Although the
role of hormones, and specifically of testosterone in the development of fetal
organs, including the brain, is unquestionable, the psychological correlates of
these influences are uncertain. The book by Baron-Cohen and co-workers presents
empirical research on this subject and claims to provide evidence that the
levels of fetal testosterone (FT) correlate with eye contact at 1 year of age
and with language ability at 1.5 and 4 years.

This 100-page
book is basically divided in two parts of equal weight: the first half is
devoted to a reappraisal of long known embryological, biochemical and
physiological data, whereas the second half describes the research carried out
by the authors. Overall, the whole work is organized as a long scientific
article and would probably better figure in a specialized journal than in a
book.

Fetuses in utero
are exposed to substantial levels of female sex hormones (i.e. estrogens and
progesterone) of maternal origin; testosterone levels in the mother and in the
female fetus are low. In the male fetus testosterone is produced in substantial
amounts by the testicles. As a consequence, FT levels measured in the amniotic
fluid are on average two to three times higher in male than in female fetuses,
although the individual variability is high.

FT is the
major determinant of the anatomic sex of the fetus or the newborn (i.e. whether
the external and internal genitalia are of the male or female type), and
defects in testosterone biosynthesis or recognition by peripheral organs often
cause a chromosomically male individual to develop female genitalia.

The brain,
though not a genital organ, presents subtle sex linked differences, the best
demonstrated of which is the extent of lateralization. In men the corpus callosum
(i.e. the major connection between the left and right cerebral hemispheres)
would be on average less extended than in women, and this difference might
contribute to the small but reliably measurable sex-related differences in some
cognitive tasks (e.g. language or geometry tests) or neurological abilities
(e.g. the incidence of left handedness or the linguistic competence of the
non-dominant cerebral hemisphere). As the authors recognize, FT levels are only
one of the possible causes of the sexual differences in the brain anatomy, and
genetics might play an equal or more important role: after all an entire
chromosome distinguishes men from women.

The second
half of the book describes a research carried out by the authors to assess the
relationship between FT levels at the end of the fourth month of gestation and
emotional or cognitive performances at 1, 1.5 and 4 years age. The methodology
of the study is straightforward: the authors access a bank of frozen amniocentesis
specimens and, after a careful and ethically sensible screening, unthaw the
specimens of normal babies, measure FT together with other relevant biochemical
parameters, and test the babies (or children) for psychological performances.
In the case of 1 year old babies the frequency of eye contact to the mother is
measured, and it is found that female babies make more frequent contacts than
male ones. At 18 months and four years of age, the language abilities are
measured.

Unfortunately,
the data fall below the expectations and the most relevant differences observed
are the well known sex-related behavioral differences between male and female
children. In some case the authors are forced to recognize this point explicitly
(e.g. p. 87-88); in other cases the authors do so only implicitly.

Sex
differences in behavior may be assigned to several biological and social
factors, and FT levels are but one of these; genetics may be expected to play a
role as well as the possible differences in parental behavior. Therefore, the
authors concede that within sex differences are better indicators of the
relevance of FT levels than between sex ones (p. 77). In spite of this
consideration, the relationships between FT levels and within sex behavioral
differences are granted minor emphasis in the book: none of the 6 figures and
14 tables that illustrate the results of this research is so constructed to
allow the reader to get a clear picture of the relationship between FT levels
in either boys or girls and their performance in the psychological tests. In
particular, the single most relevant figure, i.e. a graph bearing FT levels on
the abscissa and psychological performance of boys or girls on the ordinate, is
to be found nowhere in the book (although it is confusedly described in the
text), and the nearest information, the tables of regression parameters, do not
distinguish between boys and girls.

In
conclusion, in spite of its obvious merits in dealing with methodological (e.g.
non random sampling of mothers undergoing amniocentesis; short time window
available for this medical procedure) and ethical issues, and in reviewing
previous contributions, this work places an ungranted emphasis on FT levels and
fails to provide convincing empirical evidence; the authors’ thesis may well be
true, but is not demonstrated.

©
2005 Andrea Bellelli

Andrea Bellelli
has an MD and a degree in psychology, and teaches biochemistry in theMedical School
of the University of Rome, Italy.

Categories: Psychology