“…It has been found that generalization following TUF is greatest to linguistically related sentences (Ballard & Thompson, 1999;Jacobs & Thompson, 2000;Thompson & Shapiro, 2005;Thompson, Shapiro, & Roberts, 1993;Thompson, Shapiro, Tait, Jacobs, & Schneider, 1996;Thompson et al, 1997), lending support for specific deficit hypotheses that suggest that specific syntactic operations lie at the root of deficits. Furthermore, generalization occurs frequently from complex to simple structures but less frequently from simple to complex structures (Thompson, Ballard, & Shapiro, 1998;Thompson & Shapiro, 2005Thompson, Shapiro, Kiran, & Sobecks, 2003;Thompson et al, 1993Thompson et al, , 1997. Together, these observations have led to the establishment of the complexity account of treatment efficacy (CATE), which states that training complex sentences leads to generalization to less complex sentences with shared underlying linguistic properties (Ballard & Thompson, 1999;Jacobs & Thompson, 2000;Thompson & Shapiro, 1995Thompson et al, 1997Thompson et al, , 1998Thompson et al, , 2003.…”