Diverse multi-modal behaviors provide important cues in establishing and maintaining interactional rapport. However, these behaviors are often subtle and culture-specific. In this paper, we focus on two forms of backchannel behavior: vocal backchannels and non-verbal headnods. We employ a corpus of quasi-monologic story-telling interactions elicited from three distinct language/cultural groups: American English, Mexican Spanish, and Iraqi Arabic speakers. Through this corpus, we investigate prosodic cues associated with these two different types of verbal feedback. We identify both similarities and differences in the cues exploited by the speakers of these diverse language/cultural groups. Although both typically classed as backchannels, we observe substantial differences in cues associated with verbal and non-verbal feedback head nods across these languages. These contrasts argue for a more fine-grained analysis of the use and role of diverse social resonance behaviors.

Index Terms: prosody, backchannels, multilingual analysis, multimodal