Archives of Sexual Behavior, Vol. 4, No. 3, 1975
Cognitive and Hormonal Factors Affecting Coital Frequency 1 Cathy J. Spitz, M.A.,2'3 Alice R. Gold, Ph.D.,2 and David B. Adams, Ph.D.2,4
Sexual behavior o f female undergraduates was assessed by daily questionnaires. Of the 24 subjects, 13 were taking oral contraceptives ("pill" subjects) and I1 were using other methods o f birth control ("nonpill" subjects), primarily diaphragm or male prophylactic methods. Three main results were obtained: (1) Intercourse rates were lowest during menstruation and highest immediately following menstruation. (2) Self-rated sexual arousal on a given day correlated with the type o f heterosexual encounters on that day rather than with period o f the menstrual cycle. (3) Pill subjects reported intercourse on more days than nonpill subjects but reported a lower number o f intercourse sessions on days with intercourse than nonpill subjects. These results are interpreted within a general framework o f sexual behavior which recognizes the sexual behavior o f humans as primarily influenced by cultural and cognitive factors. The possibility is discussed that female sexual behavior might also be found to be affected by hormones if more sensitive measures were used. KEY WORDS: coitus; hormones; cognition; contraceptives; menstrual; intercourse.
INTRODUCTION Sexual b e h a v i o r in m o s t m a m m a l s is c h a r a c t e r i z e d b y p e r i o d i c sexual r e c e p t i v i t y o n t h e p a r t o f t h e female c o r r e s p o n d i n g to h o r m o n e f l u c t u a t i o n s at
This research was made possible by Grant M 72.0125C from The Population Council to Cathy J. Spitz and Faculty Research Grants from Wesleyan University to Alice R. Gold and David B. Adams. 1This paper is based on a thesis submitted by the first author in partial fulfillment for the Master of Arts in Psychology at Wesleyan. 2 Department of Psychology, Wesleyan University, Middletown, Connecticut. 3Present address: Division of Psychiatry, Waterbury Hospital, Waterbury, Connecticut. 4 Reprint requests should be addressed to David B. Adams, Department of Psychology, Wesleyan University, Middletown, Connecticut 06457. 249 @1975 Plenum Publishing C o r p o r a t i o n , 2 2 7 West 1 7 t h Street, New Y o r k , N . Y . 1 0 0 1 1 . No part o f this p u b l i c a t i o n m a y be r e p r o d u c e d , stored in a retrieval system, or t r a n s m i t t e d , in a n y f o r m or b y a n y means, e l e c t r o n i c , m e c h a n i c a l , p h o t o c o p y i n g , m i c r o f i l m i n g , recording, o r o t h e r w i s e , w i t h o u t w r i t t e n permission o f the publisher.
