International Journal of Psychophysiology 75 (2010) 291–294

Contents lists available at ScienceDirect

International Journal of Psychophysiology

j ourna l homepage: www.e lsev ie r.com/ locate / i jpsycho

Hormonal background of physiological aggressiveness in psychologicallyhealthy women

Francesca Brambilla a,⁎, Azzurra Speca b, Isabella Pacchiarotti c, Massimo Biondi b

a Department of Mental Health, Sacco Hospital, Milan, Italyb Department of Psychiatric Sciences and Psychological Medicine, Sapienza University, Rome, Italyc Department of Psychiatry, St. Andrea Hospital, II Medicine and Surgery Faculty, Sapienza University, Rome, Italy

⁎ Corresponding author. Centro di Psiconeuroendocrin20129, Italy. Tel.: +39 02 717350; fax: +39 02 701228

E-mail address: francesca.brambilla4@tin.it (F. Bram

0167-8760/$ – see front matter © 2010 Elsevier B.V. Aldoi:10.1016/j.ijpsycho.2009.12.012

a b s t r a c t

a r t i c l e i n f o

Article history:Received 16 August 2009Received in revised form 12 December 2009Accepted 20 December 2009Available online 13 January 2010

Keywords:AggressivenessEstrogenProgesteroneFree testosteroneHealthy women

The aim of this study was to see whether or not physiological hormonal fluctuations during the menstrualcycle modulate normal aggressiveness in psychophysically healthy women. In 15 probands estrogens (E),progesterone (PROG) and free testosterone (FT) plasma levels were measured by immunochemilumines-cence and levels of global aggressiveness and its subitems “verbal aggressiveness”, “suspiciousness” and“resentments” were measured by the Buss–Durkee Rating Scale in the early follicular, midluteal andpremenstrual phases of the cycle. E and PROG levels varied significantly along the menstrual cycle, whilethose of FT, of global aggressiveness (GA) and of its subitems did not change. Values of global aggressivenessdid not correlate with any of the hormonal parameters studied. However, E values correlated positively with“verbal aggression” scores in the follicular phase and positively with “resentment” in the premenstruum,while PROG levels correlated negatively with “suspiciousness” and “resentment” in the premenstrual phaseof the cycle. Hormonal and psychological changes from one phase to the next (Δ) revealed that Δ E in thesecond half of the cycle correlated negatively with “verbal aggressiveness”, while Δ PROG from follicular toluteal and from luteal to premenstrual phases correlated negatively with “resentment”. Thus, althoughaggressiveness did not seem to vary along the menstrual cycle, nor to correlate with hormonal changes,hormone secretions and fluctuations might possibly modulate some of the physiological aspects of thebehavioral parameter.

ologia, Piazza Grandi 3, Milano89.billa).

l rights reserved.

© 2010 Elsevier B.V. All rights reserved.

1. Introduction

Physiologically and pathologically aggressive behaviors in humanshave been mostly linked to central function of the serotonin (5-HT)system, including 5-HT secretion and responses of the multipleserotonin receptors to 5-HT stimulation (Virkkunen et al., 1987;Siever and Trestman, 1993; Coccaro et al., 1997; Bjork et al., 1999;Stanley et al., 2000; Bond et al., 2001; Tiihonen et al., 2001; Dolanet al., 2002; Klaassen et al., 2002; Marsh et al., 2002; New, 2005), andof central catecholaminergic and GABAergic functions (Korzan et al.,2000; Coccaro and Petty, 2003). Also, it is well known thatperipherally secreted hormones, including androgens, glucocorticoidsand growth hormone and centrally secreted opioids influence thedevelopment and the degree of physiological and pathological

