Demonstrated by Natalie Person and Peter Weimer-Hastings

AutoTutor is a fully automated computer tutor that assists students in learning about particular topics that are typically covered in a computer literacy course (i.e., hardware, operating systems, and the internet). The current version of AutoTutor was designed to simulate the effective tutoring strategies of normal human tutors. AutoTutor is a two-dimensional, embodied agent that communicates with the student via gestures, facial expressions, and a synthesized voice that has intonation variations. The six major components of AutoTutor are curriculum scripts, language extraction modules, latent semantic analysis (LSA), topic selection rules, dialogue move generation rules, and the embodied agent. Each of these components will be described in the demonstration.

The structure of a tutoring session with AutoTutor is as follows. AutoTutor begins the session by introducing himself to student and describing the topics that will be covered in the session. After the introduction, AutoTutor asks the student a very general question about computers (i.e., ``What are the parts and uses of a computer?''). The student then types a response that is assessed by the latent semantic analysis (LSA) component. In real time, LSA computes values for a number of factors that dictate what dialogue move AutoTutor will supply next. Some of the LSA factors include student ability, topic coverage, good answerness, bad answerness, and student initiative. A set of fuzzy production rules then operate on the LSA values to determine AutoTutor's next dialogue move. AutoTutor's current repertoire of dialogue moves simulate the dialogue moves that are frequently employed by human tutors. AutoTutor currently has 9 dialogue moves; they are:

(1) Positive immediate feedback.  "That's right"  "Yeah"
(2) Neutral immediate feedback.  "Okay" "Uh-huh"
(3) Negative immediate feedback.  "Not quite" "No"
(4) Pumping for more information.  "Uh-huh" "What else"
(5) Prompting for specific information.  "The primary memories of the CPU
are ROM and _____"
(6) Hinting.  "The hard disk can be used for storage" or ``What about the
hard disk?''
(7) Elaborating. ``CD ROM is another storage medium.''
(8) Splicing in correct content after a student error.  This is a
(9) Summarizing.  "So to recap," .

After AutoTutor delivers a dialogue move, the student types a response. This turn-taking exchange continues until AutoTutor is satisfied that the topic has been sufficiently covered. A different set of fuzzy production rules then determine which topic in the curriculum script AutoTutor will cover next.