250
Spitz, Gold, and Adams
the time of ovulation. In the laboratory rat, for example, the female becomes receptive (i.e., exhibits lordosis when mounted by the male) every fourth or fifth evening and remains receptive until the next morning. At first it would seem that human sexual behavior is unlike that of other mammals. Intercourse takes place throughout the menstrual cycle, and there is not a single period of heightened sexual arousal as obvious as that of most mammals. Careful research suggests, however, that the difference may be only one of degree. Nonhuman primates show peculiarities in sexual receptivity similar to some aspects of human behavior. Initial sexual advances are made fhroughout the menstrual cycle by both male and female rhesus monkeys in the laboratory, a pattern which parallels human behavior. However, during the late follicular phase (i.e., ovulation) both the male and the female are more likely to accept the advances of the other partner (Michael, 1968; Michael and Zumpe, 1970). From other studies it would appear that this increase in acceptance of advances is due to the presence of estrogen and the androgen testosterone (Trimble and Herbert, 1968; Everitt and Herbert, 1971). Research on human sexual receptivity has suggested that coital frequency may peak at three different times during the menstrual cycle. The most generally reported peak immediately follows menstruation (McCance et al., 1937; James, 1971). Other investigators have reported finding additional peaks at midcycle (i.e., ovulation) and immediately preceding menstruation (McCance et al., 1937; Udry and Morris, 1970), although results of a reanalysis of data and a replication of one of the studies question the existence of both these peaks (James, 1971; Udry and Morris, 1968). The midcycle peak is the only one for which hormonal explanations have been offered. At this time estrogen and an androgen (androstendione), two hormones which have been found to be positively related to sexual behavior in lower animals, are thought to be high (Speroff and Van de Wiele, 1971). It is difficult to advance hormonal explanations for the peaks preceding and following menstruation, at which times estrogen and androstendione are secreted by the ovary in very small amounts. Rather, cognitive factors stemming from the anticipation of abstinence from intercourse and actual abstinence from intercourse during menstruation would seem to be involved. James (1971), for example, has suggested that the postmenstrual peak cannot be accounted for by heightened sexual arousal on the part of the female, but may be due to a "husband effect," i.e., male initiation of intercourse because of the preceding period of abstinence. Hormonal influences on sexual arousal in the human female have been investigated in studies involving the use of oral contraceptives, which prevent normal fluctuation of sex hormones during the menstrual cycle. One known study has investigated cyclic effects. Udry and Morris (1970) found that the coital frequency of subjects assigned to take oral contraceptives did not differ significantly from that of subjects assigned placebos except during the luteal
Factors Affecting Coital Frequency
251
phase of the cycle. Subjects taking placebo pills (hence having a normal fluctuation of hormones) showed a decline in rate of intercourse during the luteal phase, while the intercourse rate of subjects taking oral contraceptives remained
high. Many studies have compared overall coital frequency, regardless of cycle day, of women taking oral contraceptives with that of women using less sophisticated means of contraception, such as the diaphragm or prophylactic foam. No known study of this type has found a lower frequency of intercourse among women taking birth control pills. Some have reported equivalent rates for the two groups (e.g., Morris and Udry, 1971) and others have reported a higher level for women using oral contraceptives (e.g., Westoff et al., 1969). The present study was designed to attempt to clarify some of the issues by instituting several procedures not used by previous investigators. Daily questionnaire forms asked for reports of all types of sexual experiences (not just intercourse), the initiator of each experience, and the frequency of rejection of sexual advances. In addition, different methods of analyzing intercourse rates were used. We hoped that these procedures would help in interpreting causes for cyclic changes in coital frequency, such as the proposed "husband effect," and overall differences between pill-taking and non-pill-taking women.
METHOD
