The Development of Sociality in Language and Thought
Talkat Understanding Higher-Level Intelligence from AI, Psychology, and Neuroscience Perspectives · 2024-06-24..06-28
summary
Claim: everyday rapport mechanisms (hedges, turn-taking, pitch/speech-rate alignment) materially change task outcomes. Methods: dense video annotation with ELAN/Praat, Amazon Mechanical Turk 30s ratings, temporal association and SHAP models, fNIRS hyperscanning, and a deployed agent (SARA) used by ~300 World Economic Forum leaders (100+ repeat users). Key results: hedges increase tutee persistence and success; mixed-personality groups show more alignment than homogenous ones; AAVE brainstorming then code-switching boosts science learning, but low achievers sometimes do better with task-only agents. Current LLMs communicate but fail at grounding and short-turn conversationality.
AI-written summary · about
chapters (14)
- 0:00Sociality in conversation
- 1:08Language vs conversation
- 3:37Sociality gaps in childhood
- 8:20Social bond building in children
- 12:14Rapport in teenage interactions
- 20:10Nonverbal social cues
- 23:22Conversational strategies
- 30:04LLM limits in conversation
- 38:31Socially aware AI systems
- 42:10Rapport in education
- 49:43Neural mechanisms of interaction
- 54:23Inter-brain synchrony research
- 55:48Social bonding aids tasks
- 56:51Ethics in human-like tech