aggressiveness in experimental animals (Grieng et al., 2001) and inhumans (Olweus et al., 1988; Coccaro et al., 1997; Gerra et al., 1996,1997, 1998; Finkelstein et al., 1997; Archer et al., 1998; Aromaki et al.,1999; Van Herringen et al., 2000; Pope et al., 2000; Grieng et al., 2001;Thiblin and Parlklo, 2002; O'Connor et al., 2002; Eriksson et al., 2003;Rowe et al., 2004). Little is known about the influence that femalesexual hormones exert on this behavioral parameter, even though inexperimental animals of both sexes, especially in major primates,aggressive behavior has been linked to variations of prenatal andpostnatal estrogen secretions (Simon and Whalen, 1986; Finkelsteinet al., 1997; Girolami et al., 1997; Palanza et al., 1999; Toda et al.,2001). Data from humans are sparse and contradictory. A significantpositive correlation has been reported between estradiol blood levelsand degree of pathological aggressive behavior in women, as relatedto verbal aggressiveness and physical violent sexual aggressiveness(Jnoff-Gormain et al., 1981; Gladue, 1991; Leibenluft et al., 1994).Physical aggressiveness has been reported to increase in hypogonadaladolescent girls given conjugated estrogens at low–medium doses(Finkelstein et al., 1997). In men, alcohol-related violent behavior hasbeen reported to positively correlate with estradiol blood levels(Eriksson et al., 2003), while physical violent sexual aggressions havebeen successfully suppressed by estradiol administration (Bradford,

Table 1Aggressiveness scores in normal women during the menstrual cycle phases.

Aggressiveness scores

Early follicular Midluteal Premenstrual

Global 50.4±14.4 52.4±14.9 51.7±14.8Direct aggressiveness 48.2±11.3 50.5±11.8 51.8±11.3Indirect aggressiveness 51.7±10.6 53.4±12.1 48.9±11.1Irritability 45.9±11.1 49.1±11.6 47.2±11.8Negativism 47.2±8.1 44.5±10.5 45.6±9.7Resentment 45.8±10.5 49.4±11.9 46.7±9.5Suspiciousness 45.5±12.4 46.1±13.0 47.8±13.2Verbal aggressiveness 46.2±8.2 47.5±10.6 48.6±9.9Guilt 50.6±8.8 48.5±6.7 45.3±13.9

M±SD.ANOVA for repeated measures p=non-significant for any of the parameters.

Table 2Gonadal hormone concentrations in normal women during the menstrual cycle phases.

Hormonal concentrations

Early follicular Midluteal Premenstrual P

Estrogens (pg/ml) 52.6±25.3 157.8±56.28 78.1±13.9 0.0002Progesterone (ng/ml) 0.8±0.3 11.1±5.6 5.7±5.3 0.00004Free testosterone(ng/ml)

0.9±0.7 1.3±0.5 1.1±0.7 0.15

Mean±Standard Deviations.p=ANOVA for repeated measures.

292 F. Brambilla et al. / International Journal of Psychophysiology 75 (2010) 291–294

1983; Kyomen et al., 1991; Kay et al., 1995; Shelton and Brooks, 1999).As for progesterone secretion, there are no data on its relationship withaggressive behavior.

In a very preliminary and exploratory study in a group of physicallyand psychologically healthy women, we investigated whether or notthe degree of normal aggressiveness varies along the menstrual cycle,in connection with the follicular, luteal and premenstruum phasesand with the cyclic variations of gonadal hormone concentrations. Wewanted to see whether or not physiological cyclic fluctuationsin estrogen and progesterone levels, and values of free testosteroneduring the cycle may affect normal aggressive behavior and some ofits multiple aspects in humans as observed in experimental animals,thus intervening in the physiological regulation of this behavioralparameters.

2. Methods

Fifteen women blind to the purpose of the study, recruited fromthe hospital staff were enrolled in the study (mean age=27.3±2.6).All of the probands were totally asymptomatic, without any sign oflate luteal–premenstrual phase dysphoric or anxious disorders orpremenstrual syndrome or any other psychopathology which couldinterfere with our investigation. Physical examination and routinelaboratory tests revealed no abnormalities. Mean weight of ourprobandswas kg 57.5±4,mean height cm 167.3±4.6 andmean BodyMass Index 21.3±1.8. Menstrual histories were negative for pathol-ogies, cycles occurring regularly every 28–30 days. None of theprobands used contraceptive medication. Exclusion criteria com-prised present medical diseases, endocrinopathies, immunopathies,mental disorders as defined by the criteria of the DSM IV, analyzed bythe SCID interview (American Psychiatric Association, 1994; Spitzerand Williams, 1983), previous cerebral trauma, past short- or long-lasting or present alcohol and drug abuse, administration of pharma-cotherapies of any kind.

