Speakers Raise their Hands and Head during Self-Repairs in Dyadic Conversations

People often encounter difficulties in building shared understanding during everyday conversation. The most common symptom of these are self-repairs, when a speaker restarts, edits or amends their utterances mid-turn. Previous work has focused on the verbal signals self-repair, i.e. speech disfluences (filled pauses, truncated words and phrases, word substitutions reformulations), computational tools now exist that can automatically detect phenomena. However, face-to-face conversation also exploits rich non-verbal resources previous research suggests self-repairs associated with distinct hand movement patterns. This paper extends those results by exploring head movements both speakers listeners using two motion parameters: height (vertical position) 3D velocity. show sequences containing distinguishable from fluent ones: raise hands more (and move rapidly) self-repairs. We obtain analysing data corpus 13 unscripted dialogues, we discuss how findings could support creation improved cognitive artificial systems for natural human-machine human-robot interaction.