Experimental raising or lowering of testosterone level affects mood in normal men and women
Main Article Content
In 2 randomized double-blind placebo-controlled experiments, 63 normal nonclinical men and women received treatments designed to raise or lower their testosterone levels. 16 men and 17 women applied testosterone or placebo gel to their skin for 5-day periods, and 13 men and 17 women ingested capsules of licorice extract (which lowers testosterone) or placebo for 4-day periods. Participants described their feelings on daily check lists, wrote paragraphs summarizing their experience, and completed implicit-attitude measures. Testosterone increased arousal and hostility, and licorice decreased energy and positive affect. Reactions were stronger among women than compared to men.
There have been few experimental studies of administering testosterone to normal, nonclinical human participants. Exceptions include Bhasin et al. (1996), Tricker et al. (1996), and Pope, Kouri, and Hudson (2000), who studied muscle development, mood, and aggression following testosterone injections. Experiments with animals indicate that testosterone increases persistence, boldness, and focused attention (Andrew, 1978; Boissy & Bouissou, 1994). Human correlational studies support the view that testosterone is associated with dominance and success in interpersonal encounters (Mazur & Booth, 1998). Testosterone is related to activities as diverse as choosing an occupation (Dabbs, 1992), getting married (Booth & Dabbs, 1993), committing homicide (Dabbs, Riad, & Chance, 2001), and greeting a stranger (Dabbs, Bernieri, Strong, Campo, & Milun, 2001). It is likely to have antisocial or prosocial effects, depending upon other motivational and socializing conditions (Dabbs & Dabbs, 2000).
The present study explored effects of exogenous testosterone treatments on mood, thought, and behavior. Researchers experimenting with testosterone in clinical populations (Yates, 2000) report positive effects among hypogonadal men (Yates, Perry, MacIndoe, Holman, & Ellingrod, 1999) and postmenopausal women (Davis, 1999), but they do not tell us what would happen with normal, nonclinical populations. The present study included two experiments with normal male and female participants, one on increasing and the other on decreasing testosterone levels.
The experiment on increasing testosterone was guided by the work of Bhasin et al. (1996) and Pope et al. (2000), who treated normal men with 200-600 mg intramuscular (IM) injections of testosterone per week for up to 10 weeks. Pope et al. found that 600 mg/wk increased serum testosterone levels from about 700 to 2700 ng/dl. To place these treatments in perspective, hypogonadal patients typically receive about 200 mg every two weeks via IM injection or 5-10 mg/day via transdermal patch or gel. Transdermal or oral testosterone treatments for postmenopausal women are in the range of 1-3 mg/day. Little is known about how injection, ingestion, and transdermal treatments might affect serum testosterone levels differently.
The experiment on decreasing testosterone was guided by reports that glycyrrhizic acid (GA) in licorice can lower testosterone levels. Consuming 500 mg/day of GA (the equivalent of 3-4 oz of licorice, an amount that many people eat daily) has been reported to reduce serum testosterone levels by about 40% (Armanini, Bonanni, & Palermo, 1999), although more recent work suggests a smaller decrease (Josephs, Guinn, Harper, & Askari, 2001). GA blocks the production of enzymes required for the metabolic conversions from progesterone to androstenedione to testosterone. Although GA can also increase cortisol (Josephs et al.) and decrease potassium, which in turn can increase water retention and blood pressure, it nonetheless provides a means, albeit less than perfect, of lowering testosterone.
Participants in both experiments received treatment and placebo conditions in counterbalanced order. All procedures were carried out with the approval of an Institutional Review Board and with the adequate understanding and written consent of the participants. Participants and experimenters were blind as to which condition was treatment and which was placebo. In both experiments, participants completed daily adjective check lists and wrote posttreatment statements to describe how they felt. In the licorice experiment, participants also completed two measures of implicit attitudes, the Picture Story Exercise (Smith, 1992) and an Implicit Association Test (Greenwald, McGhee, & Schwartz, 1998).
We did not monitor subjects’ serum testosterone levels in these experiments. In prior studies, serum levels were monitored by Bhasin et al. (1996) but not by Pope et al. (2000). Our treatments were comparable to treatments reported to affect serum testosterone in prior studies, and because we wished to examine only indications of effects rather than precise dose-response relationships, we did not monitor subjects' serum levels.
Method Experiment 1: Raising Testosterone Levels
Adult male and female participants applied testosterone or placebo gel to their skin each morning during two 5-day periods. Each evening participants completed adjective check lists describing their feelings, and at the end of each 5-day period they wrote paragraphs describing their reactions to the treatment.
Participants were 16 men and 17 women, mean age 20.2 years (SD = 1.5), recruited from among psychology students and acquaintances of the researchers, participating for $100 payment or credit toward a psychology course requirement. Female participants had to be taking birth control pills, to limit endogenous hormone production and avoid problems that might be associated with pregnancies. Participants reported to a pair of female experimenters on four occasions. On the first occasion they learned about the procedure, signed consent forms, and completed screening questionnaires to report their current health status, history of depression or other psychiatric disorders, medical history relevant to reproductive disorders, and current medications. We dropped one additional participant because of problems indicated on the screening questionnaire. Participants returned once to get experimental materials for the first 5-day period, again to turn in materials and receive new materials for the second 5-day period, and a final time to turn in materials and be paid. We dropped four additional participants who failed to complete the procedure.
Testosterone
Each participant received two plastic syringe-type applicators, each containing 1 ml of fast-absorbing transdermal gel (micronized testosterone in a vehicle of Pluronic F-127 mixed with lecithin; see Cutter, 2001). Placebo applicators contained only gel, and testosterone applicators contained gel combined with either 200 mg (for males) or 50 mg (for females) of testosterone. We labeled the placebo and testosterone applicators “A” and “B” and kept experimenters and participants blind as to this identification. Each applicator was graduated in 0.1 ml units, and participants were instructed to apply 0.2 ml per day, an amount that contained 40 mg testosterone for males and 10 mg for females. Participants applied the gel each morning while getting dressed by depositing it on a fingertip and rubbing it into a fatty portion of the stomach, thigh, or buttocks.
Adjective Check List
Participants completed the 300-item Gough and Heilbrun (1980) Adjective Check List (ACL) each evening to describe their feelings for the day. We analyzed data from the last two days of each 5-day treatment period, because pretests in our laboratory and the clinical experience of one author (A.K.) indicated that participants usually notice effects of testosterone only after 2-3 days of treatment.
From the ACL, we scored the semantic differential dimensions of potency, activity, and goodness (Osgood, Suci, & Tannenbaum, 1957) and the Big Five personality dimensions of neuroticism, extraversion, openness, agreeableness, and conscientiousness (Costa & McCrae, 1985). We did this by computing means, across all items a participant checked, of the item values on each of the eight dimensions, as provided by Williams et al. (Williams & Best, 1990; Williams, Satterwhite, & Saiz, 1998). We did this for each day of the ACL and then computed each participant's mean scores for each dimension in the placebo and testosterone conditions.
In addition to examining the eight dimensions, we analyzed how often the participant checked each of the 300 items in the placebo and testosterone conditions. For each item we computed testosterone minus placebo difference scores for each participant, and across participants we computed the mean and z-score (the mean divided by its standard error) for the testosterone-placebo difference on each item. We sorted the items by z-score to rank them from most characteristic to least characteristic of the testosterone condition. We did this for the sample as a whole and for males and females separately.
