Search

Annotations of word difficulty by readers provide invaluable insights into lexical complexity. Yet, there is currently a paucity of tools allowing researchers to gather such annotations in an adaptable and simple manner. This article presents PADDLe, an online platform aiming to fill that gap and designed to encourage best practices when collecting difficulty judgements. Studies crafted using the tool ask users to provide a selection of demographic information, then to annotate a certain number of texts and answer multiple-choice comprehension questions after each text. Researchers are encouraged to use a multi-level annotation scheme, to avoid the drawbacks of binary complexity annotations. Once a study is launched, its results are summarised in a visual representation accessible both to researchers and teachers, and can be downloaded in .csv format. Some findings of a pilot study designed with the tool are also provided in the article, to give an idea of the types of research questions it allows to answer.

Reducing the complexity of texts by applying an Automatic Text Simplification (ATS) system has been sparking interest in the area of Natural Language Processing (NLP) for several years and a number of methods and evaluation campaigns have emerged targeting lexical and syntactic transformations. In recent years, several studies exploit deep learning techniques based on very large comparable corpora. Yet the lack of large amounts of corpora (original-simplified) for French has been hindering the development of an ATS tool for this language. In this paper, we present our system, which is based on a combination of methods relying on word embeddings for lexical simplification and rule-based strategies for syntax and discourse adaptations. We present an evaluation of the lexical, syntactic and discourse-level simplifications according to automatic and human evaluations. We discuss the performances of our system at the lexical, syntactic, and discourse levels.

International audience; Deaf educational methods have been the subject of controversy among advocates of the oralist and the bilingual approaches for centuries. Over the past decades, the bilingual-bicultural method has proved its effectiveness in facilitating formal school learning and downscaling a higher rate of illiteracy compared to the hearing population. The bilingual-bicultural model in Western countries is designed and implemented in predominantly monolingual contexts or multilingual contexts with a dominant majority language. It aims at providing deaf learners with a simultaneous dual access to the deaf and hearing cultures through sign language and the written form of the majority spoken language. The objective of this dual access is to create a balanced form of bilingualism which will reinforce literacy development. In the Western context, the relative proximity of the written and spoken forms of the majority language allows the written form to function as a means of access to the socio-cultural heritage of the hearing community and to develop a sufficient degree of autonomy in a world where literacy became crucial. The application of the Western bilingual-bicultural model may at first glance seem tempting to mitigate a significant rate of illiteracy affecting 98% of the deaf Tunisian population. However, the diglossic situation in Tunisia, and in the Maghreb countries in general, rests upon the existence of two linguistic forms exhibiting considerable linguistic differences. On one hand, the Tunisian Dialectal Arabic (TDA) is the spoken form, and is the vehicle for the Tunisian socio-cultural heritage transmission. On the other hand, the written form, Modern Standard Arabic (AMS), assumes the role of institutional and literacy language. This particular situation requires a specific educational framework different from the classical bilingual-bicultural approach. We hypothesize that without taking into account Tunisian Dialectal Arabic, learners will not access the Tunisian hearing culture. This situation will potentially hinder literacy development in Modern Standard Arabic. Our article puts forward a trilingual-bicultural educational model adapted to the Tunisian diglossic situation. It includes TSL, and written ADT, as representatives of the deaf and hearing cultures which will both contribute to a more fluid development in a third language, written MSA, as the literacy language.

Background: For normally sighted readers, word neighborhood size (i.e., the total number of words that can be formed from a single word by changing only one letter) has a facilitator effect on word recognition. When reading with central field loss (CFL), however, individual letters may not be correctly identified, leading to possible misidentifications and a reverse neighborhood size effect. Here we investigate this inhibitory effect of word neighborhood size on reading performance and whether it is modulated by word predictability and reading proficiency. Methods: Nineteen patients with binocular CFL from 32 to 89 years old (mean ± SD = 75 ± 15) read short sentences presented with the self-paced reading paradigm. Accuracy and reading time were measured for each target word read, along with its predictability, i.e., its probability of occurrence following the two preceding words in the sentence using a trigram analysis. Linear mixed effects models were then fit to estimate the individual contributions of word neighborhood size, predictability, frequency and length on accuracy and reading time, while taking into account patients’ reading proficiency.Results: For the less proficient readers who quit reading on a daily basis, we found that the effect of neighborhood size was reversed compared to normal readers and of higher amplitude than the effect of frequency. Furthermore, this inhibitory effect is of greater amplitude (up to 50% decrease in reading speed) when a word is not easily predictable because its chances to occur after the two preceding words in a specific sentence are rather low.Conclusion: Severely impaired patients with CFL often quit reading on a daily basis because this task becomes simply too exhausting. Based on our results, we envision lexical text simplification as a new alternative to promote decisive rehabilitation in these patients. By increasing reading accessibility for those who struggle the most, text simplification might be used as an efficient readaptive tool and daily reading assistive technology, fostering overall reading ability and fluency through increased practice.

