Personality and Individual Differences 70 (2014) 39–44

Contents lists available at ScienceDirect

Personality and Individual Differences

journal homepage: www.elsevier.com/locate/paid

Personality and sexually deviant behavior

Jennifer Lodi-Smith ⇑, Kayleigh Shepard, Samantha Wagner
Canisius College, 2001 Main Street, Buffalo, NY 14208, United States

a r t i c l e i n f o a b s t r a c t

Article history: The present research expands on the relationship between subclinical paraphilia and Big Five and Dark
Received 28 April 2014 Triad personality traits in an online sample of 585 adults. In this sample, sexually deviant courtship
Received in revised form 9 June 2014 behaviors such as exhibitionism and voyeurism were related to high levels of narcissism while transves-
Accepted 10 June 2014
tism related to greater openness to experience. Further, there were gender differences in the relationship
between sadomasochism and personality with sadomasochism reflecting greater openness in men and
Machiavellianism in women. These findings suggest that deviant sexual behaviors represent both a
Keywords:
degree of antisocial personality functioning and a degree of personal openness that depends on the
Big Five
Dark Triad
specific behavior being examined and that more research is needed in this area.
Paraphilia Ó 2014 Elsevier Ltd. All rights reserved.
Sexual deviance

1. Personality and subclinical paraphilia 1.1. What is sexual deviance?

Personality traits are quickly coming to be recognized as impor-
tant predictors of sexual behavior. For example, personality traits
are related to sexual promiscuity (e.g. Jonason, Luevano, &
Adams, 2012; Markey & Markey, 2007), risky sexual behavior
(e.g. Bogg & Roberts, 2004; Kastner & Sellbom, 2012; Schmitt,
2004; Visser, Pozzebon, Bogaert, & Ashton, 2010), sexual fantasies
(e.g. Baughman, Jonason, Veselka, & Vernon, in press), and with
general sexuality (e.g. Bourdage, Lee, Ashton, & Perry, 2007; Lee
et al., 2013; Visser et al., 2010). In addition, preliminary evidence
suggests that personality traits may be indicative of deviant as well
as normative sexuality (e.g., Brown et al., 1996; Williams, Cooper,
Howell, Yuille, & Paulhus, 2009; Wismeijer & van Assen, 2013).
However, within this literature, the relationship between personal-
ity traits and sexually deviant behavior is inconsistent with some
research suggesting sexually deviant behavior may be related to
antisocial personality traits and anxiety (Williams et al., 2009)
while other research suggests that certain sexually deviant behav-
iors are unrelated to personality traits (Brown et al., 1996) and still
other research suggests that sexually deviant behaviors may actu-
ally be related to greater emotional stability (Wismeijer & van
Assen, 2013) as well as a high degree of personal openness
(Williams et al., 2009). The present research attempts to address
these inconsistencies through assessing a breadth of both
personality traits and sexual deviant behaviors in adults.
⇑ Corresponding author. Tel.: +1 716 888 2513; fax: +1 716 888 3244.
E-mail address: lodismij@canisisu.edu (J. Lodi-Smith).
The term sexual deviance characterizes any arousal or sexual
preference directed towards objects or activities outside of societal
norms. At the most extreme, interest in such behavior is labeled as
paraphilia by the Diagnostic and Statistical Manual of Mental Disor-
ders (DSM-5, American Psychiatric Association, 2013). The DSM
groups paraphilia in two broad organizing frameworks: (1) anom-
alous activity preferences include the courtship disorders of
voyeurism and exhibitionism as well as the algolagnic disorders
of masochism and sadism and (2) anomalous target preferences
include preferences directed at other humans such as pedophilia
and preferences directed elsewhere such as fetishes and transves-
tism (American Psychiatric Association, 2013).
1.2. The relationship between personality traits and sexual deviance
The evidence regarding the relationship between personality
traits and sexually deviant behavior in the literature to date is
not clear cut with multiple different and sometimes contrasting
relationships between personality traits and sexually deviant
behavior present in the literature.
First, some research suggests that subclinical deviant sexuality
may be linked to antisocial personality traits. Specifically, a broad
assessment of paraphilia fantasies and behaviors in college men
suggests that sadistic fantasies, sadistic behavior, and sexual
assault behaviors are related to narcissism and masochistic fanta-
sies, sadistic fantasies, sadistic behaviors, and sexual assault
behaviors are related to psychopathy (Williams et al., 2009). How-
ever, recent research on a large sample of Dutch individuals who

