Cel
Adult speakers of a language know several tens of thousands of words. Unless they suffer from some neurological disorders, those words can be readily used on a daily basis. This is done by retrieving lexical information from long term memory, and selecting its most relevant aspects. Cognitive models of word selection distinguish stages of processing concerned with semantic, lexical, and form properties of the words. Contrastive hypothesis have been considered to describe how appropriate lexical items are uniquely identified among all known words. Various sections of temporal cortex are known to play a prominent role in lexico-semantic processing, whereas frontal cortex is known to act as a controller of memory retrieval. More specifically, posterior left lateral and medial areas are capable to detect and resolve conflict among candidate words in cases where uncertainty arises.
Despite detailed accounts, current descriptions of lexical information processes are rather static. Discussions of cognitive processing models have often been framed on structural, rather than dynamical, arguments. In addition, a vast majority of studies characterizing lexical information processing are based on low temporal resolution brain imaging techniques. The main objective of this project is to go beyond these descriptions by characterizing the spatio-temporal dynamics of word selection processes.
The evidence will come from electro-encephalographic (EEG) and magneto-encephalographic (MEG) recordings of brain activity elicited in well-defined cognitive tasks. Innovative temporal pre-processes should allow discriminating brain activity from articulation artefacts. The evidence will also come from intra-cranial event related potentials, recorded in patients suffering from pharmaco-resistant forms of frontal and temporal lobe epilepsy. These data have high spatial and temporal resolution, and will provide strong constrains on lexical information processing models.
A description of the dynamic interactions between brain regions during word selection will change the way we think about this basic behaviour. Besides this intrinsic interest, word selection provides a very natural way to connect relatively simple decision processes (e.g. those engaged in basic visuo-motor tasks) with more integrative processes involved in information retrieval from long term memory. Better understanding the spatio-temporal dynamics of lexical information processes will also be highly valuable for improving pre-surgical evaluation procedures in pharmaco-resistant epilepsy.
Dziedzina nauki
Zaproszenie do składania wniosków
ERC-2010-StG_20091209
Zobacz inne projekty w ramach tego zaproszenia
System finansowania
ERC-SG - ERC Starting GrantInstytucja przyjmująca
75794 Paris
Francja