Subjects Twenty-four female undergraduate students, aged 18-24, served as subjects. Thirteen subjects (pill subjects) had been taking low estrogen dosage oral contraceptives (0.05-0.08 mg) for at least 6 months preceding the onset of the study. Eleven subjects (nonpill subjects) were currently using non-oral-contraceptive methods of birth control (seven subjects used diaphragms, the partners of three subjects used male prophylactics, and one subject used an intrauterine device). All subjects met the following requirements: (1) were engaged in an ongoing heterosexual relationship in which the subject had an opportunity to spend time with boyfriend(s) at least five times per week; (2) had at some time attained a level of sexual experience which included intercourse and orgasm; (3) had demonstrated in a preliminary interview a frank and apparently healthy attitude toward her own sexuality. Procedure
All subjects l~dled out daily questionnaire sheets on which they recorded all sexual experiences including intercourse, caressing, masturbation, and fanta-
252
Spitz, Gold, and Adams
sy, the initiator of each heterosexual experience (mutual, self, or partner), whether subject rejected an initiation by the partner, and self-rating of overall sexual arousal. Subjects also monitored their own bodily responses to autosexual and heterosexual experiences by means of a checklist of 12 physiological/psychological indices of sexual arousal. Subjects deposited completed questionnaires with the experimenter at the end of each week, at which time they received additional questionnaires for the upcoming week. Twenty-two subjects completed questionnaires for three cycles and two subjects completed them for two cycles. Subjects completed the questionnaire on 1775 cycle days, 95% of all possible cycle days. Cycle lengths included in the study ranged from 24 days to 35 days. Four cycles, from two subjects (20, 21, 40, 41 days), could not be calculated accurately according to the method chosen for cycle standardization and were not included in the data analysis. Interviews
All subjects were individually interviewed twice during the study. The preliminary interview acquainted the subject with the nature of the study and the procedure to be used in f'dling out the daily questionnaire. This interview allowed the experimenter to gather demographic data on each subject, level of sexual experience, and contraceptive history. At this time subjects were advised as to the confidentiality of all information collected. A final interview at the end of subjects' participation was conducted to record subjects' insights and reactions to the study and to pay subjects for their participation. In addition to the two scheduled interviews, the experimenter was available to answer questions throughout the study. Cycle Standardization A new method was used to combine data from menstrual cycles of different lengths. Methods of previous investigators were rejected for reasons given by James (1971). Instead, information collected by James (1972) on ovulation time as estimated by conception and by temperature rise was used to divide differentlength menstrual cycles into comparable lengths. The following periods were distinguished: menstrual I, menstrual II, postmenstrual, ovulatory, postovulatory, luteal, and premenstrual. They are shown for various cycle lengths in Fig. 1. The divisions are centered around menses and estimated time of ovulation. For subjects who were not taking the pill, menses was divided into two periods corresponding to heavy and light flow. The menstrual I period was considered as the first 3 days of the cycle, while the menstrual II period comprised all other succeeding menstrual days. The duration of the menstrual II period varied from cycle to cycle but averaged 3 days for nonpill
Factors Affecting Coital Frequency
253 DAY OF MENSTRUAL CYCLE
al .J
I
5
10
15
i--
28
0 T 0 T 0 T 0 T 0 T
uJ
30
G
32
0 ~ 0
24 o ,~
z~ uo
26
c
50
35
T TI T T
IT
IT
o 0 0
36
25 | '~
0
g 34 "~
ZO
T
I TI
g <[ z
..', :E
a.
0
a.
PERIOD OF M E N S T R U A L
CYCLE
Fig. 1. Calculation of periods of the menstrual cycle for menstrual cycles of different durations. Ovulation day (0) for each cycle duration was estimated from data collected by James (1972) based on conception dates. Day of basal temperature rise (T) for each cycle duration was estimated from data collected by James (1972).
subjects. Menses for pill subjects was usually shorter and lighter and was always considered as menstrual II period. Ovulation day for each cycle was estimated from the data collected by James (1972) on conception dates, and the ovulation period was considered as the 4-day period of which ovulation estimated by conception was the second day. The luteal period was distinguished following the suggestion by Udry et al. (1973) that sexual activity during this period is affected by use of contraceptive pills. It was considered to take full effect on the third day after the mean day of basal body temperature rise, as reported by James (1972), and to extend until 3 days prior to menses. The postmenstrual and postovulatory periods were bounded by the menstrual II and ovulatory periods and the ovulatory and luteal periods, respectively. The premenstrual period consisted of the 3 days preceding menstruation.
