Archives of Sexual Behavior

, Volume 42, Issue 1, pp 117–127

Alexithymia, Negative Emotions, and Sexual Behavior in Heterosexual University Students from Italy

  • Giuseppe Scimeca
  • Antonio Bruno
  • Gianluca Pandolfo
  • Umberto Micò
  • Vincenzo M. Romeo
  • Elisabetta Abenavoli
  • Adriano Schimmenti
  • Rocco Zoccali
  • Maria R. A. Muscatello
Original Paper

DOI: 10.1007/s10508-012-0006-8

Cite this article as:
Scimeca, G., Bruno, A., Pandolfo, G. et al. Arch Sex Behav (2013) 42: 117. doi:10.1007/s10508-012-0006-8

Abstract

Alexithymia is a construct which denotes thought characterized by pragmatic content, an inability to recognize and verbally express emotion, a difficulty in distinguishing between feelings and bodily sensations, and a limitation in fantasy life. Research has revealed a role for alexithymia in different kinds of sexual dysfunctions; it was also associated with reduced frequency of penile-vaginal intercourse but not with sexual behaviors—like masturbation—which do not include an emotional interaction in normal individuals. The aim of this research was to further investigate the association between alexithymia scores and sexual behavior in a sample of normal individuals, taking into account the role of gender differences and the possible effect of negative emotions (depression, anxiety, and anger). Participants were 300 university students (142 men and 158 women); sexual behavior was measured by the Sex and the Average Woman (or Man) Scale while alexithymia was measured with the Toronto Alexithymia Scale. The findings of the study showed that higher alexithymia scores were associated with lower levels of sexual satisfaction and higher levels of sexual detachment for females, and with sexual shyness and sexual nervousness for both genders. Results also suggested that the correlations between alexithymia scores and sexual behavior are partially influenced by the effect of negative emotions. Overall, it seems that the same detachment which denotes the alexithymic interpersonal style also characterizes sexual behavior.

Keywords

Alexithymia Sexual behavior Negative emotions Sex differences 

Introduction

Alexithymia is a construct denoting thought characterized by pragmatic content, an inability to recognize and verbally express emotions and feelings, a difficulty in distinguishing between feelings and bodily sensations, and a limitation in fantasy life (Taylor, 1984). Alexithymia has been commonly reported among patients suffering from somatoform disorders, revealing an important role in predicting mind–body associations (Sifneos, 1996). In spite of the role of alexithymia in research addressing mind–body associations, limited attention has been paid to the relationship between alexithymia and sexual behavior. Nevertheless, there are a number of reasons to suspect a possible association between this construct and sexual behavior, because sexuality implies a strong interconnection between the mind and the body.

Alexithymia and Sexual Behavior

The ability to fantasize and to experience and express emotions adaptively is an essential element of sex life both from a psychological (Everaerd, 1988; Pasini & Crépault, 1987; Purifoy, Grodsky, & Giambra, 1992) and neuropsychological point of view (Van de Poll & van Goozen, 1992). Consequently, it has been hypothesized that alexithymia could be associated with different kinds of sexual dysfunctions. Madioni and Mammana (2001) found an association between alexithymia and different kinds of sexual disorders; in particular, they found higher levels of alexithymia in individuals suffering from hypoactive sexual desire disorder and in males with erectile dysfunction when compared with a sample of healthy participants. Wise, Osborne, Strand, Fagan, and Schmidt (2002) found higher levels of alexithymia in individuals with sexual diagnoses (drive disorders, arousal disorders, orgasmic dysfunction, and sexual pain disorders) as well as paraphilias, when compared with prevalence of alexithymia in the general population. Michetti et al. (2007) found an association between alexithymia and premature ejaculation. More recently, Brotto, Knudson, Inskip, Rhodes, and Erskine (2010) documented normal levels of alexithymia in a sample of asexuals (although recent research has shown that lack of sexual attraction can hardly be considered a sexual dysfunction) (Brotto & Yule, 2011). These studies, on the whole, suggest that alexithymia is associated with different kinds of sexual disorders.

The association between alexithymia and normal sexual behavior has been understudied. The only study addressing this topic was undertaken by Brody (2003), who showed that alexithymia scores were inversely correlated with the frequency of penile-vaginal intercourse (only in females) but not with other sexual behaviors—like masturbation—which do not include an emotional interaction.

The aim of the present research was to investigate the association between alexithymia scores and sexual behavior in a sample of people without psychopathological conditions. In order to explore in more detail the association between alexithymia scores and sexuality, we assessed various kinds of sexual behaviors.

