Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
The neural architecture of compositional generalization in language: how do we infer the meaning of “un-reject-able-ish”
There is a Poster PDF for this presentation, but you must be a current member or registered to attend SNL 2023 to view it. Please go to your Account Home page to register.
Poster E110 in Poster Session E, Thursday, October 26, 10:15 am - 12:00 pm CEST, Espace Vieux-Port
Xiaochen Zheng1, Mona Garvert2, Hanneke den Ouden1, Roshan Cools1,3; 1Donders Institute for Brain, Cognition and Behaviour, Nijmegen, the Netherlands, 2Max-Planck-Institute for Human Cognitive and Brain Sciences, Leipzig, Germany, 3Radboud University Medical Center, Nijmegen, the Netherlands
The ability to generalize previously learned knowledge to novel situations is essential for adaptive behavior. For example, when encountering the word “un-reject-able-ish” for the first time, one can easily infer its meaning by integrating knowledge of its constituent parts based on abstract structural rules (such as the sequential order of word parts). To investigate the neural mechanisms involved in inferring novel compositional word meanings, we developed a behavioral paradigm to probe compositional inference in language. In a pre-registered fMRI study, we trained 43 participants on the meanings of compositional words made up of known stems (“good”) and novel affixes (“kla”) from an artificial language. The meaning of the compositional words depended on the position of the novel affix (“goodkla = bad”, “klahorse = pony”). We then asked them to infer the meaning of novel compositional words (“whitekla =?”, “klacat =?”) that were either congruent or incongruent with the established rule (“klawhite” is incongruent because a small version of “white” does not exist). In the scanner, participants performed a semantic priming task in which the novel words served as either congruent or incongruent primes (“whitekla”) and their synonyms (“black”) served as targets. After the scanning session, they were asked to indicate whether these novel words they had been presented held any meaning and, if so, to provide their interpretation. Our results demonstrated that participants were able to generate novel compositional meanings on the fly, successfully inferring meanings of congruent versus incongruent words. Univariate analysis of fMRI data at target words revealed a greater repetition suppression effect when primed with congruent than incongruent words in the left inferior frontal gyrus, which suggests that novel meanings are derived at this linguistic “building” hub. Further analysis of congruent versus incongruent prime-related activity revealed a broad frontal-parietal network, including the hippocampus, a brain area commonly associated with the generalization process of structural relationships. Furthermore, we employed multivariate representational similarity analysis to demonstrate structural rule and word meaning representations in hippocampus and left-lateralized core language areas. Together, these findings suggest that compositional generalization in language recruits a domain-general network shared with action planning, compositional vision and constructive relational memory, while newly inferred meanings are represented in more language-specific regions.
Topic Areas: Control, Selection, and Executive Processes, Meaning: Lexical Semantics