Toegepaste Taalwetenschap in Artikelen - Volume 75, Issue 1, 2006

In a study encompassing 103 students in the sixth grade of the Dutch primary educational system, the effect of written text in combination with spoken text in a digital educational application on young readers is examined. The central question is whether the combination of these two information channels supports automatisation of the reading process of young readers so that it improves text comprehension or whether it causes overloading of the verbal channel. Two versions of the digital educational application WebQuests were used in this research. One version consisted of written texts accompanied by identical spoken texts; the other consisted of written texts only. To each version, half of the participants were randomly subscribed. The texts were informative of nature and their theme was 'the medieval city'. Textual open and multiple choice questions are used to measure text comprehension. The experiment shows that text comprehension of different types of young readers did not show any significant differences between the two versions of presentation. This implies that the study did not find any proof that addition of speech causes an overload of the verbal channel. Neither does it retrieve any significant evidence that it yields a surplus value to the automatisation of the reading process of young readers in digital educational applications.

The role of WM in syntactic processing has been discussed extensively in the literature. Some researchers claim that a single working memory capacity is used during processing (Just & Carpenter, 1992), and there are others who claim that language comprehension and syntactic processing can be divided into two stages (Caplan & Waters, 1995). This study examines the relationship between age, working memory capacity and on-line /off-line syntactic processing, using Dutch stimuli in self-paced reading task and a reading span task. The results showed that WM capacity significantly declines with age. The self-paced reading times of the Dutch subject/object ambiguities showed a significant quantitative difference between the two age groups, but not a qualitative difference: the elderly group read significantly slower than the young group but the pattern of the reading times was identical for both age groups. Off-line, the results of the comprehension questions show that comprehension was affected for the elderly group in the items that demanded less WM capacity compared to the young group. No age-effect was found in the results of the comprehension questions in the WM-demanding items. Overall, this study cannot support the claim that a single WM capacity is used for syntactic processing and comprehension.

Articles and pronouns have already been extensively researched in various languages, including Dutch. Much less is known, however, about the pragmatic functions of these morphosyntactic forms during acquisition. The question to be investigated in the present research addresses the question of how Dutch children between 2;0 and 3;3 combine indefinite articles, definite/demonstrative articles and pronouns with pragmatic functions in reference.The research is located at the interface of morphosyntax and pragmatics. The choice of the form for a certain function is influenced by the child's ability to take on the listener's perspective. This ability is still developing in 2-year-olds, as is the acquisition of various morpho syntactic forms. The question is then, how do children combine forms with functions if they supposedly do not do this on the basis of the listener's knowledge of the referent. Three other possibilities are also investigated: specificity, discourse function and (visual) perceptibility. Both the form-function correspondences as well as errors in linking form to functions will be investigated for three Dutch children from the CHILDES-database.The results show that the children do not randomly distribute the morphosyntactic forms over pragmatic functions in reference. The patterns found can be explained on the basis of limited semantic use of forms and perceptual availability of the referent, followed by a growing sensitivity to discourse patterns.

In Dutch education, children of Dutch parents with a low educational level lag behind peers with higher educated parents. This can be explained by the language input the former group receives at home, which is thought to prepare them less well for the academic language used in schools than that of the latter group. In this study, interactions between four mothers and their three-year-olds in five speech situations were analyzed to investigate whether mothers with different educational levels did indeed differ in their use of characteristics of academic language and whether the use of these characteristics varied in different speech situations. The language input of the higher educated mothers was found to show more characteristics of academic language than that of the mothers with a lower educational level. For all mothers, their language input showed most characteristics of academic language when they talk about a picture.

Do children's representations of complex words in their mental lexicon have an internal structure, with the stem as a separate unit? De Jong et al (2000) found that adults recognize words with a large Family Size, i.e. words occurring in many derivations and compounds, faster than equally frequent words with a small Family Size. This result is an indication that the occurrence of a stem in complex words facilitates the recognition of this stem. This article investigates whether the Family-Size-effect extends to children's reaction times as well. Using a lexical decision task, the effect was observed in 9-10 year old children (N=69) in Dutch. A large vocabulary and good reading ability shortened general reaction times, but had no influence on the difference between items with a small or large Family Size. Monolingual and bilingual children performed similarly on this task.

This paper describes a study of native language (Dutch) word identification skills conducted in two groups of high school foreign language learners. The study focused on the differences between the groups with two main aims. The first was to examine the Linguistic Coding Difficulties Hypothesis (LCDH), a prevailing theory of individual learner differences allocating weaker foreign language achievement to weaker native language skills. The second aim was to determine the nature of these differences and inspect the claims that native language word identification skills differentiate weak and normal foreign language learners. This was done by investigating the effect of context on weak FL learners, who over-rely on context in compensation for weaker phonological and orthographical skills, according to the LCDH. In pursuing its aims, the study became involved in an ongoing debate surrounding the issue of native language reading.

In this article, the authors combine the insights from two linguistic sub disciplines: language contact and sign language linguistics. After a short introduction about some possible outcomes of language contact the choice for code switching and an analysis along the lines of Myers-Scottons Matrix Language model is made.First, the characteristics of bimodal language contact are discussed, and then the dataset is presented: a bimodal corpus by a deaf mother and her hearing son Jonas at the ages of approximately 3;0 and 6;0. Analyses show that a large part of the utterances is code blended, this means simultaneously signed and spoken. Even after a closer reanalysis it turns out to be difficult in many instances to decide what the matrix language is, although in many instances it is NGT.

In view of its sociolinguistic situation and its mother tongue language policy Eritrea qualifies as an excellent site for comparatively investigating the acquisition and use of literacy in nine different languages (Tigrinya, Tigre, Afar, Saho, Bidhaawyeet, Kunama, Nara, Bilen and Rashaida) using three different scripts (syllabic Geez, alphabetic Latin and consonantal-alphabetic Arabic) within one and the same cultural and educational context. This contribution presents first results of a literacy survey with 670 Eritreans (which is part of a larger NWO-WOTRO research project). It goes into the respondents' level of literacy, their use of reading and writing in a number of societal domains, their use of the different scripts, and their preferences for the different scripts. A main outcome of the survey is that both, religion and ethnolinguistic group membership, play an important role in the use and positive evaluation of specific scripts. The majority of respondents report a preference for either Geez or Arabic, the scripts of the holy books of the Coptic Orthodox Church and Islam respectively, whereas the Latin alphabet, notwithstanding the official support it gets and the fact that it is used for six out of nine Eritrean languages, hardly plays a societal role.

Usage of Dutch regional languages and dialects is very much in decline in favour of usage of standard Dutch. This paper analyses the developments in usage in the period 1995-2003 using data from five measurement points of the national cohort study Primai y Education (PRIMA). A total of 35,000 pupils and their parents were involved in this study. In addition, this paper analyses the relationship between usage of regional languages and dialects on the one hand and a number of family demographical characteristics and the children's Dutch language proficiency on the other. The results show that it is imperative to treat the Netherlands not as a whole but to differentiate between language areas. In contrast to the other regional languages and dialects the future of Limburgish seems less gloomy.