Participants’ Reports
At the end of each 5-day period, participants wrote paragraphs describing their experiences on sheets with the following instructions at the top. “In the space below, write a brief paragraph describing your mood, thoughts, and feelings during the last five days. For example, did you feel good, energetic, happy, calm, relaxed, indifferent, nervous, sad, dispirited, impatient? Have you felt different from the way you normally feel? If so, how have your feelings changed over the past five days? Write down whatever impressions you have of the experiment. Thank you for your help.” One subject failed to provide these paragraphs. Two undergraduate judges, blind as to treatment condition, read each participant's paragraphs and scored whether they differed between testosterone and placebo conditions in arousal and in hostility. We instructed the judges that “aroused” meant “energetic, aroused, excited, motivated, important, forceful” and “hostile” meant “angry, aggressive, irritable, moody, upset, unfriendly.” We assigned scores of -1, 0, or +1 to indicate whether each judge scored each of the two qualities as greater with placebo, equal in the two conditions, or greater with testosterone. The correlations between judges on these testosterone-placebo difference scores across the 32 subjects were r = .50 for arousal and r = .49 for hostility. The Spearman-Brown formula, which predicts reliability of a set of measurements from a single measurement, indicated reliability of mean judgments of the two judges to be .67 for arousal and .66 for hostility. We scored participants as more aroused or hostile in one condition or the other only when the judges agreed; when judges did not agree, we scored participants as showing no detectable effect.
Results
Adjective Check List
Mean semantic differential scores and Big Five personality scores were almost identical in placebo and testosterone conditions, and we did not analyze these scores further. However, some individual-item responses appeared to differ in ways not reflected in the overall scores. Among all participants, the three items most associated with placebo were “leisurely, sexy, spunky,” and the three most associated with testosterone were “argumentative, tough, complaining,” suggesting a casual-combative contrast. This pattern differed slightly between men and women. Among women, the extreme items were “sexy, leisurely, conservative” for placebo and “argumentative, impulsive, tough” for testosterone. Among men, they were “indifferent, careless, reserved-dull” for placebo and “aggressive, intelligent, confident" for testosterone. This suggests more of a happiness-antagonism contrast for women and boredom-assertiveness contrast for men. Examining so many items is likely to produce some chance effects, but two factors suggest that what we observed involved more than just chance. First, the mean absolute z-scores for the above items was 2.49, representing areas of the normal distribution where one would expect to find only three of the 300 items by chance alone, not six. Second, the items evoke connotations of energy often associated with testosterone, rather than simply reflecting a random collection of meanings.
Participants’ Reports
Coding of participants' summary paragraphs revealed more arousal and hostility in the testosterone than in the placebo condition (Table 1). We used sign tests to determine whether the treatment effect was significant, ignoring subjects who did not differ between the two conditions (i.e., whose scores had no sign). Subjects reported significantly more arousal and hostility in the testosterone than in the placebo condition. This treatment difference was more noticeable among women than among men.
Table 1. Number of Subjects Reporting More Arousal or More Hostility in the Placebo or Testosterone Condition
* p < .05, ** p < .01, *** p < .001.
Notes: “Aroused” includes “energetic, aroused, excited, motivated, important, forceful;” “Hostile” includes “angry, aggressive, irritable, moody, upset, unfriendly.”
Subjects reporting no difference are excluded.
Asterisks indicate significant differences between conditions between placebo and testosterone, as determined by sign test.
Method – Experiment 2: Lowering Testosterone Levels
Adult male and female subjects consumed GA or placebo capsules each morning daily for two 4-day periods. Each evening they completed adjective check lists describing their feelings, and at the end of each treatment period they wrote paragraphs describing their reactions. Subjects also completed a story-writing exercise and an Implicit Association Test (IAT) (Greenwald et al., 1998), both designed to measure effects on their implicit, or nonverbal, attitudes.
Subjects were 13 men and 17 women, mean age 20.9 years (SD = 3.2), recruited from among psychology students and acquaintances of the researchers, participating for $20 payment or credit toward a psychology course requirement. They reported to a female experimenter on three occasions. On the first occasion they learned about the procedure; signed consent forms; completed screening questionnaires regarding their current health, history of depression or other psychiatric disorders, medical history relevant to potassium metabolism (which might be affected by licorice), and current medications; and received instructions and experimental materials for the first 4-day treatment period. We dropped two additional subjects because of problems indicated on the screening questionnaire. Subjects returned to the experimenter once to turn in materials and receive new materials for the second 4-day period, and they returned again to turn in materials and be paid.
Licorice
We gave each participant two packets, each containing four 00-size gelatin capsules. Each capsule contained 500 mg placebo (avicel cellulose) in one packet and 500 mg GA in the other packet, the latter being the amount reported by Armanini et al. (1999) to lower testosterone levels within 4 days. We labeled placebo and licorice packets “A” and “B” and kept participants and the experimenter blind as to this information. We asked participants to take one capsule each morning upon rising.