http://dx.doi.org/10.1016/j.paid.2014.06.012
0191-8869/Ó 2014 Elsevier Ltd. All rights reserved.
40 J. Lodi-Smith et al. / Personality and Individual Differences 70 (2014) 39–44

engage in sadomasochistic behaviors provides contradictory evi- antisocial behaviors characterized by anomalous courtship behav-
dence regarding the supposition that deviant sexuality may be iors, algolagnic behaviors, and anomalous target preferences direc-
linked to antisocial personality traits. Specifically, participants ted at human targets may relate to an antisocial personality as
who engaged in sadomasochistic behaviors reported higher levels evidenced by low conscientiousness and agreeableness and high
conscientiousness but lower levels of agreeableness relative to scores on all three Dark Triad Traits. Next, subclinical paraphilia
control participants, a finding that is intriguing given that the may be related to low emotional stability reflecting underlying
blend of conscientiousness and agreeableness is thought to reflect anxiety about holding sexually deviant fantasies or may be related
psychoticism (McCrae & Costa, 1985). to high emotional stability reflecting comfort with one’s sexual
Next, subclinical deviant sexuality is inconclusively linked to preferences and identity. Finally, subclinical paraphilia may be
emotional stability. In college men, masochistic fantasies are corre- related to a more exploratory personality as evidenced by higher
lated with low emotional stability (Williams et al., 2009) while openness and extroversion.
Dutch sadomasochists report higher levels of emotional stability We anticipate that these patterns may vary within the type of
relative to control participants with the authors suggesting that sexual behavior being examined. Specifically, we do not include
these findings, along with higher levels of subjective well-being anomalous target preferences directed elsewhere (fetishism and
and lower levels of rejection sensitivity, need for approval, and transvestism) in the hypothesized relationship with antisocial per-
anxious attachment, are evidence that ‘‘BDSM practitioners are sonality traits as the direction of such behaviors is not necessarily
characterized by greater psychological and interpersonal strength antisocial. Further, given the recent evidence that individuals who
and autonomy, rather than by psychological maladaptive charac- engage in sadomasochistic behavior may actually be better
teristics’’ (Wismeijer & van Assen, 2013, p. 1949). adjusted than control individuals (Wismeijer & van Assen, 2013),
Finally, sexually deviant behaviors may be linked to a more algolagnic behaviors may not correspond to antisocial personality
curious, exploratory personality. In college men, transvestic fanta- or anxiety and instead may be negatively related to such traits.
sies and behaviors were related to extraversion while pedophilic
fantasies were related to openness (Williams et al., 2009). Further,
2. Methods
in Dutch sadomasochists, individuals who engage in sadomasoch-
istic behaviors were higher in extroversion and openness when
2.1. Participants
compared to control participants (Wismeijer & van Assen, 2013).
However, in 188 male transvestites, variability in types of cross-
595 participants (244 male, 334 female, and 17 individuals who
dressing behavior related only to minor differences in the Big Five
described their gender fluidly) completed the online survey. The
trait of openness to experience and when the sample was reduced
sample was predominantly Caucasian (87.9%) and ranged in age
to the 139 individuals who had never received treatment for psy-
from 18 to 59 years (M = 23.64, SD = 5.92).
chological dysfunction, these differences were attenuated and the
remaining men were ‘‘virtually indistinguishable’’ (Brown et al.,
1996, p. 265) from non-cross-dressing men in their personality 2.2. Personality measures
traits (Brown et al., 1996).
Thus, the literature to date is inconclusive regarding the rela- Big Five personality traits were assessed with the Big Five
tionship between personality traits and sexually deviant behavior. Inventory (John, Donahue, & Kentle, 1991). Dark Triad personality
As with any psychological dysfunction, in order to receive a diag- traits were assessed with the Dirty Dozen, a short index of the Dark
nosis of a paraphilic disorder, an individual must clinically present Triad (Jonason & Webster, 2010). Table 1 displays statistics and
not just with a paraphilic interest but also with a concern that reliabilities for all personality measures.
these interests cause significant clinical impairment at some level.
In the case of paraphilia, the distress accompanying the diagnosis 2.3. Subclinical paraphilia measures
may be due to the socially deviant nature of the sexual interest.
If this is the case, subclinical deviant sexual interests may corre- Nine subscales of sexually deviant behavior were adapted from
spond to variations in personality indicative of antisocial personal- the Multidimensional Inventory of Development, Sex, and Aggres-
ity traits such as low conscientiousness and agreeableness and sion (MIDSA; Augur Enterprises, 2011). Participants reported the
high levels of all three Dark Triad traits of narcissism, psychopathy, frequency (from 0 = never to 5 = almost every day) with which
and Machiavellianism. Further, the presence of deviant sexual they engaged in each of the following sexual behaviors: voyeurism,
interests may, across the board, create some degree of anxiety in exhibitionism, transvestism, scatologia (obscene phone calls),
the individual and thus be linked to lower levels of emotional sta- fetishism, sadistic fantasies and behaviors, child sexual arousal,
bility. On the other hand, engaging in sexually deviant behaviors and child sexual sadism. In addition, four subscales of sexually
may be related to higher levels of emotional stability if it is affirm- deviant behavior not tapped by the MIDSA (masochism, urophilia,
ing one’s sexual identity. Subclinical deviant sexual interests may biastophilia, and bestiality) were developed by the authors for the
also be behavioral markers of an underlying curious, explorative
personality as characterized by higher levels of openness and
Table 1
extroversion. Indeed, a combination of these trait profiles is possi-
Descriptive and reliability statistics for personality trait measures.
ble and the relationships may vary based on the specific sexual
behavior being considered. Variable M SD a
Big Five Traits
1.3. The current research Openness 3.90 .57 .77
Conscientiousness 3.30 .66 .81
Extroversion 3.00 .91 .89
The present study extends the research into the relationship Agreeableness 3.50 .65 .77
between personality and subclinical paraphilia by examining the Emotional stability 3.20 .84 .86
covariance between a variety of different sexually deviant behav- Dark Triad
iors and both Big Five and Dark Triad personality traits within a Narcissism 5.10 1.65 .73
Psychopathy 3.66 1.61 .82
large sample of adults. In doing so, we address three ways person-
Machiavellianism 4.60 1.77 .82
ality traits may relate to deviant sexual behavior. First, more
 J. Lodi-Smith et al. / Personality and Individual Differences 70 (2014) 39–44 41

