Arch Sex Behav

DOI 10.1007/s10508-010-9639-7

ORIGINAL PAPER

Are There Different Types of Female Orgasm?

Robert King • Jay Belsky • Kenneth Mah •

Yitzchak Binik

Received: 22 July 2009 / Revised: 22 January 2010 / Accepted: 13 March 2010

 Springer Science+Business Media, LLC 2010

Abstract In attempt to identify and validate different types Keywords Female orgasm
Evolution
of orgasms which females have during sex with a partner,
data collected by Mah and Binik (2002) on the dimensional
phenomenology of female orgasm were subjected to a typo- Introduction
logical analysis. A total of 503 women provided adjectival
descriptions of orgasms experienced either with a partner There remains great debate about the biological function of
(n = 276) or while alone (n = 227). Latent-class analysis female orgasm. The fact that many sexual species reproduce
revealed four orgasm types which varied systematically in successfully in the absence of (apparent) female orgasm cer-
terms of pleasure and sensations engendered. Two types, tainly suggests that female orgasm is not necessary for repro-
collectively labelled ‘‘good-sex orgasms,’’ received higher duction. Yet some evolutionary scholars contend that female
pleasure and sensation ratings than solitary-masturbatory orgasm is an adaptation sculpted by natural selection to increase
ones, whereas two other types, collectively labelled‘‘not-as- reproductive fitness in some way (Baker & Bellis, 1993a,
good-sex orgasms,’’ received lower ratings. These two 1993b; Thornhill, Gangestad, & Comer, 1995), with fitness
higher-order groupings differed on a number of psycholog- defined in terms of the dispersion of genes in future gener-
ical, physical and relationship factors examined for purposes ations (Cronin, 1991). Others argue that female orgasm exists as
of validating the typology. Evolutionary thinking regarding a mere by-product of another (male) adaptation; whereas strong
the function of female orgasm informed discussion of the selection created the sensitive male penis, the clitoris is sim-
findings. Future research directions were outlined, especially ply its (inadvertent) physical homologue (Gould, 1987; Lloyd,
the need to examine whether the same individual experiences 2005; Symons, 1979). By this latter account, female orgasms
different types of orgasms with partners with different only exist because males were selected to have orgasms as a
characteristics, as evolutionary theorizing predicts should be proximate reward for sexual activity.
the case. There has been much debate about whether female orgasm
is an evolutionary adaptation directly designed by natural
selection to promote (individual) reproductive fitness (e.g.,
Barash, 2005; Puts, 2006; Zuk, 2006). In appreciation of this,
R. King (&)
J. Belsky Judson (2005) suggested that consideration and investigation
Institute for the Study of Children, of different types of orgasm might shed light on this issue. As
Birkbeck University of London, London WC1B 3RA, UK it turns out, females, in contrast to males, report their orgasms
e-mail: spiersking@hotmail.com
differ not only in intensity but also in location, phenome-
K. Mah nology, and emotional components (Hite, 1976). Indeed, this
Behavioural Sciences Research Division, Toronto General insight is widely appreciated by sex researchers (Bentler &
Hospital, University Health Network, Toronto, ON, Canada Peeler, 1979; Levin, 1981, 1998, 2001, 2004; Levin & Wag-
ner, 1985; Singer & Singer, 1972) and sex therapists (Fisher,
Y. Binik
Department of Psychology, McGill University, 1973; Sundahl, 2003). It provides one basis for the work
Montreal, QC, Canada presented herein which sought to (1) identify and (2) validate