Plasma concentrations of estradiol (E), progesterone (PROG) andfree testosterone (FT) were measured three times during onemenstrual cycle, in the early follicular phase (cycle days 7–8), in themidluteal phase (cycle days 21–22) and in the premenstruum phase(cycle days 26–27). The early follicular, midluteal and premenstruumphases were calculated chronobiologically on the days of the cyclerelated to the previous and following menstrual bleeding, on theestrogen and progesterone concentrations and on the ratio betweenthe two hormones.

Probands came to the laboratory for blood drawing for hormonalassays at 8.30 a.m., after overnight fast. A 19-gauge butterfly needlekept patent by saline infusion was immediately inserted into anantecubital vein, and blood samples were drawn into EDTA-contain-ing tubes, plasma was separated immediately by centrifugation andfrozen at −20 °C until assayed. Hormones were assayed by immuno-chemiluminescence, all together to avoid variations due to differentbatches of reagents. Interassay coefficients of variations were 12.5%for E, 8% for PROG and 7% for FT, intraassay coefficients of variationswere 7.3% for E, 5% for PROG and 7.6% for FT. Reproducibilities of theresults were 4.3% for E, 4.9% for PROG and 5.1% for FT. Sensitivities ofthe assays were 7.3 pg/ml for E, 8.57 pg/ml for PROG, and 5.9 pg/mlfor FT. Specificities were 0.3% for E, 0.002% for PROG, and 0.2 for FT.

Global aggressiveness and its subitems were measured by theBuss–Durke Rating Scale (Buss and Durkee, 1957), in parallel withblood drawing for the hormonal assays. We used the Buss–DurkeRating Scale in spite of the fact that it measures trait and not stateaggressiveness, because we wanted to see whether or not hormonalchanges along the menstrual cycle could influence trait-relatedaggressive attitude and feelings, even though not violent, actuallymanifested or externalized. Since our subjects were selected asnormal, without signs of overt aggressive behavior either all along themenstrual cycle or in specific phases of it, we decided to investigate, a

part from global values of aggressiveness, 3 non-violent aggressivesubitems “suspciousness”, “ resentment” and “verbal aggressiveness”,that is 3 parameters which may be present and active in non-overtlyaggressive individuals, to see if they could be influenced by thenormal hormonal fluctuations in non-pathlogical subjects.

2.1. Staistical analysis

Exploratory statistical analysis was carried out using ANOVA forrepeated measures (Friedman analysis with Kendall concordancecoefficient) to control for differences in degree of global aggressivenessand its 3 subitems taken into special consideration and in hormonalsecretions along cycle. Spearman analysis was used to investigate therelationship between global aggressiveness or its subitems “resent-ment”, “suspiciousness”, “verbal aggressivity” and hormonal secretionsat the various phases of the cycle. The Spearman analysis was used alsoto correlate changes (Δ) of values from follicular to midluteal and frommidluteal to premenstruum phases between hormonal and psycholog-ical values.

3. Results

Aggressiveness total scores and subitems (Table 1) showedfluctuations all along the cycle, the difference between each point,however, being not statistically significant (one-way ANOVA forrepeated measures).

Hormone concentrations (Table 2) showed the expected sharpincreases of E and PROG from the early follicular to the midlutealphase and decreases from the midluteal to the premenstrual phase,the changes being statistically significant, while no significant changeswere seen for FT (one-way ANOVA for repeated measures: E:p=0.0002, r=0.6; PROG: p=0.00004, r=0.76; FT: non-significant).

Spearman analysis revealed no correlations of E, PROG and FTvalues with global aggressiveness in the 3 phases of the cycle.However, E values correlated positively with “verbal aggression”scores (p=0.03, T=2.38, r=0.58) in the follicular phase of the cycle,and with “resentment” scores (p=0.04, T=2.25, r=0.56) in thepremenstrual phase of the cycle. PROG values correlated negatively

293F. Brambilla et al. / International Journal of Psychophysiology 75 (2010) 291–294

with “suspiciousness” scores (p=0.05, T=−2.19, r=−0.55) andwith “resentment” scores (p=0.02, T=−2.69, r=−0.43) in theluteal phase.