Table 2
Reliability and descriptive statistics for subclinical paraphilia measures.

Variable Highest factor % M SD Min Max a Sample item

loading
Courtship 72.40 .31 .36 .00 1.95
Exhibitionism .714 33.70 .27 .57 .00 3.20 .73 I have thought about exposing my genitals. (4 items)
Voyeurism .442 60.50 .45 .57 .00 3.20 .73 I have masturbated while watching someone secretly. (5 items)
Scatologia .281 25.20 .22 .45 .00 3.00 .37 I have made obscene or ‘‘dirty’’ phone calls. (2 items)
Algolagnic 92.80 .71 .68 .00 3.83
Sadistic behavior .900 74.20 .73 .93 .00 5.00 .88 I have tied someone up while we were having sex. (7 items)
Sadistic fantasy .843 79.60 .65 .72 .00 3.86 .81 I have thought about burning someone during sex. (7 items)
Biastophilia .661 42.30 .34 .54 .00 .34 .35 I act out rape fantasies with my partner. (3 items)
Masochism .563 80.80 1.60 1.58 .00 5.00 .94 I enjoy being physically hurt during sex. (5 items)
Urophilia .473 25.70 .25 .63 .00 .25 .77 I enjoy being urinated on during sex.(3 items)
Anomalous sentient target 31.30 .19 .41 .00 3.38
Child sexual arousal .843 15.00 .16 .56 .00 4.50 .91 I have thought about having sex with a child. (4 items)
Bestiality .447 25.80 .21 .44 .00 .21 .37 I have had sexual encounters with animals. (2 items)
Anomalous non-sentient 80.70 .64 .67 .00 3.83
target
Fetishism .997 48.70 .57 .86 .00 5.00 .65 I have gotten excited while thinking about shoes or feet. (3 items)
Transvestism .282 71.20 .70 .82 .00 5.00 .64 I have worn the opposite gender’s clothes or tried them on. (3
items)
Average 97.80 .51 .41 .00 2.57

