Amina Ignatius
Research Mentor: Jonathan Brennan
Mentor Department: Linguistics, LSA
Author(s): Amina Ignatius, Jonathan Brennan
Session: Session 3 (11:00 AM – 11:50 AM)
Presentation Type: Poster 42
Abstract
EEG research conducted on what are known as event-related potentials (ERPs) has found two time-locked components to be relevant for linguistic error detection, known as the N400 and the P600, which are respectively associated with semantic and syntactic error detection. This project takes advantage of a large open-source dataset, provided by research conducted by Nieuwland et al. in the study ‘Large-scale replication study reveals a limit on probabilistic prediction in language comprehension,’ consisting of ERP data corresponding to error detection in nouns and articles. We focus on the error detection in expected vs. unexpected nouns, which has been shown repeatedly to have a significant N400 effect, and extend our scope to the subsequent P600 band to perform an analysis of individual variance in these two components, following a precedent set by Tanner et al. in the study ‘Brain-based individual differences in online L2 grammatical comprehension’ and Tanner & Hell in the study ‘ERPs reveal individual differences in morphosyntactic processing’. Our hypothesis is twofold: 1) the data will show an inverse correlation between the strength of the N400 and P600 effects, and 2) the data will show enough individual variance in the brain response to produce three clusters; one with relatively high P600 and low N400 magnitudes, one with relatively high N400 and low P600 magnitudes, and one with a relatively neutral effect in both.
