Papers
8
Total Citations
55
H-Index
5
About
Sayanti Roy is a human-robot interaction researcher whose work spans robot trust theory, robotic teaching and learning, and social robotics. She is perhaps best known for developing **Deconstructed Trustee Theory** (2021, 14 citations), a groundbreaking framework that disaggregates human-robot trust by separating a robot's physical body from its identity, enabling more nuanced modeling of perceived trustworthiness — validated through a rigorous 210-participant study. A significant thread of Roy's research explores robots as educators. Beginning in 2017, she pioneered semantic labeling approaches to robotic learning from demonstration, and by 2018 had developed reinforcement learning models enabling robots to autonomously train human participants on complex tasks. Her 2019 work on Mutual Reinforcement Learning further advanced this paradigm by positioning both humans and robots as empathetic co-learners providing reciprocal feedback. Roy has also examined robots in safety-critical and socially consequential contexts, including maintaining human situational awareness during space exploration collaboration and investigating robotic persuasion for COVID-19 social distancing compliance. Her collective body of work — spanning over 50 citations — reflects a sustained commitment to making human-robot collaboration safer, more effective, and more deeply understood.
Research Focus
Key Achievements
Top Papers
- 1Deconstructed Trustee Theory14 citations · 2021
- 2I Need Your Help... or Do I?9 citations · 2023
- 3Using Human Reinforcement Learning Models to Improve Robots as Teachers9 citations · 2018
- 4A Reinforcement Learning Model for Robots as Teachers8 citations · 2018
- 5Mutual Reinforcement Learning with Robot Trainers6 citations · 2019
- 6Can Robots Be Used to Encourage Social Distancing?3 citations · 2021
- 7Teaching and Learning using Semantic Labels3 citations · 2017
- 8Semantic structure for robotic teaching and learning3 citations · 2017