online survey using a parallel phrasing to the questions from the
existing MIDSA subscales.1 Given the low incidence of any reporting
of child sexual sadism behaviors (1.6% of the sample reported any
behaviors with a maximum score of .86 across the 7-item index),
we do not consider this behavior further in our analyses. For each
of the remaining 12 behaviors, Table 2 contains: a sample item,
the number of items used to assess it, and descriptive and reliability
statistics.
A maximum likelihood factor analysis with direct Oblimin rota-
tion of the sexually deviant behaviors revealed a four component
structure to the deviant behaviors that closely parallels the struc-
ture put forward in the DSM-5 (American Psychiatric Association,
2013). This structure was replicated across other methods of
extraction and rotation. Based on these analyses, the following
composite indices of behaviors were computed: (1) courtship: exhi-
bitionism, voyeurism, and scatologia; (2) algolagnic: sadistic
behaviors, sadistic fantasies, biastophilia, masochism, and urophil-
ia; (3) anomalous sentient target: bestiality and child sexual
arousal; and (4) anomalous non-sentient target: transvestism and
fetishism. Loadings for each behavior on its respective factor are
located in Table 2. Finally, an average composite of all of the
sexually deviant behaviors was computed.
2.5. Analysis notes
Because of the multiple analyses conducted and the large sam-
ple size, each table below provides the effect size for a p-value of
.01 and .05 as well as 95% confidence intervals for both correlations
and standardized regression coefficients. Given a significance level
of .01 and a power of .80 we would need 287 participants to detect
an approximate target effect size of .20 (based on the effect sizes in
Williams et al., 2009). With a sample size of 595, the current study
is suitably powered to detect target effects.
Given the low number of participants (n = 17) who responded to
the open-ended gender prompt with a response that could not be
coded as male or female, gender difference analyses focus on only
male and female genders. Within these 17 participants, gender
responses were quite variable (ranging from simple ‘‘androgynous’’
and ‘‘none’’ and ‘‘transgender’’ responses to the response ‘‘I don’t
know. I honestly don’t know. I’m not good with putting labels and
all that stuff. I feel both like a woman and a man, and like none of
them at the same time.’’) further preventing meaningful comparison
within this group. As neither age nor ethnicity significantly corre-
lated with any of the sexually deviant behaviors, these demographic
variables were not considered further (rs ranged from .05 to .06).

2.4. Procedure
The survey was administered online and recruitment advertised
a study about ‘‘Personality and Sexual Behaviors’’ on social media
sites including personal Facebook accounts, Tumblr accounts, and
OKCupid accounts. In order to reach a broad sampling of behaviors,
the survey was also posted in FetLife (a fetish community); on the
Facebook pages ‘‘Fetishes’’ and ‘‘BDSM’’; under Tumblr tags of #fet-
ish, #BDSM, and #sexuality; and finally on Reddit under the sub-
reddit ‘‘BDSM Community’’. Participants were directed to a secure
website to complete the study. To protect the confidentiality of the
participants, all data was downloaded and deleted from the secure
webserver once a week and stored in a password protected spread-
sheet on a second secure server. No identifying information was
collected as part of the survey, including the forum in which they
heard about the study, and survey results could in no way be
traced back to individual respondents.
1
A copy of the items used is available on request.
3. Results
3.1. Courtship
Gender was unrelated to sexually deviant courtship behaviors
as male (M = .31, SD = .35) and female (M = .31, SD = .36) partici-
pants reported engaging in equivalent levels of sexually deviant
courtship behaviors, t (576) = .13, p = .90, d = .00. Thus, gender
was not addressed in the analysis of sexually deviant courtship
behaviors. As shown in Table 3, openness and all three Dark Triad
traits had significant positive zero-order correlations with sexually
deviant courtship behaviors while agreeableness was negatively
related with these behaviors. Narcissism and Machiavellianism
remained the only significant indicators of sexually deviant
courtship behaviors in simultaneous multiple regression.
3.2. Algolagnic
Because there were significant gender differences between men
(M = .60, SD = .66) and women (M = .76, SD = .63) in the extent to
42 J. Lodi-Smith et al. / Personality and Individual Differences 70 (2014) 39–44

Table 3 Table 5
Zero-order correlations and regression coefficients with 95% confidence intervals Zero-order correlations and regression coefficients with 95% confidence intervals
relating Big Five and Dark Triad personality traits to sexually deviant courtship relating Big Five and Dark Triad personality traits to sexually deviant behaviors
behaviors across genders. directed at anomalous non-sentient targets across genders.

