ARIES (Acquiring Research Investigative and Evaluative Skills) is a learning environment with conversational agents that hold trialogs with the human learner who is acquiring critical reasoning skills on science. The learner holds conversations with two animated pedagogical agents while solving a number of engaging problems in the social and physical sciences. ARIES is embedded in a game environment with an electronic textbook, multiple choice questions, and the trialogs. Science teachers in both high school and higher institutions could assign ARIES as homework, if they decide not to devote class time to it.

PI (UM): Art Graesser

Funding Agency: Institute of Education Sciences (subcontract from Northern Illinois University; Keith Millis is PI)

Dates: 2007-2010

Amount: $640,000 (Memphis allocation)

Significant Publications:

Graesser, A. C., Jeon, M., & Dufty, D. (2008). Agent technologies designed to facilitate interactive knowledge construction. Discourse Processes, 45, 298-322.

Storey, J. K., Kopp, K. J., Wiemer, K., Chipman, P., & Graesser, A. C. (in press). Using AutoTutor to teach scientific critical thinking skills. Behavior Research Methods.


AutoCommunicator

This project is developing a question answering system that allows faculty, students, and the public to learn about technology transfer and the relevant research projects at the University of Memphis. The system will be one of the Web faculties for the FedEx Institute of Technology site. The user accesses relevant information to their queries by asking a question in natural language and engaging in a brief dialogue with AutoCommunicator until an answer is found. An animated conversational agent is available to guide the dialog.

PIs: Art Graesser and Xiangen Hu

Funding Agency: FedEx Institute of Technology

Dates: 2008-2009

Amount: $80,000


AutoTutor Emotions

AutoTutor simulates human tutorial dialog with an animated conversational agent that helps students learn qualitative physics or computer literary by holding conversations in natural language. This project tracks the emotions and knowledge of the learner by dialogue patterns, speech intonation, facial expressions, and body movements. It integrates advances in discourse processes, education, multimedia, psycholinguistics, computational linguistics, and artificial intelligence. The project investigates strategies, processes, practices, and environments that are likely to assist the learners in interactive knowledge construction, particularly at deeper levels of comprehension and problem solving.

PI: Art Graesser

Co-PIs: Stan Franklin, Rosalind Picard, Robert Reilly, Barry Kort

Funding Agency: National Science Foundation

Dates: 2003-2009

Amount: $1,256,000

Significant Publications:

Craig, S., D'Mello, S., Witherspoon, A., & Graesser, A. (2007). Emote aloud during learning with AutoTutor: Applying the Facial Action Coding System to cognitive-affective states during learning. Cognition and Emotion, 22, 777-788.

D'Mello, S. K., Craig, S. D., Sullins, J., & Graesser, A. C. (2006). Predicting affective states through an emote-aloud procedure from AutoTutor's mixed-initiative dialogue. International Journal of Artificial Intelligence in Education, 16, 3-28.

D'Mello, S. K., Picard, R., & Graesser, A. C. (2007). Toward an affect-sensitive AutoTutor. IEEE Intelligent Systems, 22, 53-61.

Graesser, A. C., Jackson, G. T., & McDaniel, B. (2007). AutoTutor holds conversations with learners that are responsive to their cognitive and emotional states. Educational Technology, 47, 19-22.

D'Mello, S. K., Craig, S. D., Witherspoon, A., McDaniel, B., & Graesser, A. C. (2008). Automatic detection of learner's affect from conversational cues. User Modeling and User-Adapted Interaction, 18, 45-80.

Graesser, A. C., D'Mello, S. K., Craig, S. D., Witherspoon, A., Sullins, J., McDaniel, B., & Gholson, B. (2008). The relationship between affect states and dialogue patterns during interactions with AutoTutor. Journal of Interactive Learning Research, 19, 293-312.


Guru

Guru models the strategies and dialogue of expert human tutors and is a logical progression from AutoTutor, which models novice human tutors. The Guru expert tutor, by using expert human tutor strategies, actions, and dialogue, should promote larger learning gains than previous novice computer tutors. Guru could have a big impact on Memphis City Schools because it is designed to improve educational outcomes on the Tennessee Gateway Science Test, which high school students must pass to receive a diploma.

PI: Andrew Olney

Co-PIs: Art Graesser, Natalie Person, Betsy Williams

Funding Agency: Institute of Education Sciences

Dates: 2008-2011

Amount: $1,858,176

Significant Publications:

Olney, A. M. (2007). Latent semantic grammar induction: Context, projectivity, and prior distributions. Proceedings of TextGraphs-2: Graph-Based Algorithms for Natural Language Processing (pp. 45-52). Rochester, NY: Association for Computational Linguistics.

