Female sexual arousal: a behavioral analysis

Transcription

1 FERTILITY AND STERILITY VOL. 80, NO. 6, DECEMBER 2003 Copyright 2003 American Society for Reproductive Medicine Published by Elsevier Inc. Printed on acid-free paper in U.S.A. Female sexual arousal: a behavioral analysis Mary Lake Polan, M.D., Ph.D., M.P.H., a John E. Desmond, Ph.D., b Linda L. Banner, Ph.D., c Michelle R. Pryor, B.A., b Stewart W. McCallum, M.D., d Scott W. Atlas, M.D., b Gary H. Glover, Ph.D., b and Bruce A. Arnow, Ph.D. e Stanford University School of Medicine, Stanford, California Received May 25, 2003; revised and accepted July 3, Supported by a grant from Pfizer, Inc. (Groton, Connecticut) and the National Institutes of Health National Center for Research Resources (Bethesda, Maryland), no. RR09784 Reprints will not be available. Correspondence: Mary Lake Polan, M.D., Ph.D., Department of Obstetrics and Gynecology, Stanford University School of Medicine, 300 Pasteur Drive H333, Stanford, California (FAX: ; E- mail: polan@stanford.edu). a Department of Obstetrics and Gynecology. b Department of Radiology. c Private practice, San Jose, California. d Department of Urology. e Department of Psychiatry /03/$30.00 doi: /s (03) Objective: This study was designed to assess female sexual arousal by using a combination of physiologic measures and self-reported level of arousal. Design: Twenty subjects viewed a 23-minute sequence of randomly ordered relaxation and erotic tapes, both with and without auditory stimulus. The physiologic parameters of vaginal blood flow, galvanic skin resistance, respiration, pulse, and blood pressure, as well as self-reported level of arousal, were simultaneously recorded and correlated with video segments. Setting: An academic teaching hospital. Patient(s): The 20 subjects (mean age SD: years) included Caucasian (10), Hispanic (2), Asian-American (4), and African-American (4) women. All women were screened for normal sexual function with the Female Sexual Function Index (FSFI) and with the Beck Anxiety Inventory and Beck Depression Inventory. Intervention(s): Randomly ordered sequences of erotic and relaxation tapes with and without sound. Main Outcome Measure(s): Physiologic and behavioral data, as well as subjective arousal rating, were acquired. The resulting set of multichannel data was correlated with erotic segments and analyzed for sound vs. no sound and time to maximal physiologic arousal. Result(s): Four independent variables were found to have values that were significantly different from 0: respiration (mean 0.239, SD 0.177, range , t 6.04), VPP (mean 0.158, SD 0.37, range , t 1.91), rvpp (mean 0.161, SD 0.35, range , t 2.075), and erotic marker (mean 0.582, SD 0.191, range , t 13.6). Neither heart rate nor galvanic skin resistance values approached significance. Respiration period was correlated negatively with arousal rating, indicating that subjects breathed faster when aroused. Auditory stimuli during erotic segments did not increase subjective arousal, and for both subjective arousal rating as well as VPP measurement, maximal response occurred within 2 minutes. Conclusion(s): Simultaneous measurement of vaginal blood flow, respiration, pulse, and a variable accounting for the onset and offset of erotic video segments accounts for approximately 50% of the variance in predicting subjective female arousal. Regardless of the presence or absence of audio input, 2 minutes was the average minimum time required to reach maximal arousal in young, sexually functional women. (Fertil Steril 2003;80: by American Society for Reproductive Medicine.) Key Words: Female sexual arousal, vaginal plethysmography, female physiologic arousal patterns Human sexual arousal responses are central to a broad array of behavioral patterns as well as being critical for reproduction. Recent reports of significant sexual dysfunction in women (43%) and men (31%) (1) have encouraged investigation into the physiologic processes of sexual arousal. Several reports in men have emphasized the significant correlation between peripheral sexual arousal and central nervous system activation of the temporal, frontal, and cingulate cortices in the brain (2 4). Most investigations have relied on correlations of self-reported arousal with central nervous system activation as assessed by positron emission tomography scans (2, 3) or functional magnetic resonance imaging (MRI) (4). We have recently reported a close correlation between objective measurements of penile tumescence and central nervous system activation of the right subinsular region, including the claustrum, as measured by functional MRI (5). Far less is known about the physiologic parameters of either central nervous system or 1480

