Word learning in 14-month-old monolinguals and bilinguals: Challenges and methodological opportunities

Abstract

Infants can learn words in their daily interactions early in life, and many studies have demonstrated that they can also learn words from brief in-lab exposures. While most studies have included monolingual infants, less is known about bilingual infants’ word learning and the role that language familiarity plays in this ability. In this study we examined word learning in a large sample (up to N = 148) of bilingual and monolingual 14-month-olds using a preferential looking paradigm. Two novel words were presented within sentence frames in one language (single-language condition) or two languages (dual-language condition). We predicted that infants would learn both words, and would exhibit better learning when they were more familiar with the sentence frame language. Using a traditional analytic approach (t-tests) and a standard linear regression, we found weak evidence that children learned one of the two words. However, contrary to our prediction, in a minority of conditions infants may have learned better when stimuli were presented in sentence frames in a less familiar language. We also conducted updated analyses using mixed-effects linear regression models, which did not support the conclusion that infants learned any of the words they encountered, regardless of the familiarity of the sentence frame language. We discuss these results in relation to prior work and suggest how open science practices can contribute to more reliable findings about early word learning.

Keywords

• Open Science
• Bilingualism
• Infants
• Word learning