Séminaire invité

The N400 is one of the most well-known neuropsychological ERP components associated with language processing. Although its discovery dates back to studies by Marta Kutas and colleagues more than half a century ago, its real nature still eludes our scientific knowledge. A lively debate about its very origin (long-term memory? semantic knowledge? Lexicon? probabilistic associations?) has been ongoing since its discovery, with no clear winners. In this talk, I will argue for a view that maintains that the N400 reflects the overall working/triggering of the entire semantic network, which can be defined by a specific set of brain regions in the pre-frontal, insular, antero-temporal, middle-temporal and temporo-parietal cortical areas. This accounts for why this component can be affected by a wide range of linguistic and non-linguistic experimental manipulations, which nonetheless share the feature of contributing to the interpretation of the stimulus.

The talk is divided into three parts. First, I will introduce the main features of the investigation of language through ERPs. Second, I will illustrate three ERP studies that I have conducted, which aimed to explore the online processing of two linguistic phenomena: a) the interpretation of pragmatic inferences such as scalar implicatures in adults and children, and the reaction of the cognitive system to their violation; b) the comprehension and online processing of the Italian negative word/Negative Polarity Item mai (ever/never). Third, I will discuss the results from these studies along the lines of the hypothesis that I advanced above, claiming that they are an example of how different semantic and pragmatic phenomena wind up affecting the same neuropsychological component, the N400.
To conclude, bringing in the picture more psycholinguistic and neuroscientific evidence coming from processing, acquisition and lesional studies, I will offer a sketch of how the semantic cortical network looks, which regions are part of it and how they are interconnected.