The digit ratio and autism: ALSPAC says no

"In this population-based study, there was no strong evidence of an association between 2D:4D [second-to-fourth digit ratio] and ASD [autism spectrum disorder] diagnosis or traits, although the CIs [confidence intervals] were wide. These results are not consistent with the extreme male brain theory."

So said the study results from Anna Louise Guyatt and colleagues [1] (open-access) who investigated the 2D:4D ratio as part of the ALSPAC (Avon Longitudinal Study of Parents and Children) initiative. The 2D:4D ratio in case you didn't know, is the idea that the subtle difference in the length of the index finger and ring finger might be a marker for androgen exposure during the nine months that made you/us.

"The index to ring finger ratio (second-to-fourth digit ratio, 2D:4D) has been widely used as a proxy for fetal testosterone exposure in autism research" is the link back to autism and onwards the idea that testosterone exposure might be part and parcel of the extreme male brain (EMB) hypothesis ("that children with ASD exhibit an exaggerated form of the male cognitive profile... and proposes that prenatal androgens are plausible biological candidates.")

ALSPAC data was used to "examine the association between 2D:4D and ASD and various autistic trait measures in a population-based cohort in the UK." Further: "The primary hypothesis being tested was that lower 2D:4D would be associated with ASDs and ASD traits."

Analysis of the 2D:4D ratio was undertaken in 56 ASD cases, 616 "high risk trait group" participants and some 5300 other children who were part of ALSPAC. Aside from an autism diagnosis (or not) and various other measures looking at ASD traits, other data were also collected on things like socio-economic status (SES) as possible confounders. An IQ measure was also included for "44/56 of included participants with ASDs."

As per the opening statement, the results were not exactly convincing that the 2D:4D is associated with an increased risk for ASD. There was some weak evidence that 2D:4D ratios might be linked to autistic traits but, and it is an important point: "These associations were mostly directionally discordant with the extreme male theory of autism, and given the number of comparisons made, and the lack of consistent association between ASD diagnosis and 2D:4D in this study, we consider that they were due to chance."

ALSPAC as an initiative, carries some significant methodological strengths insofar of the hows and whys of data collected. I've mentioned previous results from the cohort on this blog before (see here and see here for example). Obviously ALSPAC is not perfect (no cohort ever is) and pertinent to the Guyatt results, the authors acknowledge that there may have been a bias towards the applicability of their results "being weighted towards Asperger syndrome and other high-functioning ASD cases" in light of their participant group.

That being said, I do think these results add to other research suggesting that once again, we should be guarded against making too many sweeping generalisations when it comes to the autism spectrum. In a previous post on the finger length ratio (see here) some of the other 'correlates' of the 2D:4D ratio were mentioned, excluding the 'specificity' of any findings to just autism. Other research on cord blood testosterone levels and autistic traits (see here) has been similarly critical of the EMB hypothesis.

But that might not mean that is all bunk though [2]...

Music: Phats & Small - Turn Around.

----------

[1] Guyatt AL. et al. Digit ratio and autism spectrum disorders in the Avon Longitudinal Study of Parents and Children: a birth cohort study. BMJ Open. 2015; 5: e007433.

[2] Baron-Cohen S. et al. The “Reading the Mind in the Eyes” Test: Complete Absence of Typical Sex Difference in ~400 Men and Women with Autism. PLoS ONE. 2015; 10(8): e0136521.

----------

Guyatt, A., Heron, J., Knight, B., Golding, J., & Rai, D. (2015). Digit ratio and autism spectrum disorders in the Avon Longitudinal Study of Parents and Children: a birth cohort study BMJ Open, 5 (8) DOI: 10.1136/bmjopen-2014-007433

The finger length ratio

Apparently 2012 has been forecast to be quite a year for us humans. Depending on what you read (and believe), we are either destined for total annihilation or some population-wide transformation. For those of you who saw the recent Hollywood blockbuster 2012, it might be a good idea to book that space vacation fairly soon. Or maybe not...

