On the Science of Changing Sex

Why not take part of me?

Posted in Transsexual Theory by Kay Brown on January 27, 2011

narcissus♫…Take my arms… I’ll never use them…♫

As I had discussed earlier, autogynephilia is a phenomena, not a theory… but has been theorized to be caused by Erotic Target Location Error.  If this theory is true, it predicts that men who are both gynephilically attracted to female amputees and have a deep desire to have a limb, such an arm or leg, removed to be like their erotic target, would also be likely to want to be female and have other autogynephilic arousal patterns.  In one of the few studies to test this prediction, we see exactly that, as described by Lawrence:

Recently, First (2005) used semistructured telephone interviews to survey 52 persons who had expressed a wish to become an amputee or who had succeeded in doing so. About two thirds of participants were recruited from Internet discussion groups and about one third from referrals by other participants. Forty-seven (90%) participants were male, 4 (8%) were female, and 1 was intersexed and raised as a male; 32 (62%) participants reported that they were heterosexual and the remainder reported that they were homosexual or bisexual. First (2005) noted that the high proportion of nonheterosexual participants was “partly explained by the fact that nine of the subjects were referred to the study by one subject who was himself homosexual, eight of whom were also homosexual” (p. 921). Nine (17%) participants had undergone a major limb amputation. Forty-eight (92%) participants had pretended to be an amputee and 45 (87%) acknowledged sexual attraction to amputees. Fifteen (29%) participants reported other paraphilic interests, including 8 (15%) with transvestic fetishism. Ten (19%) participants reported they had sometimes wished to be the opposite sex or felt that they were in the body of the wrong sex; of these, 7 (13%) participants had crossed-dressed, not including the 8 participants who had cross-dressed in connection with transvestic fetishism. Six (12%) participants had considered sex reassignment and 1 (2%) had undergone sex reassignment.

How likely are we to find in any random group of only 52 people, a post-op transsexual and six more that have thought about it?  How about finding 15% who report erotic arousal to cross-dressing?  And add to 13% more that cross-dress and experience some gender dysphoria?  Looking at the above description, at least 15 of the 52 showed some level of gender dysphoria.  Though we can’t tell from the paragraph, I’d predict that all 15 (possibly 18) were heterosexual, which if true, would mean that 15 out of 32  people (47%) reported gender dysphoria or autogynephilic arousal.  But, at the very least, 15 out of 52 “wannabe” people (29%) reported such.

So, Erotic Target Location Error theory’s prediction that “wannabes” will also be more likely to be autogynephilic and gender dysphoric than would be expected by random chance, is very strongly supported.  Thus, the theory of Erotic Target Location Errors explaining autogynephilia is also supported by the data.

Addendum 3/3/2014:

Some researchers have suggested that “wannabes” are suffering from a disorder akin to somatoparaphrenia, where people deny ownership of a paralysed limb.  Somatoparaphrenia can be treated using a technique in which caloric vestibular stimulation (CVS), involving pouring cold water into the ear canal induces sensations in the brain. The flow of water induces an illusion of motion, and is thought to stimulate regions of the brain that create a mental map of the body.  The hypothesis is that the parts of the brain that form the somatosensory map of the limb that the individual wishes to have removed is somehow disordered in a similar manner as somatoparaphrenia.  If CVS works to relieve the desire for limb amputation, this would contradict the very strong evidence that “wannabeism” is an erotic target location error.  A recent paper showed that CVS does NOT reduce the desire, supporting the paraphilic hypothesis and not supporting the somatoparaphrenic hypothesis.


Lawrence, A., “Clinical and Theoretical Parallels Between Desire for Limb
Amputation and Gender Identity Disorder”

Leggenhager, B. et al., “Vestibular stimulation does not diminish the desire for amputation”

Comments Off on Why not take part of me?


Posted in Transsexual Theory by Kay Brown on January 16, 2011

…on a hypothesis concerning bisexual  and asexual transsexuals.

androgynous faceIn the past, I’ve tried to keep from too much speculation, trying to describe and interpret the research into transsexuality.  But in this post, I wish to take a chance and speculate a bit, perhaps generate a hypothesis to test that speculation.

A bit of background:  Blanchard tested a hypothesis that there were only two types of transsexuals, rather than three or four as others had proposed.  The earlier researchers / commentators that had proposed that there were three basic types usually settled on Asexual, Homosexual, and Transvestite (Person and Oversey, Bentler, and even Green), though they didn’t always agree on the labels, nor on which group was the “real”  or “primary” transsexual.  Hirschfeld had used four types of people in general, not just for TS/TG folks;  Homosexual, Bisexual, Heterosexual, and Asexual.  Since this four grouping also would encompass the three grouping, it seemed prudent to start with four and see if there really were differences.  From this, Blanchard tested for autogynephilia and other characteristics, such as life arc (age of transition, level and quality of childhood gender atypicality, etc.).  From his study, he noted that three groups were very much alike in that the majority reported autogynephilia and had similar life arcs (statistically speaking) such as late transitioning, while one of the four was quite different, showing very, very low percentage reports of autogynephilia, greater childhood gender atypicality, and very early transitioning.  One group was “Homosexual” and the other was “Non-Homosexual”.  Thus, his data showed only two groups as being statistically separable, but with a non-statistically-significant (meaning, it could be noise in the sample) trend for the bisexual and asexual groups to report less transvestic autogynephilic individuals.  This was the state of things in the 1980’s and ’90s.

Then came the Nuttbrock study with many more subjects (N=571) in the late 2000’s.  The data exactly duplicated Blanchard’s study from twenty years previous.  And having enough subjects meant that the trend of having fewer bisexual and asexual individuals reporting transvestic autogynephilia (in this case, erotic arousal to cross-dressing) was shown to be statically valid, at least for bisexuals; There were so few (N=12) asexuals, that the trend might still be random noise, but when combined with Blanchard’s subjects, which reported about the same way, it gives us some additional confidence that it is real.  So, as I wrote earlier last year, we are now left with trying to generate new hypotheses to test that might explain this trend.

So, what is the trend?  Let’s use the Nuttbrock data:

Self-reported       Homosexual     Heterosexual    Bisexual   Asexual
Sexuality               (androphilic)   (gynephilic)
Number:                  (n=391)               (n=71)               (n=96)       (n=12)
Autogynephilia       23.0                     81.7                    67.7            66.7

And looking at age of transition:

Hormone Therapy                                                                                AGP
Adolescent(n=171) 91.8                 0.6                         7.6            14.0
Adult (n=242)           64.5               13.2                       22.2           42.6
None (n=158)            54.2               25.5                      20.2           59.5

And finally, we should add, from the Doorn study (N=31), that 85% of adult cross-dressers (TV/CD) reported erotic arousal to cross-dressing.  This number is vitally important, as it tells us the number that will recognize and admit to such arousal.  Many researchers have looked into the issue of why less than 100% this group does report such arousal.  Those that don’t report ever having had such arousal, none-the-less report feeling the same exact type of satisfaction and most importantly, compulsive need, to cross-dress as those that do report such arousal… and in fact, the same type of feelings that those who found cross-dressing erotically arousing at one time, but no longer experience it as such.  For our purposes here, I think we can assume that they all experience autogynephilia in some form, in desire, if no longer such intense arousal.  That being so, we can note the amazing similarity to the percentage reporting AGP arousal found in the gynephilic transsexual group, who most resemble adult cross-dressers in life arc (up to but not including transition itself).

