Study: fMRI outperforms behavioral measures of developmental learning
Functional MRI (fMRI) and diffusion tensor MRI (DTI) more accurately predict the development of reading skills in dyslexic children than standard behavioral testing, according to a study published in the Proceedings of the National Academy of Sciences.
Developmental dyslexia occurs in 5 to 17 percent of children, creating a persistent difficulty in learning to read despite an absence of sensory or cognitive deficits, normal motivation and access to adequate reading instruction. "About one-fifth of individuals with developmental dyslexia manage to compensate for their underlying learning difficulties and develop adequate reading skills by the time they reach adulthood," according to Fumiko Hoeft, MD, PhD, of Stanford University School of Medicine in Palo Alto, Calif., and co-authors. Nevertheless, Hoeft and colleagues wrote, "the mechanisms by which this [developmental compensation] occurs remain largely unknown." Moreover, standardized measures of reading and language are poor predictors of dyslexic children's development of the capacity to read.
As a result, Hoeft and colleagues investigated whether neuroimaging indices of brain function could be used to predict long-term reading improvements in dyslexic children. The researchers conducted fMRI and DTI on 25 dyslexic and 20 non-dyslexic children, in addition to standard behavioral examinations, to explore the predictive capacity of both measures. All children, average age 14, were followed-up after 2.5 years.
The authors found that children with dyslexia that showed greater activation in the right inferior frontal gyri during a rhyme-judgment task and stronger white matter integrity in the right superior longitudinal fasciculus, including in the arcuate fasciculus, also demonstrated greater reading improvement upon followup. The authors noted that this "brain mechanism appears to be specific to dyslexia rather than reflecting growth in reading ability more generally, because typical readers did not show this pattern."
The researchers also discovered that dyslexic readers who showed gains in reading depended upon a right-hemisphere pathway, in contrast to the left-hemisphere pathway relied upon by non-dyslexic readers. This finding agreed with the increased activation observed in the right inferior frontal gyri. "Further, the right IFG [inferior frontal gyri] activation correlated positively with age in the dyslexic group, consistent with findings by Shaywitz et al., suggesting that this region plays an increasing role over time in dyslexic children."
The authors emphasized the capacity of modeling their brain activity findings to predict future improvements in reading, arguing that the "power of the brain measures to predict future reading gains in children with dyslexia (over 90 percent accuracy) may be contrasted with the finding that none of 17 widely used standardizes measures of reading and language predicted reading gains in these children."
"This study takes an important step toward realizing the potential benefits of combining neuroscience and education research by showing how brain scanning measures are sensitive to individual differences that predict educationally relevant outcomes," commented co-author Bruce McCandliss, PhD, of Vanderbilt University in Nashville, Tenn.
The authors discussed several limitations to their study. First, they noted that they could not determine the age at which these neurological findings could be considered relevant, suggesting that researchers should attempt to replicate these findings in younger age groups. Further, the authors did not randomly select children for controlled educational interventions, which may have biased their findings. Statistical analyses did not, however, reveal any differences in reading improvement based on age, IQ or initial task performance.
Hoeft and co-authors considered several educational implications to their findings, especially given that children who received intervention/instruction exhibited no additional progress compared to children not receiving the interventions. "First, it appears that gains in reading for dyslexic children reflect, at least in part, different neural mechanisms and pathways than those that support gains in reading for typically developing children. This encourages consideration of intervention approaches that capitalize on alternative reading strategies in addition to current interventions that build on typical reading instruction."
Additionally, Hoeft and colleagues proposed that this study demonstrates the importance of fMRI not just as a research tool, but as a practical method for producing improved clinical outcomes, saying that the "MVPA [multivariate pattern analysis] findings in the present study, that predicted which particular children would or would not advance in reading over the next 2.5 years, also suggest that such brain measures have promise for identifying individual children's future reading trajectory."
Developmental dyslexia occurs in 5 to 17 percent of children, creating a persistent difficulty in learning to read despite an absence of sensory or cognitive deficits, normal motivation and access to adequate reading instruction. "About one-fifth of individuals with developmental dyslexia manage to compensate for their underlying learning difficulties and develop adequate reading skills by the time they reach adulthood," according to Fumiko Hoeft, MD, PhD, of Stanford University School of Medicine in Palo Alto, Calif., and co-authors. Nevertheless, Hoeft and colleagues wrote, "the mechanisms by which this [developmental compensation] occurs remain largely unknown." Moreover, standardized measures of reading and language are poor predictors of dyslexic children's development of the capacity to read.
As a result, Hoeft and colleagues investigated whether neuroimaging indices of brain function could be used to predict long-term reading improvements in dyslexic children. The researchers conducted fMRI and DTI on 25 dyslexic and 20 non-dyslexic children, in addition to standard behavioral examinations, to explore the predictive capacity of both measures. All children, average age 14, were followed-up after 2.5 years.
The authors found that children with dyslexia that showed greater activation in the right inferior frontal gyri during a rhyme-judgment task and stronger white matter integrity in the right superior longitudinal fasciculus, including in the arcuate fasciculus, also demonstrated greater reading improvement upon followup. The authors noted that this "brain mechanism appears to be specific to dyslexia rather than reflecting growth in reading ability more generally, because typical readers did not show this pattern."
The researchers also discovered that dyslexic readers who showed gains in reading depended upon a right-hemisphere pathway, in contrast to the left-hemisphere pathway relied upon by non-dyslexic readers. This finding agreed with the increased activation observed in the right inferior frontal gyri. "Further, the right IFG [inferior frontal gyri] activation correlated positively with age in the dyslexic group, consistent with findings by Shaywitz et al., suggesting that this region plays an increasing role over time in dyslexic children."
The authors emphasized the capacity of modeling their brain activity findings to predict future improvements in reading, arguing that the "power of the brain measures to predict future reading gains in children with dyslexia (over 90 percent accuracy) may be contrasted with the finding that none of 17 widely used standardizes measures of reading and language predicted reading gains in these children."
"This study takes an important step toward realizing the potential benefits of combining neuroscience and education research by showing how brain scanning measures are sensitive to individual differences that predict educationally relevant outcomes," commented co-author Bruce McCandliss, PhD, of Vanderbilt University in Nashville, Tenn.
The authors discussed several limitations to their study. First, they noted that they could not determine the age at which these neurological findings could be considered relevant, suggesting that researchers should attempt to replicate these findings in younger age groups. Further, the authors did not randomly select children for controlled educational interventions, which may have biased their findings. Statistical analyses did not, however, reveal any differences in reading improvement based on age, IQ or initial task performance.
Hoeft and co-authors considered several educational implications to their findings, especially given that children who received intervention/instruction exhibited no additional progress compared to children not receiving the interventions. "First, it appears that gains in reading for dyslexic children reflect, at least in part, different neural mechanisms and pathways than those that support gains in reading for typically developing children. This encourages consideration of intervention approaches that capitalize on alternative reading strategies in addition to current interventions that build on typical reading instruction."
Additionally, Hoeft and colleagues proposed that this study demonstrates the importance of fMRI not just as a research tool, but as a practical method for producing improved clinical outcomes, saying that the "MVPA [multivariate pattern analysis] findings in the present study, that predicted which particular children would or would not advance in reading over the next 2.5 years, also suggest that such brain measures have promise for identifying individual children's future reading trajectory."