RESULTS
Overall Comparisons of Pill and NonpiU Subjects Table I presents summary data for measures of intercourse, caressing experiences, initiators of intercourse, and female rejection of male sexual advances. Two separate measures of intercourse are shown. The first measure, hereafter
254
Spitz, Gold, and Adams
Table I. Mean Rates of Sexual Intercourse, Caressing, Initiation, and Rejection of Partner Advances
Intercourse
Initiation
Percent of women reporting intercourse/day ("intercourse-days") Number of intercourse experiences/ intercourse-day Percent of women reporting mutual initiation of intercourse]day Percent of women reporting male initiation of intercourse/day Percent of women reporting female initiation of intercourse/day
Caressing without intercourse
Percent of women reporting caressing/ day Number of caressing experiences/ caressing-day
Rejection
Percent of women reporting rejection of partner advances/day
Nonpill subjects (N= 11)
Pill subjects (N= 13)
Total sample (N = 24)
0.19
0.38
0.29
1.44
1.13
1.27
0.15
0.22
0.19
0.03
0.11
0.07
0.02
0.05
0.03
0.30
0.30
0.30
1.68
1.48
1.57
0.06
0.07
0.07
called intercourse-days, represents the percentage of subjects reporting at least one intercourse experience on a particular day; the second measure represents the number of intercourse experiences on days on which intercourse occurred. Similar measures for caressing without intercourse are also shown. The data indicate that pill-taking subjects had a significantly higher number of intercourse-days than nonpill subjects (t = 3.55, df = 22, p < 0.01). The nonpill subjects, however, demonstrated a significantly higher number of intercourse experiences per intercourse-day (t = 2.39, df = 21, p < 0.05). The two measures of caressing as well as the number of days on which advances by the male partner were rejected and the number of days of mutually initiated intercourse showed no differences between the two subject groups. The significant difference between pill and nonpill subjects for number of days of male-initiated and female-initiated intercourse reflects the already noted difference between the groups on intercourse days (t = 3.49 and 2.01, respectively, df = 22, p < 0.01 and 0.10). The subjects were asked to make daily recordings of general sexual arousal on a 5-point scale, " 1 " representing no arousal and " 5 " representing very intense arousal. Preliminary inspection of the data suggested that the ratings were related to the daily pattern of sexual experiences. Therefore, arousal scores were broken down into three categories: degree of arousal on "intercourse-days," degree of arousal on "caressing-but-no-intercourse-days," and degree of arousal on "other-days." The means are presented in Table II. No significant differences between the pill and nonpill subjects were found. It is, however, noteworthy
Factol~ Affecting Coital Frequency
255
Table II. Mean Self-rated Sexual Arousal as a Function of the Nature of Sexual Experience Nonpill subjects (N = 11)
Pill subjects (N = 13)
Total sample (N = 24)
3.47
3.38
3.42
2.91 1.63 2.35
2.92 1.82 2.61
2.92 1.73 2.47
Intercourse-days Caressing-but-nointercourse-days Other-days Total
that self-reported sexual arousal was highest on intercourse-days, next highest on caressing-but-no-intercourse-days, and lowest on the other-days. This rank order of means was exhibited by all 24 subjects (p < 0.01). Cyclic Variations Each cycle was standardized according to the manner described in the Method section. Figure 2 shows the data for number of intercourse-days plotted for pill and nonpill subjects separately. It suggests that, while the pill subjects had consistently higher numbers of intercourse-days, the cyclic pattern of intercourse was similar for the two groups. Both pill and nonpill subjects had the fewest number of intercourse-days during menses and the greatest number dur50
O.---,-O N O N P I L L SUBJECTS O---e PILL SUBJECTS
7",. t
4O w
/
d
30
z~
zo
~"
10
uJ
~'O
O
I
=
I
I
I
• :t
z
~
"
2
PEBrOD
I
>=
OF M E N S T R U A L
CYCLE
Fig. 2. Percent of subjects reporting at least one intercourse experience on a day ("intercourse-days") as plotted by standardized menstrual cycle.