Here is where I am going to speculate, possibly rather wildly, but such is the nature of hypothesis generation.  If we assume that gynephilic transsexuals are all autogynephilic, and that we can take that 82% number as a baseline of how many autogynephilic transsexuals, regardless of self-reported sexuality, report that fact, we can use it as a means of making an estimate of how many autogynephilic transsexuals there are in a given sample:

Self-reported       Homosexual     Heterosexual    Bisexual   Asexual
Sexuality               (androphilic)   (gynephilic)
Number:                  (n=391)               (n=71)               (n=96)       (n=12)
Autogynephilia       28.0%                  100%                 82.7%         81.6%

And looking at percentage at age of transition:

Hormone Therapy                                                                                AGP
Adolescent(n=171) 91.8                 0.6                         7.6            17.1
Adult (n=242)           64.5               13.2                       22.2           52.1
None (n=158)            54.2               25.5                      20.2           72.8

Note that we still have an interesting anomaly in that the estimated percentage of adolescent “early” transitioners is still significantly fewer autogynephilic individuals than would be predicted controlling for sexuality from the total study.  And, as expected, most of those who have not started HRT, would be autogynephilic.

So, if this estimation is correct, we are left with a question, why do fewer bisexual and asexual individuals report autogynephilia?  Several possible hypotheses come to mind:

1)  Fewer actual autogynphilic individuals who are bisexual and asexual recognize and admit to autogynephilia.

2)  Fewer self-reported bisexual and asexual individuals are autogynephilic (at least of the transvestic type).

3)  Some combination of the above two hypotheses will be found.

The safer bet is always to look at a combination of reasons, since real human lives are messy… but, if I had to take a guess regarding the bisexual group, I will go with more of the second hypothesis, as it fits with what we know of human sexuality in general, that many male-bodied people who self-report as bisexual do so because they honestly report that at sometime in their lives, they had some sexual experience(s) with women, but that their primary sexual orientation is androphilic.  If our estimate is correct, then 17.3% of the bisexual group is primarily androphilic.  I’d also bet that more of them transitioned at a younger age, than those who were autogynphilic (and experience pseudo-androphilia).  The asexual group is more of a question mark, but my guess would be the same, out of the twelve individuals in this study who report being asexual, two of them are primarily androphilic, but find their libido is low at the present time, or they are afraid of being emotionally hurt.

So, how do we test this hypothesis?

Addendum 1/22/2012:

I thought of a way to test this hypothesis using the reclassification from ‘homosexual’ to ‘non-homosexual’ that Lawrence performed on the Smith data set.  Originally, Smith et al. dichotomously sorted, by self-reported sexuality into “homosexual” and “heterosexual” (based on natal sex).  Lawrence then reviewed that original sort and resorted based on sexual history with natal females, as indicated by prior marriage to women.  That is, she looked at those who self-reported being exclusively androphilic and, in essence, assigned an implied assessment of being ‘bisexual’, based on their actual history combined with their stated sexual orientation.  She noted that the original ‘heterosexual’ and her resorted group had similar self-reported levels of autogynephilic arousal to cross-dressing.  However, when I look at the data, I note that though the are similar, they are not identical:

Percentage Reporting Cross-dressing Arousing
Self-Identified Heterosexual:  63%
Resorted Behavioral “Bisexual”: 52%

If we examine this as a ratio of Bisexual to Heterosexual self-report of autogynephilic arousal to cross-dressing we get 0.83.  The Nuttbrock ratio of Bisexual to Heterosexual self-reported arousal is 0.83.  They are identical!!!  Thus, this analysis supports the first, rather than the second, hypothesis, since in the Lawrence re-designated Bisexual group, we know that they are in fact non-HSTS because of their sexual history.  So, my original guess was wrong.  Nuttbrock’s Bisexual Group does not include etiologically homosexual transsexuals.

If the ratio of the two groups’ percentage reporting is the same, why aren’t the raw percentages?  Nuttbrock’s data was gathered in the open community, not in a clinical setting, as was the Smith data.  This may have led to more honesty, less social desirability bias.  The bias works equally on both self-reported heterosexual and bisexual groups, so that the ratio remains the same.