Adjective Check List
The ACL procedure was the same as in the testosterone experiment, except that responses were recorded for four rather than five days. We analyzed data from the last two days of each 4-day period.
Participants’ Reports
At the end of each 4-day period, participants wrote paragraphs describing their experiences. We perused these paragraphs and saw no systematic difference between placebo and licorice conditions, and because of this we did not pursue scoring the paragraphs further.
Implicit Attitudes: The Picture Story Exercise
On the evening of the last day of each 4-day period, participants completed a Picture Story Exercise (PSE), following instructions described by Smith (1992). The test included five commonly used pictures: couple at dinner, women in a laboratory, men on a ship, captain, and boxer. Pictures were stapled into a booklet with instructions on the cover asking participants to spend five minutes writing a story about each picture and a lined page after each picture on which to write the story. On a final page in the booklet participants were asked to write, in free-association style, whatever was on their minds at that time. Two undergraduate judges, blind as to treatment conditions, read each participant’s stories in pairs (e.g., a participant’s pair of licorice and placebo writings about the boxer); scored each story on the presence or absence of activity, positive affect, and negative affect; and recorded the number of sentences in each. For each judge we produced scores to indicate the number of stories for each participant that contained activity, positive affect, and negative affect, and to indicate how many stories were shorter in the licorice than in the placebo condition. The correlations between judges on the activity, positive affect, negative affect, and length scores were, respectively, .88, .59, .73, and .85. We combined the data from the judges and assigned each participants mean scores on activity, positive affect, negative affect, and story length in the licorice and placebo conditions.
Implicit Attitudes: IAT
A Palm Pilot version of the IAT became available after the first participants had been run (Dabbs, Bassett, & Dyomina, in press), and five men and eight women participants completed this measure. The IAT measures linkages between two concepts by examining the readiness with which the concepts elicit the same response (Greenwald et al., 1998). For example, if one views “self” and “happiness” as having something in common, one will be able to identify more quickly items belonging to a set that includes “self” and “happiness” words than items belonging to a set that includes “self” and “sadness” words. The Palm IAT examined whether licorice made participants regard themselves as bold (active, energetic, bold) or shy (quiet, shy, cautious) and as helpful (love, help, forgive) or hostile (kill, hurt, attack). It did this by measuring reaction times in associating words about the self with words indicating bold, helpful, shy, and hostile. Each participant kept a Palm Pilot for the duration of the study and took the test once each evening. The Palm Pilot recorded all reaction times, along with the date and time the participant took each test, and we downloaded the information for analysis at the end of the experiment.
Results
Participants’ Reports
As indicated above, participants’ reports showed no indication of difference between licorice and placebo, and we did not analyze these reports further.
Adjective Check List
As with the testosterone experiment, we examined scores from the last 2 days of experimental treatment. Big Five personality scores and semantic differential scores, except for the activity dimension, were almost identical in the placebo and licorice conditions. For the activity dimension, a participants by condition analysis of variance indicated significantly lower activity scores in the licorice than in the placebo condition, means = 48.2 vs. 50.6, F(1,26) = 4.37, p < .05. Examined at the level of individual ACL items, there appeared to be no meaningful difference between items endorsed more in placebo and licorice conditions.
Implicit Attitudes: PSE
Table 2 shows the mean number of stories containing activity, positive affect, and negative affect in each condition. Participants by condition univariate analyses of variance of these scores showed significantly less activity (p < .05) and marginally significantly less positive affect (p < .06) in the licorice than in the placebo condition. Negative affect did not differ between the conditions. Twenty-five of the 30 participants had shorter stories in one condition than in the other; 15 of these had more stories shorter in the licorice condition and 10 had more stories shorter in the placebo condition (nonsignificant by sign test).
* p < .05
Note: Values represent mean number of picture story exercise stories (out of 6) containing activity, positive affect, and negative affect in placebo and licorice conditions.
