Asymmetrical switch costs in bilingual language production induced by reading words.

Abstract

We examined language-switching effects in French-English bilinguals using a paradigm where pictures are always named in the same language (either French or English) within a block of trials, and on each trial, the picture is preceded by a printed word from the same language or from the other language. Participants had to either make a language decision on the word or categorize it as an animal name or not. Picture-naming latencies in French (Language 1 [L1]) were slower when pictures were preceded by an English word than by a French word, independently of the task performed on the word. There were no language-switching effects when pictures were named in English (L2). This pattern replicates asymmetrical switch costs found with the cued picture-naming paradigm and shows that the asymmetrical pattern can be obtained (a) in the absence of artificial (nonlinguistic) language cues, (b) when the switch involves a shift from comprehension in 1 language to production in another, and (c) when the naming language is blocked (univalent response). We concluded that language switch costs in bilinguals cannot be reduced to effects driven by task control or response-selection mechanisms.

If you stay, it might be easier: Switch costs from comprehension to production in a joint switching task.

“…However, Peeters et al (2014) showed comprehension to production switch costs using a paradigm in which participants performed an explicit task on input words. The authors assume that comprehension of input words activates language nodes automatically (i.e., it operates as an exogenous factor, in BIA's terminology).…”

“…But since longer preparation times lead to reduced switch costs in language production (Declerck, Philipp, & Koch, 2013), we decided to avoid the use of cues (note that the language in which a word would appear was not pre-cued in Peeters et al, 2014).…”

“…However, this study only looked at off-line measures (i.e., whether interlocutors made similar choices, not how long it took them to make such choices). Therefore, it cannot provide evidence that activation of a language node in comprehension affects the time it takes to select the same language node during subsequent production.Most importantly, a recent study (Peeters, Runnqvist, Bertrand, & Grainger, 2014) found that reading an L2 (English) word (as part of a language decision or semantic categorization task) led to a switch cost when subsequently naming a picture in L1 (French), even though speakers named all the pictures in a block in L1. There was no switch cost from L1 comprehension to L2 production.…”

“…According to the authors, comprehending words in L2 automatically raised the activation level of the L2 language node, which in turn inhibited the L1 language node and made selection of L1 more difficult on the subsequent naming trial. This study therefore provides direct support for the assumption of the BIA model that language nodes are shared between comprehension and production.However, Peeters et al (2014) showed comprehension to production switch costs using a paradigm in which participants performed an explicit task on input words. The authors assume that comprehension of input words activates language nodes automatically (i.e., it operates as an exogenous factor, in BIA's terminology).…”

A review of control processes and their locus in language switching

“…This refers to switch costs being larger when switching into L1 than when switching into L2 (e.g., Jin, Zhang, & Li, 2014;Macizo, Bajo, & Paolieri, 2012;Meuter & Allport, 1999;Philipp et al, 2007;Peeters, Runnqvist, Bertrand, & Grainger, 2014;Verhoef et al, 2009; for a review, see Bobb & Wodniecka, 2013). The most prevalent explanation of this effect is based on the ICM (Green, 1998), which assumes reactive, persisting inhibition to occur during language control.…”