So why then do bisexual transsexuals report lower amounts of erotic arousal to cross-dressing?  Perhaps they experience other forms of autogynephilic arousal, as reported by Blanchard?  In any case, the Smith data set adds yet more evidence to the two type taxonomy, as already noted, but also adds to the observation that ‘bisexual’ transsexuals report slightly less arousal to cross-dressing than do ‘heterosexual’ transsexuals.

This area needs further research.

Addendum 1/23/2012

Lawrence resorted 23 out of 61 self-reported ‘homosexual’ transsexuals as not being exclusively so (23/61=38%).  If this percentage is similar, minus those who already sorted themselves as ‘bisexual’,  in the Nuttbrock data set, and that they reported erotic cross-dressing is the same percentage as Nuttbrock’s ‘bisexual’ group, then we can explain the reported figure of 23% of the androphilic group experiencing cross-dressing to be autogynephilically arousing.  That is to say, that about 33% of the “androphilic” group are not actually exclusively so.  Note that most of these started HRT as adults or are not on HRT.  This exercise reinforces the need for careful sorting based on actual sexual history, not simply by self-report.


A Further Assessment of Blanchard’s Typology of Homosexual versus Non-Homosexual or Autogynephilic Gender Dysphoria, Nuttbrock, et al. Archives of Sexual Behavior

Comments Off on Speculation…

Seeing the world in Grey and White…

Posted in Brain Sex, Female-to-Male by Kay Brown on January 8, 2011

…brain matter

It is exciting to see that neuroimaging science is getting to the point where we needn’t wait until subjects are dead before we can examine their brains in better detail. Exciting recent developments include a pair of papers from Spain in which Diffusion Tensor Imaging (DTI), a form of MRI, was used to compare FtM and MTF transsexuals to control men and women. The results were that transsexual brains exhibit white matter features, that are known to be sexually dimorphic, mid-way between male and female morphologies. The studies are doubly interesting, because the subjects had not yet begun exogenous Hormone Replacement Therapy (HRT) eliminating the possibility that the features were the result of activating effects of HRT.

So, does this mean that we have proven that transsexuals have “intersexed brains”?

Yes… and at the same time… No.

First, in the case of Female-to-Male transsexuals, it has always been noted that the majority were both gynephilic and masculine behaving since early childhood. So, these results are hardly surprising. However, the exact same thing can be said for butch lesbians, who do not necessarily identify as men. So, if we were to image the brains of a population of such butch lesbians, would we see the same masculinized neurological features? Would we see that transmen were more masculinized than butch lesbians, who would in turn be more masculinized than average lesbians? Also, if we imaged the brains of a population of gay identified transmen (“transfags”), would we then see that they do not share such masculinized features? I’m betting we would, and eagerly await the results of such imaging.

Which brings us back to the Male-To-Female subjects in the Spanish DTI study. As Lawrence has pointed out, different cultures have varying prevalence of non-homosexual transsexuality. Spain is one of those cultures where very few of the MTF transsexuals are non-exclusively-androphilic, with only 9% of the MTF transsexuals being non-exclusively-androphilic. In this case, as reported in the paper, there were no such non-exclusively-androphilic subjects. All were described as “early onset” and androphilic. Thus, this study only tells us about one type of transsexual.

Sadly, there wasn’t a gay male control group. It would have been interesting to see if the MTF transsexuals exhibited the same or more feminization (non-masculinization) of these sexually dimorphic brain structures. One hypothesis regarding exclusively androphilic transsexuals is that they are the extreme end of a range of feminine (hypo-masculine) homosexual males, who find that their innate femininity sufficient that living life as women is a better rational as well as emotional choice. (That is to say, that they are so feminine, like women, that they are women at heart.) I eagerly await DTI imaging studies that can test this hypothesis.

