Multimodal language in child-directed versus adult-directed speech

Abstract

Speakers design their multimodal communication according to the needs and knowledge of their interlocutors, phenomenon known as audience design. We use more sophisticated language (e.g., longer sentences with complex grammatical forms) when communicating with adults compared to children. The present study investigates how speech and co-speech gestures change in adult-directed speech (ADS) vs. child-directed speech (CDS) for three different tasks. Sixty-six adult participants (Mage=21.05, sixty female) completed three different tasks (story-reading-task, storytelling-task, and address-description-task) and they were instructed to pretend to communicate with a child (CDS) or an adult (ADS). We hypothesized that participants would use more complex language, more beat gestures, and less iconic gestures in the ADS compared to the CDS. Results showed that, for CDS, participants used more iconic gestures in the story-reading task, and story-telling task compared to ADS. On the other hand, participants used more beat gestures in the story-telling task for ADS than CDS. Additionally, language complexity did not differ across conditions. Our findings indicate that how speakers employ different types of gestures (iconic vs. beat) according to the addressee’s needs and across different tasks. Speakers might prefer to use more iconic gestures with children than adults. Results are discussed according to audience design theory.