The Cognitive Basis of Linguistic Illusions

Ohio State University, The, Columbus OH

Investigators

Abstract

Our ability to use and understand language in everyday conversation is a feat that seems to be accomplished both rapidly and effortlessly. Just as the impression that our visual interpretation of the world around us is robust and complete, our impression of language is similarly rich. Behind the perceived ease of language understanding is a complex set of neuro-cognitive mechanisms that must work in tandem to give us a rich linguistic experience. However, these systems are not always foolproof and can be misled, arriving at unintended interpretations or the perception of ill-formed sentences as if they were well-formed. These errors trigger what are known as "linguistic illusions", and just as visual illusions have transformed our understanding of how our visual system interprets the world around us, the systematic study of linguistic illusions can transform our understanding of how we interpret language. Importantly, by studying the conditions under which we do and do not fall for linguistic illusions, we can improve language technologies, provide better language learning instruction, and aid in assessments of cognitive impairments. The goal of this project is to better understand the scope and cause of linguistic illusions. Current language processing models assume that illusions are the product of two distinct cognitive systems: one system for fast, efficient communication (the 'parser'), and another, slower system that classifies grammaticality (the 'grammar'). A competing account claims that illusions reflect a single grammar system embedded in a noisy cognitive architecture that creates the opportunity for errors. The current project addresses this debate by extending research to a broader set of linguistic phenomena using behavioral and computational methods to better understand the exact conditions that trigger illusions. This project will inform our understanding of (i) the cognitive architecture of language, (ii) the mental mechanisms for encoding, accessing, and manipulating sentence structures, and (iii) the role of the grammar in sentence processing. The project will have a direct impact on the broader scientific community, the undergraduate researchers funded through project, and the general public. Specifically, the project will expand the pool of participants that are typically included in psycholinguistic studies to non-college students, leading to a more comprehensive study of linguistic abilities across populations, and will bridge research and education by providing meaningful research experiences to undergraduate researchers, including under-represented minorities, through undergraduate training boot camps, research assistantships, and outreach activities at K-12 schools. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.