Melissa’s research primarily uses EEG/ERPs to ask how variation in background knowledge (gleaned from reading as well as multi-sensory experiences) influences the real-time temporal (and neural) dynamics of written language processing. In her dissertation and postdoc, she has been focusing on the narrative world of Harry Potter (HP). She has found that variation in HP domain knowledge has a quick influence on understanding words that are supported in HP sentence contexts. Moreover, degree of knowledge seems to have a rapid impact on the extent to which individuals pre-activate related/relevant semantic relations (including category and thematic relations) in the normal course of sentence processing. In her postdoc, she is also examining how semantic relations and domain knowledge are related to information captured by distributed semantic models (DSMs) directly trained on texts from domains of interest (e.g., HP).