2 peripheral genital arousal in women. Only two studies reporting central nervous system activation have been reported in women using BOLD functional MRI. Park and colleagues (6) reported significant activation of the inferior frontal lobe, cingulate gyrus, thalamus, caudate nucleus, and inferior temporal lobe of women who were viewing sexually arousing video material. Confirmatory evidence was provided by Karama et al. (7), who showed increased activation of the anterior cingulate, frontal insular, and occipital temporal cortices in women. Neither study reported correlations between central nervous system activation and peripheral response. A consistent finding in studies investigating female sexual arousal is the low correlation between measures of subjective arousal and genital response (8, 9). The low concordance does not appear to be due to diminished conscious perception of changes in genital blood flow (9), nor to the use in a variety of studies of male-oriented erotic material. A comparison of female response to male- vs. female-oriented erotic material revealed increased subjective arousal in response to the latter but no enhancement of the correlation between subjective and physiological indices (10). Female sexual response patterns appear to be more complex than male arousal response, perhaps because women s perception of sexual arousal is less directly related to an awareness of genital congestion or other physiologic responses (11). In fact, Meston (12) has argued that in women, external stimulus information more strongly influences subjective arousal than internal, physiologic cues. Vaginal plethysmography (VPP), which assesses changes in vaginal blood flow, has been used as the primary method to assess female genital arousal by documenting increases in vaginal blood flow. Such methodology has provided information on a variety of factors that affect objectively measured female genital engorgement, including anxiety (13), vasodilators such as ephedrine (14) and clonidine (15), and acute exercise (16, 17), in conjunction with erotic stimuli. In addition, functional MRI has been used to measure total clitoral volume and genital blood flow to assess sexual arousal in response to erotic video material (18). However, heart rate, respiration, and skin conductance also undergo change during sexual arousal. The present study of young, sexually functional women was designed to contribute to the development of an objective experimental paradigm for female sexual arousal by focusing on two questions. First, we sought to determine the optimal combination of physiological measures associated with subjective arousal, including heart rate, respiration, galvanic skin response, and vaginal blood flow. Second, we examined the influence on sexual arousal of two stimulus parameters that to our knowledge have not been previously investigated. These include the optimal time of exposure to erotic material and whether the presence of sound enhances response. MATERIALS AND METHODS Participant Selection and Screening This study was approved by the Stanford Medical School institutional review board, and participants were recruited through local newspapers and a popular local Web community message board. Participants completed a three-part screening process that was administered by female researchers who scheduled participation and who also obtained informed consent for participation in the study. The first stage of the screening process was a phone interview that was conducted by a licensed clinical psychologist. Inclusion criteria included the following: aged between years, heterosexual (no history of sexual activities with women), sexual activity within the month preceding the interview, experience watching pornographic movies, no known sexually transmitted diseases, no history of mental illness or substance use, and no evidence of sexual dysfunction. Of the 263 interested women, 182 study participants who met the criteria were provided with information about the study and continued the screening process. During the subsequent screening process, participants completed and returned by mail a series of standard psychological and sexual function questionnaires, along with a signed consent form that provided information about the study. The Beck Anxiety Inventory (19) and the Beck Depression Inventory (20), two widely used standard measures of psychological function, were included in the packet to screen for anxiety and/or depressive disorders. Normal sexual function was assessed using the Female Sexual Function Index (FSFI) (21). Inclusion criteria on the FSFI required a score within 0.5 standard deviation of the mean score for controls (21) on all six subscales (desire, arousal, lubrication, orgasm, satisfaction, pain). Entry criteria scores on the Beck Depression Inventory and Beck Anxiety Inventory were, respectively, 9 and 7. Of the 84 women who completed the mail screening process, 52 met the above criteria and were scheduled for an experimental session. After scheduling, participants were ed a detailed overview of