No correlations were observed for FT. Spearman analysis betweenΔ hormonal values and global aggressiveness or the three subitemscores revealed that E changes correlated negatively with “verbalaggressiveness” changes (p=0.05, T=−2.19, r=−0.55), whilePROG changes between follicular and luteal phases and betweenluteal and premenstrual phases correlated negatively with changes of“resentment” (follicular–luteal phase: p=0.03, T=2.36, r=0.58;luteal–premenstrual phase: p=0.05, T=2.17, r=0.54). No correla-tions were observed between FT values and the 3 subitem scores.

4. Discussion and conclusions

Our preliminary data show that in physically and psychologicallyhealthy adult women degrees of global aggressiveness do not varysignificantly but just fluctuate along the various phases of themenstrual cycle, in spite of the expected significant secretory changesof E and PROG. Moreover, aggressiveness did not seem to correlatewith any of the gonadal secretion in our probands. All together, thesedata do not seem to reveal cycle-related fluctuations of physiologicalaggressiveness, in accord with the report of Dougherthy et al. (1998).This seems to contradict the data from the literature reportingrelationships between sexual hormone concentrations, in particularFT and E, and aggressive behavior in women (Jnoff-Gormain et al.,1981; Simon and Whalen, 1986; Olweus et al., 1988; Gladue, 1991;Leibenluft et al., 1994; Gerra et al., 1996, 1997, 1998; Finkelstein et al.,1997; Aromaki et al., 1999; Van Herringen et al., 2000; Grieng et al.,2001; Thiblin and Parlklo, 2002). However, these data refer toelevated or pathological or provoked aggressiveness related tohormonal secretions or administration, while we investigated thepossible hormonal background of fluctuations of normal aggressive-ness during the normal menstrual cycles in subjects in restingconditions. Fluctuations of non-pathological aggressiveness alongthe menstruum cycle have been reported in women with premen-struum dysphoria, with a significant increase of aggressiveness inpremenstruum (Dougherthy et al., 1998; Bond et al., 2003,), eventhough this has not been always confirmed, since women withpremenstruum dysphoria have also been reported to be moreaggressive than controls independently of the phase of the cycle(Dougherty et al., 1997). In normal women, Ritter (2003) observedsignificantly higher values of aggressive behavior (both physical andverbal) during menses than during the midluteal phase.

In our study, in spite of the absence of any correlation betweennon-pathological aggressiveness and hormonal secretions, the anal-yses on the links between hormonal parameters and aggressivesubitem scores were intriguing, suggesting that they might correlateone with the other in the different phases of the cycle. In fact, thepositive correlations between E concentrations with “verbal aggres-siveness” in the follicular phase of the cycle and with “resentment” inpremenstruum, and the negative correlations of PROG concentrationswith “suspiciousness” and “resentment” in the luteal phase of thecycle suggest that, in spite of no significant changes in aggressivebehavior along the menstrual cycle, some of its specific aspects mightbe related to hormonal changes. The data on the Δ values, reflectinghormonal and psychological changes from follicular to luteal and fromluteal to premenstruum phases, suggest that hormonal secretorychanges from low to high values and vice-versa more than theabsolute values per sè may be responsible for the modulation ofspecific aggressive behavior. In particular, in the second half of thecycle Δ “verbal aggressiveness” correlates negatively with Δ E, and allalong the cycleΔ PROG values correlate negatively with “resentment”.In all, absolute ovarian hormone values and their changes during themenstrual cycle seem to be related not to global aggressiveness but tosome of its specific aspects. The fact that FT absolute and Δ values did

not correlate with global aggressiveness or its 3 non-violent subitemsis also intriguing, since FT secretion has always been connected toaggressiveness. This apparent contradiction may be due to the factthat FT secretion is linked to violent aggressiveness, which was notpresent in our probands. Obviously, the word “correlations” does notmean necessarily dependence of one parameter from the other, andmoreover our probands were too small a group to state that a cause–effect link exists between hormonal and psychological aspects in oursubjects. Our data might at least suggest the necessity to furtherinvestigate this pathway, as important for investigating and under-standing the biological bases of aggressiveness.

As concern the possible mechanisms of action of sexual hormonesin modulating aggressive behavior subitems, it is well known that inexperimental animals and humans E, PROG and FT act on their specificgenomic and non-genomic receptors in the brain (Halbreich, 1997;McEwen, 1999; Moss and Gu, 1999; Fink et al., 1999) and modulatecentral serotonin and catecholamine secretions, receptor and secondmessenger functions (McEwen, 1999; Moss and Gu, 1999; Fink et al.,1999), even though the hormonal influence on neurotransmitterfunction has not yet proved to be involved in the development of thevarious aggressive behavior aspects in humans.