r 95% CI b 95% CI r 95% CI b 95% CI

Big Five Traits Big Five Traits
Openness .09 .0097 .1692 .06 -.0170 .1457 Openness .21 .1318 .2856 .22 .1421 .3044
Conscientiousness .07 .1495 .0104 .03 .1136 .0495 Conscientiousness .10 .1789 .0198 .06 .1434 .0194
Extroversion .07 .0104 .1495 .03 .0575 .1245 Extroversion .01 .0903 .0704 .05 .1394 .0422
Agreeableness .09 .1692 .0097 .03 .1270 .0677 Agreeableness .07 .1495 .0104 .02 .1144 .0799
Emotional stability .08 .1594 .0004 .07 .1632 .0142 Emotional stability .08 .1594 .0004 .07 .1544 .0227
Dirty Dozen Dirty Dozen
Narcissism .20 .1216 .2759 .13 .0342 .2164 Narcissism .14 .0603 .2179 .01 .0807 .1011
Psychopathy .12 .0400 .1985 .04 .0646 .1381 Psychopathy .09 .0097 .1692 .04 .0657 .1365
Machiavellianism .21 .1318 .2856 .11 .0094 .2072 Machiavellianism .09 .0097 .1692 .09 .0108 .1867

Note: p < .01 for effect sizes greater than |.11|. p < .05 for effect sizes greater than Note: p < .01 for effect sizes greater than |.11|. p < .05 for effect sizes greater than
|.08|. |.08|.

which they reported engaging in algolagnic behaviors, t
(576) = 3.07, p = .002, d = .25, Table 4 presents analyses broken
down by gender. For men, openness and Machiavellianism were
significant zero-order correlates, with only openness remaining a
robust indicator in simultaneous multiple regression. In contrast,
for women, low conscientiousness and high levels of all three Dark
Triad traits were correlated with algolagnic behavior with
openness, low conscientiousness, and Machiavellianism remaining
significant indicators in simultaneous multiple regression.
3.4. Anomalous non-sentient target
Men (M = .64, SD = .74) and women (M = .57, SD = .53) reported
statistically equivalent amounts of sexually deviant behaviors
directed at anomalous non-sentient targets t (576) = 1.36, p = .16,
d = .11. Table 5 reports such behaviors were negatively related to
conscientiousness and positively related to openness, narcissism,
psychopathy, and Machiavellianism in zero-order correlations
with openness being the only significant indicator in simultaneous
multiple regression.

3.3. Anomalous sentient target

While men reported more sexually deviant behaviors directed
at anomalous sentient targets (M = .25, SD = .50) relative to women
(M = .12, SD = .27), t (576) = 4.02, p < .001, d = .25, deviant sexual
behaviors directed at a sentient target had no significant relation-
ship to Big Five or Dark Triad traits in zero-order correlations (rs
range from .10 to .09) or in simultaneous multiple regression
(bs range from .12 to .08) in either gender. In the interest of
space, the specific effect sizes for these analyses are available on
request.
3.5. Average
There were no significant differences between men (M = .48,
SD = .41) and women (M = .51, SD = .36) in the composite index of
sexually deviant behavior, t (576) = 1.08, p = .28, d = .09. As shown
in Table 6, this composite was negatively related to agreeableness,
conscientiousness, and emotional stability in zero-order correla-
tions and positively related to openness and all three Dark Triad
traits. In multiple regression, low conscientiousness, low emo-
tional stability, high openness, and high Machiavellianism

Table 4
Zero-order correlations and regression coefficients with 95% confidence intervals relating Big Five and Dark Triad personality traits to sexually deviant algolagnic behaviors in
men and women.

r 95% CI b 95% CI
Male (N = 244)
Big Five Traits
Openness .25 .1284 .3641 .23 .0907 .3332
Conscientiousness .06 .1842 .0661 .05 .1697 .0763
Extroversion .12 .0057 .2419 .08 .0697 .2155
Agreeableness .12 .2419 .0057 .10 .2461 .0554
Emotional stability .07 .1939 .0561 .07 .0745 .2040
Dirty Dozen
Narcissism .11 .0158 .2324 .03 .1764 .1148
Psychopathy .08 .0460 .2035 .00 .1503 .1479
Machiavellianism .20 .0763 .3176 .12 .0326 .2723
Female (N = 334)
Big Five Traits
Openness .10 .0074 .2051 .14 .0281 .2401
Conscientiousness .15 .2532 .0434 .12 .2162 .0079
Extroversion .03 .0776 .1369 .03 .1431 .0873
Agreeableness .09 .1954 .0175 .00 .1225 .1206
Emotional stability .10 .2051 .0074 .02 .1052 .1381
Dirty Dozen
Narcissism .19 .0844 .2914 .04 .1527 .0710
Psychopathy .12 .0129 .2244 .04 .0945 .1683
Machiavellianism .13 .0230 .2340 .20 .0620 .3124