123
 Arch Sex Behav

different types of female orgasm in hopes of advancing crucial areas of the vagina during intercourse (Schultz, van
understanding of the female sexual experience and the evo- Andel, Sabelis, & Mooyaart, 1999).
lutionary basis of female orgasm. In sum, it is unsurprising that Masters and Johnson (1965,
Regardless of whether female orgasm is considered from 1966) did not detect any uterine peristalsis effect, regarded
an adaptationist perspective (Baker & Bellis, 1993a, 1993b; by some as a distinguishing feature of some orgasms (Fox,
Thornhill et al., 1995), an anti-adaptationist position (Gould, Wolff, & Baker, 1970) having effectively sidelined potentially
1987; Lloyd, 2005), or some other frame of reference (Masters crucial evolved physiologies of males and females. The fact that
& Johnson, 1965, 1966), it remains the case that many scholars research into copulatory orgasms by Fox et al. (1970) revealed a
implicitly or explicitly embrace the notion that all female peristaltic-related insuck effect, however, leads to the hypoth-
orgasms are essentially the same. Even if brought about esis that different types of orgasms will prove detectable, even in
by different means (Masters & Johnson, 1965, 1966), female the context of intercourse. The insuck effect involves a pressure
orgasms have been held to vary only in terms of timing rel- change in the uterus via peristaltic action which could con-
ative to male partner orgasm (Baker & Bellis, 1995; Thornhill ceivably enable females to ‘‘select’’ sperm preferentially from
et al., 1995). But Masters and Johnson’s (1965, 1966) con- particular mates and thus serve a fitness (i.e., evolutionary)
clusion that all female orgasms are more or less the same may function (Fox et al., 1970; Fox & Fox, 1971; Wildt, Kissler,
have been a result of their methods of inquiry. Not only did Licht, & Becker, 1998; Zervomanolakis et al., 2007). Baker
they use a rigid glass insertable to allow for internal filming, and Bellis’ (1993b) related and often-cited work chronicling
but in many experiments they inserted metal specula. This, upsuck following masturbation (rather than intercourse) has
crucially, covered areas of the anterior vaginal wall now regarded defied replication (Pound & Daly, 2000). For this reason, this
as especially sensitive and important in orgasm (Komisaruk & article retains the original; Fox et al. (1970) term insuck to
Sansone, 2003; Komisaruk et al., 2004). emphasize continuity with that earlier line of research.
Limited understanding of the anatomy may also have con- Wildt et al. (1998) have shown how this insuck might be
tributed to the view that all female orgasms are much the same. related to sperm transport in the uterus in mammals. Their work
Research reveals that the clitoris is a much larger and more raises the possibility that some orgasms will be experienced
complex organ than commonly assumed (Dickinson, 1949; differently than others. Oxytocin is a well-known correlate of
O’Connell, Hutson, Anderson, & Plenter, 1998; O’Connell, female orgasm (Anderson & Dennerstein, 1994, 1995; Blaicher
Sanjeevan, & Hutson, 2005). Unlike intercourse, masturbation et al., 1999; Carmichael et al., 1987; Carmichael, Warburton,
typically only involves stimulation of what should be more Dixen, & Davidson, 1994); producing anaesthetic, floating
properly termed the glans of the clitoris (O’Connell et al., sensations while making one more trusting of others (Kosfeld,
1998). This area is typically reported by females as the most Heinrichs, Zak, Fischbacher, & Fehr, 2005; Marazzitti et al.,
sexually sensitive and is a key area to be retained during gen- 2006; Zak, Kurzban, & Matzner, 2005). Oxytocin also con-
itoplasty to preserve such sexual sensitivity (Schober, Meyer- tributes to uterine contractions as experienced in orgasm or in
Bahlburg, & Ransley, 2004). But, the visible external tip of the childbirth (Russell, Leng, & Douglas, 2003). Vagino-cervical
clitoris is not the only site of female sexual sensitivity. Even stimulation has been shown to prompt oxytocin, via the Fer-
women who have had all the outer labia and clitoris removed guson reflex, to generate a range of sexually important neuro-
still, surprisingly, experience orgasm (Lightfoot-Klein, 1984, logical effects (Ferguson, 1941; Komisaruk et al., 1996, 2004;
1989). Komisaruk & Sansone, 2003). Wildt et al. (1998) described
As many have noted, while it is certainly true that stimu- the entire system as a peristaltic pump for transporting sperm
lation of the glans of the clitoris is typically the fastest and with clear links to fertility (Zervomanolakis et al., 2007).
most reliable way to produce a female orgasm, it is by no Zervomanolakis et al. found, during administration of oxytocin,
means the only way to do so; nor is it the only site at which or insuck of suitable material into that fallopian tube ipsilateral to
by which females can experience orgasm (e.g., Bentler & the dominant ovary. They further observed that pregnancy rate
Peeler, 1979; Komisaruk et al., 2004; Schober et al., 2004; was higher in those women in whom such ipsilateral transport
Singer & Singer, 1972). A variety of neural pathways in could be demonstrated; this clearly suggests that this system is
the vaginal area, including deep inside as far as the cervix, functional and could contribute to fitness, just as would be
can also trigger orgasm, quite independent of the clitoral expected of an evolutionary adaptation. Thus, there is a good
pathway. Indeed, this can occur even in (human and animal) case for a proximate mechanism for potential differential sperm
cases where the spinal cord has been completely severed selection via oxytocin-rich female orgasms, orgasms which
(Komisaruk et al., 1996, 2004; Komisaruk & Sansone, 2003). would seem fundamentally different from ones that do not
Researchers have also pointed out that the Ferguson reflex, involve as much oxytocin and thus do not induce, psychologi-
the release of oxytocin through vagino-cervical stimulation cally, floating sensations and physiologically, peristaltic action.
(Ferguson, 1941), is a key feature of sexual intercourse in In seeking to identify different types of female orgasms,
mammals (Komisaruk et al., 2004). Penises curve to fit the the present effort draws on data collected by Mah and Binik

123
Arch Sex Behav

(2002) on the phenomenology of female orgasms, obtained wide range of university disciplines), a psychology subject
by asking women about the extent to which adjectival descrip- pool, and adverts posted on Internet list-servs which were
tors (e.g., general spasming, ecstasy) accurately characterized likely to have student members. They were administered a
their orgasmic experience. Although they did not collect these questionnaire concerning either solitary masturbation (n =
data with the aim of identifying different types of orgasms or to 227) or sex with a partner (n = 276). The solitary-masturba-
advance understanding of the possible adaptive function of tory and sex-with-partner groups averaged, respectively,
female orgasms, here we use their data to test the proposition 23.0 (SD = 7.3) and 22.2 (SD = 5.6) years of age; 77.5/76.1%
that there are different types of female orgasms, with some described themselves as undergraduates, 9.3/10.5% as gradu-
characteristics to be associated with oxytocin release, insuck, ates; 79.7/83.0% described themselves as primarily hetero-
and sperm selection. Thus, in the present study, after subject- sexual, 3.1/2.5% as primarily homosexual, and 4.9/6.2% as
ing the Mah and Binik data to latent-class analysis in hopes of primarily bisexual; 12.3/8.3% did not answer this question.
identifying interpretable types of orgasms, we conducted two Solitary-masturbatory and sex-with-partner group members
sets of validational analyses. First, we asked whether any types described themselves, respectively, as 30.4/29.4% single,
of the heterosexually-induced orgasms identified differed phe- 36.6/47.1% with a partner but not living together, and 14.5/
nomenologically from what is orgasmically experienced dur- 12.7% as living together or married, while 7.5/2.9% fitted
ing masturbation. Because Mah and Binik only obtained data none of these relationship categories. Further sample details
from each respondent on orgasms experienced during inter- can be found in the original Mah and Binik report.
course or during masturbation (rather than during each), this
comparison, like the subsequent one, is necessarily a between- Procedure and Measures
subjects one. Next, we determined whether different types of
orgasms experienced during intercourse differed with respect Whether assigned to one group or another, participants were
to select aspects of heterosexual activity; we tested specific asked to rate a list of 27 adjectives in terms of the extent to
hypotheses—which are delineated when analyses and results which each characterized their most recent (1) solitary (i.e.,
are presented—based on thinking about oxytocin effects and masturbatory) orgasm or (2) orgasm with partner, however
sperm selection. These hypotheses involved orgasmic sensa- attained. Each adjective was rated on a 6-point scale (0: does
tions, emotions, sexual behavior, and location of orgasm. not describe it; 5: describes it perfectly). An initial list of 60
adjectives was reduced by means of a series of conceptual and
empirical methods, including multiple principal component
Method analyses presented in the original Mah and Binik (2002)
report, resulting in a final list of 27 adjectives that were then
Participants used to create 10 constructs displayed in Table 1. A (mean)
score (also displayed in Table 1) for each construct, which
Participants in the original Mah and Binik (2002) research, could range from 0 (lowest) to 5 (highest), was created by
some of whose data were re-analyzed for this report, were summing ratings given to each component item and dividing
undergraduate and graduate students. After carrying out a the resultant sum by the number of items included in the
pilot study to develop measurements of orgasm experience, construct. These 10 summary orgasm-experience composite
503 women were enrolled in a subsequent investigation, most scores were used in this study.
of whom were young, unmarried, and heterosexual. These In addition to having participants characterize their
women were recruited through in-class solicitation (from a orgasm, Mah and Binik (2002) queried them about their