Spitz, Gold, and A d a m s
256
ing the postmenstrual period. These results support the fmdings of James (1971). In reanalyzing data from several previous studies (McCance et aL, 1937; Udry and Morris, 1968; and others), he concluded that there was evidence only of a postmenstrual peak in coitus and not of one at midcycle or one preceding menstruation. The intercourse-day curves were broken down into three components representing the number of days on which intercourse was reported to have been initiated by both partners mutually, by the male, and by the female. The data are plotted in Figs. 3-5. Variations in mutual initiations, which comprised the largest proportion of all initiations of intercourse, were virtually identical to those found for intercourse-days (see Fig. 1). Male initiations, although constituting a much smaller proportion of intercourse initiations, followed a somewhat similar pattern. The number of female-initiated intercourse experiences was too small to detect any meaningful changes in levels throughout the cycle. The data for number of intercourse experiences per intercourse-day are shown in Fig. 6. The shape of the curve for the pill subjects was again similar to that of their intercourse-day curve. The pattern for nonpill subjects, however, was somewhat different. While the rate was low during the first part of menstruation (first 3 days), it rose during the second part of menstruation and appeared to reach its peak during ovulation rather than during the postmenstrual period. There was no evidence of systematic cyclic variations for either measure of caressing or for the number of days on which male sexual advances were rejected.
401
/JO~"~"~O~,C------O SUBJECTS O----ONONPILL PILLSUBJECTS
~ 30
10
o'""
]
I
:3
:3
z
z
~E
~E
'~
I
]
i
u.I
0
>
n_
0
CL
z
I
D
Z
~
PERIOD
OF
MENSTRUAL
CYCLE
Fig. 3. Percent o f subjects reporting at least one m u t u ally initiated intercourse experience on a day as plotted by standardized menstrual cycle.
Factors Affecting Coital Frequency
257
NONPILL
40
PILL
o---o
° w
SUBJECTS
SUBJECTS
so
~, 2 0 L~ o
10
1:
ii:
o
'~
,,z
a:
z
o
~-
o
o
o o_
ta
z w
0
-J
PERIOD OF MENSTRUAL
CYCLE
Fig. 4. Percent of subjects reporting at least one maleinitiated intercourse experience on a day as plotted by standardized menstrual cycle.
Self-reports of general sexual arousal also showed no fluctuations over the cycle. These data are shown in Fig. 7. Pill and nonpill subjects are combined for ease of presentation. The curves demonstrate dramatically the high correlation between general sexual arousal and the type of sexual experiences reported to have occurred during that day. The mean arousal levels remained constant throughout the cycle, and at no time did the rates for intercourse-days, caressing-but-no-intercourse-days, or other-days overlap. When the three curves 40
(D w
~
NONPILL
O----I
PILL
SUBJECTS
SUBJECTS
30
m
t 10-o_
H
D
H
~
I
H
~
~
o
>
i
z
~
~
~
o
~
w
z
,
~
J
D
F
~
w
w
o
>
o
D
m z w
PERIOD
OF
MENSTRUAL
l
~
D
w
CYCLE
Fig. 5. Percent of subjects reporting at least one female-initiated intercourse experience on a day as plotted by standardized menstrual cycle.
258
Spitz, Gold, and Adams
C~==O N O N P I L L SUBJECTS o - - - - e P I L L SUBJECTS
2.50
~J 2 . 0 0 z
z uJ
o
w
~ ~ 1.50 ID
1.00
I H o~ z
D z
z
LU
W
~
j
,~
o
PERIOD OF M E N S T R U A L
CYCLE
Fig. 6. Mean number of intercourse experiences per intercourse-day, plotted by standardized menstrual cycle.
4,00
C
~
I
N
T
E
R
~
3.00
CARESSING-DAYS
2,00 OTHER DAYS
1.00 w
0.00
I
I
I
J_
I
2 < J D
D
D
z w
o> J z w
z w
~
o
o
a_
PERIQD OF MENSTRUAL
D
J
CYCLE
Fig. 7. Mean rating of general arousal for days on which intercourse occurred, days on which caressing but no intercourse occurred, and other days, plotted by standardized menstrual cycle.
Factors Affecting Coital Frequency
259
were combined the data showed a postmenstrual peak in arousal simply because intercourse-days were more frequent during that period.