First, we considered sexual excitement and satisfaction, two different aspects of sexual response (Pfaus, 1999). Sexual excitement refers to the psychomotor stimulation elicited from a sexual incentive; it is a complex construct which implies a motivational/emotional state (subjective sexual desire) (Regan & Berscheid, 1999), a physiological/genital sexual arousal (Masters, Johnson, & Kolodny, 1994), and subjective sexual arousal (Green & Mosher, 1985). Sexual satisfaction refers to the general satisfaction that a person derives from his/her sex life and it can be usefully described along different individual, interpersonal, and behavioral dimensions such as sexual sensations/pleasure, the ability to focus on erotic and sexual sensations, the sexual exchange/reciprocity, emotional connection/closeness, and sexual activities (Bancroft, Loftus, & Long, 2003). Since these aspects of sexual response imply an emotional connection, and alexithymia is a construct denoting difficulties in recognizing and expressing emotions and feelings, we hypothesized that alexithymia scores would be negatively associated with both sexual excitement and satisfaction.

Furthermore, following Brody (2003), we also considered those sexual behaviors involving an emotional interaction. There are several clinical observations and scientific studies showing that alexithymic individuals have an interpersonal style characterized by coldness and distance, approaching others in an avoidant and detached fashion (Grabe, Spitzer, & Freyberger, 2001; Guttman & Laporte, 2002; Nemiah & Sifneos, 1970; Vanheule, Desmet, & Meganck, 2007; Weinryb et al., 1996). Also, alexithymic individuals show patterns of insecure attachment, since they are characterized by discomfort with being emotionally close, by the tendency to consider relationships as of secondary importance, and by the need for approval (Montebarocci, Codispoti, Baldaro, & Rossi, 2004; Troisi, D’Argenio, Peracchio, & Petti, 2001). We consequently hypothesized that alexithymia scores would be positively associated with the tendency to feel discomfort with a sexual partner (sexual shyness) and to consider the partner as a sexual object, neglecting his/her personality, and the importance of establishing a close emotional relationship with him/her (sexual detachment). We also predicted that higher alexithymic scores would positively correlate with a generic propensity to have a distressed sex life. In other words, we hypothesized that people with higher alexithymia scores would have difficulties in achieving sufficient sexual gratification because of the tendency to experience disturbing thoughts, emotions, and actions arising from their sex life (sexual nervousness). Finally, we expected to find no significant association between alexithymia scores and sexual behaviors which do not require any kind of affective interaction with a partner. Consequently, we hypothesized that alexithymia scores would not be associated with a form of solitary sex, such as pornography, defined as the tendency to appreciate sexual pleasure coming from the portrayal of explicit erotic images, both of a verbal and graphic nature (Masters & Johnson, 1970).

Gender Differences

We were also interested in testing gender differences. Several prevalence studies have found that men scores significantly higher than women on the total TAS-20 (Lindholm, Lehtinen, Hyyppä, & Puukka, 1990; Parker, Bagby, & Taylor, 1993; Salminen, Saarijarvi, Aarela, Toikka, & Kauhanen, 1999). Levant et al. (2003) proposed that this gender difference may be due to Western culture, which requires men to suppress or restrict the expression of their emotions. However, Italian studies have found no significant gender differences in alexithymia scores (Pasini, Chiaie, Seripa, & Ciani, 1996), possibly because Italian culture gives more importance to the expression and communication of emotions. If we compare gender differences in sexuality, we can see that there are several differences between men’s and women’s sexuality that may influence the relationship between alexithymia scores and sexual behavior. Women’s sexuality tends to be strongly linked to the need for a positively close relationship; they are more likely to strive for intimacy and committed relationships (Peplau, 2003), and they usually long to feel and express love for another person more than men (Regan & Berscheid, 1999).

Consequently, as found by Brody (2003), since alexithymia implies difficulties in feeling emotionally involved with a partner, and emotional involvement is more important in women’s sexual experiences, we expected that the associations we predicted between alexithymia scores and the sexual behavior scales would be stronger for women than for men. Thus, we examined gender as a potential moderator of these associations, predicting that the effect of alexithymia scores on sexual behaviors would be of different magnitude among men and women.

The Role of Negative Emotions

Another purpose of this study was to verify whether negative emotions (depression, anxiety, and anger) could affect the association between alexithymia scores and sexual behavior. Particularly, we were interested in determining whether there was a relationship between maladaptive sexual behavior and alexithymia scores over and above the relationship between negative emotions and sexual functioning. There is some consensus that positive emotions are associated with good sexual adjustment (Heiman, 1980; Koukounas & McCabe, 2001; Mitchell, DiBartolo, Brown, & Barlow, 1998; Peterson & Janssen, 2007), and that negative emotions correlate with some specific indicators of sexual maladjustment.

Depression, anxiety, and situational anger are good predictors of poor sexual motivation and sexual dysfunctions (Bancroft et al., 2003; Beck, 1967; Beck & Bozman, 1995; Bozman & Beck, 1991; Carvalho & Nobre, 2011; Nobre & Pinto-Gouveia, 2006) while trait anger is associated with negative interpersonal sexual behavior (sexual nervousness, sexual detachment, and aggressive sex) (Muscatello et al., 2010). Moreover, the association between alexithymia and negative emotions is a well—established finding (Lumley, 2000). Sexual experiences of alexithymic individuals may involve negative emotions because sexual experience with a partner implies a close emotional involvement and alexithymia is associated with several indicators of difficulties in interpersonal emotional adjustment (Montebarocci et al., 2004; Troisi et al., 2001). Based on evidence from the literature, we explored if the relationship between alexithymia scores and sexual behavior was moderated by negative emotions.

