Imageability and frequency effects on visual word recognition: evidence from a transparent orthography.

The aims of the two experiments reported here were to examine the effects of imageability on naming of very skilled as compared to skilled adult Turkish readers and on previously skilled (poor) readers.Participants were all adult native speakers of Turkish (n = 44) who were either first year undergraduate or foundation level students at the Eastern Mediterranean University, Northern Cyprus, who received course accreditation for their participation. This group were further divided according to level of skill as explained below into two groups of very skilled (n =16) and skilled (n =28) readers. The results of Experiments 1 and 2 showed an interesting pattern of findings: in Experiment 2 only frequency was found to be significant, although a similar pattern was also found in Experiment 1, i.e. there was only a significant main effect for frequency; there was, however, a significant 2-way interaction of skill by imageability and a 3-way interaction of skill by frequency by imageability. Further analysis showed that this was due to i) very skilled readers showing more imageability effects than skilled readers and ii) the effect of imageability being significant for low-frequency words.