Note: p < .01 for effect sizes greater than |.16| for men and |.13| for women. p < .05 for effect sizes greater than |.12| for men and |.11| for women.
 J. Lodi-Smith et al. / Personality and Individual Differences 70 (2014) 39–44 43

Table 6 unrelated to antisocial personality. Thus, at least in the current

Zero-order correlations and regression coefficients with 95% confidence intervals study, transvestism is largely related to an open, explorative per-
relating Big Five and Dark Triad personality traits to average level of sexually deviant
behaviors.
sonality style and not to any antisocial characteristics. This pattern
is consistent with past research on transvestism that suggests little
r 95% CI b 95% CI difference between transvestites and non-transvestites in person-
Big Five Traits ality (Brown et al., 1996).
Openness .22 .1421 .2952 .22 .1381 .2973 As a whole, subclinical paraphilia on average was related both
Conscientiousness .13 .2082 .0501 .08 .1632 .0035
Extroversion .04 .0405 .1200 .01 .0786 .0995
to a degree of antisocial personality tendencies and anxiety as well
Agreeableness .10 .1789 .0198 .03 .1282 .0624 as to being an open individual willing to try new experiences. It is
Emotional stability .10 .1789 .0198 .10 .1898 .0161 interesting to note that while antisocial personality traits and per-
Dirty Dozen sonal openness were related to different types of sexual behaviors,
Narcissism .13 .0501 .2082 .01 .0818 .0966
none of the assessed types of paraphilia by itself was related to
Psychopathy .13 .0501 .2082 .04 .0624 .1360
Machiavellianism .21 .1318 .2856 .15 .0546 .2483 emotional stability. Therefore, at least in the present sample, devi-
ant sexuality does not appear to be linked to higher levels of anx-
Note: p < .01 for effect sizes greater than |.11|. p < .05 for effect sizes greater than
iety and negative emotionality for a given behavior but is
|.08|.
indicative of neuroticism with relatively high levels of deviant
sexuality across multiple domains.
While the present research brings some intriguing insights into
remained significant indicators of high levels of sexually deviant
personality and deviant sexuality, it also has limitations that sug-
behaviors.
gest directions for future research in this area. First, since our sam-
ple was obtained using online forums we must use caution in
4. Discussion generalizing our results since, of course, all participants are com-
puter literate and may potential be biased by the data collection
The present research expands on our current understanding of method though research suggests that personality assessment via
the relationship between personality traits and subclinical para- the internet is both diverse and reliable (i.e., Gosling, Vazire,
philia in a large sample of adults. In doing so, it demonstrates that Srivastava, & John, 2004). More importantly for the current
deviant sexual behaviors may represent both a degree of antisocial research, given that some of our participants were solicited from
personality functioning and personal openness. However, as antic- websites and forums directed at sexually deviant behavior, it is
ipated, these patterns differed to some extent depending on the possible that our sample is biased towards an interest in such
type of paraphilia being considered. behaviors. As we did not collect data on where participants learned
The more interpersonally antisocial behaviors comprised by the about the study in order to protect their privacy, we cannot
sexually deviant courtship behaviors such as exhibitionism and address this potential bias and recommend that, as with all
voyeurism corresponded with an underlying antisocial personality. research, replication should be implemented within as generaliz-
Specifically, sexually deviant courtship behaviors related to low able a sample as possible.
agreeableness and the Dark Triad with narcissism being the most Next, while we measured most of the paraphilia covered in the
robust indicator of such behaviors. This relationship suggests that DSM-5, our assessment did not measure frotteurism, necrophilia,
sexually deviant courtship behaviors may be part of the behavioral coprophilia, or klismaphilia – all included as specific paraphilia in
manifestation of the entitlement and lack of empathy characteris- the DSM-5 (American Psychiatric Association, 2013). In addition,
tic of narcissistic individuals. Such behaviors may also reflect a while some constructs tapped both thoughts and behaviors, the
degree of personal openness as sexually deviant courtship behav- majority of the MIDSA (Augur Enterprises, 2011) is oriented
iors also related to openness to experience. However, this relation- towards behavior rather than fantasy and it is certainly plausible