Table 1 Adjectival component items and composite constructs of Mah and Binik’s (2002) 10 dimensions of female orgasm
Composite Building Flooding Flushing Shooting Throbbing General Pleasurable Relaxation Emotional Ecstasy
dimensions sensations sensations sensations sensations sensations spasms satisfaction intimacy

Component Swelling Flowing Spreading Spurting Pulsating Quivering Pleasurable Soothing Loving Elated
adjectives Building Flooding Flushing Shooting Throbbing Trembling Fulfilling Relaxing Tender Ecstatic
Shuddering Satisfying Peaceful Unifying Euphoric
Close Rapturous
Passionate
M 2.44 2.44 2.57 1.52 3.33 2.99 4.08 2.53 2.36 2.72
SD 1.39 1.45 1.3 1.34 1.34 1.43 0.89 1.49 1 49 1.25
Absolute 0–5 0–5 0–5 0–5 0–5 0–5 0–5 0–5 0–5 0–5
range

123
 Arch Sex Behav

sexual behavior and relationships, with each question being Table 2 BIC model fit statistics for latent-class analysis of orgasmic
responded to using a 6-point scale (0 = very low; 5 = very typology in sex with partner context
high). Of importance for this report were questions relating to No. clusters BIC(LL) BIC(L2)
happiness (M = 4.31, SD = 0.91), satisfaction (M = 4.14,
1 14729.72 10078.80
SD = 0.94), and emotional closeness (M = 4.47, SD = 0.94)
regarding the sexual relationship. Participants were asked to 2 14458.46 9807.54
rate the subjective length of the orgasm on a 6-point scale 3 14361.46 9710.54
(0 = very short; 5 = very long time; M = 2.87, SD = 1.07). 4 14344.76 9693.84
Using a different 6-point scale (0 = very weak, 5 = very 5 14346.32 9695.40
strong), participants also rated the intensity of both physical 6 14363.77 9712.85
(M = 3.65, SD = 1.04) and non-physical sensations (M = Note: LL refers to log likelihood
3.65, SD = 1.21) of the orgasm. Additional questions con-
cerned how the orgasm was brought about—irrelevant for the component per orgasm type graphically. The 41 cases whose
solitary context which constituted 45.1% of the total, other- orgasms qualified as Type I, labelled ‘‘High Pleasure and Sen-
wise through oral sex (9.7%), intercourse (18.5%), mastur- sations,’’ described them as manifesting the most building,
bation by self (1.4%), masturbation by partner (11.5%) or flooding, flushing, spurting, and throbbing sensations, as well as
other (13.0%). Of interest, too, were questions about where in involving the most general spasm, relaxation, and emotional
the body the orgasm was generated; participants were offered intimacy. The Type II orgasms of 159 participants were labelled
nine (non-mutually-exclusive) binary, yes/no options (with ‘‘High Pleasure, Medium Sensations’’ and defined by high
per cent answering yes indicated): centered around outer scores on components relating to pleasure but noticeably lower
genitals (31.8%); started in outer genitals, spread deeper on scores relating to internal sensations and feelings of relaxa-
(47.3%); centered deep inside (16.5%); centered in whole tion. Next, there were 46 Type III, labelled ‘‘Medium Pleasure
pelvic area only (2.8%); spread to whole pelvic area (22.5%); and Sensations,’’defined by relatively low scores in areas relat-
centered in other parts of the body only (2.6%); spread to ing to internal sensations and in components describing pleasure
other parts of the body (37.2%); centered in whole body and satisfaction. Finally, 30 cases described their orgasms as
(4.4%); and spread to whole body (27.2%). manifesting the least of these attributes and were labeled Type
IV,‘‘Low Pleasure and Sensation.’’

Results Validating Orgasm Typology

Identifying Types of Female Orgasm During Sex Two sets of analyses were conducted to validate the fourfold
with Partner typology of orgasms. The first sought to determine whether
any or all the types of sex-with-partner orgasms identified in
In an effort to identify types of orgasms experienced in the the latent-class analysis differed from solitary-masturbatory
partner context, scores on the 10 composite measures of par- orgasms in terms of how they were described; the second set
ticipants in the sex-with-partner group were subject to latent of analyses compared the four orgasms subtypes which
class analysis, using the latent class analysis software Latent emerged from the latent class analysis of data from the sex-
Gold (Vermunt & Magidson, 2000). Latent-class analysis can with-partner group on a set of external correlates.
be considered a probabilistic extension of K-means cluster
analysis, with the advantages that it is model-based, thereby Partnered Orgasms Versus Solitary/Masturbatory
affording use of statistical criteria for determining various dif- Orgasms
ferent cluster solutions. The analyst can choose between solu-
tions by using the Bayesian Information Criterion (BIC). The Presumably, if the types of partnered orgasms have any
model with the lowest BIC is generally preferred because it validity, they should differ from those achieved by solitary
indicates a good balance between model fit and parsimony, masturbation. To test this proposition, a MANOVA was
defined as having relatively few parameters. Models can be conducted to determine whether the average solitary mas-
created through sequentially relaxing assumptions regarding the turbatory orgasm differed from the average partnered orgasm
covariance structure of the indicators. Given that it had the (i.e., averaged across the composite of the four types) across
lowest BIC of alternatives, a four-class model appeared to be the the 10 composite orgasm-experience constructs; ANOVAs
best fit for the data (see Table 2). were carried out following the MANOVA. After comparing
Table 3 presents the mean scores of the four identified average solitary with average partnered orgasms, we exam-
orgasm types on each of the 10 composite orgasm-experience ined how each of the four identified types of partnered orgasm
constructs. Figure 1 presents the mean scores for each of differed from the average solitary masturbatory orgasms on

