Composition decomposed: Distinct neural mechanisms support processing of nouns in modification and predication contexts.

Abstract

Understanding language requires the ability to compose the meanings of words into phrase and sentence meanings. Formal theories in semantics have framed the hypothesis that all instances of meaning composition, irrespective of the syntactic and semantic properties of the expressions involved, boil down to a unique formal operation, that is, the application of a function to an argument, a view known as “Frege’s Conjecture.” We test the processing consequences of this idea using event-related potentials (ERPs) and a novel experimental paradigm where composition versus noncomposition of words from the same grammatical category (nouns) are compared in two different syntactic environments: predication and modification. We found that noun composition in a modification context, where the noun follows an adjective, elicits a reduced N400 component, whereas noun composition in a predication context, where the noun follows a verb, produces an enhanced LAN component. These data challenge the uniqueness thesis, central to formal semantics, and support instead linguistic theories and processing models that posit different composition operations for predicates and modifiers.

How (not) to look for meaning composition in the brain: A reassessment of current experimental paradigms

“…The same sequence of two words can occur as part of the same constituent, in (15a), the Compose condition, or as separated by a syntactic boundary, in (15b), the Cut condition, in this case also marked by punctuation. The first EEG study using this design, by Olstad et al (2020), removed punctuation marks in order to match the precritical (e.g., "grey") and critical ("elephants") words. Additional safeguards had to be implemented to prevent accidental composition in the Cut condition.…”

“…Similar to other paradigms, Cut-Compose also affords the possibility of investigating the compositional mechanisms involved in different semantic and syntactic contexts. Olstad et al (2020) compared modification as in ( 15), with predication constructions as in ( 16), to assess whether these two different "modes of composition"-Predicate Modification vs. Functional Application, Adjoin vs. Merge-correspond to different neural events. As the study was conducted in Norwegian, the Cut sentence was created by fronting the object:…”

“…An additional perceptual difference is capitalization of the first letter of the critical word in Cut. Olstad et al (2020) avoided the use of punctuation and capitalization, relying on the structural properties of sentences to ensure that the noun is interpreted as starting a new sentence in the Cut condition. Follow-up experiments are needed to investigate the effects of both punctuation and capitalization in the visual modality, whether they affect the detection and quantification of composition signals, and the corresponding impact of appropriate prosody or intonation around the Cut boundary in the auditory modality.…”

Compositionality in a Parallel Architecture for Language Processing

“…In any case, this body of work provides suggestive evidence for distinct neuroanatomical pathways, subserving M‐ and G‐stream processing. But much remains to be done to precisely demarcate the boundaries of composition(ality) in brain space and time (Olstad, Fritz, & Baggio, 2020).…”

Epistemic transfer between linguistics and neuroscience: Problems and prospects

“…Pylkkänen and collaborators found stronger MEG responses from the left anterior temporal lobe (LATL) in 'minimal' NPs (e.g., 'red boat') vs word lists containing the same critical noun ('cup, boat'), at about 200-250 msec after noun onset (Bemis & Pylkkänen 2011Blanco-Elorrieta & Pylkkänen 2016;for EEG data, see Flò et al 2020, Olstad et al 2020). This early LATL response was initially linked to structure-building operations, but further research by the same lab indicates that it may rather reflect combinatorial processes sensitive to specific conceptual relations between words (e.g., the specificity of the noun relative to the modifying adjective, as in 'spotted fish' vs 'spotted trout'; for a review and critical discussion, see Westerlund & Pylkkänen 2017, Pylkkänen 2019.…”