In sum, these were the hypotheses of this research:

Hypothesis 1

It was predicted that alexithymia scores would be negatively associated with sexual excitement and satisfaction.

Hypothesis 2

It was predicted that alexithymia scores would be positively associated sexual shyness, detachment, and nervousness.

Hypothesis 3

It was predicted that alexithymia scores would have no significant association with pornography.

Hypothesis 4

It was predicted that the associations between alexithymia scores and the sexual behavior scales would be stronger for women than for men.

Method

Participants

A total of 300 volunteer students (142 men and 158 women) recruited from the University of Messina participated in the study. An additional 10 had their data excluded because their MMPI-2 profiles indicated severe psychopathological conditions (defined on the MMPI-2 as T score ≥65).

Inclusion criteria were the absence of psychopathological conditions, heterosexuality, and the presence in the personal history of the participants of several acts of heterosexual intercourse throughout their life; this last criterion was selected to exclude persons who had never had sexual intercourse, since the hypothesis of this study concerned interpersonal sexual behavior.

Mean age of participants was 24.16 years ± 2.67 (range, 19–30); men and women did not differ significantly in age, t(298) = 1.72, p = .09. Mean age of first intercourse was 17 years (SD = 2.16). Mean frequency of sexual intercourse was 4.06 for men and 4.74 for women while median frequency of sexual intercourse was 4 for both men and women (4 refers to 2–3 times/week) (for more detailed sociodemographic features, see Table 1).
Table 1

Participants sociodemographics by gender

Factor

Total (n = 300)

Males (n = 142)

Females (n = 158)

N (%)

N (%)

N (%)

Age (years)

 19–21

53 (17.6)

24 (8)

29 (9.6)

 22–24

115 (38.2)

53 (17.6)

62 (20.6)

 25–27

96 (32.2)

43 (14.3)

53 (17.6)

 28–30

36 (11.9)

22 (7.3)

14 (4.6)

 Range

19–30

19–30

19–30

 M (SD)

24.16 (2.67)

24.43 (2.78)

23.90 (2.56)

 Median

24.00

24.00

24.00

Occupational status

  

300 (100)

 Employment

10 (3.3)

7 (2.3)

3 (1)

 Unemployment

290 (96.6)

135 (45)

155 (51.6)

Parents’ income (euro)

 12000–18000

54 (17.9)

26 (8.6)

28 (9.3)

 19000–25000

140 (46.6)

69 (23)

71 (23.6)

 26000–31000

74 (24.6)

35 (11.6)

39 (13)

 31000–37000

32 (11.6)

12 (4)

20 (6.6)

 Range

12000–37000

12000–37000

12000–37000

 M (SD)

24009 (6080.23)

24500 (2595.01)

23557 (2749.19)

 Median

24000

25000

23500

Marital Status

 Single

287 (95.6)

140 (46.6)

147 (49)

 Married

9 (3.0)

1 (.03)

8 (2.6)

 Divorced

4 (1.3)

1 (.03)

3 (1)

 Widowed

0 (0)

0 (0)

0 (0)

Religion

 Catholic

262 (87.3)

128 (42.6)

134 (44.6)

 Protestant

5 (1.7)

2 (.7)

3 (1)

 Jewish

4 (1.3)

1 (.03)

3 (1)

 Other

2 (0.7)

1 (.03)

1 (.03)

 None

27 (9)

9 (3)

11 (3.6)

Residence

 Living with parents

238 (79.3)

110 (36.6)

128 (42.6)

 Living with a family

9 (3.0)

2 (.7)

7 (2.3)

 Cohabiting

23 (7.6)

10 (3.3)

13 (4.3)

 Sharing an accommodation

13 (4.3)

8 (2.6)

5 (1.6)

 Living alone

11 (3.7)

10 (3.3)

1 (.03)

 Other

6 (2)

2 (.7)

4 (1.3)

Age at first sexual relation

 <14

11 (3.5)

7 (2.3)

4 (1.5)

 15–16

40 (13.2)

25 (8.3)

15 (5)

 17–18

126 (41.9)

70 (23.3)

56 (18.6)

 >19

123 (41)

40 (13.3)

83 (27.6)

Frequency of sexual intercourse

 1. >1 time/day

24 (8)

14 (8)

10 (8)

 2. 1 time/day

19 (6.3)

10 (6.3)

9 (6.3)

 3. 4–6 times/week

36 (12)

18 (12)

18 (12)

 4. 2–3 times/week

93 (31)

50 (31)

43 (31)

 5. 1 time/week

46 (15.3)

22 (15.3)

24 (15.3)

 6. 1 time/2 weeks

38 (12.6)

14 (12.6)

24 (12.6)

 7. Rarely

30 (10)

10 (10)

20 (10)

 8. Hardly ever

14 (4.6)

4 (4.6)

10 (4.6)

Participants responded to advertisements placed on notice boards and on the university website. Psychology and medical students were excluded because we suspected that their knowledge could have biased the responses of the tests.

