.Lexical insertion is a very important process in the production of language, whether verbal or written. It is often metaphorically stated that words are stored in a mental dictionary, called lexicon. The general belief is that the more entries in the lexicon, the better. But having a large number of stored data is not useful if we do not have the right tools to access the information we need. We know also from experience that the lexical choice is not always easy, and sometimes we fail to reach that precise word to express a certain idea, even though we know it exists. Psycholinguists call this fact the tip-of-the-tongue phenomenon or, more technically, dysnomia or anomia.
Access to data from the lexicon depends on two factors: a) how words are organized, and b) which mechanisms exist to access the data.
Complex Network theory (Newman, 2003) can help to both tasks: organization and navigation of the lexicon. Words can be organized by means of two-layer networks with a double dimension: syntagmatic and paradigmatic. Methods taken from DNA regulatory networks can help to design efficient and affordable navigation mechanisms capable to change the configuration while walked, in order to adjust to the necessities of the user.
Therefore, this proposal aims using mildly bio-inspired dynamic complex networks to develop a computational implementation of computer-assisted writing to help users in cases of lexical access disorder, or dysnomia.
The topic is related to complex network theory and semantics of natural languages. But the project also affects psychology and psycholinguistics, provided that it deals with a language disorder; cognitive modeling, when trying to simulate the organization of memory; natural language production, because we want to model an interactive mechanism to assist it, as well as biologically inspired theoretical computer science, when we take the model of genetic regulation to improve navigation.
Fields of science
Call for proposal
See other projects for this call