This paper reports a study conducted with French first-grade and second-grade children (mean age: 6;8 and 7;8 respectively). The first aim was to re-examine the Gough and Tunmer's (1986) Simple View in assessing the specific contribution of decoding ability and language comprehension to reading comprehension. The second one was to analyse the difficulties of children in reading comprehension. Reading and listening comprehension were assessed using both visual and auditory version of the same test. Decoding ability was assessed by means of a nonword reading test. On the basis of reading comprehension scores, skilled and less skilled comprehenders were contrasted, and then two groups of less skilled comprehenders were differentiated on the basis of the decoding scores.Hierarchical regression analyses computed on the whole sample showed that listening comprehension was a more powerful predictor than decoding ability in first-and second-grade children. In both grades, the pattern of performance in less skilled comprehenders showed a relative independence between decoding and reading comprehension. The good decoders' group and the poor decoders' group showed similar poor performance in reading comprehension and poor performance in listening comprehension. However, their difficulties could stem from different sources. Some instructional recommendations were formulated taking into account individual differences in decoding and spoken language abilities, as soon as the first months of formal reading acquisition.We are grateful to the two anonymous referees of this journal for their helpful comments on the manuscript. We would like to thank Catherine Loridant for her precious collaboration in the collect of the data. We gratefully acknowledge the children and the teachers who participated to the experiment.

The purpose of this experiment, conducted with second-grade children (mean age: 7;8), was to examine the hypothesis that less skilled comprehenders in a reading situation suffer an impairment in spoken language comprehension and, more specifically, in the on-line processing of anaphoric pronouns. Skilled and less skilled comprehenders performed a cross-modal naming task investigating the effects of pronoun gender and pragmatic inference from the verb on the integration of two successive sentences. Results revealed different patterns of effects in the two groups. The skilled comprehenders integrated on-line sentences by relying on pronoun gender and verb meaning. Pronoun gender appeared to exert a dominant influence relative to verb bias. In the less skilled comprehenders, on-line integration was not systematic, being dependent on the meaning of the verb and the proximity of the referent. Complementary analyses revealed similar patterns of effects among less skilled comprehenders, whether they were good decoders or poor decoders. These results show that less skilled comprehenders are developmentally delayed compared with their skilled peers, and extend the language impairment hypothesis to cover discourse-level processes.

How long does it take for word reading to become automatic? Does the appearance and development of automaticity differ as a function of orthographic depth (e.g., French vs. English)? These questions were addressed in a longitudinal study of English and French beginning readers. The study focused on automaticity as obligatory processing as measured in the Stroop test. Measures of decoding ability and the Stroop effect were taken at three time points during first grade (and during second grade in the United Kingdom) in 84 children. The study is the first to adjust the classic Stroop effect for inhibition (of distracting colors). The adjusted Stroop effect was zero in the absence of reading ability, and it was found to develop in tandem with decoding ability. After a further control for decoding, no effects of age or orthography were found on the adjusted Stroop measure. The results are in line with theories of the development of whole word recognition that emphasize the importance of the acquisition of the basic orthographic code.