Animacy modulates gender agreement comprehension in Hindi: An ERP study

ABSTRACT

The animate–inanimate distinction is of crucial importance cognitively, and animacy has been known to influence language comprehension. However, little is known about the role of animacy in verb agreement processing. The present study employed event-related brain potentials to examine whether the gender agreement of the verb with animate (natural gender) and inanimate (grammatical gender) subject nouns reveal similar or different processing mechanisms in Hindi. Critical stimuli were intransitive sentences of the form subject–verb–aux. Subject nouns were either animate or inanimate, and the verb either showed correct gender agreement or violated it. The violation of gender agreement with animate subjects evoked a P600 effect, whereas gender agreement violation with inanimate subjects revealed a long-latency N400-like effect. The result suggests that different underlying mechanisms are involved in the computation of gender agreement with animate and inanimate arguments in Hindi, illustrating the crucial role that animacy plays in verb agreement processing.

KEYWORDS:

• Language
• N400
• P600
• Gender
• Agreement
• Animacy

Acknowledgements

The authors wish to thank Sebastian Sauppe and Mahima Gulati for their insightful comments on earlier versions of this article. The authors would also like to acknowledge the three anonymous reviewers of this manuscript for their fruitful comment and suggestions, which went a long way in improving the manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Most animate nouns possess natural gender, which corresponds to the biological sex of the referent. In case of inanimate nouns, there is no correspondence with the biological sex of the referent. Thus, inanimate nouns possess grammatical gender.

2 In Hindi, verb agreement is contingent upon the case marking of the argument(s) in a sentence and not upon the animacy of the argument. More details on this are discussed in Section 1.3 of this paper.

3 Although the LAN has been associated with the detection of morphosyntactic violations, such as agreement, it has been a subject of debate whether it is reliably elicited by agreement violations (Molinaro et al., 2008; Tanner & Van Hell, 2014; Molinaro et al., 2015; van de Meerendonk et al., 2009).

4 Feminine arguments in Hindi have different agreement markers than masculine arguments. This would have necessitated a huge number of trials for a balanced design, leading to an unusually long experiment. Therefore, we exclusively employed masculine nouns in our critical stimuli. Filler sentences, however, included feminine arguments.

Additional information

Funding

This work was supported by Indian Institute of Technology Ropar through an ISIRD grant to Kamal Kumar Choudhary [IITRPR/Acad./2216].