A limitation to our study is that our group of probands was small.Thiswas due to the difficulty tofind youngwomen of the same agewithperfectly regular menstrual cycles, normal hormonal secretions, nottaking contraceptive pills, presenting no psychopathological aspectswhatsoever and in particular no premenstrual dysphoria, accepting togo through this study protocol. Obviously our study is only preliminary,and needs to be confirmed in younger and older regularlymenstruatingsubjects. Another bias of our investigation was that we measuredhormonal and psychological parameters only once during each phase ofthe same cycle andonly during one cycle, in themorningwithout takinginto consideration the possibility of variations related to the hormonalcircadian rhythms of secretion, in resting conditions and not underaggressiveness-provoked approaches. Another bias is represented bythe repeated use across themenstruum cycle of BDHI, a scale which is atrait measure of aggressiveness, instead than a psychological measuredesigned for repeated use, which might have shown more significantrelationships. As mentioned above, we selected this trait-related scalejust becausewewanted tomeasure trait-related aggressive attitude andfeeling actually manifested or externalized.

Our study can only disclose potential links between psychologicaland hormonal parameters and it cannot establish cause–effectrelationships, for which different protocols of investigations shouldbe used. In spite of that, and even though our data must be consideredas very preliminary and our statistical analyses as exploratory, wethink that they are worth of future investigations in order to validatetheir significance.

References

American Psychiatric Association, 1994. Diagnostic and Statistical Manual of MentalDisorders (DSM-IV), ed 4. APA, Washington.

Archer, J., Birring, S.S., Wu, F.C.W., 1998. The association between testosterone andaggression among young men: empirical findings and a metaanalysis. AggressiveBehaviour 24, 411–420.

Aromaki, A.S., Lindman, R.E., Eriksson, C.J.P., 1999. Testosterone, aggressiveness, andantisocial personality. Aggressive Behaviour 25, 113–123.

Bjork, J.M., Dougherty, D.M., Moeller, F.G., Cherek, D.R., Swann, A.C., 1999. The effects oftryptophan depletion and loading on laboratory aggression in men: time courseand a food-restricted control. Psychopharmacology 142, 24–30.

Bond, A.J., Wingrowe, J., Critchlow, D.G., 2001. Tryptophan depletion increases aggressionin women during the premenstrual phase. Psychopharmacology 156, 477–480.

Bond, A.J., Critchlow, D.G., Wingrove, J., 2003. Conflict resolution in women is related totrait aggression and menstrual cycle phase. Aggressive Behaviour 29, 228–238.

Bradford, J.M.W., 1983. Research on sexual offenders. Recent Trends, Psychiatry ClinicNorth America 6, 715–731.

Buss, A.H., Durkee, A., 1957. An inventory for assessing different kinds of hostility.Journal Consultant Psychology 21, 343–349.

Coccaro, E.F., Petty, F., 2003. CSF GABA correlates with impulsivity, but not aggression,in personality disordered subjects. Biological Psychiatry 53, 117S.

294 F. Brambilla et al. / International Journal of Psychophysiology 75 (2010) 291–294

Coccaro, E.F., Kavoussi, R.J., Cooper, T.B., Hauger, R.L., 1997. Central serotonin activityand aggression: inverse relationship of prolactin with response to d-fenfluramine,but not CSF 5-HIAA concentrations in human subjects. American Journal ofPsychiatry 154, 1430–1435.

Dolan, M., Deakin, W.J.F., Roberts, N., Anderson, I., 2002. Serotonergic and cognitiveimpairment in impulsive aggressive personality disordered offenders: are thereimplications for treatment? Psychological Medicine 32, 105–117.

Dougherthy, D.M., Bjork, J.M., Cherek, D.R., Moeller, F.G., Huang, D.B., 1998. Effect ofmenstrual cycle phase on aggression measured in the laboratory. AggressiveBehavior 24, 9–26.

