Spoken Indicators of Trust Across Cultures
Abstract
Researchers in many fields have sought to find a specific signal, or set of signals, in humanbehavior that indicates trust in others or that may be correlated with human judgments oftrustworthy behavior. However most of this research has focused on nonverbal cues of trust andtrustworthiness; what makes one person likely to be trusted by another. Since 2003 our lab hasinvestigated spoken cues to deception, first within American culture and currently across Americanand Chinese cultures. From this data we can derive hypotheses about spoken indicators of humantrust and potential spoken correlates of trustworthiness. We will also investigate how these spokencues correlate with ethnic background, gender, and personality dimensions, using the NEO-FFIpersonality inventory. We will test these hypotheses in two ways: by collecting speech data fromsubjects playing an economic “trust” game and by conducting a Wizard of Oz experiment in whichsubjects believe they are interacting in such a game with a computer system. The synthetic voiceof this system will mimic either the spoken characteristics of trustworthiness that we identify fromour human-human interactions, or a voice diametrically different from these characteristics, or astandard “robotic” voice. The ultimate goals of our work will be to identify the type of speech thatsystems should employ to be trusted by their users and also to be able to identify the characteristicsin human speech that indicate when such trust is indeed present.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Apr 09, 2018
- Source ID
- FA95501810039
Entities
People
- Julia Hirschberg
Organizations
- Air Force Office of Scientific Research
- Trustees of Columbia University in the City of New York
- United States Air Force