Stance—attitudes and opinions about the topic of discussion—has been investigated textually in conversation analysis, discourse analysis, and computational models, but little attention has focused on its acoustic-phonetic properties. It is a challenge, given the complexity of stance and the many other types of meaning that share the same acoustic channels, all overlaid on the lexical and syntactic material of the message. With the goal of identifying automatically-extractable, acoustically-measurable correlates of stance-taking, this work identified signals of stance in prosodic measures of fundamental frequency, intensity, and duration in an audio corpus of dyads engaged in collaborative conversational tasks designed to elicit frequent cha...