Dougherty, D.M., Biork, J.M., Huang, D., Moeller, F.G., 1997. The relationship betweenself-reported menstrual symptomatology and aggression measured in thelaboratory. Personality and Individual Differences 22, 141–149.

Eriksson, C.J., Von der Pahlen, B., Sarkola, T., Seppa, K., 2003. Oestradiol and humanmalealcohol-related aggression. Alcohol & Alcoholism 38, 589–596.

Fink, G., Sumner, B., Rosie, R.,Wilson, H., McQueen, J., 1999. Androgen actions on centralserotonin neurotransmission: relevance for mood, mental state and memory.Behavioural Brain Research 105, 53–68.

Finkelstein, J.W., Susman, E.J., Chinchilli, V.M., Kunselman, S.J., D'Arcangelo, M.R.,Schwab, J., Demers, L.M., Liben, L.S., Lookingbill, G., Kulin, H.E., 1997. Estrogen ortestosterone increase selfreported aggressive behaviors in hypogonadal adoles-cents. Journal of Clinical Endocrinology and Metabolism 82, 2433–2439.

Gerra, G., Avanzino, P., Zaimovich, A., Fertonani, G., Caccavari, R., Del signore, R., Gardini,T., Malarico, E., Lecchini, R., Maestri, D., Brambilla, F., 1996. Neurotransmitter andendocrine modulation of aggressive behavior and its components in normalhumans. Behavioural Brain Research 81, 19–24.

Gerra, G., Zaimovich, A., Avanzino, P., Chittolini, B., Giucastro, G., Caccavari, R., Palladino,M., Maestri, D., Monica, C., Delsignore, R., Brambilla, F., 1997. Neurotransmitter-neuroendocrine responses to experimentally-induced aggression in humans:influence of personality variables. Psychiatry Research 66, 33–43.

Gerra, G., Zaimovich, A., Giucastro, G., Folli, F., Maestri, D., Tessoni, A., Avanzino, P.,Caccavari, R., Bernasconi, S., Brambilla, F., 1998. Neurotransmitter-hormonalresponses to psychological stress in peripuberal subjects: relationship to aggressivebehavior. Life Sciences 62, 617–625.

Girolami, L., Fontani, G., Lodi, L., Lupo, C., 1997. Agonistic behavior, plasma testosterone,and hypothalamic estradiol binding in male rabbits. Aggressive Behaviour 23, 33–40.

Gladue, B.A., 1991. Aggressive behavioral characteristics, hormones, and sexualorientation in men and women. Agressive Behaviour 17, 313–326.

Grieng, H., Sendera, A., Dantendorfer, K., 2001. Naltrexone as a treatment of self-injurious behavior— a case report. Acta Psychiatrica Scandinavica 103, 234–236.

Halbreich, U., 1997. Hormonal interventions with psychopharmacological potential: anoverview. Psychopharmacology 33, 281–286.

Jnoff-Gormain, G., Arnold, G.S., Nottelman, E.D., Susman, E.J., Cutler Jr, G.B., Chrousos, G.P.,1981. Relations between hormone levels and observational measures of aggressivebehavior of young adolescents in family interactions. Developmental Psychology 24,129–139.

Kay, P.A.J., Yourkow, J., Forman, L.J., Chopra, A., Cavalieri, T., 1995. Transdermal estradiolin the management of aggressive behaviors in male patients with dementia.Clinical Gerontology 15, 54–58.

Klaassen, T., Riedel, W.J., Van Praag, H.M., Menheere, P.P.C.A., 2002. Neuroendocrineresponse to metachlorophenylpiperazine and ipsapirone in relation to anxiety andaggression. Psychiattry Research 113, 29–40.

Korzan, W.J., Summers, T.R., Summers, C.H., 2000. Monoaminergic activities of limbicregions are elevated during aggression: influence of sympathetic social signaling.Brain Research 870, 170–178.

Kyomen, H.H., Nobel, K.W., Wei, J.Y., 1991. The use of estrogen to decrease aggressivephysical behavior in elderly men with dementia. Journal of the American SociatyGeriatric Dentistry 39, 1110–1112.

Leibenluft, E., Fiero, P.L., Rubinow, D.R., 1994. Effects of the menstrual cycle on dependentvariables in mood disorder research. Archives of General Psychiatry 51, 761–781.

