AI companion robots remind goals, encourage users, and listen

Researchers at Purdue University, led by assistant professor Sooyeon Jeong, are developing AI companion robots designed to interact with humans in socially and emotionally natural ways. These robots aim to support individuals in everyday settings such as classrooms, hospitals, and therapy environments by responding to human behavior, emotions, and communication patterns. Building on prior work with cancer patients, people with aphasia, hospitalized children, and older adults, the team is exploring new applications, including a robot study companion that provides goal reminders, encouragement, and wellness prompts during study sessions. Their findings indicate that different users respond variably to reminders, encouragement, or mere companionship, suggesting future robots will need to adapt their behavior based on individual moods, personalities, and tasks. Another key focus of Jeong’s research is enhancing robots’ conversational abilities by teaching them to exhibit attentive listening through backchannel signals—small verbal and nonverbal cues like nodding or saying “right” that humans use to maintain engagement in conversations. By analyzing human dialogue using