Is inhibition implemented during bilingual production and comprehension? n-2 language repetition costs unchained

ABSTRACT

Many models assume that inhibition plays an integral role during bilingual language control, a process that restricts bilingual language processing to the target language. However, there is limited evidence for such a claim. In the current study, we set out to investigate one known marker of bilingual inhibition (n-2 language repetition costs) that has, so far, mainly been investigated with digits in a production task. Hence, we ran a n-2 language repetition study with other types of stimuli (i.e. pictures and written words) in a production and a comprehension task. The results showed that n-2 language repetition costs were found with both stimulus types in a production task. In the comprehension tasks, n-2 language repetition costs were only observed with one of the two stimulus types and in only one language. These results indicate that inhibition is implemented during bilingual production, and is possible, but not necessary, during bilingual comprehension.

KEYWORDS:

• Bilingualism
• inhibition
• comprehension
• production
• n-2 repetition costs

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 An ANOVA of the RT data revealed a significant effect of language, F(2, 34) = 47.03, p = .000,  = .734, with slower responses during French trials (1071 ms; SE: 15), t(17) = 7.81, p = .000, and English trials (1003 ms; SE: 17), t(17) = 5.41, p = .000, than in German trials (979 ms; SE: 15). Responses were also slower during English than German trials, t(17) = 5.74, p = .000. The effect of language transition was significant, F(1, 17) = 24.03, p = .000,  = .586, with slower responses during ABA trials (1026 ms; SE: 15) than during CBA trials (1011 ms; SE: 15), indicating n-2 language repetition costs of 15 ms. The interaction was also significant, F(2, 34) = 6.25, p = .005,  = .269, with larger n-2 language repetition costs observed in French trials (30 ms; t(17) = 7.42, p = .000) than in English (8 ms; t(17) = 1.2, p = .240), t(17) = 3.04, p = .007, or German trials (7 ms; t(17) = 1.47, p = .159), t(17) = 3.18, p = .005. Though, there was no difference in n-2 language repetition costs in German and English, t < 1.

An ANOVA of the error data revealed a significant effect of language, F(2, 34) = 6.92, p = .009,  = .289, with more errors during French trials (3.9%; SE: 0.8) than in English (2.4%; SE: 0.4), t(17) = 2.27, p = .036, or German trials (1.9%; SE: 0.4), t(17) = 3.16, p = .006. Though, there was no significant difference in error rates between German and English trials, t(17) = 1.53, p = .143. The effect of language transition, F < 1, and the interaction were not significant, F(2, 34) = 1.47, p = .245,  = .079.

Additional information

Funding

This project has received funding from the European Union's Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No 706128.