Bilingualism affects infant cognition: Insights from new and open data

Abstract

Bilingualism has been hypothesized to shape domain-general cognitive abilities across the lifespan, in what some have called the “bilingual advantage”. Here, we examined the replicability of a seminal study that showed monolingual–bilingual differences in infancy (Kovács & Mehler, 2009a) by collecting new data from 7-month-olds and 20-month-olds and reanalyzing three open datasets from 7–9 month-olds (D’Souza et al., 2020, Experiment 1; Kalashikova et al., 2020, Visual and Auditory conditions). All infants (total N = 181) were tested in an anticipatory eye movement paradigm, where they learned to use a visual and/or auditory cue to anticipate a visual reward presented consistently on one side of a screen during training, and on the opposite side of the screen at test. To correctly anticipate the reward at test, infants had to update their previously learned behavior. Across 3 out of 4 studies, results from a new analytic approach showed that 7–9 month-old bilinguals were better able to update the previously-learned response at test (a “bilingual advantage”), which could be related to bilinguals’ weaker initial learning of the contingency during the learning phase (a “monolingual advantage”). At 20 months, bilinguals performed better at test, though groups showed similar performance during the learning phase. Overall, these results show that bilingualism affects how infants process both new and updated information during learning.

Six solutions for more reliable infant research

“…As a first example, a recent paper (Dal Ben, Killam et al, 2021) investigated cognitive differences between 7-month-old monolingual and bilingual infants based on the seminal work of Kovács & Mehler (2009). On nine training trials, infants saw a central cue followed by a reward on one side of the screen and on nine test trials the reward switched sides such that it appeared on the opposite side of the screen (sides were counterbalanced).…”

“…Overall, this example demonstrates an interesting interplay between inclusion criterion and experimental power, due to different effect sizes. Different strategies might be optimal for different paradigms, depending on the tradeoff between gains in effect size versus the loss of participant numbers when stricter inclusion criteria are implemented (see also Dal Ben, Killam, Pour Iliaei, & Byers‐Heinlein, 2021, for another example where stricter inclusion criterion yielded more robust results). Note that for previous studies on infant‐directed speech, when stated, inclusion criteria were much stricter than the strictest criterion assessed in ManyBabies 1 (50%), often requiring infants to complete 100% of trials (e.g., Fernald & Kuhl, 1987; Inoue, Nakagawa, Kondou, Koga, & Shinohara, 2011).…”

Six solutions for more reliable infant research

“…As a first example, a recent paper (Pour Iliaei et al, 2020) investigated cognitive differences between 7-month-old monolingual and bilingual infants based on the seminal work of Kovács & Mehler (2009). On nine training trials, infants saw a central cue followed by a reward on the left side of the screen (counterbalanced) and on nine test trials the reward switched sides such that it appeared on the right side of the screen.…”

Defining bilingualism in infancy and toddlerhood: A scoping review

“…For example, in a review by Williams et al (2021), of 26 studies focusing on infants' executive function and memory skills, only 10 of them reported a significant difference between monolinguals and bilinguals-with bilinguals outperforming their monolingual peers. Some people have attributed the contradictory findings to the variability of tasks utilized (Hilchey & Klein, 2011;Valian, 2014), the analytical approaches (Iliaei et al, 2020), and even the definitions used to classify infants into various language groups (Luk & Bialystok, 2013;Marian, 2018). To help answer this question, the present scoping review will analyze whether the differences in definitions (using language exposure cutoff as a proxy) could impact the likelihood of the study reporting a significant difference between monolingual and bilingual infants.…”