As well as sampling the rioja and pinchos, I'm presenting two posters. Please do drop by and say hello if you're attending.
108.143: Oscillatory neural responses to speech and nonspeech sounds in a nonverbal child with autism
Thursday 12pm, Poster 143
|Click to enlarge|
The study involves G, an autistic girl who has never spoken. Shu used magnetoencephalo-graphy (MEG) to record the magnetic fields produced by G's brain while she listened to different sounds.
Interestingly, G showed a very weak response to speech sounds, but an extremely strong response to nonspeech sounds. This was unlike any of the other children we've tested - either typically developing children or verbal autistic children.
We're not sure what to make of the data so would love any feedback. Given G's profound language difficulties, it's tempting to assume that her differential brain response to the two sounds is related to the differences in "speechiness" of the sounds. But we don't know that for sure. It's also unclear at this stage how representative G is of other kids like her.
Nonetheless, we think these are interesting preliminary findings. At the very least they show that it's possible to conduct MEG studies with nonverbal kids, who are usually excluded from neuroimaging research.
144.148: Individual differences in homograph reading amongst Hebrew-speaking autistic children
Friday 3pm, Poster 148
|Click to enlarge|
Homographs are written words that have multiple meanings and, sometimes different pronunciations. For example, the word "tear" is pronounced differently depending on whether the context is crying or ripping.
One of the more consistent findings in autism research is that people with autism tend to be bad at working out which pronunciation of the homograph to use. This is seen as key evidence for the "weak central coherence" account, which attributes many features of autism to an inability to use contextual information.
The advantage of conducting the study in Hebrew is that there are many more suitable homographs in Hebrew than in English - mainly because written Hebrew doesn't really bother with vowels! This meant we could give the kids lots of homographs to read, so our data should be much more reliable than previous studies conducted in English.
Our main finding was that there were statistically reliable individual differences in homograph reading skills within our autism group. In other words, only a subgroup had difficulties. And importantly those differences couldn't be explained away in terms of boring things like older kids or kids with generally stronger reading skills being better at the task.
In fact, the best predictor of homograph reading was performance on a picture naming task. The numbers in the study are relatively small, so again need to be treated with caution - but they suggest (to me at least) an alternative view of homograph reading difficulties in autism. Perhaps the problem is less to do with comprehension of the sentence context and more to do with selecting the right word to say when speaking or reading aloud.
SFARI (the Simons Foundation Autism Research Initiative) will be live-blogging the conference. You can also follow the #IMFAR2013 hashtag on Twitter.