No abstract

Two experiments were conducted to see if teaching poor readers to use inner speech could improve their reading and writing. In the first experiment, there were 8 third grade children, 4 in the experimental group and 4 in the control one, matched from a pre-test of reading. In the experimental group, the children were trained to use inner speech in 27 tasks, from explicit self-speech of the adult and of the child to implicit self-speech by the child alone. The results show significant differences between the two groups in the post-test of reading. In the second experiment, there were 6 students in second grade, 3 in the experimental group and 3 in the control one. In the experimental group, the children were trained to use inner speech in 18 tasks: 6 cognitive tasks that do not requireshort term memoryas in thefirst experiment, 6 reading tasks, 6 writing tasks. In the experimental group, the children were trained to use self-speech to process the tasks whereas, in the control group, the adult's help was mostly visual. The results were that the experimental group succeeded significantly better in the post-tests of reading and writing.Research on psychogenesis of literacy (Ferreiro & Palacio-Gomez, 1988;Fijalkow J., 1989;Fijalkow & Fijalkow E., 1992) shows that, in the beginning, children look at written language as an object just as they look at other objects in the surrounding world. He/she memorizes, for example, the shape of the words, their length, the shape of some letters, the first in the word specifically. To identify a written form, he/she recognizes it visually. When he/she begins to write ("invented spelling"), he/she proceeds in the same way : letters are treated as if they were objects. Words and letters are written in this period by putting a letter after another letter so his written product looks like a written text. In time these written forms are linked to a personal meaning, a meaning related to some characteristic or attribute of the referred object. These graphical representations have a symbolic value, as a feather is an Indian or a hat is a cowboy. "Milk" is a very long word because there is much milk in the bowl. During all this time, even if he/she learnt letters, and is able to recognize or to spell them, he/she uses them as if they were common objects.

No abstract

No abstract

The role of speech and inner speech in reading is contentious and the study of inner speech in reading is fraught with methodological problems. An approach sometimes used is the 'e-cancelling' method, where subjects are instructed to mark all incidences of the letter 'e' in a text so that inferences can be made from underlining 'e's that would be silent versus those that would be sounded in speech. However, this method does not work well in the French language and, in attempting to use letter cancelling approaches with French children aged 7-8 years, we have used other vowels as targets. In this paper we present some results about the effects on rate of reading and comprehension of various tasks intended to induce attention to the sound of parts of the text. Our conclusion, in line with that of Liva (1987) and other researchers, is that, at this early stage of reading development, reference to the sound of text does not inhibit comprehension. RESUME Le langage intbrieur dans la lecture des jeunes enfants frangaisLe r61e du langage et de l'auto-langage (langage inttrieur) dans la lecture demeure controversk. Quant A l'ktude de l'auto-langage dans la lecture, elle pose nombre de problbmes d'ordre mkthodologique. La mise en kvidence de l'auto-langage dans la lecture A partir de la mtthode du 'e-cancelling', parfois utiliske avec la langue anglaise, a suggtrk une approche avec la langue franGaise qui consiste a souligner des lettres cibles. I1 s'agira par exemple de souligner dans un texte les 'a' qui se prononcent /a/ comme dans le mot 'valise'. On fait alors l'hypothhse que les enfants qui ont l'habitude de s'aider de l'auto-langage en cours de lecture vont souligner seulement les lettres 'a' qui se prononcent /a/ et ne commettront pas l'erreur de ISSN 0141-0423 0 United Kingdom Reading Association 1995. Published by Blackwell Publishers, INNER SPEECH IN YOUNG FRENCH CHILDREN 33souligner les lettres 'a' qui se prononcent diffkremment comme dans 'aime' par exemple.Des travaux antkrieurs (Liva, 1987) montrant que les enfants bons lecteurs de 26me annke de primaire (classes franpises) soulignent avec plus de rkussite les lettres cibles 'qui s'entendent' que les mauvais lecteurs, font apparaitre une relation entre la lecture et la t2che de soulignage. L'objectif de cette ktude est de montrer que la t2che de soulignage ne gene pas la comprthension du texte.Le plan expkrimental comporte 6 groupes de 14 enfants de 7-8 ans, apprentis lecteurs de 2tme annke de primaire (en France), soumis A des tsches diffkrentes contraignant les enfants A porter leur attention sur l'aspect phonique des mots pour repkrer la lettre cible. Nous prksentons les rksultats de I'effet du soulignage de la lettre cible sur la comprkhension d'un texte, sur la subvocalisation, et sur la vitesse de lecture. Conformkment aux travaux de Liva (1987) et A ceux d'autres chercheurs, les rtsultats montrent qu'A cette ktape de l'apprentissage de la lecture le fait de 'se dire' le texte pour pouvoir souligner les lettres cibles ne gene pas la comprkhension. I1 ...