123
 Arch Sex Behav

Table 3 Means of orgasm subgroups on adjectival composite constructs

Adjectival constructs Partner orgasms Solitary orgasms
Type I Type II Type III Type IV Masturbation
High pleasure/sensation High pleasure/medium sensation Medium pleasure/sensation Low pleasure/sensation
n = 41 n = 159 n = 46 n = 30 n = 227
M SD M SD M SD M SD M SD

Building sensations 3.68a 1.14 2.68b 1.10 1.22c 1.06 1.38c 1.50 2.43b 1.39
a b c c
Flooding sensations 3.74 1.04 2.72 1.26 1.60 1.33 0.87 1.14 2.39b 1.41
Flushing sensations 3.94a 0.98 2.79b 0.99 1.77c 1.19 1.67c 1.24 2.44d 1.33
Spurting sensations 2.77a 1.37 1.78b 1.26 0.70c 0.90 0.57c 0.91 1.40d 1.30
Throbbing sensations 4.18a 1.05 3.69a,d 1.01 1.99c 1.38 2.20c 1.44 3.35d 1.30
a a b b c
General spasms 3.97 0.88 3.58 1.05 1.79 1.10 1.51 1.15 2.84 1.48
Pleasurable satisfaction 4.88a 0.21 4.13b 0.63 4.53a 0.51 3.10c 1.17 3.93b 0.97
Relaxation 3.56a 1.33 2.20b 1.36 2.56b,c 1.52 0.92d 1.17 2.78c 1.42
Emotional intimacy 4.40a 0.60 2.81b 1.13 3.54c 1.06 1.15d 1.03 1.59d 1.28
Ecstasy 4.25a 0.71 2.96b 0.87 3.22b 0.91 0.99c 0.67 2.41d 1.26
Absolute range 0–5 0–5 0–5 0–5 0–5
Means in the same row that do not share superscripts differ at p\.05 in the Bonferroni post hoc test of difference

123
 Arch Sex Behav

partner orgasms, F(10, 502) = 21.03, p\.001, g2 = .30 using

Wilks’ lambda as a statistic. Comparisons of the average orgasm
with partners and the average solitary masturbatory orgasm
indicated that the latter yielded significantly less flushing sen-
sation, general spasming, pleasurable satisfaction, emotional
intimacy, and ecstasy, but more relaxation (see Tables 3, 4). No
significant differences were found with respect to building,
flooding, spurting, or throbbing sensations.

Comparing each Type of Sex-with-Partner Orgasm with

Solitary-Masturbatory Orgasms

In order to determine whether each specific type of orgasm

with a partner differed from solitary masturbatory ones on
these (and other) phenomenological characterizations, four
MANOVAs were conducted comparing each type with sol-
itary masturbation across the 10 dependent constructs before
carrying out follow-up ANOVAs (see Table 4) and follow-up
Bonferroni comparisons across all five groups (see Table 3).
Fig. 1 Comparison of the component mean scores of the orgasm types MANOVAs revealed a significant overall difference across
the 10 dependent constructs reflective of orgasmic experi-
each of the 10 composite orgasm-experience constructs. ence between average solitary-masturbatory and (1) Type I
Means scores included in all comparisons are displayed in orgasms, F(10, 267) = 20.53, p\.001, g2 = .44, (2) Type II
Table 3, with Table 4 presenting results of significance tests orgasms, F(10, 385) = 14.88, p\.001, g2 = .28, (3) Type III
of specific group comparisons across the 10 orgasm-experi- orgasms, F(10, 272) = 23.68, p\.001, g2 = .48, and (4) Type
ence dependent constructs. IV orgasms, F(10, 256) = 9.75, p\.001, g2 = .28, all using
Wilks’ lambda as a statistic.
Average Solitary-Masturbatory Versus Average As displayed in Table 3, Bonferroni comparisons yielded
Sex-with-Partner Orgasms many significant pairwise differences between each of the four
orgasm types identified by means of latent-class analysis and the
MANOVA revealed a significant overall difference across the average solitary masturbatory one. Inspection of means dis-
10 dependent constructs reflective of orgasms experience played in Table 3 reveals the following differences, among
between average solitary-masturbatory and average sex-with- many identified: High-pleasure/high-sensation orgasms with a

Table 4 Comparison of orgasm subgroups on adjectival composite constructs

Adjectival composite Type I high pleasure/ Type II high pleasure/ Type III medium Type IV low pleasure Solitary (n = 227)
constructs sensation with partner med sensation with pleasure/low sensations & sensation with vs. partner (n = 276)
(n = 41) vs. solitary partner (n = 159) vs. with partner (n = 46) vs. partner (n = 30) vs.
(n = 227) solitary (n = 227) solitary (n = 227) solitary (n = 227)
F (1, 267) g p F (1, 385) g p F (1, 272) g p F (1, 256) g p F (1, 502) g p

Building sensations 29.65** .10 .01 3.52 .01 ns 31.65** .11 .01 14.95** .06 .01 \1 .00 ns
Flooding sensations 34.40** .12 .01 5.71* .02 .02 12.19** .04 .01 32.12** .11 .01 \1 .00 ns
Flushing sensations 46.89** .15 .01 7.84** .02 .01 10.04** .04 .01 9.12** .04 .01 3.82* .01 .05
Spurting sensations 37.82** .12 .01 8.33** .02 .01 12.17** .04 .01 11.49** .04 .01 3.26 .01 ns
Throbbing sensations 14.75** .05 .01 7.69** .02 .01 40.91** .13 .01 20.14** .07 .01 \1 .00 ns
General spasms 22.18** .08 .01 29.39** .07 .01 20.84** .07 .01 22.23** .08 .01 4.61* .01 .03
Pleasurable satisfaction 39.08** .13 .01 5.57* .01 .02 16.76** .06 .01 18.41** .07 .01 11.71** .02 .01
relaxation 10.76** .04 .01 15.89** .04 .01 \1 .00 ns 46.63** .16 .01 11.75** .02 .01
Emotional intimacy 190.42** .42 .01 94.48** .20 .01 94.60** .26 .01 3.17 .01 ns 141.53** .22 .01
Ecstasy 82.88** .24 .01 23.07** .06 .01 17.43** .06 .01 36.56** .13 .01 28.03** .05 .01
* p\.05, ** p\.01

