Acceso usuarios
Hakima Megherbi, Marie-France Ehrlich
In line with Nation and Snowling's (1997, 1999) position, two experiments with first-grade children (mean age : 6;9) examined the hypothesis that less-skilled comprehenders in a reading task have language difficulties in high-level processes, such as processing of anaphors. Reading and listening comprehension were measured by means of a global test. With skilled and less-skilled comprehenders in the reading test, the on-line processing of three types of anaphor (repeated noun, general noun and pronoun) was investigated in a listening situation, by means of an error-detection paradigm. Error-detection performance was better and faster if the anaphor was solved. In experiment 1, two types of errors were used : phonological errors and phonologico- semantic errors. In experiment 2, only phonological errors were used, the phonemic distance between the phoneme- source and the phoneme-target in the non-word being low or high.
In the two experiments, reading comprehension and listening comprehension were strongly correlated. Less-skilled comprehenders showed markedly lower oral performance than skilled comprehenders. In experiment 1, the percentage of detection was lower in less-skilled comprehenders and detection time was similar in the two groups. For phondogico- semantic errors, detection time was longer for general nouns in less-skilled comprehenders. In experiment 2, the percentage of detection was lower and detection time was longer in less-skilled comprehenders, for low and high distant errors. Performance did not vary as a function of type of anaphor, for skilled or for less-skilled comprehenders. These data did not support the hypothesis that less-skilled comprehenders have an impairment in on-line processing of anaphors in a listening situation. However, a qualitative analysis showed specific difficulties in less-skilled comprehenders. Results were interpreted in terms of the « quality » of lexical representations (Perfetti & Hart, 2001) . The paradigm used for investigating the processing of anaphors was discussed.
