What is the role of the semantic system in acquired dyslexia in a transparent orthography?

Imageability has been shown to affect word naming in English where print-to-sound translation is assumed to be slow, error prone and inefficient (Strain, Patterson & Seidenberg, 1995). However, in transparent Turkish orthography imageability effect emerges only as a function of reading skill (Raman & Baluch, in press). It is thus of interest to examine how the role of semantics in word naming in acquired dyslexia in a totally transparent orthography compares to English (Saffran, Bogyo, Schwartz & Marin, 1980).
A case study of a native speaker of Turkish with acquired dyslexia (age 67) is reported. The patient was asked to read aloud a list of high and low frequency words manipulated factorially with imageability. There were significantly more errors on low frequency words than high frequency words but no difference in errors between high and low imageable words. Implications of findings are discussed in relation to orthographic transparency.