the study, including photographs of the experimental equipment and setting. To maintain the integrity of the study design, participants were not provided with information about the nature of the video content to be used during the experiment. On the day of the study, participants were interviewed in person by a licensed clinical psychologist to ensure adequate comfort with the experimental procedure and to screen for major psychological disorders. The 21 participants who kept their appointment all passed this final screening stage and participated in the experimental protocol. During one session, the VPP malfunctioned and, therefore, data were collected and analyzed for 20 subjects. The 20 subjects ranged in age from years (M SD: years). Ten subjects were Caucasian, 2 were FERTILITY & STERILITY 1481

3 Hispanic, and 4 each were Asian American and African American. Eleven subjects were taking oral contraceptives with regular cycles at the time of testing, whereas 7 subjects (3 in the luteal phase of the cycle and 4 at midcycle) reported regular, spontaneous cycles every days and used condoms for contraception. The contraceptive history of two subjects was not recorded. Experimental Procedures All studies were performed between 5 and 9 PM by female researchers for the convenience of study participants and availability of the experimental laboratory. Participants were queried about the stage of menstrual cycle and method of contraception. They were guided to the small experimental room, which was attractively furnished with a comfortable reclining stretcher and a flat-screened monitor. Control machines, computers, and the equipment operators were concealed in an adjacent room. Participants changed into a hospital gown and undressed from the waist down in privacy; they were then fitted with the physiological monitoring devices, including an abdominal belt to monitor respiration (Piezo Respiratory Belt Transducer; UFI, Morro Bay, CA), a photoplethysmograph (Piezo Electric Pulse Transducer; UFI) on the middle finger of the left hand to measure heart rate, and electrodes attached to the pointer and ring fingers of the left hand to measure galvanic skin resistance (GSR; AD Instruments, Colorado Springs, CO). A gynecologist inserted the VPP (Behavioral Technology SVG-60UL-IR Geer Gauge, Salt Lake City, UT). The VPP was inserted into the posterior vaginal fornix, with a consistent attempt to place the VPP detector at the 12 o clock position. However, preliminary testing with the VPP detector positioned at the 12, 3, 6, and 9 o clock positions had demonstrated that VPP detector orientation did not materially affect measurements of vaginal blood flow. After instrumentation, subjects relaxed in a recumbent position. Participants were instructed to minimize movement and were instructed in use of the potentiometer, a dial with a 1 to 10 scale for measuring feelings of arousal. Data from the potentiometer were recorded via PsyScope ( psyscope.psy.smu.edu). The participant was instructed to block out all distracting stimuli and to try to imagine being a participant in the activities presented. Room lights were then extinguished, and the researcher left the room. The participant started the 22-minute video using a remote control and was presented in privacy with a film containing erotic and nonerotic segments. While watching the video, the participant made continuous ratings of her perceived sexual arousal on the handheld potentiometer, which had a small reading light attached to guide ratings. At the end of the videotape, participants removed all instrumentation themselves, dressed, participated in a debriefing interview with the clinical psychologist, and received compensation for participation. TABLE 1 Sequence of video and audio segments. Video Design Sound first Sound second Video segment 1 first n 5 n 6 Video segment 2 first n 5 n 4 Selection of Video Material A female research team headed by a licensed clinical psychologist who specializes in sexual counseling selected the stimuli. The control stimuli were taken from relaxation tapes that featured nature scenes (e.g., flowers, mountains, ocean, etc.) and gentle instrumental music. The erotic segments came from pornographic videos geared toward women; the videos depicted two couples (1 per segment) who were engaged in a variety of sexual activity and intercourse, with two to three sentences of flirtatious conversation in the beginning of the segment, followed by vocal and physical sounds made during sexual activity and intercourse, as well as background music, all provided by Sinclair Intimacy Institute (Chapel Hill, NC). Each 23-minute sequence began and ended with a 1-minute presentation of a neutral black screen. Participants were then presented with four blocks of content 3-minute control segments (C) followed by 7.5-minute erotic segments (E), in the order CECE. Participants were presented with