Marsh, D.M., Dougherty, D.M., Moeller, G., Swann, A.C., Spiga, R., 2002. Laboratory-measured aggressive behavior of women: acute tryptophan depletion andaugmentation. Neuropsychopharmacology 26, 660–671.

McEwen, B.S., 1999. The molecular and neuroanatomical basis for estrogen effects inthe central nervous system. Journal of Clinical Endocrinology and Metabolism 84,1790–1797.

Moss, R.L., Gu, Q., 1999. Estrogen: mechanisms for a rapid action in CA1 hippocampalneurons. Steroids 64, 14–21.

New, A.S., 2005. Regional control of aggression: the role of serotonin. BiologicalPsychiatry 57, 1395.

O'Connor, D.B., Archer, J., Hair, W.M., Wu, F.C.W., 2002. Exogenous testosterone,aggression, and mood in eugonadal and hypogonadal men. Physiology andBehaviour 75, 557–566.

Olweus, D., Mattsson, A., Schalling, D., Low, H., 1988. Circulating testosterone levels andaggression in adolescent males: a causal analysis. Psychosomatic Medicine 42,253–269.

Palanza, P., Parmigiani, S., Liu, H., Vom Saal, F.S., 1999. Prenatal exposure to low doses ofthe estrogenic chemicals diethylstilbestrol and o, p′-DDT alters aggressive behaviorof male and female house mice. Pharmacology Biochemistry and Behaviour 64,665–672.

Pope, H.G., Kouri, E.M., Hudson, J.I., 2000. Effects of supraphysiologic doses oftestosterone on mood and aggression in normal men. A randomized control trial.Archives of General Psychiatry 57, 133–140.

Ritter, D., 2003. Effect of menstrual cycle phase on reporting levels of aggressionusing the Buss and Perry Aggression Questionnaire. Aggressive Behaviour 29,531–538.

Rowe, R., Maughan, B., Worthman, C.M., Costello, E.J., Angold, A., 2004. Testosterone,antisocial behavior, and social dominance in boys: pubertal development andbiosocial interaction. Biological Psychiatry 55, 546–552.

Shelton, P.S., Brooks, V.G., 1999. Estrogen for dementia-related aggression in elderlymen. Annals of Pharmacotherapy 33, 808–812.

Siever, L., Trestman, R.L., 1993. The serotonin system and aggressive personalitydisorder. International Clinical Psychopharmacology 8 (Suppl. 2), 33–39.

Simon, N.G., Whalen, R.E., 1986. Hormonal regulation of aggression: evidence of arelationship among genotype, receptor binding, and behavioral sensitivity toandrogen and estrogen. Aggressive Behaviour 12, 255–266.

Spitzer, R.L., Williams, B.W., 1983. Structured Clinical Interview for DSM-III-UpjohnVersion. Biometric Research Division. New York State Psychiatric Institute, NewYork, N.Y., USA.

Stanley, B., Molcho, A., Stanley, M., Winchel, R., Gameroff, M.J., Parsons, B., Mann, J.J.,2000. Association of aggressive behavior with altered serotoninergic function inpatients who are not suicidal. American Journal of Psychiatry 157, 609–614.

Thiblin, I., Parlklo, T., 2002. Anabolic androgenic steroids and violence. Acta PsychiatricaScandinavica 106, Suppl. 412, 125–128.

Tiihonen, J., Virkkunen, M., Rasanen, P., Pennanen, S., Sainio, E.L., Callaway, J., Halonen,P., Liesivuori, J., 2001. Free l-tryptophan plasma levels in antisocial violentoffenders. Psychopharmacology 157, 395–400.

Toda, K., Saibara, T., Okada, T., Onishi, S., Shizuta, Y., 2001. A loss of aggressive behaviourand its reinstatement by oestrogen in mice lacking the aromatase gene (Cyp 19).Journal of Endocrinology 168, 217–220.

Van Herringen, K., Audenaert, K., Van de Wiele, L., Verstraete, A., 2000. Cortisol inviolent suicidal behaviour: association with personality and monoaminergicactivity. Affetive Disorders 60, 181–189.

Virkkunen, M., Nuutila, A., Goodwin, F.K., Linnoila, M., 1987. Cerebrospinal fluidmonoamine metabolite levels in male arsonists. Archives of General Psychiatry 44,241–247.