Home Articles Reader Opinion Editorial Book Reviews Discussion Writers Guide About TCRecord
transparent 13
Topics
Discussion
Announcements

Two Heads May be Better than One: Learning from Computer Agents in Conversational Trialogues


by Arthur C. Graesser, Carol M. Forsyth & Blair A. Lehman — 2017

Background: Pedagogical agents are computerized talking heads or embodied animated avatars that help students learn by performing actions and holding conversations with the students in natural language. Dialogues occur between a tutor agent and the student in the case of AutoTutor and other intelligent tutoring systems with natural-language conversation. The agents are adaptive to the students’ actions, verbal contributions, and, in some systems, their emotions (such as boredom, confusion, and frustration).

Focus of Study: This paper explores several designs of trialogues (two agents interacting with a human student) that have been productively implemented for particular students, subject matters, and depths of learning. The two agents take on different roles, but often serve as peers and tutors. There are different trialogue designs that address different pedagogical goals for different classes of students. For example, students can (a) observe vicariously two agents interacting, (b) converse with a tutor agent while a peer agent periodically chimes in, or (c) teach a peer agent while a tutor rescues a problematic interaction. In addition, agents can argue with each other over issues and ask what the human student thinks about the argument.

Research Design: Trialogues have been developed for systematic experimental investigations in several studies that measure student impressions, learning gains from pretest to post-test on objective tests, and both cognitive and affective states during learning. The studies compare conditions with different pedagogical principles underlying the trialogues in order to assess the impact of these principles on student impressions, learning, emotions, and other psychological measures. Discourse analyses are performed on the language and actions in the log files in order to assess their impacts on psychological measures.

Recommendations: Tests of these agent-based systems have shown improvements in learning gains and systematic influences on student emotions. In the future, researchers need to conduct more research to empirically evaluate the psychological impact of different trialogue designs on psychological measures. These trialogue designs range from scripted interactions between agents being observed by the student, to the student helping a fellow peer agent, to the student resolving an argument between two agents. The central question is whether the learning experiences and outcomes show improvement over typical human-computer dialogues (i.e., one human and one tutor agent) and conventional pedagogical interventions.



To view the full-text for this article you must be signed-in with the appropropriate membership. Please review your options below:

Sign-in
Email:
Password:
Store a cookie on my computer that will allow me to skip this sign-in in the future.
Send me my password -- I can't remember it
 
Purchase this Article
Purchase Two Heads May be Better than One: Learning from Computer Agents in Conversational Trialogues
Individual-Resource passes allow you to purchase access to resources one resource at a time. There are no recurring fees.
$12
Become a Member
Online Access
With this membership you receive online access to all of TCRecord's content. The introductory rate of $25 is available for a limited time.
$25
Print and Online Access
With this membership you receive the print journal and free online access to all of TCRecord's content.
$210


Cite This Article as: Teachers College Record Volume 119 Number 3, 2017, p. 1-20
http://www.tcrecord.org ID Number: 21774, Date Accessed: 12/14/2017 1:54:52 PM

Purchase Reprint Rights for this article or review
Article Tools

Related Media


Related Articles

Related Discussion
 
Post a Comment | Read All

About the Author
  • Arthur Graesser
    University of Memphis
    E-mail Author
    ARTHUR C. GRAESSER is Distinguished University Professor of Interdisciplinary Research in the Department of Psychology and the Institute of Intelligent Systems at the University of Memphis and is an Honorary Research Fellow in the Oxford University Center for Educational Assessment at the University of Oxford. His primary research interests are in cognitive science, discourse processing, computational linguistics, and the learning sciences. He has developed automated tutoring systems with conversational agents (such as AutoTutor and Operation ARA) and automated text analysis systems (Coh-Metrix, QUAID). Recent publications include “Deeper Learning with Advances in Discourse Science and Technology,” in Policy Insights from Behavioral and Brain Sciences, and “Intelligent Tutoring Systems, Serious Games, and the Generalized Intelligent Framework for Tutoring (GIFT),” in Using Games and Simulation for Teaching and Assessment.
  • Carol Forsyth
    Educational Testing Service
    E-mail Author
    CAROL M. FORSYTH is now Associate Research Scientist in the Cognitive, Accessibility, & Technology Sciences Center at Educational Testing Service. She received her Ph.D. in Cognitive Psychology and Cognitive Science Certificate from the University of Memphis in 2014. Her research interests include intelligent tutoring systems, epistemic games, and discourse processes during natural language conversations for tutoring and assessment. Recent publications include “Operation ARIES! Methods, Mystery and Mixed Models: Discourse Features Predict Affect in a Serious Game,” in Journal of Educational Data Mining, and “Discourse Comprehension,” in the Oxford Handbook of Cognitive Psychology.
  • Blair Lehman
    Educational Testing Service
    E-mail Author
    BLAIR A. LEHMAN is now an Associate Research Scientist in the Cognitive, Accessibility, & Technology Sciences Center at Educational Testing Service. She received her Ph.D. in Cognitive Psychology and Cognitive Science Certificate from the University of Memphis in 2014. Her research interests include human and computer tutoring, emotions during learning, and natural language conversation for tutoring and assessment. Recent publications include “Inducing and Tracking Confusion with Contradictions During Complex Learning,” in International Journal of Artificial Intelligence in Education, and “Confusion Can Be Beneficial for Learning,” in Learning & Instruction.
Member Center
In Print
This Month's Issue

Submit
EMAIL

Twitter

RSS