From a 1999 lecture by the director of an NIH agency:
Distinguishing between disabled readers with an IQ-reading achievement discrepancy and those without a discrepancy reflects an invalid practice at the beginning stages of reading. Specifically, children with and without a discrepancy do not differ in the information processing skills (phonological and orthographic coding) that are necessary for the accurate and rapid reading of single words. Likewise, genetic and neurophysiological (functional MRI) studies have not indicated differential etiologies for reading disabled children with and without discrepancies. Converging data from several NICHD sites also indicates that the presence and magnitude of IQ-reading achievement discrepancies are not related significantly to a child's response to intervention.
One of the most hideous (I use the word deliberately) policy errors in public education is the baseless conceit that resources should not distributed on the basis of a disability, but rather focally on the discrepancy between "global IQ" and "measured outcome". One presumes this is based on a belief that:
1. Low IQ makes intervention futile.
2. There is something special about a focal deficit that makes this the preferrred tartget of intervention.
3. Children with low IQ results are best locked in a dark room, preferably without a key.
This is a perverse belief; the more one examines it the more it is seen a particularly repulsive combination of belief contradicted by evidence, and a condemnation by IQ test. And yet, it remains the fundamental basis of assigning resources for educational intervention in most school systems.