DISCUSSION
Three findings of the present study are particularly important in the search for a general understanding of human female sexual behavior: (1) the postmenstrual peak and menstrual low point in intercourse during the cycle of menstruation, (2) the correlation of sexual arousal with type of sexual encounter, and not with periods of the menstrual cycle, and (3) the effect of oral contraceptives, which led to a greater number of intercourse-days and a smaller number of intercourse sessions per intercourse-day among women taking oral contraceptives compared to women using a diaphragm or women whose partners used a male prophylactic. We believe that all three sets of results can best be interpreted within the general framework of human sexual behavior advanced by Ford and Beach (1951). After an encyclopedic survey of sexual behavior across mammalian species and various human cultures, they concluded that throughout the higher primates, and most strongly in man, eroticism has become progressively less influenced by hormones and more influenced by learning and cultural conditioning. In the following discussion, we contend that the results of this study show little sign of hormonal influence, and seem, instead, to be influenced by cultural and cognitive factors. Several lines of evidence suggest that hormonal effects played a minimal role in the results of the present research. If hormones were a significant factor, one would have expected a peak in intercourse rates and in sexual arousal during or immediately preceding ovulation, the period corresponding to relatively high levels of estrogen and testosterone in the human female and corresponding to the hormonally influenced estrus of other mammals. Instead, we found that the frequency of intercourse-days peaked at the postmenstrual period of the menstrual cycle and was lower at the ovulatory period. Also, we found no correlation between self-rated sexual arousal and the period of the menstrual cycle. In addition, the ingestion of hormones in the form of oral contraceptives should have altered the normal pattern of intercourse rates during the menstrual cycle. However, our data suggested similar cyclic variations in intercourse rates for both pill and nonpfll groups. Data from a previous study investigating the effect of hormones on coital frequency offer additional support. Morris and Udry (1971) found no difference in intercourse rates between a group of women randomly selected to use oral contraceptives and another group randomly selected to use placebos. This finding deserves special attention because hormonal factors and cognitive factors associated with different contraceptive methods were not confounded. All subjects were instructed to continue using their usual (nonpill) means of contraception.
260
Spitz, Gold, and Adams
Cognitive and cultural factors appear most useful in explaining the three major findings o f the present research. The low frequency o f intercourse-days during menstruation and the high frequency directly following menstruation have previously been interpreted by Ford and Beach in terms of the fact that in nearly every society women are prohibited from engaging in any form o f sexual activity during the period o f menstrual flow. They suggest that the increased incidence o f intercourse after menstruation may be a phenomenon o f rebound from this socially determined abstinence (Ford and Beach, 1951, p. 213). The realization that one has not been able to engage in intercourse for a number o f days may create a psychological need to engage in the activity, a phenomenon analogous to reactance (Brehm, 1966). If this is the case, our data suggest that the effect is due to b o t h partners and not just to the male partner, as in the "husband effect" proposed b y James (1971), since both mutually initiated and male-initiated intercourse exhibited the postmenstrual peak. Further research might be directed toward a clearer understanding o f the rebound effect: would it be found after any period o f enforced abstinence, such as illness or travel, or is it, for some reason, specific to the days following menstruation? It is difficult to explain the second major finding, the correlation o f selfrated sexual arousal with types o f sexual encounter, by the usual drive-reduction model. Since female initiation o f sexual behavior occurred in only a small proportion o f the cases, it is unlikely that the correlation could be interpreted as reflecting heightened sexual drive causing the woman to seek out and initiate heterosexual experiences. It seems more likely that the cause-effect direction was reversed. Women inferred their degree o f sexual arousal from the nature o f their experiences. Bem (1967) has called this kind of cognitive process "self-perception. ''s The third finding, that pill women had a higher number of intercourse days b u t a lower number o f intercourse sessions per intercourse day, is again best explained by cognitive factors. Two possible, although not mutually exclusive, explanations will be discussed. 6 The first, which we shall call the "anxiety o f unwanted pregnancy" or " a n x i e t y " explanation, hypothesizes that intercourse is Sex differences in this process would be very interesting to examine. Would males show the same tendency to infer sexual arousal from sexual experience? Past research has characterized the female as field dependent (Witkin et al., 1954) and sexually repressed (Douvan, 1970), whereas males have been seen as highly responsive to sexual drives. If this is the case, one would expect a lower correlation between self-rated sexual arousal and presence or absence of heterosexual experiences among men than among women, because the male's arousal score would be subject to additional variation due to factors of availability and willingness of partner. 6 A third possible explanation for the greater number of intercourse-days for pill as opposed to nonpill subjects is self-selection. Women who take birth control pills might choose to do so because they desire (or their partners desire) a more active and frequent sex life than women who do not take the pill. Evidence from interviews with our subjects, however, provides little support for such an interpretation. In addition, if this were the case, one would expect the frequency of intercourse sessions per intercourse-day also to have been greater for pill subjects.