123
Arch Sex Behav

partner were rated on all composite component variables con- on hypotheses pertaining to the role of oxytocin in facilitating
sistently and significantly higher than solitary masturbatory ones. sperm-selecting orgasms; depending on the nature of the
High-pleasure/medium-sensation orgasms were also rated sig- dependent variable, either MANOVA/ANOVA or v2 were
nificantly higher than solitary-masturbatory ones on all compos- used to make these comparisons.
ite measures, with the single exception of building sensations.
Interestingly, medium-pleasure/medium-sensation orgasms Measures of Sensations
with a partner scored significantly lower than the solitary
masturbatory ones on all measures except relaxation, with the The first set of comparisons focused on three sensations—
same being true of low-pleasure/low-sensation orgasms with subjective orgasm length and intensity of physical and non-
partner, though the exception this time involved emotional physical sensations. Subjectively long-lasting orgasms, as dis-
intimacy. tinct from ones that take a long time to bring about, were pre-
Further, and not surprisingly, orgasms achieved through dicted to occur disproportionately when orgasms were classified
solitary masturbation were rated as significantly less emotion- as ‘‘good sex.’’ The second prediction was that women experi-
ally intimate than the high-pleasure/high-sensation orgasms. encing good-sex orgasms would report higher intensity of
More interesting perhaps was that low-pleasure/low-sensa- physical sensations based on the theoretical premise that such
tion orgasms with a partner were characterised as no more orgasms involve internal sensations related to the peristalsis of
emotionally intimate than those achieved by oneself. Indeed, insuck creation (Fox et al., 1970; Wildt et al., 1998; Zervo-
low-pleasure/low-sensation orgasms achieved with partners manolakis et al., 2007). The third prediction was less common-
(Type IV) were rated as less relaxing and ecstatic than soli- sensical, namely, that good-sex orgasms would be associated
tary masturbatory orgasms. Additionally, in terms of spas- with higher ratings of intensity of non-physical sensations. The
ming, both the low- and medium-pleasure orgasms achieved reasoning behind this prediction was that if sperm-selecting
with partners (Types III and IV) were rated lower than soli- orgasms are mediated by oxytocin release, then along with the
tary masturbation orgasms. uterine contractions associated with the action of oxytocin, spe-
In sum, whereas some orgasms achieved with a partner— cific non-physical dreamy, floaty sensations that also accom-
Types I and II—proved to be of higher quality than those pany oxytocin should also be intensely experienced following
achieved on one’s own, perhaps meriting the label‘‘good-sex these orgasms. A MANOVA showed a significant effect for
orgasms,’’ other, seemingly ‘‘not-as-good-sex’’ orgasms orgasm type (good sex vs. not as good sex) across the three
achieved with a partner—Types III and IV—proved to be of dependent variables, F(3, 270) = 12.54, p\.001, g2 = .12 using
lower quality than those experienced all by oneself. The latter Wilks’ lambda as a statistic. Individual ANOVAs provided
observation was based also on the fact that solitary mastur- empirical support for all three predictions (see Table 5).
bation orgasms scored fairly high in terms of pleasurable
satisfaction compared to high-pleasure/medium-sensation Measures of Emotions
orgasms, falling only a little short of the high-pleasure/high-
sensation orgasms on this measure. In short, solitary mas- The emotional and affective aspects of the sexual relationship
turbatory orgasms fell roughly in the middle of the range on in which the orgasm occurred were the focus of the next three
most measures relative to the four-class model of orgasm comparisons. Commonsensically and in line with what is
with a partner identified by means of latent-class analysis. known about the affiliative effects of oxytocin, it was
expected that good-sex orgasms would occur in the context of
Further Validational Analyses (1) greater happiness in and (2) satisfaction with the current
relationship, and would (3) engender feelings of greater
On the basis of the results just reported, it was concluded that emotional closeness to the sex partner than would not-as-
rather than validating each of four distinct types of orgasms to good-sex orgasms. A MANOVA showed a significant effect
emerge from the latent-class analysis, the most confident for orgasm type (good sex vs. not as good sex), F(3,
conclusion that could be drawn was that two more general 165) = 4.063, p\.01, g2 = .07 using Wilks’ lambda as a
types of orgasms during sex with a partner could be distin- statistic. Individual ANOVAs provided empirical support for
guished, putative ‘‘good-sex’’ and ‘‘not-as-good-sex’’ ones. all three predictions (see Table 6).
Thus, Types I and II were combined to create Type A (i.e.,
good-sex orgasms) and Types III and IV were combined into Sexual Behavior During Orgasm
Type B (i.e., not-as-good-sex orgasms) for use in the second
stage of orgasm-typology validation. The validational anal- In contrast to expectations perhaps derived from Masters and
ysis focusing on the additional measurements of sexual Johnson’s (1965) conclusion that there is one type of female
behavior and relationship characteristics involved a series of orgasm brought about by stimulation of the outer genitalia, it
comparisons of good-sex and not-as-good-sex groups based was predicted that fewer good-sex orgasms would be brought

123
 Arch Sex Behav

Table 5 Comparison of means and ANOVA of external correlates of orgasm sensation descriptions by orgasm type
External correlate: Orgasm type (sex with a partner context) ANOVA results
Type A, good-sex (n = 84) Type B, not-as-good-sex (n = 187) F (1, 269) g p
M SD M SD

How long orgasm seemed to last 3.35 1.01 2.89 1.09 10.39** 0.04 .001
Intensity of physical sensations 4.04 0.87 3.70 1.00 7.06** 0.03 .008
Intensity of non-physical sensations 4.49 0.69 3.73 1.12 32.76** 0.11 .001
Absolute range 0–5 0–5
** p\.01

