Dinamiche elettrofisiologiche nella lettura di parole: dall'analisi ortografica ai processi di elaborazione semantica

The purposes of the doctoral thesis were manifold. First of all, we aimed at clarifying the neural underpinnings of single word reading and the time course of lexical processing by comparing the bioelectrical activity elicited by stimuli with different degrees of orthographic legality (letter strings, legal pseudo-words and words) and semantic denotation (flora names vs. fauna names; concrete words vs. abstract words). A strong effort was directed to avoid possible confounding effects of psycholinguistic variables like frequency of occurrence and length, which are known to affect written word processing. Source reconstruction by means of LORETA was aimed to draw inferences on the cerebral regions specifically involved in different processing stages, with the final purpose of integrate current knowledge resulting from neuroimaging and electrophysiological approach. Overall the data provided evidence that the orthographic and lexico-semantic word properties were processed in parallel between 200 and 400 ms after stimulus onset. The temporo-parietal region was sensitive to orthographic legality (250-350 ms). The latency of lexical effects (word/pseudo-word discrimination) varied as a function of the number of a word's orthographic neighbours, being faster (200-250 ms) to pseudo-words with a small number of neighbours over anterior sites. The lexical effect found over the left occipito-temporal region (300-400 ms, BA 19 and BA 37) suggested a lexical access by means of a visual route. Semantic categories were discriminated as early as 200 ms post stimulus over the left occipito-temporal areas (BA 37 and BA 20). The results are discussed in the light of the possible role played by the word age of acquisition, as well as the specific properties of living concepts (biological relevance and homomorphism). Abstract and concrete word processing differed in terms of a stronger involvement of extra-striate visual areas during concrete word processing (350-380 ms) and a stronger involvement of prefrontal cortex in response to abstract words (350-380 ms). The results suggested that concrete, imaginable concepts activate perceptually based representations not available to abstract concepts. Overall, these data show how ERPs can dissociate between lexical and higher level processes. Semantic processing may take place near-simultaneously and in different brain regions with the processing of information about the form of a word and its lexical properties.