Tuesday, June 17, 2025

Presentation Panel: Experimental Methods

The role of noun transparency in the facilitative processing of grammatical gender in Heritage Spanish

• Zuzanna Fuchs, Ph.D., University of Southern California

[Download presentation slides]

In the rich literature on grammatical gender in Heritage languages, one important question concerns the role of transparency of gender marking in Heritage Speakers’ knowledge of gender and computation of gender agreement. Transparent nouns are those whose form is strongly associated with a certain gender (ex. in Spanish, nouns ending in -o are typically masculine, and those ending in -a are typically feminine), and non-transparent nouns are those whose form is not clearly associated with one gender (ex. in Spanish, nouns ending in -e). Many studies involving written presentation of stimuli have shown that Heritage Speakers tend to show higher sensitivity to mismatches in gender agreement between a noun and an article and/or adjective when the noun is transparent than when it is not. In the present study, we investigate the role of transparency in auditory word recognition in Spanish using eye-tracking in the Visual World Paradigm: participants (52 heritage speakers, 14 Spanish-dominant speakers) heard phrases like Indica la mesa (“Indicate the table.”) and were asked to click on the corresponding (target) image either (a) when the target and competitor images on the screen were of nouns of the same grammatical gender, or (b) when they were of different genders. This experimental approach allows us to observe how gender processing and word recognition proceed moment by moment, and to determine whether noun transparency – which in Spanish is a property of the noun ending – affects these processes prior to the noun ending itself unfolding in the auditory input.

Lexical processing in Vietnamese heritage language speakers: Research utilizing the masked priming technique

• Juliet Huynh, Ph.D., University of Wisconsin-Madison

[Download presentation slides]

Masked translation priming studies have shown that priming in either the L1-L2 direction or L2-L1 direction depends on the relative dominance between the L1 and the L2 (Basnight-Brown & Altarriba, 2007; Gollan et al., 1997; Jiang, 1999). This study examines translation priming in Vietnamese heritage language (HL) speakers who are relatively more proficient in their L2 English than L1/HL. If relative proficiency plays a critical role, then L1/HL speakers should only show L2-L1/HL priming since they are L2 dominant.

L1/HL Vietnamese speakers performed lexical decision tasks that employed the regular three-field masking procedure in two experiments. Masked priming offers insight into early levels of lexical activation. In Experiment 1 (N=26), participants were presented with L1/HL Vietnamese targets (MẮT, translation: “eye”) preceded by primes that were either a repetition of the target (mắt) or an L2 English translation (eye). Conversely, in Experiment 2 (N=24), participants were presented with L2 English targets (EYE) preceded by primes that were either a repetition of the target (eye) or the L1/HL translation (mắt).

Priming was found in both experiments in the repetition condition while found only in the L2-L1/H1 direction for the translation condition. The lack of priming in the L1/HL-L2 direction does not follow the consistent L1-L2 priming found in previous literature. It seems that the L1/HL cannot prime translations in the other language when the masked priming procedure is used. However, the L2 is dominant enough to prime L1, suggesting that relative proficiency of bilinguals’ languages affect the pattern of translation priming results.

References

Using Intuition-Based Tasks to Explore Heritage Language Grammar

• Ji Young Kim, Ph.D., University of California, Los Angeles

[Download presentation slides]

Intuition informs us about many aspects of a language–whether speech sounds accented, whether a construction is grammatical, or whether it conveys particular social meanings. Heritage speakers sometimes demonstrate different patterns from non-heritage native speakers in how they speak and process their heritage languages. As a result, researchers interested in these divergent patterns often design behavioral studies in which native speakers—both heritage and non-heritage—are directly or indirectly asked to provide their intuitions about aspects such as nativeness, grammaticality, and attitudes toward these patterns.

With the help of the internet, conducting online surveys and experiments on heritage languages has become significantly easier. Researchers can now remotely collect intuitive judgments from speakers around the world without the need for in-person interaction. However, without adequate contacts, finding participants remains a challenge. To address this, researchers have turned to crowdsourcing platforms to gain access to a broader participant pool. Despite inherent limitations (e.g., limited control over testing procedures, challenges in verifying participant eligibility), crowdsourcing is a fast, cost-effective, and convenient method for reaching large and diverse populations.

In this presentation, I will discuss Prolific, a crowdsourcing platform primarily designed for academic research. Using a study on global heritage accent rating as an example (Kim et al., 2023, 2025), which gathers native listeners’ intuitions about the perceived nativeness of heritage speakers’ speech, I will demonstrate how Prolific can be effectively utilized in heritage language research. Additionally, I will present quality control measures that help ensure the reliability and validity of the collected data.

References

Heritage Language Development in Dual Language Schools: Methodological Innovations and Future Directions

• Patrick D. Thane, M.Ed., Ph.D., University of Houston

[Download presentation slides]

Research on the morphosyntactic development of Spanish heritage speakers (HSs) has continued to amass rapidly. However, very little work has compared the acquisition of Spanish by children attending dual language bilingual education (DLBE) programs with development in the context of traditional monolingual education. This presentation incorporates recent data from DLBE schools on the development of subjunctive mood (Thane, 2024b; Thane et al., 2025), differential object marking (Thane, 2024a), and direct object clitic gender (Austin et al., 2024) in children between ages 7 and 14 using elicited production and forced choice tasks. Findings across studies suggest that DLBE provides a subtle advantage in children’s acquisition of differential object marking compared to monolingual education. Older children demonstrated greater command of the structures tested, and children who self-reported using Spanish more frequently produced and recognized the morphosyntactic structures in question more frequently.

Together, these findings show that DLBE is an essential but insufficient component of heritage language acquisition and maintenance. The presentation concentrates primarily on the link between methods from formal approaches to bilingualism and heritage language education. Implications and opportunities for DLBE programs beyond language acquisition theory are emphasized for both a research audience and the general public. While these projects utilize formal approaches to bilingualism, they are methodologically innovative in that they set the stage for implementation science research in DLBE programs such as studies of the effectiveness of pedagogical interventions in bilingual classrooms.

References