sound in either the first or second control and erotic segments, with sequences counterbalanced across participants. A total of 21 subjects participated in the experiment. One subject had technical problems that precluded inclusion of her data. The remaining 20 subjects were distributed across the four possible conditions as shown in Table 1. Data Acquisition Physiological and behavioral data were acquired at a sampling rate of 100 Hz using a Powerlab system running under a Macintosh operating system (AD Instruments). Heartbeat data were converted to instantaneous beats per minute, and respiration, to seconds-per-breath measures with conversion utilities available in Powerlab. An erotic marker variable was also recorded, which consisted of a binary signal that indicated the duration of each erotic and nonerotic segment of film. A similar video marker was used to denote onset and offset of the video relative to any preceding or subsequent baseline response. The dependent variable measure, subjective arousal rating, was acquired using a potentiometer that was set to a range of 0 to 10V. To change subjective arousal ratings, subjects adjusted the potentiom Polan et al. Female sexual arousal Vol. 80, No. 6, December 2003

4 FIGURE 1 Heartbeat signal obtained from VPP is seen at Hz. FIGURE 2 Respiration (dotted line) and subjective arousal rating (solid line) for one subject. The graph illustrates a negative correlation between respiration period and subjective arousal. Because respiration period is inversely related to breathing rate, this figure indicates that faster breathing occurred when the subject was more sexually aroused and that breathing returned to a slower rate during the nonerotic video segments. To facilitate comparison of the waveforms, subjective arousal rating and respiration data were converted to Z scores. In this and subsequent figures, thick bars above the time (X) axis and below the waveforms indicate the erotic portions of the video. eter knob and selected the desired arousal rating (on a scale of 0 to 10, where 10 is maximal arousal) under visual guidance of a voltmeter display that was connected to the potentiometer. The resulting set of multichannel data was saved as a text file. Analysis All subjects participated in a debriefing interview at the end of the experimental procedure. None reported physical discomfort with placement or use of the VPP or other physiologic monitoring device. All subjects found the potentiometer that registered level of sexual arousal easy to use. All subjects were comfortable with the interview and intake process as well as with the experimental procedure and physiologic monitoring. Preprocessing Because VPP and GSR data were subject to drift over time, a linear regression with time as the independent variable was used to remove any overall linear trend from these variables. Once this was accomplished, a new variable was created from the VPP data, consisting of the power from the heartbeat signal as measured from the VPP. This new variable was labeled rvpp and was computed by first identifying the heartbeat frequency in the VPP signal (typically at Hz). Identification of the heartbeat signal was accomplished by displaying the power spectrum of the VPP and allowing the user to choose a starting and ending frequency value that corresponded to the beginning and end of the heartbeat bump in the power spectrum, as illustrated in Figure 1. The rvpp signal was then calculated by window filtering the VPP within the selected range and calculating the RMS power (standard deviation) within a sliding 10-second window. With this procedure, higher amplitude of the heartbeat measured from the VPP will result in greater values of rvpp. Because of occasional noise in heartbeat and respiration signal, data were inspected for values that were out of the normal physiological range and excluded from analyses. A nonnormal range was operationally defined for heart rate as 40 beats/min or 120 beats/min, and for respiration, as 0.1 s/breath or 20 s/breath. FERTILITY & STERILITY 1483

5 FIGURE 3 Direct current-coupled VPP response and subjective arousal ratings for two subjects. The panel on the left illustrates a positive correlation between these variables, whereas the panel on the right shows a negative correlation. Thick bars above the time (X) axis and below the waveforms indicate the erotic portions of the video. Determining Temporal Lead Lag Relationships Once the rvpp variable was created, all the data were downsampled to a 0.1-Hz sample rate for subsequent analysis. From the downsampled data, a shifted set of variables from 6 to 6 bins (i.e., 60 seconds [lead] to 60 seconds [lag]) was created for each independent variable. For each subject, a cross-correlation function was created for each independent variable, using the subject s rating of subjective arousal (continuous 0 10 scale) as the dependent