Table 6 Comparison of means and ANOVA of external correlates of relationship descriptions by orgasm type
External correlate Orgasm type (sex with a partner context) ANOVA results
Type A, good-sex (n = 61) Type B, not-as-good-sex (n = 105) F (1, 164) g p
M SD M SD

Happiness with current relationship 4.64 0.58 4.26 0.86 9.379** 0.05 .003
Satisfaction with current relationship 4.49 0.65 4.07 0.89 10.612** 0.06 .001
Emotional closeness to sexual partner 4.70 0.53 4.36 0.97 6.478* 0.04 .012
Absolute range 0–5 0–5
* p\.05, ** p\.01

about through stimulation of outer genitalia, that is, through oxytocin in terms of creating general well being and deep
oral sex or masturbation (in the presence of a partner) rather peristaltic effects. The data proved consistent with the pre-
than via intercourse. To test this hypothesis, the variable per- diction, v2(1, n = 276) = 19.46, p = .005. Whereas 15% of
taining to type of sex associated with orgasm was recoded, good-sex orgasms proved to be centered on the whole body
distinguishing all those resulting from intercourse from all (13/87), this was true of only 1.6% of not-as-good-sex orgasms
others, after dropping the ‘‘other’’ category. Thus, the three (3/189).
orgasm-through-non-intercourse conditions (i.e., oral stimu-
lation, masturbation by self or by other) were combined on
the presumption that they involved stimulation to outer geni- Discussion
talia only. As each participant was only reporting on a single
orgasm, v2 analysis was appropriate. A 2 (Orgasm Type: A vs. Although it is widely appreciated that female orgasms are not
B) 9 2 (Sexual behavior: Intercourse vs. Non-intercourse) v2 all the same, varying in a number of ways, little systematic
analysis with the 207 (of 276) cases of partnered orgasm on evidence of such has been published to date. Thus, the first goal
whom data were available revealed no significant difference in of the present study was to determine whether, using data
the mode of bringing the orgasm about, v2(1, n = 207)\1. collected for another purpose, distinct types of orgasms expe-
Thus, it did not prove to be the case that good-sex orgasms were rienced during sex with a partner could be identified. It was
more likely to occur in the context of intercourse than those expected that orgasms suggested by an evolutionary analysis to
classified as not-as-good-sex orgasms. It is worth mentioning discriminatively select sperm might prove identifiable and
that a number of orgasms (n = 69) could not be coded as distinct from others not shaped by natural selection for this
intercourse/non-intercourse because the mode of bringing the specific purpose. The second aim of the research reported
orgasm about was only described as other than those listed. herein was to validate the different types of orgasms identified
by means of latent-class analysis, using data collected on part-
Location of Orgasm ner relationships and others aspects of the sexual experience.

If it is true that some orgasms occur through activation of sen- Identifying Types of Orgasms with Partner
sory tissue deep inside, then it follows that questions about
where the orgasm was centered should follow a predictable The fact that four types of orgasms emerged from the athe-
pattern, with good-sex orgasms more likely to be centered on oretical latent-class analysis should not be read to mean that
the whole body. This would be consistent with the action of these are the only types of orgasms with a partner that are

123
Arch Sex Behav

experienced. Because the analysis was based on data col- Validating Typology of Orgasms: External Correlates
lected for another purpose, it cannot be presumed that the
same four-class typology would emerge were other studies The first stage of empirical validation of the four-type
done using lists of orgasm descriptors different than those typology derived from the latent-class analysis and involving
considered here. In sum, it would be a mistake to reify the four comparison with masturbatory orgasms suggested that it
types of orgasm identified, at least until they are replicated in made more sense to work with two types of orgasms with
future work. partner (good sex, not as good sex) rather than all four when it
Of interest nevertheless was that the four types discerned came to further validational analysis. Therefore, a series of
seemed meaningfully interpretable in terms of their face predictions were tested using data collected by Mah and
validity. Most important, perhaps, is that two merited the Binik (2002) on features of the sexual encounters and rela-
label‘‘good sex’’and two the label‘‘not-as-good-sex,’’though tionships that were not included in the latent-class analysis of
it needs to be remembered that this work did not focus upon adjectives describing the orgasms experienced. These were
sex with partner that did not result in an orgasm. Thus, the two informed by common sense as well as the evolutionary view
high pleasure and high or medium sensation orgasms, Types I that orgasms shaped by natural selection for purposes of
and II, would qualify for the former term and the two medium sperm selection—and likely involving oxytocin release—
and low pleasure orgasms for the latter. should be distinguishable from others, especially in terms of
the psychological and physiological sensations associated
with them.
Validating Orgasm Typology
Women reported significant variation in their orgasms in
terms of emotional and physiological components in addition to
Because, as indicated above, the types of orgasm identified
their scaling them in terms of pleasure. Some of the phenome-
were constrained by the data available for analysis, it was
nology described, especially in terms of general spasming,
essential to validate the typology before perhaps importing
whether the whole body was involved, and intense non-physical
too much meaning into it beyond what could be inferred in
sensations proved consistent with known effects of oxytocin.
terms of face validity. Toward this end, the first such effort
Such good-sex orgasms could not be distinguished, however,
involved determining whether the four types of orgasm with
from less-good-sex ones in terms of whether the orgasms
partners differed, first collectively, then individually, from
occurred in the context of intercourse.
how solitary masturbatory orgasms were characterized. The
Even though the fact that spasming proved more charac-
second set focused on external correlates of two higher-order
teristic of good-sex (Type A) than less-good-sex orgasms
sets of orgasms with partner.
(Type B) is consistent with insuck (Fox et al., 1970; Wildt
et al., 1998), it would be mistaken to infer that this research
Validating Typology of Orgasms: has documented insuck per se or the hypothesized effects of
With Partner Versus by Oneself oxytocin—or confirms an evolutionary interpretation of cer-
tain orgasms as potential sperm selecting ones. Nevertheless,
Not only did it prove to be the case, as common sense would the fact that women described their orgasms with such vari-
lead one to expect, that, on average, orgasms achieved with ability might be one reason why scholars have hitherto been
partners scored more highly in terms of pleasure and sensation divided regarding the nature and function of female orgasm.
than orgasms without partners but, far more interesting, per- There is now reason to suspect that scholars have not always
haps, was that this was not entirely the case. Recall that those been describing the same thing.
two types of orgasms (I, II) that could be regarded as evidence Now that a prima facie case for the distinguishing of different
of ‘‘good sex’’ consistently outscored solitary masturbatory types of female orgasms during sex with a partner has been
orgasms on nine of the ten composite measures of orgasm made, with some consistent with, though not confirming, a
experience, clearly showing them to be more pleasurable. But, sperm-selecting insuck process perhaps shaped by natural
at the same time, orgasm types III and IV scored lower on nine selection and involving oxytocin release, additional work is
of 10 adjective ratings than solitary masturbatory orgasms. needed to determine whether variation in female orgasms is
Apparently, at least in terms of the orgasm experience itself, linked to qualities of the partner, especially fitness-related ones
sometimes sex with oneself is more physically pleasurable (e.g., bilateral symmetry). It will also be important to determine,
than sex with a male partner, even when the latter provides with more precise measurements than available in this inquiry,
sufficient sexual arousal to generate an orgasm. Fundamen- whether insuck or oxytocin release is actually involved. Dating
tally, these data would seem to contradict the Masters and back as far as Freud (1932), much research has focussed on
Johnson (1965) view that masturbatory orgasms are the same female inability to orgasm, especially through sexual inter-
as those achieved through intercourse, especially in terms of course. An evolutionary analysis would seem to shift the focus
pleasure and sensation. to female orgasm as a potential response to male quality. A key

