When Grammar Meets Gender Meets Sex: A Cross-Cultural and Cross-linguistic Study on Grammatical Gender Processing

Grammatical gender processing is a lexical-syntactic feature that classifies nouns into different gender values. The noun "book" for instance, is masculine in Spanish (libro), neuter in German (Buch), and feminine in Polish (książka). Gender assignment in inanimate nouns like "book" lacks a semantic core. Yet, on the one hand, discussion remains on where gender is localized within the architecture of lexical access, and specifically, whether it is a semantic feature or more of an independent syntactic feature (see Cubeli et al., 2005; Sá-Leite et al., 2022). On the other hand, it's interesting how the gender values of many Indo-European languages establish a dichotomy that parallels our social and sexual reality: "book" can be masculine or feminine. The critical question here links grammar and social cognition, and asks whether our sex-based reality influences our cognitive representation and processing of grammatical gender. Evidence integrating the linguistic and social-cognitive dimensions of gender across languages remains limited. Building on Casado, Sá-Leite, Pesciarelli & Paolieri (2023), the present project aims to conduct a large-scale cross-linguistic replication of their gender categorization paradigm, systematically extending it to typologically and socioculturally diverse languages with gender systems that vary in the number of gender values (i.e., including a neuter value that breaks the masculine–feminine dichotomy). The original study examined gender congruency effects in a determiner categorization task in Spanish, using plural demonstrative pronouns (e.g., estos/estas) preceded by gender-marked nouns, and investigated how psycholinguistic variables (e.g., animacy, gender value) and social-cognitive measures (e.g., sexism, masculinity/femininity) modulated grammatical gender processing. For this, in addition to behavioral measures, participants completed standardized questionnaires assessing hostile and benevolent sexism and gender role traits (e.g., ASI, BSRI), using validated versions where available or carefully translated instruments otherwise. A key finding of the original study was that the feminine gender value diminished the gender congruency effect, particularly among participants exhibiting more masculine-oriented attitudes. The authors interpreted this pattern as reflecting a reduced activation of feminine gender representations, potentially limiting their facilitative role in grammatical gender processing. We replicate the core experimental manipulation while adapting stimulus selection to the morphosyntactic properties of each target language. In Romance languages with transparent plural gender marking (Spanish, European and Brazilian Portuguese, Galician, Catalan, Italian, French), plural demonstratives are used. In languages where plural forms neutralize gender distinctions (e.g., German, Polish, Russian, Ukrainian), singular demonstratives or agreement markers are used to preserve gender transparency and agreement variability. Arabic is included with language-specific adaptations consistent with its gender system. Across all languages, materials are tightly controlled to maintain equivalence in gender congruency, animacy, and lexical properties. By testing multiple languages and societies within a single experimental framework, this project aims to provide robust cross-linguistic evidence on how grammatical gender is processed during lexical access as a function of noun animacy, type of gender, sexual stereotypes, and, critically, the potential interaction of the effect with social-cognitive factors (e.g., participant sex, gender role orientation, and sexist attitudes). We do so by addressing methodological limitations related to sample composition and statistical power in previous work, primarily through the recruitment of adequately powered samples that are balanced across key demographic and social-cognitive variables.