variable. The lead or lag value corresponding to the maximum absolute value of the Pearson r was recorded for each subject and for each independent variable, and t tests were performed to test whether the lag value for each independent variable differed significantly from 0. Determining Best Set of Regressors Once the appropriate lead and lag times were determined, each independent variable was regressed on the subjective rating variable over the entire session to determine whether fluctuations in the physiological variable correlated with subjective arousal. The standardized regression weight ( ) for each independent variable was calculated for each subject, and t tests determined whether the values were significantly different from 0. From the pool of independent variables that had statistically significant values, the best set of one, two, three, and four variables was determined, where best was defined as the set that maximized the proportion of accountable variance among all possible combinations. For each set, the mean, standard deviation, and minimum and maximum proportion of accountable variance was calculated. The number of times that each independent variable appeared in the best set of variables was also counted for each set size. Sound Vs. No Sound To determine whether the presence of sound resulted in either higher subjective ratings or physiological measures of arousal, sound and no-sound segments were identified for each subject, using the erotic marker variable to identify the onset of each segment. Subjective arousal rating was integrated over the course of the sound and no-sound segments for each subject, and the difference in the integrated rating was computed for each subject. A paired t test was then computed to determine whether this difference was significantly 0. Similar tests were conducted for the integrated physiological variables. Determining Time Needed for Sexual Arousal Response To assess the minimum time required to transition from nonaroused to aroused states, and from aroused to nonaroused states, time series of subjective arousal, synchronized to the onset or offset, respectively, of the erotic segments, were averaged across subjects. Visual inspection of the time from segment onset to the time of asymptotic arousal rating was used to estimate the minimum video duration that was required to achieve sexual arousal Polan et al. Female sexual arousal Vol. 80, No. 6, December 2003

6 TABLE 2 Frequency of inclusion in the best set of regressors. No. of regressors Erotic marker VPP RVPP Respiration RESULTS Lead Lag Relationships Analysis of lead and lag times for independent variables indicated that only one variable, respiration, showed the greatest correlation with subjective rating at a nonzero lag [t(19) 2.443, P.025]. The correlation for respiration was maximized when it was shifted 1 bin (10 seconds) forward in time, indicating that changes in this variable preceded changes in subjective rating by about 10 seconds. No other variables were found to be significantly different from 0. Best Set of Regressors Four independent variables were found to have values that were significantly different from 0: respiration (mean 0.239, SD 0.177, range , t 6.04, P.0001), VPP (mean 0.158, SD 0.37, range , t 1.91, P.071), rvpp (mean 0.161, SD 0.35, range , t 2.075, P.052), and erotic marker (mean 0.582, SD 0.191, range , t 13.6, P.0001). Neither heart rate nor GSR values approached significance. Respiration period (Fig. 2) was consistently negatively correlated with arousal rating, indicating that subjects tended to breathe faster when aroused. In contrast, both positive and negative correlations were observed for VPP and rvpp variables. Figure 3 illustrates an example of each type of correlation for two different subjects. It is possible that the mixture of positive and negative correlations reflects an orientation or positional sensitivity of the VPP and that the average value for the VPP and rvpp measures underestimates the relationship between these variables with sexual arousal. Measurements of rvpp are independent of heart rate, emphasizing the fact that rvpp measures the amplitude of the vaginal heart rate signal rather than simply the rate (data not shown). To determine how frequently each variable was included in the best set of regressors, regression with all possible combinations of one, two, three, and four variables was performed for each subject. The best set was defined as the combination maximizing the multiple R value. Table 2 depicts the number of times each variable appeared in the best TABLE 3 Proportion of accountable variance for each best set of regressors. No. of regressors Mean SD Min Max set of variables. Table 3 illustrates mean, standard deviation, minimum, and maximum proportion of variance accounted for using the best set of one, two, three, and four regressors. Tables 2 and 3 illustrate that the