In this paper, we present a new parallel corpus addressed to researchers, teachers, and speech therapists interested in text simplification as a means of alleviating difficulties in children learning to read. The corpus is composed of excerpts drawn from 79 authentic literary (tales, stories) and scientific (documentary) texts commonly used in French schools for children aged between 7 to 9 years old. The excerpts were manually simplified at the lexical, morpho-syntactic, and discourse levels in order to propose a parallel corpus for reading tests and for the development of automatic text simplification tools. A sample of 21 poor-reading and dyslexic children with an average reading delay of 2.5 years read a portion of the corpus. The transcripts of readings errors were integrated into the corpus with thegoal of identifying lexical difficulty in the target population. By means of statistical testing, we provide evidence that the manual simplifications significantly reduced reading errors, highlighting that the words targeted for simplification were not only well-chosen but also substituted with substantially easier alternatives. The entire corpus is available for consultation through a web interface andavailable on demand for research purposes.

National audience; Outil de recherche numérique en insertion classe : simplification du texte pour améliorer la fluidité de lecture et la compréhension chez des élèves en CE1 et CE2

National audience; La compréhension écrite est fondamentale pour la réussite scolaire, professionnelle, et personnelle (Gaussel, 2015). Selon le modèle « simple » de la lecture proposé par W. A. Hoover et P. B. Gough (1990), la compréhension écrite est le produit de deux habiletés indépendantes : les habiletés en compréhension orale (vocabulaire et récupération de l’information sémantique) et celles dites de décodage (identification des mots écrits). L’objectif de ce projet est d’étudier l’impact de la simplification et de la nature du texte sur la compréhension en lecture en fonction du profil cognitif d’élèves entre 7 et 9 ans. Pour obtenir des données sur la vitesse de la lecture et la compréhension, un outil informatique a été développé sur iPad qui propose en version simplifiée ou originale des textes de natures différentes, littéraires et scientifiques. Chaque texte a été lu dans sa version originale ou simplifiée. Les textes étaient suivis de cinq questions de compréhension avec trois propositions de réponses. Ce dispositif a été testé auprès de 9 classes soit 170 élèves. L’expérimentation se poursuit sur 3 ans avec la même cohorte afin de déterminer les facteurs qui expliquent pourquoi un élève devient un bon ou mauvais compreneur à l’entrée du collège. La simplification des textes est lexicale, syntaxique et discursive. Elle vise à rendre la lecture plus facile. Ceci pourrait avoir une influence sur la motivation notamment chez les élèves ayant un profil de faible lecteur. Les deux natures de textes, littéraire ou scientifique, imposent des contraintes tel que la relation de cause à effet ou l’utilisation de connecteurs logiques dans les textes scientifiques ce qui peut être un avantage sur la compréhension de l’écrit. La première phase expérimentale a permis de passer des tests cognitifs, langagiers et socio-culturels en individuels. Lors de la seconde phase expérimentale, les élèves ont lu en classe, en autonomie les vingt textes. Au vu des premiers résultats, la simplification des textes littéraires améliore la vitesse de lecture et la compréhension. La simplification n’a pas d’effets sur la lecture des textes scientifiques. Bien que la vitesse de lecture soit identique pour les deux types de textes, littéraire et scientifique en version originale, la compréhension est meilleure pour les textes scientifiques originaux que les textes littéraires originaux. En revanche, la compréhension des textes littéraires simplifiés est supérieure à la compréhension des textes scientifiques en versions originales et simplifiées.

International audience; no abstract

International audience

International audience; We describe a computational lexicon which encodes verbal syntactic constructions in French. The aim of the project is to encode into a standard formalism the valency lexicon for French initiated in the eighties by M. Salkoff and A. Valli and continued since then. The whole computational lexicon will include about six thousand verbal entries with a list of the arguments of each verb, the restrictions associated with the arguments and examples found in real corpora. Such a lexicon may be a useful resource for the linguistic community as well as for natural language processing applications.