One can be some but some cannot be one: ERP correlates of numerosity incongruence are different for singular and plural

Abstract

Humans can communicate information on numerosity by means of number words (e.g. one hundred, a couple), but also through Number morphology (e.g. through the singular vs. the plural forms of a noun). Agreement violations involving Number morphology (e.g. *one apples) are well known to elicit specific ERP components such as the Left Anterior Negativity (LAN); yet, the relationship between a morphological Number value (e.g. singular vs. plural) and its referential numerosity has been scantly considered in the literature. Moreover, even if agreement violations have been proved very useful, they do not typically characterise the everyday language usage, thus narrowing the scope of the results. In this study we investigated Number morphology from a different perspective, by focusing on the ERP correlates of congruence and incongruence between a depicted numerosity and noun phrases. To this aim we designed a picture-phrase matching paradigm in Italian. In each trial, a picture depicting one or four objects was followed by a grammatical phrase made up of a quantifier and a content noun inflected either in the singular or in the plural. When analysing ERP time-locked to the content noun, plural phrases after pictures presenting one object elicited a larger negativity, similar to a LAN effect. No significant congruence effect was found in the case of the phrases whose morphological Number value conveyed a numerosity of one. Considering the LAN as an index of morpho-syntactic incongruence, these results suggest that 1) LAN-like effects can be triggered independently from the grammaticality of the utterances and irrespective the P600 component; 2) the reference to a numerosity can be partially encoded in an incremental way when processing Number morphology; and, most importantly, 3) the processing of the morphological Number value of plural is different from that of singular as the former shows a narrower interpretability than the latter.

Effects of animacy on the processing of morphological Number: a cognitive inheritance?

“…When a morpheme is used as a default, it is, in principle, not interpretable at the semantic level. Number morphology is not an exception in this regard: the singular is generally the default value for uncountable references (whereas other Number values are more likely to be semantically interpretable; Franzon, Zanini & Rugani 2018;Arcara et al 2019).…”

From meaning to form and back in American Sign Language verbal classifier morphemes

“…the singular is generally the default value for uncountable references (whereas other Number values are more likely to be semantically interpretable; Franzon, Zanini & Rugani 2018;Arcara et al 2018).…”

Cognitive and communicative pressures in the emergence of grammatical structure: A closer look at whether number sense is encoded in privileged ways

“…In our typological survey, we highlighted the parallelism between the Number values observed in morphology and the numerosities handled by the non-verbal numerical systems (Franzon et al 2019). These cognitive systems, the Object File System and the Analogue Magnitude System, respectively allow the identification of three/four units, and the estimation of larger numerosities (Rugani 2018).…”

“…Faster responses have been found on left side for words inflected in the singular and on right side for words inflected in the plural suggesting a SNARC-like effect in response to morphological Number (Roettger & Domahs 2015). Moreover, electrophysiological activity elicited after a mismatch between a visually presented numerosity and morphological Number on a referring word resembles a response to a grammatical violation (LAN-like effect; Arcara et al 2019). Neuroimaging evidence suggests that neural areas associated with non-verbal numerical processing are activated following the presentation of a Number agreement violation (Carreiras et al 2010).…”