Section: Partnerships and Cooperations

Regional Initiatives

The overall idea of this project is that current work  [78] shows that it is possible to discriminate between morphological (i.e. concatenative) and phonological activity that is produced by the brain upon linguistic stimuli. That is, the experimental setup provides an on-line diagnostic for the presence or absence of phonological computation in the production of words.

On the neuroimaging side, the long-term challenge is to reproduce Sahin et al.'s  [78] experiment with non-invasive methods (see the following section). If successful, the study will show that a processing sequence predicted on linguistic grounds is implemented in the brain in fine-grained spatiotemporally patterned activity. From the neuroimaging point of view, the development of such non-invasive methods that can accurately identify events in known regions will have an important impact on both computer science and neuroscience. Replacing deep electrode probes (implanted in the patient's brain) with algorithms to map cognitive processes onto brain activation will help developing new applications of functional neuroimaging. Note that results could also turn out to foster clinical tools in the diagnosis of patients affected by white matter abnormalities and altered structure-function relationships in the connectional anatomy of language.

This project aimed to perform a feasibility study for this research area. More precisely to investigate whether current neuroimaging technologies are able to provide the tools for the proposed linguistic analysis.