Filippi Piera, Ocklenburg Sebastian, Bowling Daniel L, Heege Larissa, Güntürkün Onur, Newen Albert, de Boer Bart
a Artificial Intelligence Laboratory , Vrije Universiteit Brussel , Brussels , Belgium.
b Center for Mind, Brain and Cognitive Evolution , Ruhr-University Bochum , Bochum , Germany.
Cogn Emot. 2017 Aug;31(5):879-891. doi: 10.1080/02699931.2016.1177489. Epub 2016 May 3.
Humans typically combine linguistic and nonlinguistic information to comprehend emotions. We adopted an emotion identification Stroop task to investigate how different channels interact in emotion communication. In experiment 1, synonyms of "happy" and "sad" were spoken with happy and sad prosody. Participants had more difficulty ignoring prosody than ignoring verbal content. In experiment 2, synonyms of "happy" and "sad" were spoken with happy and sad prosody, while happy or sad faces were displayed. Accuracy was lower when two channels expressed an emotion that was incongruent with the channel participants had to focus on, compared with the cross-channel congruence condition. When participants were required to focus on verbal content, accuracy was significantly lower also when prosody was incongruent with verbal content and face. This suggests that prosody biases emotional verbal content processing, even when conflicting with verbal content and face simultaneously. Implications for multimodal communication and language evolution studies are discussed.
人类通常会结合语言和非语言信息来理解情感。我们采用了一种情绪识别斯特鲁普任务来研究不同通道在情感交流中是如何相互作用的。在实验1中,用高兴和悲伤的语调说出“高兴”和“悲伤”的同义词。与忽略言语内容相比,参与者更难忽略语调。在实验2中,用高兴和悲伤的语调说出“高兴”和“悲伤”的同义词,同时展示高兴或悲伤的面孔。与跨通道一致条件相比,当两个通道表达的情绪与参与者必须关注的通道不一致时,准确率较低。当要求参与者关注言语内容时,若语调与言语内容和面孔不一致,准确率也会显著降低。这表明,即使语调与言语内容和面孔同时冲突,语调也会对情感言语内容的处理产生偏向。文中还讨论了其对多模态交流和语言进化研究的意义。