ship was weak and attenuated in simultaneous regression. that there may be variation between behavior and fantasy in their
Algolagnic behaviors characterized by sadomasochism also dis- relations to personality as individuals may not always have the
played a relationship to antisocial personality and personal open- opportunity to act on fantasy. Thus, future research should be
ness, though this relationship varied by gender. In men, increasingly comprehensive in its assessment of both sexually
algolagnic behaviors primarily related to high openness while deviant behaviors and sexually deviant fantasies.
Machiavellianism was the most robust indicator of sadomasochism Further, because we focused on a general population and were
in women. This finding is intriguing in light of recent research sug- assessing, by definition, low base rate behaviors, there was a low
gesting that Dutch practitioners of sadomasochism are better base rate of deviant sexuality in the present study. While nearly
adjusted than controls (Wismeijer & van Assen, 2013). Despite 98% of the sample reportedly engaged in some sexually deviant
gender differences in the Dutch sample in sadomasochistic roles behavior at some point, the overall frequencies were still quite
suggesting more dominance among men and submissiveness low and this may have impacted our results. For example, there
among women2, analyses controlled for gender rather than parsing was no relation between personality and anomalous sexual behav-
gender effects. This may have obfuscated gender differences sado- ior directed at a sentient target. This finding may be an artifact of
masochistic behavior and further investigation into the relationship the extremely low base rate of these behaviors in our sample not
between personality, adjustment, and algolagnic behaviors is the independence of such behaviors from general personality func-
warranted. tion. In order to fully map out the relationship between personality
Anomalous sexual behaviors directed at a non-sentient target traits and deviant sexual behavior, future research in this area
via transvestism and fetishism also related to both antisocial per- should also assess populations with higher rates of these behaviors
sonality and openness though openness was clearly the most such as clinical samples seeking treatment for paraphilia. Paired
robust indicator of such behavior. It is interesting to note that with non-clinical samples, such studies would provide a strong
transvestism, when analyzed independent from fetishism, was foundation for determining the full nature of the relationship
between personality traits and sexually deviant behavior in the
2
Paralleled in the current data as the driving factor of the gender differences
population.
reported above was that women engaged in more masochistic behavior (M = 2.04, The present research also cannot address nuances within a
SD = .1.58) than did men (M = .91, SD = 1.26), t (576) = 9.20, p < .001, d = .78. given sexually deviant behavior. For example, the DSM-5
44 J. Lodi-Smith et al. / Personality and Individual Differences 70 (2014) 39–44
(American Psychiatric Association, 2013) characterizes transves-
tism with two different specifiers: with autogynephilia (sexual
arousal by thoughts, images of self as female) and with fetishism
(sexual arousal to fabrics, materials, or garments). These differen-
tial specifiers may result in different relationships to personality
traits. However, more detailed data on the underlying behavior is
needed to address such questions.
In addition, while we characterize the sample as subclinical
given the rates of behavior, as we note in the introduction of the
manuscript, paraphilia are only considered of clinical significance
if they somehow impair the functioning of the individual. The cur-
rent research only assessed frequency of sexually deviant behav-
iors, not their psychological impact. Assessing the strength of
paraphilia interests relative to normal sexuality as suggested in
the DSM-5 (American Psychiatric Association, 2013) or assessing
the degree of distress caused by deviant sexual interest would
move the understanding of personality and deviant sexual behav-
ior forward in a more clinically meaningful way.
Finally, the present findings suggest that, in a subclinical popu-
lation, deviant sexual behaviors represent both a degree of antiso-
cial personality functioning and a degree of personal openness and
that the extent to which this is the case varies by the behavior
being addressed. However, the directionality of these relationships
cannot be determined in this cross-sectional study. Given the cor-
respondence demonstrated in the present research between sexu-
