AutoTutor is an intelligent tutoring system that holds conversations with the human in natural language. AutoTutor has produced learning gains across multiple domains (e.g., computer literacy, physics, critical thinking). Three main research areas are central to AutoTutor: human-inspired tutoring strategies, pedagogical agents, and technology that supports natural language tutoring. 

AutoTutor incorporates strategies of human tutors that were identified in human tutoring protocols (Graesser et al. 2009, 1995), as well as ideal strategies derived from fundamental learning research (e.g., modeling- scaffolding-fading, learning progressions), with the basic research goal of determining which of the features help learning and student motivation (Graesser 2011; Graesser et al. 2012a, b). Overall, AutoTutor has been very effective as a learning technology. AutoTutor has produced learning gains that are on average about 0.8σ (standard deviation units) above controls who read static instructional materials (e.g., textbooks) for an equivalent amount of time (Graesser 2011; Graesser et al. 2012b). Learning gains are on par with expert human tutors in computer mediated conversation. On a “by- stander Turing test,” AutoTutor was indistinguishable from a human tutor when individual conversational turns were evaluated by third-person bystanders who examined transcripts of human-tutor interactions (Person et al. 2002).