Implicit Attitudes: IAT
A participants by condition analysis of variance of reaction times in the Bold-Shy IAT indicated that participants were significantly slower (mean = 34 msec) in characterizing themselves as bold than as shy in the licorice than in the placebo condition, F(1, 53) = 6.09, p < .05. A similar analysis of the Helpful-Hostile IAT showed no significant difference (mean = 6 msec) between licorice and placebo conditions in participants' characterizing themselves as helpful or hostile, F(1, 53) = 0.28, ns.
Discussion
The two studies generally complemented each other, although their measures and findings were not identical. Testosterone increased arousal and hostility, and licorice decreased the favorability of implicit attitudes about activity, positive affect, and boldness. Although licorice can also decrease potassium and increase blood pressure and cortisol, its present effects were generally opposite to those of the testosterone gel and are plausibly attributed to lowered testosterone levels. Previous findings that testosterone has only slight effects among normal, nonclinical subjects may have been due to the use of relatively insensitive measures.
Women are usually not included in studies of testosterone, except for studies of sexual behavior and libido following hormone replacement therapy (Sherwin, Gelfand, & Brender, 1985; Shifren et al., 2000). The physiological effects of testosterone are similar in men and women, however, and the present findings indicate that its psychological effects are similar. The apparent stronger reaction among women than among men is puzzling, and two explanations come to mind. The first is that women have less affinity than men for the thoughts, feelings, and actions triggered by testosterone, and as a result they are less comfortable with the way testosterone makes them feel. The second concerns the fact that men and women received different amounts of testosterone. We intended to produce comparable serum increases for the two sexes, relative to their own baselines, but women may have increased proportionally more than men. We did not monitor serum testosterone levels, but the small amounts of testosterone needed to increase women’s levels may have passed through the skin more readily than the large amounts needed to increase men’s levels.
Monitoring of testosterone levels before and after treatment is needed in future studies. Such monitoring could tell us how much participants' reactions depend upon pretreatment testosterone levels and how well testosterone passes through the skin. Skin permeability is an important factor in transdermal application of medical substances (Cutter, 2001; Johnson, Blankschtein, & Langer, 1995; Meikle et al., 1996), and high surface treatment levels may not affect serum levels in all individuals. At present, it is probably better to monitor treatment effects by using serum rather than salivary assays. Unpublished work in our laboratory and published reports elsewhere (Dollbaum & Duwe, 1997) indicate that, for unknown reasons, salivary assays give spuriously high readings following exogenous steroid hormone treatments.
Some effects were observed with implicit but not with explicit measures. This is consistent with what we found in another preliminary study, in which participants reported feeling no difference between testosterone and placebo conditions, although observers could see a difference in their behavior. The implicit PSE and IAT findings regarding activity, affect, and boldness indicated licorice effects, even though the explicit ACL and self-report findings did not. Implicit measures deserve further exploration in hormone research. Rosen and Neugarten (1960) found that the number of characters and amount of conflict, activity, and affect scored in TAT responses decreased with age, consistent with the fact that testosterone decreases with age (Dabbs, 1990). Schultheiss, Campbell, and McClelland, (1999) found that implicit attitudes regarding power and competition interacted with participants’ testosterone levels. Examining implicit attitudes along with other measures will help to free testosterone research from reliance on questionnaire reports. In future research we might examine experimental effects of testosterone on such variables as attention, memory for various kinds of details, sensitivity to sexual stimuli (cf. Alexander & Sherwin, 1991; Anderson, Bancroft, & Wu, 1992), sensitivity to expressions of anger (van Honk et al., 2000), categorization of faces on the basis of sex (Macrae, Alnwick, Milne, & Schloerscheidt, in press), self-presentation strategies, and the attractiveness of potential mates. We can examine these variables in both correlational studies of endogenous testosterone levels, which can employ large numbers of participants and last for extended periods of time, and in experimental studies of exogenous testosterone treatments, which can determine whether it is testosterone, rather than something else, that causes the effect.
Alexander, G. M., & Sherwin, B. B. (1991). The association between testosterone, sexual arousal, and selective attention for erotic stimuli in men. Hormones and Behavior, 25, 367-381.
