On the interpretation of scalar implicatures in first and second language

Abstract We investigated the effect of presenting items in a foreign language (L2) on scalar-implicatures computation. To ensure that L2 processing was more effortful than the processing of the native language (L1), participants were late learners of L2 immersed in an L1 environment and they were presented with oral stimuli under time constraints. If scalar-implicatures computation requires cognitive effort one should find that people are more likely to compute scalar implicatures in L1 than in L2. In two experiments, participants were asked to perform a Sentence Evaluation Task either Italian, their native language, or in a foreign language (English or Spanish). The task included underinformative statements such as “Some dogs are animals” that, if interpreted in a pragmatic way (i.e., “Some but not all dogs are animals”) should be rejected as false. In both experiments, we found more rejections in the native language condition than in the foreign language conditions. These results provide support for models that maintain that scalar-implicature computation is effortful.