ally deviant behaviors and personality traits, a number of processes
may be at work. It may be that more antisocial individuals are
attracted to antisocial sexually deviant behaviors through a pro-
cess of selection wherein behaviors that are inherently antisocial
and directed at other individuals without their consent such as
exhibitionism and voyeurism are attractive to an individual with
an underlying antisocial personality. It may also be the case that
sexual preferences shape personality through socialization processes
such that sexual interests that require a degree of self-exploration
and movement away from gender norms such as transvestism
drive a greater degree of personal openness in general. Indeed, in
terms of development, these processes are likely quite dynamic
and reciprocal with both sexual preferences and personality traits
reinforcing each other over time.
Author’s note
The collection of data and preparation of this manuscript was
supported by grants awarded to Kayleigh Shepard and Samantha
Wagner under the auspices of the Canisius College Earning
Excellence Program.
References
American Psychiatric Association (2013). Diagnostic and statistical manual of mental
disorders (5th ed.). Washington, DC: American Psychiatric Publishing.
Augur Enterprises. (2011). MIDSA clinical manual. Augur Enterprises.
Baughman, H. M., Jonason, P. K., Veselka, L., & Vernon, P. A. (in press). Four shades of
sexual fantasies linked to the Dark Triad. In press at Personality and Individual
Differences. doi: 10.1016/j.paid.2014.01.034.
Bogg, T., & Roberts, B. W. (2004). Conscientiousness and health behaviors: A meta-
analysis. Psychological Bulletin, 130, 887–919. http://dx.doi.org/10.1037/0033-
2909.130.6.887.
Bourdage, J. S., Lee, K., Ashton, M. C., & Perry, A. (2007). Big Five and HEXACO model
personality correlates of sexuality. Personality and Individual Differences, 43(6),
1506–1516. http://dx.doi.org/10.1016/j.paid.2007.04.008.
Brown, G. R., Wise, T. N., Costa, P. T., Herbst, J. H., Fagan, P. J., & Schmidt, C. W.
(1996). Personality characteristics and sexual functioning of 188 cross-dressing
men. The Journal of Nervous and Mental Disease, 184, 265–273.
Gosling, S. D., Vazire, S., Srivastava, S., & John, O. P. (2004). Should we trust Web-
based studies? A comparative analysis of six preconceptions about Internet
questionnaires. American Psychologist, 59, 93–104.
John, O. P., Donahue, E. M., & Kentle, R. L. (1991). The Big Five Inventory – Versions 4a
and 54. Berkeley: University of California at Berkeley, Institute of Personality
and Social Research.
Jonason, P. K., Luevano, V. X., & Adams, H. M. (2012). How the Dark Triad traits
predict relationship choices. Personality and Individual Differences, 53, 180–184.
http://dx.doi.org/10.1016/j.paid.2012.03.007.
Jonason, P. K., & Webster, G. D. (2010). The Dirty Dozen: A concise measure of the
Dark Triad. Psychological Assessment, 22, 420–432. http://dx.doi.org/10.1037/
a0019265.
Kastner, R. M., & Sellbom, M. (2012). Hypersexuality in college students: The role of
psychopathy. Personality and Individual Differences, 53, 644–649. http://
dx.doi.org/10.1016/j.paid.2012.05.005.
Lee, K., Ashton, M. C., Wiltshire, J., Bourdage, J. S., Visser, B. A., & Gallucci, A. (2013).
Sex, power, and money: Prediction from the Dark Triad and honesty–humility.
European Journal of Personality, 27, 169–184. http://dx.doi.org/10.1002/
per.1860.
Markey, P. M., & Markey, C. N. (2007). The interpersonal meaning of sexual
promiscuity. Journal of Research in Personality, 41, 1199–1212. http://dx.doi.org/
10.1016/j.jrp.2007.02.004.
McCrae, R. R., & Costa, P. T. (1985). Comparison of EPI and psychoticism scales with
measures of the five-factor model of personality. Personality and Individual
Differences, 6, 587–597.
Schmitt, D. P. (2004). The Big Five related to risky sexual behavior across 10 world
regions: Differential personality associations of sexual promiscuity and
relationship infidelity. European Journal of Personality, 18, 301–319. http://
dx.doi.org/10.1002/per.520.
Visser, B. A., Pozzebon, J. A., Bogaert, A. F., & Ashton, M. C. (2010). Psychopathy,
sexual behavior, and esteem: It’s different for girls. Personality and Individual
Differences, 48, 833–838. http://dx.doi.org/10.1016/j.paid.2010.02.008.
Williams, K. M., Cooper, B. S., Howell, T. M., Yuille, J. C., & Paulhus, D. L. (2009).
Inferring sexually deviant behavior from corresponding fantasies: The role of
personality and pornography consumption. Criminal Justice and Behavior, 36,
198–222. http://dx.doi.org/10.1177/0093854808327277.
Wismeijer, A. A. J., & van Assen, M. A. L. M. (2013). Psychological characteristics of
BDSM practitioners. Journal of Sexual Medicine, 10, 1943–1952. http://
dx.doi.org/10.1111/jsm.12192.