There are, in fact, studies (Biemiller, 1970; Barr, 1972) which indicate that oral reading errors can be analyzed to reveal whether the reader is using predominately context or graphic information for recognizing words. Readers who are depending primarily on context for "decoding" tend to make errors which are semantically consistent with the story being told. Readers who are using the graphic information to decode print-to-speech tend to make no response or to make responses consistent with the graphic information ("house" read as "horse").

By examining their oral reading errors, Biemiller identified three phases of development of reading skills in first-grade students-the first characterized by a predominant use of contextual information, the second by a predominant use of graphic information, and the third by a mixed usage of context and graphic information. Biemiller concluded that:

Data presented in this study indicate that the child's first task in learning to read is mastery of the use of graphic information, and possibly, of the notion that one spoken word corresponds to one written word. The child's early use of contextual information does not appear to greatly facilitate progress in acquiring reading skill. The longer he stays in the early, context-emphasizing phase without showing an increase in the use of graphic information the poorer the reader he is at the end of the year. Thus, the teacher should do a considerable proportion of early reading training in situations providing no context at all, in order to compel children to use graphic information as much as possible. (p. 95)

For present purposes, the point of Biemiller's and Barr's research is that the learning of decoding skills is not likely to be improved by improving languaging by oracy skills. Decoding skills may be, and perhaps should be, taught using nonmeaningful grapheme-phoneme correspondences. Reading comprehension on the other hand should be improved by training in oracy skills (see Hypothesis 4). Hence, auding ability, whether acquired naturally or as a result of schooling in oracy skills, ought to be predictive of reading comprehension ability after reading decoding skills have been acquired-and the data reviewed under Hypothesis 2 indicate that it is. fact that, in the present model, auding and reading utilize the same languaging and conceptualizing systems. Hence, the limiting factors underlying both auding and reading rate are skill in languaging and in conceptualizing.

HYPOTHESIS 3

Hypothesis 3 states that performance on measures of maximal rates of auding and reading will be comparable, assuming fully developed reading decoding skills. This hypothesis follows from the fact that, in the present model, auding and reading utilize the same languaging and conceptualizing systems. Hence, the limiting factors underlying both auding and reading rate are skill in languaging and in conceptualizing.¹

1. Note that auding and reading are subsets of the more general processes of listening and looking, respectively. Hence confirmation of the present hypothesis is evidence for the hypothesis that listening and looking rates are equal, as they should be since they are simply modality names for one internal process-focal attending (see Chapter IV, pp. 50-54).