Procedure

The study was introduced to the participants as an investigation into heterosexuality and personality; they were asked to answer self-report questionnaires anonymously.

All measures were administered under anonymous conditions and the participants provided written informed consent after a full explanation of the protocol design which had been approved by the local ethics committee. The sample was divided in groups of 25–50 participants: the administration of questionnaires took place in large rooms to ensure confidentiality and each group came for a single session. Two trained psychologists, who were masked to the aims of the study, administered the semi-structured interviews to single participants in single sessions. The Minnesota Multiphasic Personality Inventory (MMPI-2) (Butcher et al., 2001; Nencini & Belcecchi, 1973; Tellegen et al., 2003) was used in order to exclude the presence of psychopathological conditions that could affect the reliability of the sample. Participants did not receive grants or other forms of payment, although they were told that they would receive a brief summary of the results of tests at the end of the research. Overall the entire protocol took about 180 min.

Measures

Sociodemographic Variables

A sociodemographic questionnaire was used to ask participants about their age, sex, marital status, occupation, habitual residence, religion, and socioeconomic status (determination of socioeconomic status was based on parents’ income data). Three questions were specifically related to sexuality: several experiences of sexual intercourse, age of the first act of sexual intercourse, and frequency of sexual intercourse.

Sexual Behavior

Sexual behavior was measured by the Italian version of the Sex and the Average Woman (or Man) Scale (SAWM), a questionnaire that assesses sexual attitudes and sexual behavior in nonclinical individuals (Eysenck & Wilson, 1977). The questionnaire has 11 scales with 159 questions requiring closed answers (yes/no) and explores the different domains of sexual behavior. For the purpose of this research, we selected the following scales: 1. Sexual Excitement (EXC: easiness and degree of arousal in response to sexual stimuli; 9 items, e.g., “I get sexually excited very easily”), 2. Sexual Satisfaction (SAT: the general satisfaction that a person derives from his/her sex life; 12 items, e.g., “All in all, I am satisfied with my sex life”), 3. Sexual shyness (SHY: the discomfort felt in the presence of the opposite sex; 6 items, e.g., “I am afraid of sexual relationships”), 4. Sexual Detachment (DET: the tendency to consider partners as a sexual object, neglecting his/her personality and the importance of establishing an emotional close relationship with him/her; 6 items, e.g., “I believe in taking pleasure where I find it”), 5. Sexual Nervousness (NERV: the inability to achieve sufficient sexual gratification due to disturbing thoughts, emotions, and actions coming from sex life; 13 items, e.g., “Thoughts about sex disturb me more than they should”), 6. Pornography (PORN: the appreciation of pleasure arising from explicit erotic images, both of verbal and graphic nature; 8 items, e.g., “I like to look at sexy pictures”).

The scales are considered as dimensional variables: there are no cut-off points suggesting pathological sexual behavior. Eysenck (1976) validated this questionnaire through the use of other sexual, personality and social attitude inventories. The SAWM has already been used to investigate the relationship between personality and sexuality in a variety of clinical (Eysenck, 1971, 1973; La Torre et al., 2003) and non-clinical populations (Eysenck, 1972, 1976; Martin, Eaves, & Eysenck, 1977; Muscatello et al., 2010; Schmitt & Buss, 2000), and has shown good psychometric properties (Eysenck, 1976). The Italian translation revealed respectable basic psychometric parameters of reliability, discriminant, and convergent validity; the scales used in this study showed acceptable levels of internal consistency (Cronbach’s alpha coefficients between .80 and .87); confirmatory factor analyses on the questionnaire supported the 11 factor structure of the SAWM (Micò, 2010).

Internal consistency reliability of the SAWM scales for our sample was estimated using Cronbach’s alpha; the six scales we used for our research showed acceptable levels of internal consistency, with alpha values of .82 for the EXC scale, .84 for the SAT scale, .81 for the SHY scale, .86 for the DET scale, .84 for the NERV scale, and .87 for the PORN scale.

Alexithymia

Alexithymia was measured with the Italian version of the Toronto Alexithymia Scale (TAS-20; see validation details in Bressi et al., 1996). The TAS-20 consists of 20 items rated on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree), assessing the different aspects of alexithymia such as the difficulty in identifying and verbally expressing feelings (e.g., “When asked which emotion I’m feeling, I frequently don’t know the answer”; “I’m unsure of which words to use when describing my feelings”), the difficulty in distinguishing between feelings and bodily sensations (e.g., “I sometimes experience confusing sensations in my body”), the limitation in imaginary life (e.g., “When other people are hurt or upset, I have difficulty imagining what they are feeling”), thought characterized by pragmatic contents (e.g., “I prefer doing physical activities with friends rather than discussing each other’s emotional experiences”), and problematic interpersonal relationships (e.g., “People I’ve been in close relationships with have complained that I neglect them emotionally”).

