Recently, a study by Champoux-Larsson and Dylman (2018) showed that the bilingual advantage previously found in the use of emotional prosodic cues in children to infer a speaker’s emotional state (e.g., Yow & Markman, 2011) was driven by a bias towards prosody. Namely, the higher level of bilingualism the participants in ChampouxLarsson and Dylman (2018) had, the more they had difficulty ignoring prosodic emotional cues in spoken words even when they were asked to focus on the semantics of the words. While Misono et al. (1997) found that monolingual adults rely on both semantic and prosodic cues to determine emotion in speech equally, it is not known yet whether this also is true for bilingual adults. In other words, it is unclear whether the prosodic bias found in bilingual children withstands even in adulthood for bilinguals. Thus, we present a study where adults with varying levels of bilingualism were asked to determine the emotional valence of utterances based on the participant’s general impression (i.e., without specifying which cue to use), based on the utterance’s emotional prosody or based on its semantic content. The spoken words’ semantics was positive, negative or neutral and the words were uttered with either a congruent emotional prosody or with an incongruent emotional prosody. Data is currently being prepared for analysis and results will be available within the coming weeks.