Authors:

Norris, D. & Brown, G.

Reference:

Year of publication:

1985

CBU number:

1776

Abstract:

Seidenberg (1984) has argued that skilled reading is not influenced by the manner in which orthography encodes phonology. Orthographic cues to a word’s pronunciation are only important when naming low frequency words, even in orthographies as different as English and Chinese. Seidenberg claims that his data are best explained in terms of a modified version of McClelland and Rumelhart’s (1981) interactive activation model rather than either a dual-code (race) model or a whole-word and anology model. In the present paper we show that both the dual-code and the anology theories can provide a good account of this and other data. Furthermore, there appears to be little prospect of distinguishing these theories empirically. We also show that Seidenberg’s data do not provide adequate support for his claims about the role of phonological information in naming Ebglish words, and that the results of other studies suggest that differeent orthographies are processed differently.