Successes and failures in building learning environments to promote deep learning: The value of conversational agents
This chapter describes some attempts to promote deep learning (as opposed to shallow learning) through conversational pedagogical agents. Learning environments with agents have been developed to serve as substitutes for humans who range in expertise from novices to experts. For example, AutoTutor helps students learn by holding a dialogue in natural language with the student, whereas trialogues have two agents interacting with the student in a three-way interaction. Agents can guide the interaction with the learner, instruct the learner what to do, and interact with other agents to model ideal behavior, strategies, reflections, and social interactions. Some agents generate speech, gestures, body movements, and facial expressions in ways similar to people. These agent-based systems have sometimes facilitated deep learning more than conventional learning environments. Agents have shown learning gains on a variety of subject matters and skills, including science, technology, engineering, mathematics, research methods, metacognition, and language comprehension. Learning environments are currently being developed to improve lifelong learning and collaborative problem solving.
Graesser, A., Lippert, A., & Hampton, A. (2017). Successes and failures in building learning environments to promote deep learning: The value of conversational agents. Retrieved from https://digitalcommons.pvamu.edu/psychology-facpubs/60