Facilitative use of grammatical gender in Heritage Spanish

Abstract

This paper presents an eye-tracking study using the Visual World Paradigm that tests whether participants are able to access gender information on definite articles and deploy it to facilitate lexical retrieval of subsequent nouns. A comparison of heritage speakers of Spanish with control monolingual speakers of Spanish suggests that the heritage speakers’ performance on this task is qualitatively similar to that of the baseline. This suggests that, despite non-target-like performance in offline tasks targeting gender production and comprehension, heritage speakers of Spanish can use gender in a target-like manner in online tasks. In line with proposals put forth by Grüter et al. (2012) and Montrul et al. (2014), a preliminary comparison with previous work on L2 learners (Lew-Williams & Fernald, 2010; Grüter et al., 2012; Dussias et al., 2013) provides tentative support for the idea that the nature of early language learning is crucial in developing the ability to use grammatical gender to facilitate lexical retrieval (Grüter et al., 2012; Montrul et al., 2014).

Eyetracking evidence for heritage speakers’ access to abstract syntactic agreement features in real-time processing

“…Previous work on facilitative use of gender agreement in the processing of nouns in other populations has suggested that when the experimental method involves a gender cue located on an article that is frequent or obligatory in the language, 2 such results are consistent with two possible accounts: under a syntactic account, participants are in fact accessing abstract syntactic information on the article during processing of the noun phrase; under a probabilistic account, the results reflect a mechanism that relies on surface probabilities between frequently cooccurring article-noun pairs (van Heugten and Shi, 2009;Lew-Williams and Fernald, 2010;Melançon and Shi, 2015). Existing work on HSs in this domain (Fuchs, 2021) is consistent with either account, and therefore further work is needed to adjudicate between the accounts. Under a syntactic account, we should expect to observe HSs' facilitative use of grammatical gender when the gender cue is located on a non-frequent, non-obligatory element within the nominal phrase.…”

“…Given this, HSs patterning with adult and child controls in their ability to use gender to facilitate lexical retrieval has implications for the understanding of how the nature of the acquisition process may impact processing abilities. Following Grüter et al (2012) and Montrul et al (2014), Fuchs (2021) suggests that early and naturalistic acquisition of grammatical gender in the speech stream may be crucial for developing robust associations 10.3389/fpsyg.2022.960376 between pre-nominal gender cues and subsequent nouns, as discussed further in section "Discussion. "…”

“…However, a growing body of evidence suggests that task modality may play a role in HSs' performance on various experimental tasks that have been used to assess their knowledge of grammatical gender (Alarcón, 2011;Montrul et al, 2014), and online methods may be advantageous in providing a more nuanced understanding of HSs' knowledge (Polinsky, 2018;Bayram et al, 2020;Fuchs, 2021). An example of work using online methods in this domain controls for what HSs know about grammatical gender in the HL and instead focuses on how they use that information in real time, demonstrating that HSs' processing of grammatical gender may be target-like, counter to expectations based on HSs' divergent performance on offline tasks.…”

“…In an eye-tracking task in the Visual World Paradigm (Tanenhaus et al, 1995). Fuchs (2021) found that HSs of Spanish were able to access and deploy gender information on prenominal gender-marked articles to facilitate the processing of the subsequent noun. In the study, when viewing a display with two images representing lexical items of different genders ("mismatch" condition) and hearing a prompt that included a prenominal gender-marked cue (the masculine article el or the feminine article la), both HSs and the control group fixated on the target item faster than when viewing a display with two images of the same grammatical gender ("match" condition), for which the prenominal gender cue did not disambiguate the target item.…”

“…Nevertheless, recent evidence from studies using online methodologies suggests that despite surface differences in production and comprehension of grammatical gender agreement, when one controls for knowledge of gender categorization, processing of gender by HSs may be qualitatively target-like. In an eyetracking study in the Visual World Paradigm, Fuchs (2021) found that HSs of Spanish were able to fixate on target items faster when a pre-nominal gender-marked article was sufficient to uniquely identify the target item than when it was not. Fuchs concluded that HSs were able to use gender information in real-time to facilitate lexical retrieval, in a manner qualitatively like the control group.…”