Dyslexia Brain Differences Show Up Before Formal Reading Instruction

Last time, we talked about early behavioral differences between prereading children that predicted future reading impairment. Today, we’re continuing on the theme of early predictive differences, this time in the brain.



The question of how early brain differences arise is a worthwhile one. We want to know whether the dyslexic brain is tackling reading differently from the very beginning or if these brain differences arise after some reading experience, perhaps reflecting compensatory strategies that the children may have developed.

Specht and colleagues (Scandinavian Journal of Psychology 2008) conducted a brain imaging study on Norwegian children (a good population to study because reading instruction starts in second grade in Norway). The basic goal of their experiment was to scan 6 year olds (before they learned to read) and see if they process words differently depending on their risk for dyslexia. Unlike the Lervag study, this study was not longitudinal. Specht and colleagues determined which kids were at risk for dyslexia using a risk index that took into account factors like heredity, language development, and other factors.

Kids looked at four kinds of stimuli during an fmri scan: pictures, logos, regular words and irregular words while performing a categorization task (“Is this something you can play with?” and similar questions). I won’t spend too much time comparing between conditions because I’m not clear on what characteristics were controlled for between the stimulus types.

There were differences between the at-risk and normal reading group in all conditions. There were several interesting findings. First, risk index score correlated with increased activation when looking at words in the angular gyrus, an area that has been reported to be involved in language/phonological processing.

Our old friend, the visual word form area, also shows up. At a more liberal statistical threshold ( p<.001 uncorrected and with a small volume correction) they found that risk index score correlates negatively with left occipitotemporal activation when viewing irregular words.

So what does this mean? For one thing, differences arise early, before formal reading instruction. The two groups did not differ significantly on standardized reading measures at the time of testing (although there was a trend (p<.096) towards a difference in reading scores. So there seems to be something different about how these kids approach words from the very beginning. It would be interesting to know what is driving these differences. I wonder what strategies the kids were using to in the scanner to complete the categorization task, especially since they couldn’t read yet. For word conditions, kids only had an accuracy of 20-30%. What were they doing for the words they couldn’t read? Were the scanner differences driven by the words they could recognize, or all the words?

I’m particularly puzzled by the VWFA findings. The VWFA is usually thought to develop based on expertise with letters Does this mean that even before reading instruction there is some difference in expertise between kids at risk and not at risk? Interesting questions for future investigation.

Specht K, Hugdahl K, Ofte S, Nygård M, Bjørnerud A, Plante E, & Helland T (2009). Brain activation on pre-reading tasks reveals at-risk status for dyslexia in 6-year-old children. Scandinavian journal of psychology, 50 (1), 79-91 PMID: 18826418