Factors Affecting Coital Frequency
261
normally inhibited to a certain extent by the anxiety of one or both partners over the prospect of an unwanted pregnancy; that because of the high reliability of oral contraceptives subjects using this method would have less anxiety and hence more intercourse. The second, which we shall call the "contraceptive-convenience" explanation, hypothesizes that intercourse is normally inhibited to a certain extent by the nuisance or delay of using a diaphragm or male prophylactic, and that the use of oral contraceptives frees the subject and her partner from this inhibition. While both explanations can account for the lower number of intercoursedays among nonpill subjects, the contraceptive-convenience explanation can better account for their higher rate of intercourse per intercourse-day. Just as the failure to be prepared for intercourse might frequently inhibit intercourse among nonpiU subjects relying on use of a diaphragm or male prophylactic, so the prior preparation for intercourse might be expected to increase the probability of intercourse. Having already prepared for intercourse once, the subject and her partner might be especially influenced to take advantage of their preparation and proceed to an additional session of intercourse. This effect would not be expected for pill subjects since no special preparation is necessary for them. It is also interesting that two previous studies which compared pill-taking women specifically with non-pill-taking women who had IUDs (Herzberg e t al., 1971) or whose husbands had had vasectomies (Rodgers and Ziegler, 1968) found all three groups to show an initial increase in libido after the contraceptive method was begun. 7 Since the major features which these three methods (i.e. pills, IUDs, and vasectomies) have in common are not hormones but rather their relatively low failure rates and the fact that their use does not interrupt the intercourse experience, the similarity of results again strongly suggests the importance of either or both of the cognitive explanations suggested above. In conclusion, the results of the present study corroborate the view that human female sexual behavior is less influenced by hormonal factors and more influenced by social and cognitive factors than the sexual behavior of other primates, which in turn has progressively evolved in the same direction from the sexual behavior of other mammals. The issue remains open, however, as to whether hormones affect any aspects of human female sexual behavior. Although daily ratings of intercourse and caressing and the self-assessment of sexual arousal by a single rating scale do not appear to be substantially influenced by hormonal effects, such indices may not be the most sensitive measures for this. We have found that intercourse and caressing are highly dependent on male initiation or male participation in the initiation and that self-assessment of sexual arousal is highly dependent on heterosexual experience. The task of search7Libido, in the study by Herzberg et al. (1971), was assessed by combining data from three separate self-rating scales on sexual interests, frequency of intercourse, and frequency of sexual climax per intercourse. In the study by Rodgers and Ziegler (1968), the measure of libido was based simply on frequency of intercourse.
262
Spitz, Gold, and Adams
ing for m o r e sensitive m e a s u r e s o f v a r i a t i o n in h u m a n female sexual r e c e p t i v i t y a n d initiative, i n d e p e n d e n t o f social factors, is still a c h a l l e n g i n g one. It seems u n l i k e l y t h a t a m e c h a n i s m s u c h as estrus, w h i c h h a s p l a y e d s u c h a n i m p o r t a n t role in t h e sexual b e h a v i o r o f a n i m a l s p h y l o g e n e t i c a l l y r e l a t e d t o m a n , w o u l d b e t o t a l l y a b s e n t f r o m h u m a n sexual b e h a v i o r .
ACKNOWLEDGMENTS The a u t h o r s wish to t h a n k Gflda Oliver for h e r h e l p in a n a l y z i n g the d a t a a n d Lorelei B r u s h a n d Delores K r e i s m a n for t h e i r c o m m e n t s o n an earlier d r a f t o f this article.
REFERENCES Bem, D. (1967). Self-perception: An alternative interpretation of cognitive dissonance phenomena. PsychoL Rev. 74: 183-200. Brehm, J. W. (1966). A Theory o f Psychological Reactance, Academic Press, New York. Douvan, E. (1970). New sources of conflict at adolescence and early adulthood. In Bardwick, J., Douvan, E., Horner, M., and Gutmann, D. (eds.), Feminine Personality and Conflict, Brooks/Cole, Belmont, Calif. Everitt, B. J., and Herbert, J. (1971). The effects of dexamethasone and androgens on sexual receptivity of female rhesus monkeys. J. Endocrinol. 51: 575-588. Ford, C. S., and Beach, F. A. (1951). Patterns o f Sexual Behavior, Harper, New York. Herzberg, B. H., Draper, K. C., Johnson, A. L., and Nicol, G. C. (1971). Oral contraceptives, depression and libido. Brit. Med. J. 3: 495-500. James, W. H. (1971). The distribution of coitus within the human intermenstruum. J. Biosoc. Sci. 3: 159-171. James, W. H. (1972). Cycle day of ovulation. J. Biosoc. Sci. 4: 371-378. McCance, R. A., Luff, M. C., and Widdowson, E. E. (1937). Physical and emotional periodicity in women. J. Hyg. 37: 571-611. Michael, R. P. (1968). Gonadal hormones and the control of primate behavior. In Michael, R. P. (ed.), Endocrinology and Human Behavior, Oxford University Press, London. Michael, R. P., and Zumpe, D. (1970). Rhythmic changes in the copulatory frequency of rhesus monkeys in relation to the menstrual cycle: A comparison with the human cycle. J. Repro& Fertil. 21: 199-201. Morris, N. M., and Udry, J. R. (1971). Sexual frequency and contraceptive pills. Soc. Biol. 18: 40-45. Rodgers, D. A., and Ziegler, F. J. (1968). Changes in sexual behavior consequent to use of noncoital procedures of contraception. Psychosom. Med. 30: 495-505. Speroff, L., and Van de Wiele, R. L. (1971). Regulation of the human menstrual cycle. Am. J. Obstet. Gynecol. 109: 234-247. Trimble, M. R., and Herbert, J. (1968). The effect of testosterone or estradiol upon sexual and associated behavior of adult female rhesus monkey. J. EndocrinoL 42: 161-186. Udry, J. R., and Morris, N. M. (1968). Distribution of coitus in the menstrual cycle. Nature 220: 593-596. Udry, J. R., and Morris, N. M. (1970). Effect of contraceptive pills on the distribution of sexual activity in the menstrual cycle. Nature 220: 502-503. Udry, J. R., Morris, N. M., and Waller, L. (1973). The effect of contraceptive plUs on sexual activity in the luteal phase of the human menstrual cycle. Arch. Sex. Behav. 2: 205-214.
Factors Affecting Coital Frequency
263
Westoff, C. F., Bumpass, L., and Ryder, N. B. (1969). Oral contraceptive, coital frequency and the time required to conceive. Soc. BioL 16: 1-10. Witkin, H. A., Lewis, H. B., Hertzman, M., Machover, K., Meissner, P. B., and Wapner, S. (1954). Personality Through Pereeption, Harper, New York.