erotic marker occurs most frequently as the single best predictor of arousal rating and that the addition of one physiological variable can increase the proportion of accountable variance by another 6% 7%. The VPP variable occurs more frequently than rvpp or respiration in the best set of two regressors, although not by an overwhelming degree. Consequently, inclusion of all physiological variables would be the best method for maximizing predictability of subjective arousal. Sound Vs. No Sound Figure 4 illustrates the time course of subjective arousal and loss of arousal, averaged over subjects. As can be seen in Figure 4, there is no apparent difference in either the time course or magnitude of the subjective arousal response for sound vs. no-sound conditions. Consistent with this impression, statistical analysis of the sound vs. no-sound difference, computed by integrating the subjective rating response for the sound segment and the no-sound segment and performing a paired t test, revealed no significant difference. Because the first nonerotic video segment occurred at the very beginning of the experiment, it was possible to view the decline of rating response only after the first erotic video segment, which included sound for half the subjects and no sound for the other half. Similar computations of integrated response and paired t tests for the physiological variables also revealed no significant sound vs. no-sound difference. Determining Time Needed for Sexual Arousal Response Inspection of Figure 4 and Figure 5 reveals that for both subjective arousal ratings and VPP measures, asymptotic response occurred within 2 minutes. Decreases in arousal appear to occur more rapidly, requiring only a minute. DISCUSSION Little is known about the specific physiologic processes that are involved in integrated female sexual response. Although low correlations between genital and subjective response in women have been amply documented, few studies FERTILITY & STERILITY 1485

7 FIGURE 4 Time course of subjective arousal ratings averaged over subjects for erotic (left panel) and nonerotic (right panel) video segments. The left panel illustrates the ratings for sound (solid line) and no sound (dotted line) separately. The right panel depicts the decrease in rating relative to the onset of the second nonerotic segment (representing a mixture of sound and nonsound nonerotic segments). have investigated other physiologic markers that may enhance understanding of female sexual psychophysiology. We undertook this investigation to develop a reproducible, experimental methodology for studying female sexual function by investigating relationships between subjective arousal and several peripheral physiologic measures, in addition to vaginal blood flow, in women who viewed sexually stimulating video material. In addition, we examined relationships between subjective response and two stimulus parameters, namely length of exposure to explicit sexual material and presence or absence of sound. Our findings indicate that simultaneous measures of vaginal blood flow, respiration, pulse rate, and a variable accounting for the onset and offset of erotic video segments account for approximately 50% of the variance in predicting the subjective arousal rating. We did not document a significant advantage to including the audio track for the visual stimuli used in this study. This conclusion could be different for different types of video and audio segments; for example, those employing better actors or emphasizing the storyline rather than the visual aspects of the erotic material. Regardless of the presence or absence of audio, we found that 2 minutes was the average minimum time required to reach asymptotic arousal response. For future studies, a 3-minute minimum might be advisable to maximize correlations between subjective and physiologic parameters. Although we did not incorporate assessment of brain activation in the current study, the experimental paradigm that we described would be suitable for functional MRI and other methods examining relationships between brain activity and female sexual response. Such studies may be extremely useful in bringing about improved therapeutic options for women suffering female sexual dysfunction. Although peripheral vasodilators such as sildenafil have revolutionized the treatment of sexual dysfunction in men, a recent placebo-controlled trial found no advantage for sildenafil in a large sample of women with female sexual arousal disorder (22). Our understanding of the activation of female desire is quite limited (23). Future functional MRI studies correlating brain activation with subjective ratings of arousal and objective physiologic response could increase our understanding of female sexual function. Analyses that use subjective rating as the dependent variable implicitly assume that the subject is always cognizant of changes in sexual arousal, and immediately reflects those changes in her rating response. To the extent that this assumption is incorrect, physiologic measures alone may provide a better indicator of processes occurring in the brain. In addition, although our volunteers were free of sexual dysfunction and of varied ethnicity, we do not know how generalizable their responses are to the total population of women. With these caveats, these physiologic measurements 1486 Polan et al. Female sexual arousal Vol. 80, No. 6, December 2003