Graesser, A. C., D'Mello, S. K., & Person, N. K. (2009). Meta-knowledge in tutoring. In D. J. Hacker, J. Dunlosky, & A. C. Graesser (Eds.), Handbook of metacognition in education. Mahwah, NJ: Erlbaum.

Cade, W., Copeland, J., Person, N. K., & D'Mello, S. (2008). Dialogue modes in expert tutoring. In B. P. Woolf, E. Aimeur, R. Nkambou, & S. Lajoie (Eds.), Intelligent tutoring systems: ITS 2008 Proceedings (pp. 470-479). Montreal, Canada: Springer-Verlag.

D'Mello, S. K., Jackson, G. T., Scotty, S., Morgan, B., Chipman, P., White, H., Person, N. K., Kort, B., el Kaliouby, R., Picard, R., & Graesser, A. C. (2008). AutoTutor detects and responds to learners affective and cognitive states. Emotional and Cognitive Issues in ITS 2008 Workshop Proceedings (pp. 31-43). Montreal, Canada: Springer-Verlag.

Lehman, B. A., Matthews, M., D'Mello, S. K., & Person, N. K. (2008). What are you feeling?: Investigating student affective states during expert human tutoring sessions. In B. P. Woolf, E. Aimeur, R. Nkambou, & S. Lajoie (Eds.), Intelligent tutoring systems: ITS 2008 Proceedings (pp. 50-59). Montreal, Canada: Springer-Verlag.


iDRIVE

This project implements vicarious learning strategies wherein learners observe virtual tutoring sessions with conversational agents and multimedia learning environments. The agents ask and answer deep-level questions that facilitate constructive learning in labs and classroom instruction. Exposure to deep-level reasoning questions improves the number and quality of questions asked that are critical to establish interactive knowledge construction. Dialogs with deep-level reasoning questions and also interactive AutoTutor tutoring sessions improved learning over equivalent content presented at a monolog for middle- and high-school aged students.

PI: Barry Gholson

Co-PIs: Art Graesser, Trey Martindale

Funding Agency: Institute of Education Sciences

Dates: 2005-2009

Amount: $1,050,000

Significant Publications:

Gholson, B., Witherspoon, A., Morgan, B., Brittingham, J., Coles, R., Graesser, A. C., Sullins, J., & Craig, S. D. (in press). Exploring the deep-level reasoning questions effect during vicarious learning among eighth to eleventh graders in the domains of computer literacy and Newtonian physics. Instructional Science.

Craig, S. D., Sullins, J., Witherspoon, A., & Gholson, B. (2006). Deep-level reasoning questions effect: The role of dialog and deep-level reasoning questions during vicarious learning. Cognition and Instruction, 24(4), 565-591.

Gholson, B., & Craig, S. D. (2006). Promoting constructive activities that support vicarious learning during computer-based instruction. Educational Psychology Review, 18, 119-139.

Craig, S. D., Driscoll, D., & Gholson, B. (2004). Constructing knowledge from dialog in an intelligent tutoring system: Interactive learning, vicarious learning, and pedagogical agents. Journal of Educational Multimedia and Hypermedia, 13, 163-183.

Driscoll, D., Craig, S. D., Gholson, B., Ventura, M., Hu, X., & Graesser, A. (2003). Vicarious learning: Effects of overhearing dialog and monolog-like discourse in a virtual tutoring session. Journal of Educational Computing Research, 29, 431-450.


Language across Cultures
(Collaborations with the University of Texas at Austin and Cornell University)

This project investigates the language and discourse patterns of English and Arabic texts using computerized text analysis tools. Specifically, the researchers are interested in analyzing discourse patterns in various corpora such as newspapers, speeches, and conversations to elucidate the leadership style, personality, and social status of leaders. In addition to English and Arabic, analyses will be performed on Korean, Chinese, and other languages. We will use computational tools that automatically analyze texts on hundreds of measures of language and text cohesion (using Coh-Metrix), including word characteristics, syntax complexity, lexical diversity, readability, connectives, latent semantic analysis, co-referential cohesion, mental model dimensions, and genre.

