Alexa as an Active Listener: How Backchanneling Can Elicit Self-Disclosure and Promote User Experience

Active listening is a well-known skill applied in human communication to build intimacy and elicit self-disclosure support wide variety of cooperative tasks. When conversational UIs, active from machines can also greater by signaling the users that they are being heard, which have positive outcomes. However, it takes considerable engineering effort training embed skills at scale, given need personalize active-listening cues individual their specific utterances. A more generic solution needed increasing use agents, especially growing number socially isolated individuals. With this mind, we developed an Amazon Alexa provides privacy-preserving pseudo-random backchanneling indicate listening. User study (N = 40) data show improves perceived degree smart speakers. It results emotional disclosure, with participants using words. Perception speakers as listeners positively associated support. Interview corroborate feasibility provide These findings important implications for speaker interaction design several domains work social computing.