8 FIGURE 5 Time course of VPP response for erotic (left) and nonerotic (right) video segments, averaged over subjects. The left panel illustrates the VPP response for sound (solid line) and no-sound (dotted line) separately. The right panel depicts the decrease in VPP relative to the onset of the second nonerotic segment (representing a mixture of sound and nonsound nonerotic segments). may provide a baseline with which to compare women who are sexually dysfunctional and to break down the component parts of sexual arousal so that clear diagnostic criteria can be defined and therapeutic interventions entertained. References 1. Laumann EO, Park A, Rosen RC. Sexual dysfunction in the United States. JAMA 1999;281: Stoleru S, Gregoire M-C, Gerard D, Decety J, Lafarge E, Cinotti L, et al. Neuroanatomical correlates of visually evoked sexual arousal in human males. Arch Sex Behav 1999;28: Redoute J, Stoleru S, Gregoire M-C, Costes N, Conotti L, Lavenne F, et al. Brain processing of visual sexual stimuli in human males. Hum Brain Mapping 2000;11: Park K, Seo JJ, Kang HK, Ryu SB, Kim HJ, Jeong GW. A new potential of blood oxygenation level dependent (BOLD) functional MRI for evaluating cerebral centers of penile erection. Int J Impot Res 2001;13: Arnow BA, Desmond JE, Banner LL, Glover GH, Solomon A, Polan ML, et al. Brain activation and sexual arousal in healthy, heterosexual males. Brain 2002;125: Park K, Kang HK, Seo JS, Kim HJ, Ryu SB, Jeong GW. Bloodoxygenation-level-dependent functional magnetic resonance imaging for evaluating cerebral regions of female sexual arousal response. Urology 2001;57: Karama S, Lecours AR, Leroux JM, Bourgouin P, Beaudoin G, Joubert S, et al. Areas of brain activation in males and females during viewing of erotic film excerpts. Hum Brain Mapping 2002;16: Rosen RC, Beck JG. Patterns of sexual arousal. Psychophysiological processes and clinical applications. New York, NY: Guilford Press, Laan E, Everaerd W, Van der Velde J, Geer JH. Determinants of subjective experience in sexual arousal in women: feedback from genital arousal and erotic stimulus content. Psychophysiology 1995;32: Laan E, Everaerd W, van Bellen G, Hanewald G. Women s sexual and emotional responses to male- and female-produced erotica. Arch Sex Behav 1994;23: Basson R. The female sexual response: a different model. J Sex Marital Ther 2000;26: Meston CM. Sympathetic nervous system activity and female sexual arousal. Am J Cardiol 2000;86:30F 4F. 13. Beggs VE, Calhoun KS, Wolchik SA. Sexual anxiety and female sexual arousal: a comparison of arousal during sexual anxiety stimuli and sexual pleasure stimuli. Arch Sex Behav 1987;16: Meston CM, Heiman JR. Ephedrine-activated physiological sexual arousal in women. Arch Gen Psychiatry 1998;55: Meston CM, Gorzalka BB, Wright JM. Inhibition of subjective and physiological sexual arousal in women by clonidine. Psychosom Med 1997;59: Meston CM, Gorzalka BB. Differential effects of sympathetic activation on sexual arousal in sexually dysfunctional and functional women. J Abnorm Psychol 1996;105: Meston CM, Gorzalka BB. The effects of immediate, delayed, and residual sympathetic activation on sexual arousal in women. Behav Res Ther 1996;34: Deliganis AV, Maravilla KR, Heiman JR, Carter WO, Garland PA, Peterson BT, et al. Female genitalia: dynamic MR imaging with use of MS-325 initial experiences evaluating female sexual response. Radiology 2002;225: Beck AT, Epstein N, Brown G, Steer RA. An inventory for measuring anxiety: psychometric properties. J Consult Clin Psychol 1988;56: Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch Gen Psychiatry 1961;4: Rosen R, Brown C, Heiman J, Leiblum S, Meston C, Shabsigh R, et al. The Female Sexual Function Index (FSFI): a multi-dimensional selfreport instrument for the assessment of female sexual function. J Sex Marital Ther 2000;26: Basson R, McInnes R, Smith MD, Hodgson G, Koppiker N. Efficacy and safety of sildenafil citrate in women with sexual dysfunction associated with female sexual arousal disorder. J Womens Health Gender Based Med 2002;11: Levin RJ. The physiology of sexual arousal in the human female recreational and procreational synthesis. New perspectives in the management of female sexual dysfunctions. Grand Master Lecture. Boston, MA: Boston University School of Medicine and the Department of Urology, October 22, FERTILITY & STERILITY 1487