Now that we have DTI data on the HSTS population, we must gather data on the non-HSTS population to see if they do or do not exhibit such partially feminized (hypo-masculinized) features. Once again, we have no data that would show, conclusively, one way or the other, that both types of MTF transsexuals share neurologic features that would serve as the basis of a unifying etiology.

It’s not much of a limb to go out on, given the very strong data that shows that there are two types with two separate etiologies, but I’d be willing to bet that that when we do have such DTI images of gynephilic MTF transsexuals, that they will be no different than typical straight men, at least with respect to these particular sexually dimorphic features.

Which brings us to another study (Luders) looking at grey matter instead of white sexually dimorphic brain structure differences between MTF transsexuals and control men and women, this time one that suggests that for a very localized spot, they found a feature in MTF transsexuals that is different and possibly ‘more feminine’ than male controls. However, globally, the MTF brains were shown to more like men, and if anything, perhaps more masculine than control men, as they showed less grey matter then men, who show less grey matter then women. But one spot, the right putamen, is definately different in MTF transsexuals than in control men, showing more grey matter, in fact, more grey matter than the control women.

Sadly, as the researchers themselves point out, they may have included both exclusively androphilic and non-exclusively androphilic transsexuals without analyzing them separately; Of the 24 subjects, six self-reported to be androphilic and 18 reported to be gynephilic. It is heartening that the researchers recognize that in the future, they need to make the distinction and analyze the two types separately. Further, as these TS folk are pre-HRT, their ages may be used as a statistical proxy for our purposes, as MTF transkids usually (>90%) begin transition and HRT before age 25: the mean age was 47 years old, standard deviation of 13 years, with the range from 23 to 72. We know from Lawrence’s re-sorting from the Smith data set that a fair number of MTF transsexuals who self-report being androphilic are not exclusively so, thus it is likely that less than six of the subjects are exclusively androphilic.  A little math will show that that with a normal distribution, only one of the 24 subjects would have been expected to have been 25 or younger, so this 23 year old may be the only one.  Therefor, as Luders et al did find a statistically significant signal, we might infer that it is more likely that it came from the non-exclusively androphilic type, and thus likely also autogynephilic.

While it may be tempting to declare that this feature found in the right putamen proves that MTF transsexuals, in-fact, that gynephilic MTF transsexuals, have a part of their brains that is femininized, this conclusion should not be drawn from this data, at this time. Although it may be a marker of transsexual neurology, it may not be from a feminization of this region, especially given that it shows more grey matter than both control men and women. It may be a marker of an unusual neurological development altogether unrelated to sexual dimorphism. It may in fact be a marker for autogynephilia, or proneness to erotic target location errors. In fact, such a marker is expected to be found. We need further studies.

First, we need to compare the two types of transsexuals to test if this is a marker of a unifying neurological feature of transsexuality. Second, we need to compare both of them to individuals who are sexually aroused at the thought of becoming amputees, to test the hypothesis that this may be a marker of proneness to erotic target location errors.

But, in the mean time, the world is not so much to be understood as black and white, but shades of grey and white.

For more essays on trans-brains see Brain Sex.

Addendum 1/28/2011:

A good link to explore the first two papers further:

Addendum 12/19/2011:

You may wish to read a new blog entry on an additional paper on this topic.


Rametti G, Carrillo B, Gómez-Gil E, Junque C, Segovia S, Gomez A, Guillamon A., “White matter microstructure in female to male transsexuals before cross-sex hormonal treatment. A diffusion tensor imaging study.”

Rametti G, Carrillo B, Gómez-Gil E, Junque C, Zubiarre-Elorza L, Segovia S, Gomez A, Guillamon A., “The microstructure of white matter in male to female transsexuals before cross-sex hormonal treatment. A DTI study.”

Luders E, Sánchez FJ, Gaser C, Toga AW, Narr KL, Hamilton LS, Vilain E., “Regional gray matter variation in male-to-female transsexualism.”

Tagged with: ,

Comments Off on Seeing the world in Grey and White…