“…This makes language switching a double‐edged sword: One the one hand, it is more ecological and makes greater use of the learner's linguistic resources, but on the other hand, it is likely to induce a cognitive cost in both perception and production (see Gullifer, Kroll, & Dussias, , for a review). When bilinguals are asked to switch between languages while speaking, each language switch is accompanied by an increase in the time required to start speaking and by greater chances of making errors as compared to the nonswitching conditions (see, among many others, Costa & Santesteban, ; see also Branzi, Martin, Abutalebi, & Costa, ; Verhoef, Roelofs, & Chwilla, , for electrophysiological and neuroimaging data). Similarly, when bilinguals are presented with written input that involves language switches, they typically display longer recognition times and larger electrophysiological effects for switch as compared to nonswitch trials (e.g., Macizo, Bajo, & Paolieri, ; Van der Meij, Cuetos, Carreiras, & Barber, ), even when switches are not perceived consciously, as is the case in masked priming experiment (e.g., Casaponsa, Carreiras, & Duñabeitia, ; Chauncey, Grainger, & Holcomb, ; Duñabeitia, Dimitropoulou, Uribe‐Etxebarria, Laka, & Carreiras, ).…”