Individuals are considered as showing pathological levels of alexithymia if their score is 61 or above. Previous studies (Bressi et al., 1996; Taylor, Bagby, & Parker, 1997) have shown that the Italian version of the TAS-20 has good internal consistency (Cronbach’s alpha of .75 and .82 in normal and clinical groups, respectively) and test–retest reliability over a 3-week interval (r = .77). The TAS-20 Scale has a three-factor structure congruent with the concept of alexithymia. Factor 1 measures Difficulty Identifying Feelings (DIF), distinguishing them from bodily sensations of emotional arousal; Factor 2 assesses Difficulty Communicating Feelings (DCF), and Factor 3 determines the extent of Externally Oriented Thinking (EOT). A confirmatory factor analysis showed the same factor structure as the original English version and adequate internal consistency of the subscales, with α coefficients equal or greater than .70 (Bressi et al., 1996). Internal consistency reliability of the total alexithymia score for our sample was also estimated using Cronbach’s alpha; the TAS-20 total score showed a good level of internal consistency, with an alpha value of .89.

Negative Emotions

The Hamilton Depression Scale (Ham-D) (Hamilton, 1960) and the Hamilton Anxiety Scale (Ham-A) (Hamilton, 1959) were used to assess symptoms of depression and anxiety. Scores on the items were based on a semi-structured interview of the participants about his or her experiences over the previous week: there is a first question for each item and follow-up questions are provided for further exploration or additional clarification of symptoms. The Ham-D is a 17-item semi-structured interview that assesses depressive symptoms like depressed mood, health concerns, loss of interests, insomnia or psychomotor retardation. The items were rated on 3-or 5-point scales and scores on the scale can range from 0 to 53, with higher scores indicative of higher levels of depression. Scores ranging from 0 to 7 suggest no or minimal symptoms of depression, 8–17 indicate mild depression, 18–25 suggest moderate depression, and scores of 26 and above are associated with severe depression. Pancheri, Picardi, Pasquini, Gaetano, and Biondi (2002) confirmed the validity of the Italian version of the scale. Internal consistency of the scale in this study was acceptable (alpha = .79).

The Ham-A is also a semi-structured interview that consists of 14 items, each defined by a series of symptoms including anxiety, fear or phobias, insomnia, depression, palpitations, breathing difficulty, and restlessness. Each symptom of these two scales is rated on a 5-point scale, ranging from 0 (absent) to 4 (severe). Scores on this scale have a possible range of 0–56, with higher scores associated with higher symptoms of anxiety. Scores ranging from 0 to 7 suggest no or minimal symptoms of anxiety, while scores of 8 and above indicate the presence of anxiety. Previous research documented high reliability and diagnostic concordance for the Italian version of this scale (Albert, Maina, Forner, & Bogetto, 2004; Maina, Albert, Gandolfo, Vitalucci, & Bogetto, 2005). The inter-rater reliability of Ham-D and Ham-A was repeatedly tested during the study period with results ranging from .72 to .90. Internal consistency of the scale in this study was good (alpha = .87).

Anger was measured using the Italian adaptation of the State-Trait Anger Expression Inventory (STAXI) (Spielberger, 1994). The STAXI is a 57-item self-report questionnaire that measures experience, expression, and control of anger. The STAXI produces six scales and two subscales; we used the Trait Anger scale (T-Ang, which measures one’s general propensity to experience anger) which is divided into angry temperament (tendency toward anger without provocation) and angry reaction (anger upon specific provocation). It consists of 10 items, such as “I have a fiery temper” and “I am a hotheaded person,” that have to be rated on a 4-point scale (from 1: almost never to 4: almost always). The STAXI has been validated on a variety of normal and clinical populations and both the original and the Italian version have good psychometric properties (Spielberger, 1994). Internal consistency of the STAXI Trait Anger scale for our sample was .82.

Statistical Analyses

One-tailed Pearson correlations and partial correlations were used to control for the effect of socioeconomic status, since alexithymia is associated with older age, male sex, lower SES, and fewer years of education (Lane, Sechrest, & Riedel, 1998). We did not consider partialing age and years of education because our sample consisted of university students and hence was quite homogeneous with regards these two variables; the other variable—gender—was considered in the hypothesis. Regression analyses were conducted to analyze the associations between alexithymia and sexual behavior. The hypotheses concerning gender and negative emotions were tested in two hierarchical multiple regression analyses. Statistical analysis was performed on SPSS for Windows 16.0.

