Autism is a fascinating topic and one that I have a personal interest vested in. I found two papers, a review of the literature on neurochemical correlates of autism disorder and an empirical paper investigating the predictive values of whole brain structural MR scans in autism.

I’ll address the latter paper first because it was dense, but I’ll be as quip and as intelligible as I can. It is stated in both papers that historically/currently the model of diagnosing autism is done by analyzing behavioral phenotypes (i.e. impaired social communication, social reciprocity and repetitive and stereotypic behavior). However, this process is time consuming and problematic in adults (who’ve developed coping skills). This paper examines the etiology and pathogenesis of autism by looking at brain structure and function via neuroimaging techniques.

There has been research on the physiology of the brain of autistic patients (in this paper ASD patients). Research has revealed a difference in grey and white matter in ASD patients. Using MR scans is not “new” however, the authors were critical of the methods of past research that analyzed brain structures and regions. The technical term is Voxel Base Morphology (for better explanation: http://www.tina-vision.net/docs/memos/2003-011.pdf). The problem was that this past procedure were univariate and looked at one region at a time and does not take into account the interrelationship of different brain regions. The new approach used a multivariate system of analyzing brain morphology.

The idea of this study was to look at the difference in grey and white matter using MRIs and using the multivariate support vector machines (SVM) (http://research.microsoft.com/en-us/um/people/cburges/papers/svmtutorial.pdf) because this method takes into account inter- regional correlations.I think by now we understand that regions of the brain don’t necessarily act alone. Individuals with the disorder most likely have abnormalities in the development of neural systems (rather than isolated regions) leading to differences in “intra-regional correlations” of brain metabolism, function and anatomy.

Anyways, the result is that two major systems in ASD were found to be significant predictors the fronto-striatal and cerebellar regions and the limbic system. The SVM and MRI technique also display results that can predict severity of ASD and therefore used to categorized group members.

The second paper was a review of the literature on the different neurochemicals including; serotonin, dopamine, norepinephrine, acetylcholine, oxytocin, endogenous opioids, cortisol, glutamate, and gamma-aminobutyric acid (GABA). The problem with past research has been the issue of subject quantity which makes it difficult to come to conclusions. Most research on the different chemicals are in their infancy however a few of the aforementioned chemicals appear to be significant.

Serotonin for example appears to have the most empirical evidence for a role in autism according to the literature and with the development of sophisticated imaging and genetic techniques researchers are beginning to identify potential abnormalities in central serotonergic functioning. The neuropeptide oxytocin may also play a role in the pathogenesis of the disorder.  The reviewers also conclude that future work need to take into account variables such as gender, race, pubertal status, and spontaneous chemical stress that are associated with the experience of neuro-testing.

There are plenty of avenues we could approach this topic anywhere from clinical and applied research behavioral to the neurodevelopment and etiology .

FYI: today, the newest edition of the DSMV-R is out! Go get it for your night stand. Also, a great read or for a casual peruse for those who take their time in the John.