Email Helen Gent at hmgent2@illinois.edu for more details and Zoom information.
Talk’s title: Nuances of knowing: Using a fictional world to study how the brain makes meaning during word-by-word reading
Abstract:
Across cognitive systems, world knowledge allows individuals to organize raw sensory information into meaningful experiences. Language processing is no exception: how much and how well individuals know things should matter in determining how language is understood in real time. Yet this knowledge-based variability receives little attention in most studies of language processing. In this talk, I will describe a line of research addressing this issue using a linguistically rich, yet constrained, popular domain: the fictional world of Harry Potter (HP), by J.K. Rowling. Across four studies, we recorded event-related brain potentials (ERPs) while young adults who varied in their knowledge of HP read sentences describing fictional HP “facts” (e.g., “Harry has a patronus. It takes the form of a {stag / lion}.”) as well as sentences describing general topics. As a measure of real-time access to knowledge, we focused on N400 amplitudes (negative-going brain potentials occurring about 400 ms after word onset) to sentence-final words.
Individuals’ domain knowledge of HP (assessed using a trivia-style quiz) was moderately-to-strongly correlated with the amplitude of N400 brain potentials to contextually supported (i.e., accurate) words completing sentences about HP, but not to unsupported (inaccurate) endings, nor to N400 effects of more vs. less supported endings to sentences describing everyday/general topics. Single-trial regression analyses pitting domain knowledge against participants’ reports of knowledge about individual HP “facts” (i.e., judgments of whether or not they had known each fact ahead of reading it) revealed trial-level knowledge was a strong, but not the sole, predictor of N400 amplitudes to supported words. Even after accounting for trial-level knowledge reports, and especially when retrieval conditions were more difficult (i.e., for trials reported as unknown by an individual or those generally less likely to be known across participants) but not impossible (e.g., unknown initially, but familiar seeming after input of the critical word), N400 amplitudes were modulated by individuals’ domain knowledge of HP.
We speculated that degree of domain knowledge might modulate real-time semantic retrieval by virtue of the differential organization of that information in semantic memory. We therefore manipulated the relationship between the final (critical) words of HP sentences and their sentence contexts. Critical words were either correct, incorrect-and-unrelated, or incorrect but related either to the events described by the sentence or via sharing the (fictional) category of the critical word. As predicted, domain knowledge influenced N400 amplitudes not only to correct words, but also to the related words, suggesting the knowledge was necessary for this information to be readily available in real time.
Our results provide the first empirical demonstration that rapid, real-time retrieval of knowledge during reading is determined by how much domain-relevant information is known and which facts are readily available to an individual comprehender. Domain knowledge seems to influence implicit retrieval of partial (i.e., incompletely available) information and the availability of information related via events/themes and categories. These findings demonstrate how using a knowledge-based individual differences approach can shine a new lens on how the brain makes meaning from language.