Results

Table 2 shows the sexual behavior, the negative emotions, and TAS-20 descriptive statistics by sex. Alexithymia scores were in the normal range (Bressi et al., 1996); we found 10 % of the total sample to be alexithymic (TAS-20 scores ≥61), 7.7 % in males and 12 % in females. SAWM scores were also in the normal range, both for males and females, when considering the Italian normative sample (Micò, 2010). Also, as expected in a sample of healthy individuals, our sample showed no or minimal symptoms of anxiety and depression, and normal levels of anger. No effect of socioeconomic status was found. Multicollinearity was assessed using the variance inflation factor (VIF). VIF scores ranged between 1.29 and 1.69 and the largest Condition Index (CI) was less than 10, suggesting a lack of significant multicollinearity (Belsley, 1991).
Table 2

Descriptive statistics of alexithymia scores, sexual behaviors and negative emotions by gender

 

Males (n = 142)

Females (n = 158)

M (SD)

M (SD)

TAS-20

44.56 (10.87)

46.33 (12.70)

Sexual excitement

4.95 (2.11)

2.87 (2.04)

Sexual satisfaction

7.68 (2.78)

7.03 (2.99)

Sexual detachment

3.14 (2.72)

.95 (1.67)

Sexual shyness

.54 (.95)

.33 (.57)

Sexual nervousness

2.44 (1.88)

1.77 (1.61)

Pornography

4.5 (2.14)

2.69 (2.09)

Ham-A

3.18 (2.18)

3.56 (2.33)

Ham-D

3.43 (1.53)

4.57 (1.61)

T-Ang

18.90 (6.20)

18.40 (7.69)

TAS-20 Toronto Alexithymia Scale, Ham-A Hamilton Anxiety Scale, Ham-D Hamilton Depression Scale, T-Ang Trait Anger

To test our hypotheses, we performed a hierarchical multiple regression analysis in which sexual variables were entered on Step 1, gender was added on the second step, and the interaction terms between the sexual variables and gender (calculated using partialed products based on centered variables) were added on Step 3.

Table 3 shows the results of the hierarchical multiple regression analysis: sexual behavior (sexual satisfaction, sexual detachment, sexual shyness, and sexual nervousness) was a significant predictor of alexithymia scores, accounting for 15.3 % of the variance (Model 1); adding gender in the second step significantly increased by 2.1 % the explained variance in alexithymia scores (Model 2), and the interaction between predictors and gender further increased the explained variance by 2.8 % (Model 3).
Table 3

Hierarchical multiple regression analyses predicting alexithymia from gender

 

Model 1

Model 2

Model 3

β

t

p

r

sr

β

t

p

r

sr

β

t

p

r

sr

Sexual excitement

−.12

−1.67

.095

−.05

−.09

−.07

−1.05

ns

−.05

−.06

−.11

−1.5

ns

−.06

−.07

Sexual satisfaction

−.17

−2.74

.007

−.27

−.15

−.15

−2.54

.012

−.27

−.13

−.15

−2.56

.021

−.27

−.12

Sexual detachment

.13

2.00

.046

.19

.11

.18

2.61

.009

.19

.14

.19

2.87

.002

.19

.16

Sexual shyness

.14

2.00

.046

.29

.11

.15

2.27

.024

.19

.12

.18

2.65

.009

.29

.14

Sexual nervousness

.13

1.92

.056

.26

.10

.13

1.84

.066

.26

.09

.14

2.02

.091

.26

.09

Pornography

.03

<1

.037

.09

.03

.06

<1

ns

.09

.05

.07

<1

ns

−.09

.05

Gender

     

.17

2.70

.007

.07

.14

.16

2.51

.015

.08

.13

Interaction

          

−.18

−3.28

.001

−.09

−.17

Model R2

.153

    

.174

    

.182

    

R2 Change

.153

    

.021

    

.028

    

F

(6, 293) = 8.85****

(1, 292) = 7.30**

(1, 291) = 10.79****

Gender with male coded as 1 and female coded as 2

p < .05, ** p < .01, *** p < .005, **** p < .001

Thus, the first hypothesis of this study, concerning the association between alexithymia scores and sexual response, was partially verified, because regression analyses showed a significant negative association between the TAS-20 and sexual satisfaction but not with sexual excitement. As concerns Hypothesis 2, the study also verified the association between alexithymia scores and those sexual behaviors involving an emotional interaction, since, according to the regression analyses, alexithymia scores were positively associated with sexual detachment, shyness, and sexual nervousness. Hypothesis 3 was also verified since no association was found between alexithymia scores and pornography (Model 1). Furthermore, the results of the hierarchical regression analysis supported Hypothesis 4: gender significantly added to the prediction, with being female increasing the power of the association between alexithymia scores and the sexual behavior scales (Model 2).

