Title: Lcriture est un mode de reprsentation du langage oral donc les habilets de lecture et dcriture sont
1Lécriture est un mode de représentation du
langage oral donc les habiletés de lecture et
décriture sont conditionnées par des facteurs
biologiques mais ces habiletés ont un effet
en retour sur le traitement du langage oral
2Aires spécifiques pour le traitement des lettres
Polk et Farah (1998)Polk et al. (2002) avec
RMNf, activation spécifique du gyrus fusiforme
gauche
3(No Transcript)
4Lorganisation cérébrale de la lecture est
fonction dune plasticité neurale qui se
maintient bien au-delà de lenfance Le maintien
dune certaine organisation dépend dune
expérience plus ou moins constante
5Aires spécifiques pour le traitement des
configurations orthographiques (mots et
pseudo-mots)
6Aire didentification des mots possibles
(triangle-2) mots et pseudo-mots vs séquences
de consonnes, PET (Petersen et al., 1990)
7Cohen et al. (2000)RMNf et PEsaire temporale
postérieure et inférieure gauche (plus exactement
portion médiane du gyrus fusiforme), 180 à 200
msec. après le début de la présentationnon
observé chez les enfants de 2e année primaire
(McCandliss Posner, non publ.)
8This area of the extrastriate cortex is lateral
to an area called visual word form area
(Warrington Shallice, 1980), which is activated
by orthographic sequences. Dehaene and Cohen
proposed that visual lexical forms and more
generally orthographic forms are represented in a
region of the left fusiform gyrus defined by the
Talairach coordinates x-43, y-54 and z-12.
9The activation of the visual word form area takes
place from 150 to 200 milliseconds after the
onset of the letter sequence exposure (e.g.,
Tarkianene et al., 1999).
101. The visual word form area is activated by both
words and pseudo-words, but not by strings of
consonants like xjpqt (Dehaene et al., 2002
Petersen et al., 1990). Thus, it is concerned
by the processing of letters as such, but only by
legal, phonotactically acceptable structures of
letters.
112. The VWFA area is not activated by the auditory
presentation of words and pseudo-words (Dehaene
et al., 2002), Thus, it seems that it is not
part of a purely phonological system.
123. Physical modifications such as case (AGE
age - aGe) or font (age) do not affect the
amplitude of its activation (Polk Farah, 2002
Dehaene et al., 2001).This suggests that the
knowledge represented in this area is not the
knowledge of a physical form but of an abstract
linguistic structure.
134. Its activation occurs even when, using masking
techniques, the subject is unable to recognize
the word presented (Dehaene et al., 2001). This
suggests that the processing carried out in this
area is not, or not necessarily, conscious.
14Our word identification system has to compute an
invariant representation from the visual input.
It has to discard irrelevant variations such as
differences in case or font or size (eight
EIGHT), and maintain very small details (eight
sight) that support different pronunciations and
meanings.
15This functional specialization of the reading
system results implies that some visual neurons
respond to A and a in the same way, but
differently to a and e.
16These neurons tend to be grouped together in some
fixed regions of the visual cortex.This is
reproducible localization. Thus, whenever
subjects read a word, a reproducible portion of
the left occipito-temporal sulcus is activated
and hosts functionally specialized circuits for
letter and word recognition (Cohen and Dehaene,
2004).
17The regional selectivity for word recognition is
probably more relative than absolute, with some
intermixing of selective groups of neurons, or
with groups of neurons showing different degrees
of responsiveness to different classes of stimuli.
18Dehaene et al. (2004) at least three
functionally different subareas. In the
posterior subpart of the the mid-fusiform cortex,
bilateral priming (reduction of activity) in the
response to repetition of letters at the same
absolute location, regardless of whether they
were presented in the same word (reflet
REFLET) or in an anagram (r e f l e t
TREFLE). This seems to be a subarea of
letter detection.
19In a somewhat more anterior subarea of the left
hemisphere, priming was similar when the same
word was repeated, even when shifted by one
letter location (r e f l e t REFLET),
and when a word was followed by its anagram
(reflet TREFLE). Thus, letters, or fragments
of words, are represented in this area
independently of location.
20Finally, in an even more anterior subarea,
priming tended to be larger for same words than
for anagrams. This subarea may be involved in
word coding.
21Shaywitz et al. (2002) positive correlation
between amplitude of activation of left
mid-fusiform and level of performance in tests of
phonological decoding in individuals aged 7 to 18
years, even when age was taken into account.
Temple et al. (2003) a phonological remediation
program led to increased activity in regions
including visual word form area in a group of
children who presented dyslexia.
22 Connaissances cognitives et
linguistiques
Identification
des mots écritsCompréhension de
l écrit
23représentation phonologique représentation de
quelque aspect ou propriété de la structure
abstraite de la forme de la langue
24Phonèmes les plus petits éléments constitutifs
de la parole qui permettent des distinctions
sémantiques
25 bal /b/, /a/ et /l/ cal bol
bar
26Dans le système alphabétique, les lettres (ou
des groupes de lettres) correspondent grosso modo
à des phonèmes
27Différences dans le degré de transparence
(opacité) du système orthographique
28(No Transcript)
29En anglais 1120 manières de représenter 40
phonèmesEn italien 33 manières de représenter
25 phonèmesPaulesu et al. (2000)
P PP(It) PP(Ang)Anglais
443 526 529Italiens
410 437 486Diff.
33 89 43
30Système cérébral commumgyri temporaux
supérieur, moyen et inférieurgyrus fusiforme
gauchecortex prémoteur et frontal
inférieurSystèmes spécifiquesanglais
temporal inférieur postérieur gauche et partie
plus antérieure du gyrus frontal
inférieuritalien dans la jonction du gyrus
temporal supérieur gauche et le cortex pariétal
inférieur (planum temporale)
31Principales caractéristiques du système
didentification des mots écrits, chez le
lecteur expert
32fixations durent en général de 200 à 300
millièmes de seconde saccades durent de 25 à
30 millièmes de seconde
33(No Transcript)