1) Equality of variances was not assumed Functions of the Dream Self as aggressor or victim and its emotions In the central group of Figure 1, we can see that the characters and the word intervals are distributed in an ordered fashion along the length of Dimension 1. Positive values on Dimension 1 are associated with reports with more words, more characters, with aggressions or social victimisation against the Dream Self, and with the feeling of anger. Dimension 2 is associated with the presence of the elements of the group composed of the functions of the Self as aggressor and physical victim. Dimension 2 is also associated with an absence of anger in the report. Anger is an emotion limited to reports in which the Dream Self acts as a social aggressor, r=.10, p=.002, or as a social victim, r=.15, p<.001. Anger was not related to the function of the Dream Self as a physical aggressor, r=-.004, p=.904, or as a victim of physical aggression, r=.016, p=.608. Apprehension, happiness, sadness and confusion were not found to be important in the correspondence analysis. This can be observed by how close these variables are to the coordinate origin. However, these emotions also produced significant differences in the number of words in the report (Table 1). A clear differentiation between physical and non-physical aggressions was also observed. Conclusions: The presence of more characters, more aggressions, and emotions were related to an increase in the length of dream reports. Report length was also increased with aggressions and social victimisation, and the emotions in which the Dream Self was involved. Anger was found to be a key emotion in social aggressions received or inflicted by the Dream Self. Notably, it showed almost no relationship with physical aggressions. Meanwhile, the data of the Dream Self showed a simple underlying structure which suggests that physical and social aggressions should be analysed separately. It also suggests that future research should specifically analyse the data which exclusively concern the Dream Self. We could summarise all of the observations by saying that more complex dreams required more words in order to report them. References Blagrove, M. & Pace-Schott, E. F. (2010). Trait and neurobiological correlates of individual differences in dream recall and dream content. International Review of Neurobiology, 92, 155-180. DOI: 10.1016/S0074-7742(10)92008-4 Hall, C., & Van de Castle, R. (1966). The Content Analysis of Dreams. New York: Appleton-Century-Crofts. Saez-Uribarri, I. (2008). La ansiedad al despertar y las características de los personajes de los sueños. Vigilia-Sueño, 20(2) 61-72. Text translated at: http://cogprints.org/7069/4/Dream_Characters.pdf Saez-Uribarri, I. (2012). Biases in the relationship between dream threats and the level of anxiety upon awakening. Paper presented to 21st Congress of the European Sleep Research Society, Paris. Schneider, A., & Domhoff, G. W. (1999). DreamBank. http://www.dreambank.net/. _______________________ * E-mail: inisaezu@bilbao.com