Finally, Table 3 shows significant interactions between gender and three sexual variables—sexual satisfaction, sexual detachment, and sexual shyness (Model 3)—thus suggesting different potential moderator effects of gender on the above associations. To interpret these interactions, simple slopes analyses were performed following the guidelines of Aiken and West (1991). Regressions of sexual behavior variables on alexithymia scores were conducted for males and females; slopes resulted in a significant interaction between gender and sexual detachment, and gender and sexual satisfaction. As concerns the interaction between gender and sexual detachment, it was significant for females, β = .33, t(157) = 4.70, p < .0001, whereas for males β was −.10 (it was not significant), t(141) = − 1.17. As regards the interaction between gender and sexual satisfaction, it was also significant only for females, β = .22, t(157) = −3.13, p < .002, whereas for males it did not result significant: β was .08, t(141). These moderating effects are plotted in Figs. 1 and 2. As concerns the interaction between sexual shyness and gender, the simple slopes of the regression line did not differ significantly from zero both for males and females.
Fig. 1

Regression lines showing significant interaction between sexual satisfaction and gender predicting alexithymia

Fig. 2

Regression lines showing significant interaction between sexual detachment and gender predicting alexithymia

Finally, to verify whether the negative emotions affected the association between alexithymia scores and sexual behavior, we performed a hierarchical multiple regression analysis, adding the three emotional variables on the first step and the sexual behavior scales on the second step, to see whether the sexual behavior scales would predict alexithymia scores over and above the negative emotions variables identified in the first step. Following our predictions, as can be seen in Table 4, the sexual behavior scales significantly predicted variance in alexithymia scores (Model 2), beyond that predicted by the negative emotions (anxiety, depression, and anger) (Model 1). It is worth noting that adding the sexual behavior scales in the second step significantly increased by 13.3 % the explained variance in alexithymia scores (Model 2), with sexual satisfaction (sr = −.16) and sexual shyness (sr = .12) contributing significant unique variance to the prediction of alexithymia.
Table 4

Hierarchical multiple regression analyses predicting alexithymia from sexual behavior controlling for negative emotions

 

Model 1

Model 2

β

t

p

r

sr

β

t

p

r

sr

Ham-A

.01

.06

.955

.16

.01

.01

.01

.996

.16

.01

Ham-D

.21

2.52

.012

.21

.14

.20

2.54

.012

.21

.13

T-Ang

.20

3.67

.001

.20

.20

.18

3.32

.001

.20

.17

Sexual excitement

     

−.082

−1.20

.232

−.06

−.06

Sexual satisfaction

     

−.181

−3.06

.002

−.27

−.16

Sexual detachment

     

.120

1.84

.067

.19

.09

Sexual shyness

     

.148

2.23

.027

.29

.12

Neurotic sex

     

.071

1.05

.296

.26

.05

Pornography

     

.059

.89

.375

.09

.05

Model R2

.085

    

.218

    

R2 Change

.085

    

.133

    

F

(3,296) = 9.220****

(9,290) = 9.001****

* p < .05, ** p < .01, *** p < .005, **** p < .001

Discussion

The aim of this research was to investigate the association between alexithymia scores and sexual behavior in a sample of individuals without specific sexual or psychopathological disorders, taking into account the role of gender and negative emotions.

The first hypothesis of this study concerned the association between alexithymia scores and sexual response, the regression analysis showed that higher alexithymia scores were associated with low levels of satisfaction but not with sexual excitement. Furthermore, gender acted as a moderator variable for this association, revealing that only females with higher alexithymia scores showed reduced levels of sexual satisfaction. It seemed, therefore, that the association between alexithymia scores and sexual response was limited, as it exclusively implied the reduced ability of women to feel satisfied by one’s own sex life. This result was congruent with studies suggesting the importance of emotional contact and intimacy for female sexual satisfaction (Ellison, 2001; Heiman & LoPiccolo, 1988; Schnarch, 1991).

Regarding Hypothesis 2, the results of the study also suggested an association between higher scores on the alexithymia scale and different aspects of sexual behavior. First, higher scores on the alexithymia scale were associated with a higher likelihood that individuals would report some tendencies to feel discomfort, fear, and embarrassment in the presence of the opposite sex (sexual shyness). Higher scores on the alexithymia scale were also associated with a higher probability that individuals would report some tendencies to feel easily worried and troubled about sex because of disturbing thoughts, emotions, and actions arising from their sex lives (sexual nervousness). The association between alexithymia scores and sexual detachment was instead moderated by gender, indicating that only females with higher alexithymia levels showed a higher likelihood to report some tendencies to consider the partner as a sexual object, neglecting his personality and the importance of establishing a gratifying interpersonal relationship with him.

As concerning pornography (Hypothesis 3), following our predictions, no significant association was found between alexithymia scores and this kind of sexual behavior. These results were consistent with Brody’s (2003) findings, highlighting that alexithymia scores were associated with sexual intercourse but not with masturbation. Pornography—like masturbation—is a simpler sexual behavior than sexual intercourse; it is, in fact, a form of solitary sex since excitement and pleasure are aroused by the portrayal of verbal or graphically explicit erotic images (Masters & Johnson, 1970). As such, although not devoid of emotions, pornography does not imply a real emotional interaction with a partner and, probably for this reason, its occurrence is less conditioned by individual differences in ability to be aware and express emotions.

