Feedforward, -backward, and neutral transparency measures for British English

Orthographic transparency metrics for opaque or deep languages, such as French and English, have tended to focus on feedforward and/or feedback directions, with claims made for the influence of both on reading. In the present study, data for five transparency metrics for southern British English, three of which are neither feedforward nor feedback, are presented, demonstrating the complex relationships between the metrics and offering an explanation for feedback effects in children's reading accuracy. The structure of such metrics from a variety of corpus sizes and origins is investigated, and it is concluded that large corpus sizes do not make a substantial contribution to the value of such metrics, when compared with smaller samples, and that adult and child corpuses have very similar profiles. Probabilities of occurrence for the phonemes, graphemes, and sonographs in this study may be downloaded from brm.psychonomic-journals.org/content/supplemental.