Edited by Fergus Lyon, Guido Möllering and Mark N.K. Saunders
Chapter 23: A Voice is Worth a Thousand Words: The Implications of the Micro-coding of Social Signals in Speech for Trust Research
Benjamin Waber, Michele Williams, John S.Carroll and Alex ‘Sandy’ Pentland INTRODUCTION Self-report measures of trust reflect an important and often highly reliable tool for researchers interested in trust (Mayer and Davis, 1999). However, self-report measures require subjects to stop and think about how much they trust others or are trusted by others. Researchers are not able to use these methods when subjects cannot stop to fill out surveys in real time. In our setting, medical conversations or handoffs, one member of the pair must quickly receive critical information about a patient’s current medical condition and then immediately begin caring for that patient. The rushed and technical nature of these conversations also makes qualitative research difficult because most of the social signals embedded in these conversations are non-verbal. During a transition in care, such as those we observed, medical personnel rarely stopped to relay social information verbally, making transcripts of their conversations useless for retrieving social content. Although video recording and coding of non-verbal behaviour such as eye contact is an option, it is more invasive than audio recording. The social signals embedded in the non-linguistic elements of conversation reflect a source of relational information that has received little research attention from trust scholars (Curhan and Pentland, 2007; Pentland, 2004). Thus in this chapter we seek to understand how the social signals embedded in non-linguistic elements of conversations are related to interpersonal trust. Non-linguistic elements of conversations include voiced utterances, which are vowel sounds like /o/, and unvoiced utterances, which are...
You are not authenticated to view the full text of this chapter or article.