Much to the chagrin of my Vietnamese parents, and probably also a few French teachers from my school years, I never picked up a language beyond English well enough to comfortably call myself bilingual. I think the closest thing to a bilingual experience I’ve ever had was when I attempted to order in French at a restaurant in Montreal (during Psychonomics 2019!), which greatly impressed my non-French-speaking friends until the waiter’s gentle suggestion that we stick to English for the rest of the night.
It’s possible that my cowardly adherence to monolingualism has cost me some opportunities for improved problem-solving, social-emotional development, and long-term success, all of which are frequently attributed to bilingualism or multilingualism, including by the U.S. Department of Education.
Given the advantages afforded by learning more than one language, you might begin to wonder if knowing multiple languages could potentially lead speakers to develop overall stronger executive functioning abilities, like controlling attentional focus or making appropriate cognitive selections. And indeed, this has been a rich area of research (and debate) among psycholinguists, with many assuming that bilinguals tend to be better at adapting their attentional focus to better suit their goals, an ability referred to as “adaptive control”.
Adaptive control abilities are regularly exhibited when people engage in a Stroop task, where names of colors are presented on a screen, and participants are instructed to say what color the font is (see the image below). This is easy enough when a word like RED is displayed in the color red; the difficulty of the task arises if the word RED is presented in a color that’s not red. So if the word RED is shown in the color green, for example, the participant needs to say “green”. It turns out that while suppressing the urge to simply read the name of a color when the name itself is presented in a different color isn’t terribly difficult, it does lead to consistent slowdowns and lower accuracy in people’s responses.
Moreover, these slowdown effects tend to get larger if participants see more congruent trials (where the color word and the actual color match) than incongruent trials (where the color word and the color of the word differ). In other words, the hallmark sign of adaptive control patterns is an interaction of congruency and whether more items in a Stroop task are congruent or incongruent (see below). This interaction, shown below, is key to the claim that this kind of attentional control is adaptive—when participants are exposed to many incongruent Stroop items, they seem better prepared to deal with the conflict between mismatched words and colors.
In the bilingualism research space, it’s sometimes thought that people who speak more than one language use adaptive control in situations where their languages might have different words for the same object. Giacomo Spinelli and Simone Sulpizio (pictured below) test this assumption in a recent brief report published in Psychonomic Bulletin & Review titled “Is adaptation involved in bilingual language production? A fresh look at the assumptions motivating potential bilingual‑monolingual differences in adaptive control”.
To investigate whether bilinguals show adaptive control patterns in situations where their two languages conflict in how to name an object, the authors designed a simple picture-naming task, where some of the pictures were of objects with cognate labels, which are similar in two languages (like a picture of an elephant, which is “elefante” in Italian), and pictures with non-cognate labels (e.g., a picture a horse, which in Italian is “cavallo”). See procedural diagram below. Participants, who spoke Italian primarily and knew English as a second language, had to provide the English label for each picture, which the experimenters hoped would be more sensitive to interference from their primary language of Italian.
Since adaptive control presents as a statistical interaction between congruency and the proportion of congruent items, the experimenters also manipulated whether participants saw more cognate pictures than non-cognate pictures or vice versa, with half of participants seeing three times as many cognate pictures as non-cognates (in a 75/25 split) and the other half seeing the reverse (three times as many non-cognate pictures as cognates).
Their results showed that while bilingual speakers were overall slower at naming pictures whose English labels were very different from their Italian equivalents, there was no interactive influence on whether participants saw mostly cognate or non-cognate objects. However, the results of a Stroop task, which participants completed after the picture-naming task, showed a clear interaction, thereby indicating that participants could still engage in adaptive control to handle the conflict between color and word—they just didn’t use it when their languages were in conflict.
More broadly, these results suggest that the assumption that bilingualism affords better adaptive control is unfounded. This plays into broader issues in bilingual research, which the authors argue is too heavily focused on comparisons between monolingual and bilingual data and not enough direct investigations of the kinds of mechanisms uniquely involved in bilingual processing. With this shift, they argue, bilingual research will become better equipped with “the right kind of map” in its investigations of what specific kinds of advantages are derived from learning multiple languages. Likely to be among them? Not embarrassing yourself at a restaurant.
Featured Psychonomic Society article
Spinelli, G., & Sulpizio, S. Is adaptation involved in bilingual language production? A fresh look at the assumptions motivating potential bilingual-monolingual differences in adaptive control (2024). Psychonomic Bulletin & Review. https://doi.org/10.3758/s13423-024-02503-6
