, 1991 and Tallal et al., 1994). Although the correspondence between the two sets of studies is impressive the pattern of abnormalities in SLI also differs from that seen in the KE family

in several ways. In the current study, grey matter in the posterior temporal cortex in SLI is significantly RAD001 order decreased relative to controls, whereas it was increased in affected KE family members. Similarly, the putamen was found to have increased grey matter in affected KE family members, whereas we found no structural differences in the putamen in SLI. Finally, the caudate nucleus was found to be significantly reduced in volume in affected KE family members relative to their unaffected relatives and functionally overactive in a PET study of word repetition (Watkins et al., 1999). In our SLI study and the functional MRI study of the KE family, the caudate nucleus was not reliably activated by the task used and no group differences in function were detected. Also, the unaffected siblings in our study had significantly less grey matter in the caudate nucleus bilaterally relative to the typically developing controls

and did not differ significantly from their siblings with SLI. The latter suggests that reduced caudate nucleus volume can be considered a heritable risk factor for SLI but requires additional deficits Androgen Receptor Antagonist to affect language development. Alternatively, the siblings in our study have some protective factors, plasticity or compensatory mechanisms available to them that are unavailable to their affected

siblings. The increased grey 3-mercaptopyruvate sulfurtransferase matter of the left central opercular cortex in the unaffected siblings relative to the SLI and control groups might reflect such compensatory mechanisms. The similarities between the functional and structural abnormalities in this group of people with SLI and the affected members of the KE family are likely a reflection of the similarities in their behavioural deficits. Both groups have impairments in nonword repetition and oromotor function. Whether any of the individuals with SLI that we studied also have a mutation in FOXP2 is unknown, but is unlikely, given the rarity of such mutations in individuals with SLI ( Newbury & Monaco, 2010). In a larger population of SLI, however, allelic variation in a downstream target gene of FOXP2, CNTNAP2 was found to correlate with performance on nonword repetition ( Vernes et al., 2008), so investigations of this gene in our participants are warranted. Previous developmental studies measuring grey matter volume and cortical thickness have revealed gradual linear and nonlinear reductions in these measures that continue into the second decade (e.g., Giedd et al., 1999, Giorgio et al., 2010 and Gogtay et al., 2004).