Overall, these findings provided support to the importance of studying sexuality as an emotion (Everaerd, 1989). Scientific research has already demonstrated the association between alexithymia and different kinds of sexual disorders, such as drive disorders, arousal disorders, orgasmic dysfunction, and sexual pain disorders (Madioni & Mammana, 2001; Michetti et al., 2007; Wise et al., 2002). Our findings showed that even individuals who were not alexithymic but, nevertheless, reported a reduced ability to be aware of and to express emotions, were more likely to report specific problematic sexual behaviors, which refer to sexual response (reduced sexual satisfaction) and interpersonal sexual behavior (higher levels of detachment, shyness and sexual nervousness). It is worth underlining that our sample did not consist of extreme groups bearing clinical diagnoses, such as alexithymia, depression or sexual disorders. Consequently the conclusions of this research are limited to the constraints of normal behavior. Surely, it would be useful to determine whether the findings we observed are also typical of individuals with psychopathological conditions because this may give some insights regarding the treatment of sexual disorders and their relationships with other aspects of psychopathology.

These results also highlighted the strong connection between personality and sexuality. Indeed, individuals with higher scores on the alexithymia scale showed a tendency to live their sex life in the same cold and detached way that characterizes interpersonal relationship styles (Spitzer, Siebel-Juerges, Barnow, Grabe, & Freyberger, 2005). This finding was in line with studies stressing how individual differences in sexual behavior are not independent from individual differences in personality (Schmitt & Buss, 2000) and that, consequently, research on this topic could make important contributions to the field of sexuality.

The predictions concerning Hypothesis 4 were partially confirmed, since the results showed that gender acted as a moderator variable only for the association between alexithymia scores and two different sexual behaviors: sexual satisfaction and detachment. However, no significant differences were found for the remaining variables; both men and women with higher alexithymia scores showed higher levels of sexual shyness and nervousness. This result was consistent with recent research showing that emotional closeness, although more relevant for women (Peplau, 2003), may also be important for male sexuality (Byers, 2005; Carpenter, Nathanson, & Kim, 2009; Štulhofer et al., 2004). Nevertheless, it is important to stress that reduced levels of sexual satisfaction—which is an important component of well-being for most individuals (Mulhall, King, Glina, & Hvidsten, 2008)—were associated with alexithymia scores only for females, suggesting that, overall, difficulties in the awareness and expression of emotions are associated with worse outcomes for females than for males.

Finally, the results of this study suggested that the relationship between alexithymia scores and sexual behavior was partially influenced by the effect of negative emotions. Generally, alexithymia scores remained associated with sexual behavior when controlling for the effect of anger, depression, and anxiety. Nevertheless, there was a reduction of the magnitude of this association, since some specific interpersonal sexual behaviors ceased to be associated with alexithymia scores (sexual detachment and nervousness), while sexual response (satisfaction) and sexual shyness were not influenced by negative emotions. It was not easy to understand the reason for these different effects, given the absence of a shared theoretical framework explaining the relationship between alexithymia and negative emotions. However, from these findings, it seemed that the influence of negative emotions in the association between alexithymia scores and sexuality was not pervasive but limited to specific sexual behaviors.

Undoubtedly, this study had several limitations. The research sample consisted of undergraduates without relevant psychopathological conditions taken from an urban area of southern Italy: results would have been more valid if the sample was more heterogeneous regarding age and cultural/socioeconomic variables. There were other limitations arising from methodological issues. This study had a correlational design: although we checked for the effect of some negative emotions, the possible effect of other third variables could not be excluded. It was also not possible to know the direction of causality between the variables we studied. Moreover, most of the findings yielded only small correlations. Finally, we only applied self-report instruments: given that alexithymia is characterized by a difficulty of self-reflexivity, and that self-report measures require self-reflection skills, the validity of our conclusions is not guaranteed.

Beyond these limitations, the results of this research suggested the importance of studying the relationship between alexithymia and sexual behavior. It is worthwhile stressing that the sexual problems we have studied were not properly sexual dysfunctions but involved the quality of one’s sex life: sexual nervousness, sexual detachment, sexual shyness, and satisfaction all implied emotional discomfort from sexual activities but they were not per se indicators of sexual disorders. Accordingly, we think that clinicians should not simply look for sexual dysfunctions, but they should also evaluate the quality of the patient’s sex life, bearing in mind the possibility that the difficulties in identifying and describing feelings, as well as the presence of negative emotions, may be important variables to consider in the evaluations of sexual problems.

Copyright information

© Springer Science+Business Media, LLC 2012

Authors and Affiliations

  • Giuseppe Scimeca
    • 1
  • Antonio Bruno
    • 1
  • Gianluca Pandolfo
    • 1
  • Umberto Micò
    • 1
  • Vincenzo M. Romeo
    • 1
  • Elisabetta Abenavoli
    • 1
  • Adriano Schimmenti
    • 2
  • Rocco Zoccali
    • 1
  • Maria R. A. Muscatello
    • 1
  1. 1.Section of Psychiatry, Department of Neurosciences, Psychiatric and Anesthesiologic Sciences, Policlinico di MessinaUniversity of MessinaMessinaItaly
  2. 2.Department of Human and Social ScienceKore University of EnnaEnnaItaly

Personalised recommendations