123
 Arch Sex Behav

area of interest is therefore whether women experience different Fisher, S. (1973). The female orgasm. New York: Basic Books.
types of orgasm with different partners. Research addressing Fox, C. A., & Fox, B. (1971). A comparative study of coital physiology,
with special reference to the sexual climax. Journal of Reproduc-
this issue would complement and extend that showing that the tion and Fertility, 24, 319–336.
frequency of female orgasm is associated with partner genetic Fox, C. A., Wolff, H. S., & Baker, J. A. (1970). Measurement of intra-
quality (Thornhill et al., 1995) or resources (Pollet & Nettle, vaginal and intra-uterine pressures during human coitus by radio-
2009). Indeed, just because there exists evidence that orgasmic telemetry. Journal of Reproduction and Fertility, 22, 243–251.
Freud, S. (1932). Female sexuality. International Journal of Psycho-
capacity is heritable (Dawood, Kirk, Bailey, Andrews, & Analysis, 13, 281–297.
Martin, 2005; Dunn, Cherkas, & Spector, 2005) and related Gould, S. J. (1987). Freudian slip. Natural History, 96, 14–21.
to enduring psychological characteristics of females (Burri, Hite, S. (1976). The Hite Report: A nationwide survey of female sexu-
Cherkas, & Spector, 2009; Cohen & Belsky, 2008) does not ality. London: Bloomsbury.
Judson, O. (2005). Anticlimax [Review of the book The case of the
mean that partner characteristics are unimportant. female orgasm: Bias in the science of evolution, by E. Lloyd].
Nature, 436, 916–917.
Komisaruk, B. R., Bianca, R., Sansone, G., Gomez, L. E., Cueva-Rolon,
R., Beyer, C., et al. (1996). Brain-mediated responses to vaginoc-
References ervical stimulation in spinal cord-transected rats: Role of the vagus
nerves. Brain Research, 708, 128–134.
Anderson, H. M., & Dennerstein, L. (1994). Increased female sexual Komisaruk, B. R., & Sansone, G. (2003). Neural pathways mediating
response after oxytocin. British Medical Journal, 309, 929. vaginal function: The vagus nerves and spinal cord oxytocin.
Anderson, H. M., & Dennerstein, L. (1995). Oxytocin and female sexuality. Scandinavian Journal of Psychology, 44, 241–250.
Gynecological and Obstetric Investigation, 40, 217–221. Komisaruk, B. R., Whipple, B., Crawford, A., Liu, W. C., Kalnin, A., &
Baker, R., & Bellis, M. A. (1993a). Human sperm competition: Mosier, K. (2004). Brain activation during vaginocervical self-
Ejaculate adjustment by males and the function of masturbation. stimulation and orgasm in women with complete spinal cord
Animal Behaviour, 46, 861–885. injury: fMRI evidence of mediation by the vagus nerves. Brain
Baker, R., & Bellis, M. A. (1993b). Human sperm competition: Research, 1024, 77–88.
Ejaculate manipulation by females and a function for the female Kosfeld, M., Heinrichs, M., Zak, P., Fischbacher, U., & Fehr, E. (2005).
orgasm. Animal Behaviour, 46, 887–909. Oxytocin increases trust in humans. Nature, 435, 673–676.
Baker, R., & Bellis, M. A. (1995). Human sperm competition: Copulation, Levin, R. (1981). The female orgasm-a current appraisal. Journal of
masturbation, and infidelity. London: Chapman & Hall. Psychosomatic Research, 25, 119–133.
Barash, D. (2005). Let a thousand orgasms bloom! [Review of the book Levin, R. (1998). Sex and the human female reproductive tract-what
The case of the female orgasm: Bias in the science of evolution, by really happens during and after coitus. International Journal of
E. Lloyd]. Evolutionary Psychology, 3, 347–354. Impotence Research, 10(Suppl.1), S14–S21.
Bentler, P. M., & Peeler, W. H. (1979). Models of female orgasm. Levin, R. (2001). Sexual desire and the deconstruction and reconstruc-
Archives of Sexual Behavior, 8, 405–423. tion of the human female sexual response model of Masters and
Blaicher, W., Gruber, D., Bieglmayer, C., Blaicher, A. M., Knogler, W., Johnson. In W. Everard, E. Laan, & S. Both (Eds.), Sexual appetite,
& Huber, J. C. (1999). The role of oxytocin in relation to female desire and motivation: Energetics of the sexual system (pp. 63–93).
sexual arousal. Gynaecologic and Obstetric Investigation, 47, 125– Amsterdam: Royal Netherlands Academy of Arts and Sciences.
126. Levin, R. (2004). An orgasm is…who defines what an orgasm is? Sexual
Burri, A. V., Cherkas, L. M., & Spector, T. D. (2009). Emotional and Relationship Therapy, 19, 101–107.
intelligence and its association with orgasmic frequency in women. Levin, R. L., & Wagner, G. (1985). Orgasm in women in the labo-
Journal of Sexual Medicine, 6, 1930–1937. ratory—Quantitative studies on duration, intensity, latency, and
Carmichael, M. S., Humbert, R., Dixen, J., Palmisano, G., Greenleaf, vaginal blood flow. Archives of Sexual Behavior, 14, 439–449.
W., & Davidson, J. M. (1987). Plasma oxytocin increases in the Lightfoot-Klein, H. (1984). Prisoners of ritual: An odyssey into female
human sexual response. Journal of Clinical Endocrinology and genital circumcision in Africa. New York: Harrington Park Press.
Metabolism, 64, 27–31. Lightfoot-Klein, H. (1989). The sexual experience and marital adjust-
Carmichael, M. S., Warburton, V. L., Dixen, J., & Davidson, J. M. ment of genitally circumcised and infibulated females in the Sudan.
(1994). Relationships among cardio-vascular, muscular and oxy- Journal of Sex Research, 26, 375–392.
tocin responses during human sexual activity. Archives of Sexual Lloyd, E. (2005). The case of the female orgasm: Bias in the science of
Behavior, 23, 59–79. evolution. Cambridge, MA: Harvard University Press.
Cohen, D., & Belsky, J. (2008). Avoidant romantic attachment and Mah, K., & Binik, Y. M. (2002). Do all orgasms feel alike? Evaluating a
female orgasm: Testing an emotion-regulation hypothesis. Attach- two-dimensional model of the orgasm experience across gender
ment and Human Development, 10, 1–11. and sexual context. Journal of Sex Research, 39, 104–113.
Cronin, H. (1991). The ant and the peacock. Cambridge: Cambridge Marazzitti, D., Bana, A., Casamassima, F., Catena, M., Consoli, G.,
University Press. Gesi, C., et al. (2006). Oxytocin: An old hormone for new avenues.
Dawood, K., Kirk, K. M., Bailey, J. M., Andrews, P. W., & Martin, N. G. Clinical Neuropsychiatry, 3, 302–321.
(2005). Genetic and environmental influences on the frequency of Masters, W. H., & Johnson, V. E. (1965). The sexual response cycles of
orgasm in women. Twin Research and Human Genetics, 8, 27–33. the human male and female: Comparative anatomy and physiol-
Dickinson, R. L. (1949). Human sex anatomy. New York: Robert ogy. In F. A. Beach (Ed.), Sex and behavior (pp. 512–534). New
E. Krieger. York: Wiley.
Dunn, K. M., Cherkas, L. F., & Spector, T. D. (2005). Genetic influences Masters, W. H., & Johnson, V. E. (1966). Human sexual response.
on variation in female orgasmic function: A twin study. Biology London: Churchill.
Letters, 1, 260–263. O’Connell, H. E., Hutson, J. M., Anderson, C. R., & Plenter, R. J. (1998).
Ferguson, J. K. W. (1941). A study of the motility of the intact uterus at Anatomical relationship between urethra and clitoris. Journal of
term. Surgical Gynecology and Obstetrics, 73, 359–366. Urology, 159, 1892–1897.