PI (UM): Art Graesser

Funding Agency: U.S. Department of Homeland Security Counterintelligence Field Activity (subcontracts from University of Texas at Austin; James Pennebaker is PI)

Dates: 2007-2009

Amount: $167,990 (Memphis allocation)

Significant Publications:

Graesser, A. C., Jeon, M., Yang, Y., & Cai, Z. (2007). Discourse cohesion in text and tutorial dialogue. Information Design Journal, 15, 199-213.


MetaTutor

MetaTutor is a new multi-agent, hypermedia-based intelligent tutoring system that is designed to improve the effectiveness of animated pedagogical agents (APAs) as external regulatory agents in the learning of the circulatory system. A mixed-initiative intelligent tutoring system similar to AutoTutor simulates the discourse patterns and pedagogical strategies of human tutors. The underlying assumption of MetaTutor is that students should regulate key cognitive, metacognitive, motivational, social, and affective processes to learn complex science topics. The design of MetaTutor is based on extensive research by Azevedo and colleagues showing that adaptive human scaffolding that addresses both the content of the domain and the processes of self-regulated learning enhances students' learning of challenging science topics with hypermedia.

PI: Roger Azevedo

Co-PIs: Art Graesser, Vasile Rus, Danielle McNamara

Funding Agency: National Science Foundation

Dates: 2006-2009

Amount: $904,581


Sandia Labs

Working in collaboration with researchers at Sandia National Laboratories and the University of Notre Dame, UM researchers will identify skills that may differentially affect performance of individual humans in cognitive tasks relevant to flying airplanes and communicating with team members. The project will either identify or develop measures to quantify individual ability with respect to each identified skill. A battery of tests will be administered to experimental test participants to assess their relative abilities to predict task performance.

PI: Art Graesser

Funding Agency: Sandia National Laboratories

Amount: $135,000


The Writing Pal

This project develops a new automated intelligent tutoring system that provides interactive and adaptive strategy training that encourages students to use independent writing techniques. The W-Pal will be evaluated with high school students and English teachers from urban and suburban schools in Memphis. The goal is to provide a tool that provides writing strategy instruction via automated technologies, which offer tutoring that mimics human one-on-one tutoring. The W-Pal allows teachers to provide adaptive one-on-one tutoring, not to a few students in the classroom, but to all of the students in the classroom. As such, this research will significantly impact the educational community by providing an automated instructional writing tool that can potentially benefit students across the nation.

PI: Danielle McNamara

Co-PIs: Art Graesser, Phil McCarthy, Loel Kim

Funding Agency: Institute of Education Sciences

Dates: 2008-2011

Amount: $2,015,456

PAST


AutoTutor [PI: Art Graesser]

AutoTutor is an intelligent tutoring system that helps students learn about computer literacy or physics by holding a conversation in natural language. AutoTutor appears as an animated agent that acts as a dialog partner with the learners. The animated agent delivers AutoTutor's dialog moves with synthesized speech, intonation, facial expressions, and gestures.

Coh-Metrix [PI: Danielle McNamara]

Coh-Metrix is a system for computing computational cohesion and coherence metrics for written and spoken texts, using advanced methods that are widely used in computational linguistics. Coh-Metrix allows readers, writers, educators, and researchers to instantly gauge the difficulty of written text for the target audience.

iMAP [PI: Max Louwerse]

The iMAP (Intelligent MapTask Agent) project investigates multimodal communication in humans and agents, focusing on linguistic modalities (prosody and dialog structure) that reflect major communicative events, and nonlinguistic modalities (eye gaze, facial expressions, and gesture).


iSTART [PI: Danielle McNamara]

iSTART (Interactive Strategy Trainer for Active Reading and Thinking) is an automated strategy trainer designed to help students become better readers via multi-media technologies. Pedagogical agents provide students with interactive and adaptive training to use active reading strategies.


Plate Tectonics [PI: Art Graesser]

This project investigates the impact of a Web tutor on helping college students' identify true versus false bodies of knowledge while exploring Web pages to research the causes of the eruption of Mt. St. Helens. The Web tutor (called SEEK, an acronym for Source, Evidence, Explanation, and Knowledge) was designed to improve a critical stance through several facilities in a computer environment: spoken hints on a mock Google search page, on-line ratings on the reliability of particular Web sites, and a structured note-taking facility that prompted them to reflect on the quality of particular Web sites.


Quaid Tool [PI: Art Graesser]

A computer model of human question understanding (called QUEST) helps survey designers identify problems with questions on a Web-based tool called QUAID (Question Understanding Aid). QUAID is a software tool that assists survey methodologists, social scientists, and designers of questionnaires in improving the wording, syntax, and semantics of questions. QUAID is being used by six government agencies.