Publication Date


Document Type


Degree Name

Master of Science in Psychology (MS)

Committee Member

Mott, Daniel

Committee Member

Munoz, Monica

Committee Member

Cieslicka, Anna B


The purpose of the present study investigates how bilinguals comprehend gender, especially when one of their languages is grammatically marked in gender. This study explored influences of world knowledge, lexical representations, and language activation in a bilingual population. Eye-movements were recorded as bilinguals read sentences containing gender-stereotyped nouns (e.g., surgeon) followed by a male or female pronoun (he/she) that served as an anaphoric referent. Language dominance in Spanish and English dominant and balanced bilinguals, the gender of the stereotype noun (i.e., female or male), and congruency between noun and anaphor (match or mismatch) were included as variables. The following eye movements were measured: gaze duration, regressions, and right-bounded durations. Gaze duration showed slower reading times for Spanish dominant bilinguals. Regressions to the antecedent also showed a greater effect for Spanish dominant bilinguals. Right-bounded duration showed longer reading times in the anaphor region and also was found higher for the Spanish dominant bilinguals. The results are supported by an inferential view that states readers infer gender through a mental representation of previously stored gender information. The associations made between Spanish, a language that specifies gender explicitly (e.g., la mesa; Heredia, Cieślicka, & Altarriba, 2016), and English, a language that does not contain a grammatical gender in nouns. The conditions manipulated the pronoun to match or mismatch its stereotyped antecedent. Reading difficulties were quantified as mismatch effect by tracking eye-movements as the participants read each sentence. The anaphor and antecedent regions of interest (ROI) were measured for early- and late-stage effects through eye measurement recordings. Analyses were used to separate the effects in each of the two ROIs. Effects are reported for the early-stage processing through gaze duration and skip percentages, and for late-stage processes through right-bounded duration, first-pass, and regression times. The role of language dominance of bilingual participants in relation to the comprehension of gender in discourse is reported. Bilinguals who were Spanish-dominant showed more processing difficulty effects than English-dominant and balanced bilinguals.