MRI, brain differences, and autism

MRI: Sagittal view of the brain. Photo courtesy of Wikipedia commons.

You may have read the news reports blaring the finding of an “autism test” that could lead to early and definitive diagnosis of autism. The new evaluation, which has proved worthy of its own name, the Lange-Lainhart test, uses magnetic resonance imaging (MRI) techniques to image brain areas to detect changes associated with autism.

I’ve been unable to find the complete paper, reported to have been published in Autism Research on Nov. 29; the journal has only papers through October available on its Website as of this writing. According to reports, however, the authors state that the new test was 94% accurate in identifying who was autistic and who wasn’t among 60 males tested. The participants in the study were ages  8 to 26; 30 were diagnosed with what the researchers call “high-functioning autism,” and 30 were typically developing.

The imaging technique the authors used involves tracing water diffusion along axons, the long connectors that link neurons to other neurons or tissues. This diffusion tensor imaging process yields an image that can highlight variations in the patterns of these connective pathways in different areas of the brain. This study focused on six brain areas associated with language, social, and emotional functioning, all of which are traditionally considered to be problematic among people with autism.

In the brains of non-autistic participants, the flow patterns were organized in a typical way that indicated connectivity among the brain regions. In the participants diagnosed with high-functioning autism, the flow was disorganized in a pattern common to the autistic group, indicating less connectivity and interaction and thus less exchange of information in the network. The researchers repeated the test on another, smaller set of participants, 12 with autism and 7 without, and produced similar results.

These findings imply that autistic brains may operate like a set of computer hardware components that cannot communicate very well with each other while still functioning perfectly well separately. There may be camera that captures a visual image without trouble or a microphone that captures a voice clearly, but the system lacks the network necessary to integrate the two inputs into a unified perception.

The news reports I’ve read on the study make a big deal out the prospect that this imaging breakthrough could lead to earlier diagnosis of autism, something that most experts believe is key to ameliorating some of its negative manifestations. But experts also urge the standard cautious optimism, and rightfully so.

For one thing, the participants in this study were ages 8 to 26, not within the time frame for early diagnosis of autism, and all of them were male. The study findings can’t tell us whether their brains present with these differences as a result of developing with autism, or whether they have autism because their brains are built this way. Before there can be talk of “early diagnosis” and linking these changes to manifestations of autism, we’d need studies showing these differences in much younger children. Further, given the frequent findings of differences between males and females on the spectrum, investigations involving autistic girls and women are necessary.

This study is not the first to use imaging to identify distinctions between autistic and non-autistic people. Other studies have also done so, finding pattern variations in the neuronal tracts of children with autism compared to children without it, in critical areas relevant to the clinical symptoms of autism.

While I find these results intriguing, I note one thing that no one seems to have commented on. In the reports I’ve read about this study, the researchers observe that currently, the only way to diagnose autism is based on a symptom checklist, questionnaires, screenings, and so on—any autism parents reading this will know that drill—and the ultimate call relies on the expertise of the medical professional conducting the evaluations. The implication of these comments is that we need some better, more unequivocal, less-subjective methods of identifying autistic people.

Yet, presumably the autistic boys and men they used for this study were diagnosed using just such subjective evaluation, and their autism diagnoses appear to have been confirmed in 94% of cases by similarities of MRI findings. In my mind, this outcome suggests that the process of subjective evaluation seems to be working pretty well. Of course, we’re a visual species and like our decisions to be given literally in black and white. Such MRI results may fulfill a need that has less to do with correct outcomes than it does with a dose of visual confirmation–and satisfaction.

About ejwillingham
Sciwriter/editor/autism-ADHD parent. SciMaven @ I speak my pieces @ & @

7 Responses to MRI, brain differences, and autism

  1. I rarely see accurate non-inflating science reporting in news media.

    Can you get current articles through the univ library – loan program? I mean maybe one of the other UT schools have that journal in their database….(like here).

    “The study findings can’t tell us whether their brains present with these differences as a result of developing with autism, or whether they have autism because their brains are built this way.” Astute and probably beyond the interpretive ability of your average reporter.

    I think there is more benefit than visual affirmation for identifying brain activity/function related to autistic behavior. Backtracking from lack of connective-ness between areas of the brain – does that not prompt more hypotheses about earlier development? Personally I hope for a more organic diagnostic method than behavior, eventually.

    Thanks for reviewing this, Emily.

    • biolotrix says:

      I’ve read the JAMA paper via a different source, and the Autism Research paper just wasn’t available at their Website. Don’t know why. I don’t have trouble obtaining full-text when I need it…but this one couldn’t be obtained.

      Thanks for your comments. The behavioral approach is time consuming and problematic, but it seems to be working OK, at least based on this report. Not everyone can have MRIs, and it’s particularly difficult for little ones, I know from experience, so we’ll probably have to retain behavioral eval in some form or another regardless.

      • I am glad you can access full text. I would be surprised if you couldn’t.

        My hopes for an organic diagnostic tool were not meant to disparage what is currently used. I agree that MRI is not an ideal diagnostic solution but where it can be used in research to learn more about all brains – well, I think that’s good.

  2. Ali says:

    Do you know what the researchers might have meant by “high functioning autism”? What a useless phrase–I’ve seen it used to mean people on the spectrum who don’t stim, or speak fluently, or have IQs above 70 (or 85 or 100 or whatever), among other things. Any of those could be a confounding variable, especially the language or IQ ones.

  3. Pingback: Questions of Race and the Identification of Students with Autism « Glenbowden's Blog

  4. Jon Brock says:

    Thanks for a great write-up. I’ve linked to this on my own blog post on another recent MRI study claiming to aid autism diagnosis. The same issues and limitations are relevant I think

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: