The aim of our project is to understand forgetting in adults and children, more especially short-term forgetting. Memory is the central cognitive characteristic of human cognition. All of our complex cognitive activities, such as learning, reasoning, and planning, require both the integration and the maintenance of new information and the retrieval of previously acquired knowledge. Within cognitive architecture, one specific structure, working memory, interfaces between these two types of information. Working memory (WM), the system that holds, organizes, and manipulates the contents of our current thoughts, can be seen as the central piece of deliberate cognition. Historically, the concept of WM emerges from the fusion of the attentional system to a short-term memory conceived as a simple store of information. Its limited capacity is a major determinant of our success in complex tasks such as text comprehension, reasoning, and problem solving, and measures of WM capacity have therefore been identified as major determinants of cognitive development in childhood and in old age as well as of individual differences in intellectual abilities. Although the amount of research focused on working memory has increased during the last 30 years, one of its most intriguing mechanisms is still unclear. Indeed, forgetting information stored at short term, that is when a small amount of time separates the acquisition of information and its retrieval for recall, has huge consequences for learning. This phenomenon, so incredibly frequent in our everyday life, is still poorly understood. Within the joint approaches of experimental cognitive psychology and computational science, our aim is to develop a model explaining the underlying mechanisms of forgetting and their development through life.

This project in cognitive psychology aims at describing the relationship between the temporal organization of stimuli and memory processes. Two experimental and theoretical research subjects are developed: 1) the study of the effect of the manipulation of presentation orders of stimuli, and the modeling of categorization processes 2) the modeling of immediate serial-recall tasks and the evaluation of the memory span using sequence alignment algorithms. A single model will be then evaluated in order to account for the discriminability of stimuli in sequential presentations in the experiments of both programs. First, we aim at investigating the mechanisms by which categorical knowledge is abstracted by manipulating the presentation order in which stimuli are presented to subjects. We hypothesize that presentation orders which respect the subject's internal conceptual organization are beneficial to acquiring knowledge. We aim at showing that orders based on rules substantially facilitates learning as compared with previously known beneficial orders such as similarity-based orders. This year, we develop a solution to the issue raised in our last year ANR review. Also, a more careful examination of an exemplar-based model (GCM-gamma) taking into account the temporal organization of stimuli will be provided. A second part of the project aims at reexamining Miller's (1956) estimation of short term memory capacity. The rather high evaluation of Miller (i.e., 7±2) results from the fact that simple items can be recoded by subjects into chunks in memory span tasks Miller examined. Recent research designed to prevent recoding has given evidence for a lower estimation of 3 to 4 chunks. Recently, we showed that both evaluations can simultaneously be found correct in a given task. Nevertheless, this project goes beyond the study of chunking mechanisms and proposes an analysis of scoring systems. Classical scoring produces a debatable estimation of the number of items correctly recalled : it is most often simple and rigorous, but maybe too crude. We aim at applying sequence alignment algorithms applied in genetics in order to develop a scoring system offering a typology of the errors arising in memory and their statistics. In comparison to genetics, these errors will be identified as deletions, mutations, translocations and duplications. This method will be applied to a series of replications of classical memory span tasks, beginning by the pioneer studies related by Miller.

« Sensunique » is a project based on the concept of a Controlled Language applied to technical writing (protocols, instructions, alert messages). This project aims to optimize a computerized writing assistant and to validate its operational reality on a corpus of medical and healthcare texts. The objective is to demonstrate its usefulness, applicability and effectiveness as an innovative Information and Communication Technology which meets the needs of public or private structures that require high quality technical texts. This writing assistant already exists as a prototype and serves two main purposes : a) for technical writers or professionals that are required to write technical texts : its goal is to optimize the time needed to write and the quality of the texts; b) for the users or readers, may they be professional or general public : it goal is to ensure a thorough understanding of the texts, whether those texts, whether the transmission of information is collective or individual, with or without safety dimension. Yet, for the prototype to be transferred to a company and marketed, it has to be optimized (in terms of linguistics methodologies implemented, ergonomics and hardware development) and validated by being used by real users, in its operational environment and on a large corpus. This motivates our application. This project takes root on the ANR project LiSe (« Linguistique, normes, traitement automatique des langues et Sécurité : du « data et sense-mining » aux langues contrôlées », CSOSG 2006) that was carried out by Professor Sylviane Cardey (University of Franche-Comté), in partnership with Airbus-France and The Cell and Tissue Engineering and Biology Federative Research Institute IFR133 (University of Franche-Comté). Building on the success and scientific benefits of this project , the University of Franche-Comté is planning to enhance the value of some of the results (Controlled Language and writing assistant), with the aim of creating a spin-off. The spin-off project is supported by the Incubator for innovative companies of Franche-Comté and has won the following awards : - Concours Numérica 2010, catégorie émergence, - Grand Prix de la Prévention Médicale 2010, - Concours National d’Aide à la Création d’Entreprises Innovantes 2010, catégorie émergence. The results that are expected at the end of the project will be valued through: - A patent on the Controlled Language methodology applied to technical writing; - A deposit with the Agency for the Protection of Programs; - The transfer of technology to the spin-off of the project LiSe, that will support the marketing of the product (if the spin-off project fails, the technology will be transferred to another company).

The proposal aims, firstly, at solving several concrete questions in the modern theory of conservation laws and related convection-diffusion problems. These questions (Tasks) are concerned with actual techniques of nonlinear analysis (some of the questions seemed not accessible only a few years ago); they originate from recently identified of from long-standing important applications. The Tasks are all relevant of the domains of expertise of one of the members of the project and at the same time, there is a strong connection to at least one of the other members and to our national and international collaborations. The second goal is to make advance our understanding of the theory and the crucial tools (including the most modern ones) for analysis of conservation laws and related nonlinear PDEs. Indeed, the tasks we have selected are representative of several fundamental issues relevant to the problems in hand, such as: notions of solution and well-posedness; analysis of non-local terms in conservation laws; cooperation of semigroups in evolution equations; convection-diffusion problems of mixed type; boundary and interface problems for conservation laws; (ir)regularity and qualitative behaviour of solutions; converging and efficient numerical approximations. Our third goal is, while advancing on the solution of these concrete questions, to form progressively an internationally recognized group in Besancon working on a wide spectrum of modern approaches to conservation laws problems. Indeed, the proposal team consists of five MCF, all of us belong to the Laboratoire de Mathematiques de Besancon CNRS UMR 6623. For three of us, conservation laws or systems and related nonlinear PDEs constitute the main field of research. Yet at the present time, everyone focuses on a different domain and specific subjects: nonlocal (fractional) laws / degenerate parabolic equations, boundary and interface problems for scalar laws / systems and viscous limits. The project is born from our intention to bring our different cultures to interaction. These three persons share the main load of work on the project. The implication of the two other team members is a starting point for more ample collaboration in the future; they joined the group because of their recent interest to the subject, and they contribute an expertise of different kind, the one on linear PDE techniques and the one on numerical analysis and scientific computing. We ask for funding for a post-doc to join the team during the second year of the project and bring an experience of his own in the quickly growing field of control of hyperbolic conservation laws. The topics we treat were all developed in strong interaction between french, italian, american, russian, norwegian, polish, german, spanish, chinese researchers. We already have strong connections with some of the leading groups in the subjects we treat, and we are intended to maintain and develop such contacts. The work on the Tasks will be complemented by three workshops or schools (one per year), plus a final meeting with participation of exterior experts and of many collaborators of the project ,on selected aspects of the modern theory of nonlinear PDEs including actual topics in control of systems of conservation laws.