123
Arch Sex Behav

O’Connell, H. E., Sanjeevan, K., & Hutson, J. M. (2005). Anatomy of the Sundahl, D. (2003). Female ejaculation and the G-spot. London: Hunter
clitoris. Journal of Urology, 174, 1189–1195. House.
Pollet, T., & Nettle, D. (2009). Partner wealth predicts self-reported Symons, D. (1979). The evolution of human sexuality. New York:
orgasm frequency in a sample of Chinese women. Evolution and Oxford University Press.
Human Behavior, 30, 146–151. Thornhill, R., Gangestad, S., & Comer, R. (1995). Human female
Pound, N., & Daly, M. (2000). Functional significance of human female orgasm and mate fluctuating asymmetry. Animal Behaviour, 50,
orgasm still hypothetical. Behavioral and Brain Sciences, 23, 620–621. 1601–1615.
Puts, D. A. (2006). [Review of the book The case of the female orgasm: Vermunt, J. K., & Magidson, J. (2000). Latent GOLD’s user’s guide.
Bias in the science of evolution, by E. Lloyd]. Archives of Sexual Boston: Statistical Innovations Inc.
Behavior, 35, 103–108. Wildt, L., Kissler, S., Licht, P., & Becker, W. (1998). Sperm transport in
Russell, J. A., Leng, G., & Douglas, A. J. (2003). The magnocellular the human female genital tract and its modulation by oxytocin as
oxytocin system, the fount of maternity: Adaptations in pregnancy. assessed by hysterosalpingoscintigraphy, hysterotonography,
Frontiers in Neuroendocrinology, 24, 27–61. electrohysterography and Doppler sonography. Human Repro-
Schober, J. M., Meyer-Bahlburg, H. F. L., & Ransley, P. G. (2004). Self- duction Update, 4, 655–666.
assessment of genital anatomy, sexual sensitivity and function in Zak, P. J., Kurzban, R., & Matzner, W. T. (2005). Oxytocin is associated
women: Implications for genitoplasty. British Journal of Urology, with human trustworthiness. Hormones and Behavior, 48, 522–527.
94, 589–594. Zervomanolakis, I., Ott, H. W., Hadziomerovic, D., Mattle, V., Seeber,
Schultz, W. W., van Andel, P., Sabelis, I., & Mooyaart, E. (1999). B. E., Virgolini, I., et al. (2007). Physiology of upward transport in
Magnetic resonance imaging of male and female genitals during the human female genital tract. Annals of the New York Academy of
coitus and female sexual arousal. British Medical Journal, 319, Science, 1101, 1–20.
1596–1600. Zuk, M. (2006). The case of the female orgasm [Review of the book The
Singer, J., & Singer, I. (1972). Types of female orgasm. Journal of Sex case of the female orgasm: Bias in the science of evolution, by
Research, 8, 255–267. E. Lloyd]. Perspectives in Biology and Medicine, 49, 294–298.

123