The Incredable Shrinking Brain…

As I’ve noted before, the brain responds to sex hormones.  Androgens tend to cause the brain to grow larger.  It appears that the absence of androgens in males, allows the brain size to shrink to closer to female norms over time.  In fact, such changes can occur quite rapidly.  Significant changes occur in the volume of the hypothalamus.

Lawrence has already critiqued two earlier studies of sexually dimorphic brain structure in (or near, depending on definition) the hypothalamus, in which she questions the author’s assertion that the BSTc is sexually dimophic and is an organizational effect (“locked in” during development) and is sex-reversed in transsexuals.  Lawrence argues strongly that this could not possibly be an organizational effect, since the BSTc is not sexually dimorphic until adulthood.  These studies were conducted in connection to a group of subjects from the Nederlands, with a common author, Dick Swaab.  Here I would like to explore a more recent study involving the very same subjects, joined by a few additional subjects, of another area, the hypothalamic uncinate nucleus, which is known to be sexually dimorphic.  Again, Swaab and his co-author claim this is not only sex reversed, but is again, to some degree, an organizational effect.  I question this conclusion, based on an analysis of their data.

Interestingly, the authors seem to have responded to Lawrence’s criticism regarding including only “Non-Homosexual” transsexuals in their study, in that in this third paper, they include categorization of the subjects into Blanchard’s two groups, “Homosexual” and “Non-Homosexual”.  All of the subjects that were in the earlier two studies are now either clearly identified as “Non-Homosexual” or “Not known”.  Only one new subject is identified as “Homosexual”.  As Lawrence notes, we can infer the likelihood of the not knowns as being “Non-Homosexual” based on the age at which they transitioned, as indicated by when they began HRT.  The youngest “Not known” is 36.  This strongly suggests that they are all “Non-Homosexual”.

The authors also categorized the subjects by “GID type: onset“.  For the life of me, I can’t imagine what meaning this has, as they list a subject who began treatment at age 64 as “early”, and even one who never sought any treatment at all, who died at age 84, as “early”.  I suspect this is a reference to self-reported awareness of their gender dysphoria.

This study uses a group of men who were castrated as part of treating prostate cancer as controls for reduction in circulating androgens, which is a better control group than the previous studies used.

If we accept, uncritically, the analysis of the authors, there appears to be a significant difference between the castrated men and the MTF transsexuals, while at the same time, no difference between the MTF transsexuals and natal females.  However, there is a catch, a BIG CATCH !

When one is studying these anatomic details of the brain, we are not simply slicing and dicing the brain and recording how big a clearly differentiable nodule of the brain is.  Actually, these parts of the brain are largely indistinguishable to the casual observer.  In order to even see these structures, one applies various stains to the thin slices of neural tissue where one hopes to find the structure in question.  These first slices are done literally blind, based on the distances from anatomic landmarks from maps of the “average” human brain.  Sometimes, they miss!  (It’s only about the size of a grain of sand.)  And sometimes, the stains don’t “take”.  And even when the stains do take, the boundaries between the clump of nerves of interest, and the surrounding nerves are often very, very fuzzy!  It takes an experienced bio-technician, working very carefully, to trace the boundary using computer software of photos of the region in question.  To get an idea of how difficult this can be, please examine the photos included in the paper.  Honestly, can you clearly tell where the borders are?  I sure can’t.

So, here’s the catch… Swaab and his co-author have chosen to include those subjects where they couldn’t identify the nucleus in question.  They give the numerical volume of the missing nucleus as “zero”.  This is akin to a school teacher misplacing a pupil’s exam paper and giving that student a score of “zero”!  An assertive student would then argue that they could also be given a score of 100%, given that the teacher can’t prove that they got any of them wrong.  Giving a missing data point a numeric value is very questionable.

Why do I argue that this is questionable?

Note that this makes several of the data groups bimodal, with a cluster at some non-zero value, and a cluster at zero, for each group that had such missing data.  This skews the mean downward.  The authors argue this is an acceptable analysis, noting that the MTF and natal female groups have approximately the same number of subjects in which they couldn’t identify the nucleus.  However, note that one of the control males also had no identifiable nucleus.  Are we to accept that such a bimodal distribution is real?  I don’t.

So, if we throw out these null results, given that we can’t reasonably give them a numerical volumetric value, is there still a significant difference between the groups?  Once again, we note that there is no difference between the MTF and natal female data.  But now, there is also no difference between these two groups and those men who were castrated!

Simply put, the data clearly shows that though the hypothalamic uncinate nucleus of the average MTF transsexual is the same as natal female, after years of HRT, so is that of castrated non-transsexual men.  Thus, the data clearly shows that this is purely an activational effect, and in no way gives us any clues as to the etiology of MTF transsexuals.

Addenda 3/2/2010

If the size of the INAH3 is an activational effect of the absence of androgens in the MTF transsexuals, it would follow that the time spent on HRT might correlate with the size.  It was always possible that it happens too fast to be in the useful time window of this data set, but I plotted the data anyway, just to see what would happen.  Interestingly it looks like the shrinkage of the INAH3 takes on the order of years, as suggested by this graph:

For more essays on trans-brains see Brain Sex.

References:

Hulshoff Pol, H. E., Cohen-Kettenis, P. T., Van Haren, N. E., Peper, J. S., Brans, R. G., Cahn, W., et al. (2006). Changing your sex changes your brain: Influences of testosterone and estrogen on adult human brain structure. European Journal of Endocrinology, 155(Suppl. 1), S107-S114.
http://eje-online.org/cgi/content/full/155/suppl_1/S107

Garcia-Falgueras A, Swaab DF. A sex difference in the hypothalamic uncinate nucleus: relationship to gender identity. Brain. 2008
http://brain.oxfordjournals.org/cgi/reprint/131/12/3132