I have to say that I have never really been one for the various methods suggested to predict our future. With my science hat on, the notion that we can 'see beyond' has never really excited me despite my openness to things like precognition (see the recent study from Daryl Bem for a more science-y take on this). Indeed, even if anyone were to be able to provide some experimental evidence for such an ability (and it is a big 'if'), I would have thought the first thing they would have done is pick the winning lottery numbers or failing that at least make some money on predicting things like the winner of the Eurovision Song Contest (now that would be some feat). If that person didn't personally want to financially benefit from the proceedings, they could always give the money to charity; in the current economic climate, there is most definitely a need.

One area in the whole 'future-reading' game has however been of perhaps slightly more interest to me: the idea that our hands may provide some clues to our future, at least our future health. I say all of this because of this study recently published, and picked up here, which suggests that finger length (ratio) may correlate with a form of motor neuron disease (amyotrophic lateral sclerosis, ALS). The study is an interesting one, in that researchers found that a low index finger to ring finger (2D:4D) ratio was significantly associated with ALS compared to controls (p=0.007). They further go on to link the findings back to an old friend, testosterone, and more specifically high prenatal testosterone exposure.

I have touched upon testosterone in relation to autism a while a back but did not go in to the detail. Testosterone and autism have been in the news of late following some focus on the use of the 'lupron protocol'. I will say no more on this particular issue. The basic theory is that autism is an example of an 'extreme male brain', whereby people with autism are hyper-systemisers, as opposed to empathisers. Systemisers are great at things like computers, engineering and map-reading abilities but not exactly great material for being a Priest / Vicar. Likewise I wouldn't necessarily put that great empathiser in charge of building your new skyscraper.

The reason put forward for such a male brain: elevated testosterone levels in-utero.

How do you measure such exposure? Well the digit ratio among other things has been put forward as quite a good correlate. What are the findings in relation to autism I hear you say? There is some evidence that autism follows the trend of low 2D:4D finger ratio potentially indicative of elevated prenatal testosterone exposure. I can see why the whole theory is so attractive. It is quite a simple theory (testosterone exposure = different way of thinking) and importantly, easily testable. The evidence base is quite large; and in terms of autism, bridges the gap between listed psychological/behavioural symptoms, physiological findings and even providing some information about possible transmission of autistic symptoms.

Having said that, there are a few issues with the theory which perhaps merit some discussion. First and perhaps foremost is the 'universality' of the theory to 'all' cases of autism. I think most researchers who set out to research/investigate/examine autism start in the same naive way that I did. You read the various research papers and textbooks which more often that not give you this impression that autism is autism (one condition with a few diagnostic variants) and from that you organise your research studies to look at autism as some kind of singular entity, comparing autism for example against not-autism controls (even this has issues), and if you are lucky a group like learning disability. The reality however as you progress through your career and wisdom starts to seep in, is that autism is very much not a singular autism, but rather autisms; and accompanying those autisms are as wide a variety of co-morbidities as you would expect to find in the not-autism population.

I started this post talking about finger ratio and risk of ALS. Wait a minute, low finger ratio and ALS, low finger ratio and autism. Does this mean that people with autism are at greater risk of motor neuron disease? Or on the flip-side, people with motor neuron disease for autism? There might be a familial possibility but I am pretty sure that ALS is not a significant co-morbidity to autism unless you perhaps factor in environmental exposure. Aside from the various papers suggestive of a relationship between (female) sexuality and finger ratio, finger ratio has also been linked to stock trading, athletic ability and risk of arthritis (the osteo- variety). Quite out of the blue, I also found this paper suggesting that latent toxoplasma infection might also show some involvement following on from my previous posts on T gondii and its potential 'mind controlling' effects (at least in the rat model).

I don't want to seem to negative on the whole finger ratio hypothesis. I am not saying that the 2D:4D ratio might not represent some effect in some cases of autism. The primary problems I have is when such effects are generalised to an entire population and potentially at the expense of an effect outside of the core condition.

Horoscope anyone?