Anderson, R. A., Bancroft, J., & Wu, F. C. W. (1992). The effects of exogenous testosterone on sexuality and mood in normal men. Journal of Clinical Endocrinology and Metabolism, 75, 1503-1507.
Andrew, R. L. (1978). Increased persistence of attention produced by testosterone, and its implications for the study of sexual behavior. In J. B. Hutchison (Ed.), Biological Determinants of Sexual Behavior. (pp. 255-275). New York: Wiley.
Armanini, D., Bonanni, G., & Palermo, M. (1999). Reduction of serum testosterone in men by licorice. New England Journal of Medicine, 341, 1158.
Bhasin, S., Storer, T. W., Berman, N., Callegari, C., Clevenger, B., Phillips, J., et al. (1996). The effects of supraphysiologic doses of testosterone on muscle size and strength in normal men. New England Journal of Medicine, 335, 1-7.
Boissy, A., & Bouissou, M. F. (1994). Effects of androgen treatment on behavioral and physiological responses of heifers to fear-eliciting situations. Hormones and Behavior, 28, 66-83.
Booth, A., & Dabbs, J. M. Jr. (1993). Testosterone and men's marriages. Social Forces, 72, 463-477. Costa, P. T., Jr., & McCrae, R. R. (1985). The NEO Personality Inventory. Odessa, FL: Psychological Assessment Resources.
Cutter, C. B. (2001). Compounded percutaneous testosterone gel: Use and effects on hypogonadal men, Journal of American Board of Family Practice, 14, 22-32.
Dabbs, J. M. Jr. (1990). Age and seasonal variation in serum testosterone concentration among men. Chronobiology International, 7, 245-249.
Dabbs, J. M. Jr. (1992). Testosterone and occupational achievement. Social Forces, 70, 813-824.
Dabbs, J. M. Jr., Bassett, J. F., & Dyomina, N. V. (in press). The Palm IAT: A portable version of the Implicit Association Test. Behavior Research Methods, Instruments, and Computers.
Dabbs, J. M. Jr., Bernieri, F. J., Strong, R. K., Campo, R., & Milun, R. (2001). Going on stage: Testosterone in greetings and meetings. Journal of Research in Personality, 35, 27-40.
Dabbs, J. M. Jr., & Dabbs, M. G. (2000). Heroes, rogues, and lovers: Testosterone and behavior. New York: McGraw-Hill.
Dabbs, J. M. Jr., Riad, J. K., & Chance, S. E. (2001). Testosterone and ruthless homicide. Personality and Individual Differences, 31, 599-603.
Davis, J. (1999). Androgen replacement in women: A commentary. Journal of Clinical Endocrinology and Metabolism, 84, 1886-1891.
Dollbaum, C. M., & Duwe, G. F. (1997). Absorption of progesterone after topical application: Plasma and saliva levels. Presented at the 7th Annual Meeting of the American Menopause Society.
Gough, H. G., & Heilbrun, A. B. Jr. (1980). The Adjective Check List Manual. Palo Alto, CA: Consulting Psychologists Press.
Greenwald, A. G., McGhee, D. E., & Schwartz, J. L. K. (1998). Measuring individual differences in implicit cognition: The Implicit Association Test. Journal of Personality and Social Psychology, 74, 1464-1480.
Johnson, M. E., Blankschtein, D., & Langer, R. (1995). Permeation of steroids through human skin. Journal of Pharmaceutical Sciences, 84, 1144-1146.
Josephs, R. A., Guinn, J. S., Harper, M. L., & Askari, F. (2001). Licorice consumption and salivary testosterone concentration. The Lancet, 358, 1613-1614.
Macrae, C. N., Alnwick, K. A., Milne, A. B., & Schloerscheidt, A. M. (in press). Person perception across the menstrual cycle: Hormonal influences on social-cognitive functioning. Psychological Science.
Mazur, A., & Booth, A. (1998). Testosterone and dominance in men. Behavioral and Brain Sciences, 21, 335-363.
