The aim of this presentation is to present a part of a larger group of studies devoted to analyzing the different ways writers manage different levels of linguistic and psycholinguistic constraints. Our study focuses on the temporality of writing, as it constitutes a key parameter of writing, influencing the management of constraints by the writers. First, we shall present the system of linguistic and psycholinguistic constraints we have defined for methodological purposes in order to organize a description of the production output of writing that incorporates idiosyncratic variations and allows us to define styles of writing. In a second part, we will present the different levels of temporality that affect writing, and the register problems they pose for the analyst. The third part is devoted to the presentation of an experiment being conducted in classes of primary and secondary school in order to observe the effects of time frames imposed on the writers, on the properties of written language and textual products. This experiment brings together a large number of researchers who examine the entire corpus, each from a specific point of view in order to analyze the effect of different variables. In the experiment we shall present, young writers had to produce texts in conditions with three different sets of parameters: the properties of the text used as the prompt, the relationship between the source text and the text to write, and the temporal patterns of making the task: * properties of the source texts: the source texts for supporting the experiment came from an African tale. Two versions were developed for the purpose of experimentation that differ in how the words of characters were transcribed : one version uses direct discourse, the other has transpositions in the form of indirect discourse and narrative speech. * relationship between source text and text to write: some students had to reproduce in writing the story that had been read by the teacher, others had to produce a continuation of the narrative. * temporal patterns: for some groups of students, the writing task immediately followed listening to the text, for others, this task was done later. The factors on which we focus in this presentation concern only the discursive modality used to transcribe the characters' manner of speaking. This is to assess whether students are sensitive not only to the content of what was said but also to how language was used in the source text to translate what was said and to measure the effect of time lag on these properties. These elements help us to better understand the diversity of young writers' ways of developing and representing a text, and to better understand how writers, while writing, use different strategies in order to built representation of their own text. The initial results suggest that the time between hearing the text and writing the second text accentuates the differences between writers, which attests to differences in their memorization strategies