The impact of language switching on statistical word learning – Experiment 2

It is currently unclear whether language switching hurts, facilitates, or has no impact on word learning. In a first experiment, we observed an advantage of language switching in cross-situational word learning. To replicate and extend these findings, we will conduct a second experiment examining the learning of 2:1 mappings, in which two words map to the same object, as in translations or synonyms. The mappings will be created using English-like and German-like nonwords across three learning conditions: an English-only condition, a German-only condition, and a mixed-language condition. In the mixed condition, each 2:1 mapping will consist of one English-like and one German-like nonword, consistent with translation equivalents. In contrast, in the pure language conditions, both words in a 2:1 mapping will belong to the same language, consistent with synonym learning. We expect to replicate the results of Experiment 1, such that participants will show better learning in the mixed learning condition than in the pure learning conditions, indicating a global benefit of language mixing. Within the mixed-language condition, we will further examine the effects of language switching and repetition. We predict higher accuracy on repetition trials than on switch trials, reflecting local switch costs. Finally, the experiment will include a novel remapping phase in which each object from the initial learning phase is assigned a new nonword: half in English and half in German. We expect participants to remap objects more easily when the new label is in the language that was not encountered for that object during the initial learning phase (only for the pure learning conditions).