Meikle, A. W., Arver, S., Dobs, A. S., Sanders, S. W., Rajaram, L., & Mazer, N. A. (1996). Pharmocokinetics and metabolism of a permeation-enhanced testosterone transdermal system in hypogonadal men: Influence of application site ‒ a clinical research center study. Journal of Clinical Endocrinology and Metabolism, 81, 1832-1840.
Osgood, C. E., Suci, G. J., & Tannenbaum, P. H. (1957). The Measurement of Meaning. Urbana, IL: University of Illinois.
Pope, H. G. Jr., Kouri, E. M., & Hudson, J. L. (2000). Effects of supraphysiologic doses of testosterone on mood and aggression in normal men. Archives of General Psychiatry, 57, 133-140.
Rosen, J. L., & Neugarten, B. L. (1960). Ego functions in the middle and later years: A thematic apperception study of normal adults. Journal of Gerontology, 15, 62-67.
Schultheiss, O. C., Campbell, K. L., & McClelland, D. C. (1999). Implicit power motivation moderates men's testosterone response to imagined and real dominance success. Hormones and Behavior, 36, 234-241.
Sherwin, B. B., Gelfand, M. M., & Brender, W. (1985). Androgen enhances sexual motivation in females: A prospective, crossover study of sex steroid administration in the surgical menopause. Psychosomatic Medicine, 47, 339-351.
Shifren, J. L., Braunstein, G. D., Simon, J. A., Casson, P. R., Buster, J. E., Redmond, G. P. et al (2000). Transdermal testosterone treatment in women with impaired sexual function after oophorectomy. New England Journal of Medicine, 343, 682-688.
Smith, C. P. (1992). Motivation and personality: Handbook of thematic content analysis. New York: Cambridge University Press.
Tricker, R., Casaburi, R., Storer, T. W., Clevenger, B., Berman, N., Shirazi, A. et al (1996). The effects of supraphysiological doses of testosterone on angry behavior in healthy eugonadal men A clinical research center study. Journal of Clinical Endocrinology and Metabolism, 81, 3754-3758.
van Honk, J., Tuiten, A., van den Hout, M., Koppeschaar, H., Thijssen, J., de Haan, E. et al. (2000). Conscious and preconscious selective attention to social threat: Different neuroendocrine response patterns. Psychoneuroendocrinology, 25, 577-591.
Williams, J. E., & Best, D. L. (1990). Measuring sex stereotypes: A multination study. New York: Sage.
Williams, J. E., Satterwhite, R. C., & Saiz, J. L. (1998). The importance of psychological traits: A cross-cultural study. New York: Plenum.
Yates, W. R. (2000). Testosterone in psychiatry: Risks and benefits. Archives of General Psychiatry, 57, 155-156.
Yates, W. R., Perry, P. J., MacIndoe, J., Holman, T., & Ellingrod, V. (1999). Psychosexual effects of three doses of testosterone cycling in normal men. Biological Psychiatry, 45, 254-260.
Table 1. Number of Subjects Reporting More Arousal or More Hostility in the Placebo or Testosterone Condition
* p < .05, ** p < .01, *** p < .001.
Notes: “Aroused” includes “energetic, aroused, excited, motivated, important, forceful;” “Hostile” includes “angry, aggressive, irritable, moody, upset, unfriendly.”
Subjects reporting no difference are excluded.
Asterisks indicate significant differences between conditions between placebo and testosterone, as determined by sign test.
* p < .05
Note: Values represent mean number of picture story exercise stories (out of 6) containing activity, positive affect, and negative affect in placebo and licorice conditions.
This research was supported by a grant from the Harry F. Guggenheim Foundation. The authors thank Jessica Frankowski
Mary Smith
and Oliver Schultheiss for help with the Picture Story Exercise
and Cathy Crowley
Mal Wilkinson
and Lou Norton of the Pavilion Compounding Pharmacy for providing the testosterone and licorice treatment materials.
Appreciation is due also to reviewers including
Dr. Alan Booth
Department of Sociology
Penn State University
University Park
PA 16802
USA
and Robert Josephs
PhD
Professor of
Psychology
University of Texas at Austin
Austin
TX 78712
USA.