Arthur C. Graesser

Is it an Agent, or Just a Program?: A Taxonomy for Autonomous Agents

The advent of software agents gave rise to much discussion of just what such an agent is, and of how they differ from programs in general. Here we propose a formal definition of an autonomous agent which clearly distinguishes a software agent from just any program. We also offer the beginnings of a natural kinds taxonomy of autonomous agents, and discuss possibilities for further classification. Finally, we discuss subagents and multiagent systems.

Coh-Metrix: Analysis of text on cohesion and language

Advances in computational linguistics and discourse processing have made it possible to automate many language- and text-processing mechanisms. We have developed a computer tool called Coh-Metrix, which analyzes texts on over 200 measures of cohesion, language, and readability. Its modules use lexicons, part-of-speech classifiers, syntactic parsers, templates, corpora, latent semantic analysis, and other components that are widely used in computational linguistics. After the user enters an English text, Coh-Metrix returns measures requested by the user. In addition, a facility allows the user to store the results of these analyses in data files (such as Text, Excel, and SPSS). Standard text readability formulas scale texts on difficulty by relying on word length and sentence length, whereas Coh-Metrix is sensitive to cohesion relations, world knowledge, and language and discourse characteristics.

Question Asking During Tutoring

Whereas it is well documented that student question asking is infrequent in classroom environments, there is little research on questioning processes during tutoring. The present study investigated the questions asked in tutoring sessions on research methods (college students) and algebra (7th graders). Student questions were approximately 240 times as frequent in tutoring settings as classroom settings, whereas tutor questions were only slightly more frequent than teacher questions. Questions were classified by (a) degree of specification, (b) content, and (c) question-generation mechanism to analyze their quality. Student achievement was positively correlated with the quality of student questions after students had some experience with tutoring, but the frequency of questions was not correlated with achievement. Students partially self-regulated their learning by identifying knowledge deficits and asking questions to repair them, but they need training to improve these skills. We identified some ways that tutors and teachers might improve their question-asking skills.

AutoTutor: an intelligent tutoring system with mixed-initiative dialogue

AutoTutor simulates a human tutor by holding a conversation with the learner in natural language. The dialogue is augmented by an animated conversational agent and three-dimensional (3-D) interactive simulations in order to enhance the learners engagement and the depth of the learning. Grounded in constructivist learning theories and tutoring research, AutoTutor achieves learning gains of approximately 0.8 sigma (nearly one letter grade), depending on the learning measure and comparison condition. The computational architecture of the system uses the .NET framework and has simplified deployment for classroom trials.

Affect and learning: an exploratory look into the role of affect in learning with AutoTutor

The role that affective states play in learning was investigated from the perspective of a constructivist learning framework. We observed six different affect states (frustration, boredom, flow, confusion, eureka and neutral) that potentially occur during the process of learning introductory computer literacy with AutoTutor, an intelligent tutoring system with tutorial dialogue in natural language. Observational analyses revealed significant relationships between learning and the affective states of boredom, flow and confusion. The positive correlation between confusion and learning is consistent with a model that assumes that cognitive disequilibrium is one precursor to deep learning. The findings that learning correlates negatively with boredom and positively with flow are consistent with predictions from Csikszentmihalyis analysis of flow experiences.

Autotutor: A tutor with dialogue in natural language

AutoTutor is a learning environment that tutors students by holding a conversation in natural language. AutoTutor has been developed for Newtonian qualitative physics and computer literacy. Its design was inspired by explanation-based constructivist theories of learning, intelligent tutoring systems that adaptively respond to student knowledge, and empirical research on dialogue patterns in tutorial discourse. AutoTutor presents challenging problems (formulated as questions) from a curriculum script and then engages in mixed initiative dialogue that guides the student in building an answer. It provides the student with positive, neutral, or negative feedback on the student’s typed responses, pumps the student for more information, prompts the student to fill in missing words, gives hints, fills in missing information with assertions, identifies and corrects erroneous ideas, answers the student’s questions, and summarizes answers. AutoTutor has produced learning gains of approximately .70 sigma for deep levels of comprehension.

Recognition memory for typical and atypical actions in scripted activities: Tests of a script pointer + tag hypothesis

A script pointer+tag hypothesis assumes that the conceptual representation which interrelates actions in scripted activities, for example, going to a restaurant , consists of a “pointer” to a generic script as a whole, plus a “tag” for each action that is not typical of the generic script. The hypothesis predicts, and two experiments confirmed, that (a) memory discrimination is better for atypical actions in a passage than for typical script actions and (b) there is no memory discrimination for very typical actions. The typicality of actions robustly predicted false alarm rates but not hit rates. The results suggested that discriminative accuracy is best explained by properties of a passages representation rather than the amount of cognitive resources allocated at acquisition.

AutoTutor: A simulation of a human tutor

AutoTutor is a computer tutor that simulates the discourse patterns and pedagogical strategies of a typical human tutor. AutoTutor is designed to assist college students in learning the fundamentals of hardware, operating systems, and the Internet in an introductory computer literacy course. Most tutors in school systems are not highly trained in tutoring techniques and have only a modest expertise on the tutoring topic, but they are surprisingly effective in producing learning gains in students. We have dissected the discourse and pedagogical strategies these unskilled tutors exhibit by analyzing approximately 100 hours of naturalistic tutoring sessions. These mechanisms are implemented in AutoTutor. AutoTutor presents questions and problems from a curriculum script, attempts to comprehend learner contributions that are entered by keyboard, formulates dialog moves that are sensitive to the learners contributions (such as short feedback, pumps, prompts, elaborations, corrections, and hints), and delivers the dialog moves with a talking head. AutoTutor has seven modules: a curriculum script, language extraction, speech act classification, latent semantic analysis, topic selection, dialog move generation, and a talking head.

Coh-Metrix: Providing Multilevel Analyses of Text Characteristics

Computer analyses of text characteristics are often used by reading teachers, researchers, and policy makers when selecting texts for students. The authors of this article identify components of language, discourse, and cognition that underlie traditional automated metrics of text difficulty and their new Coh-Metrix system. Coh-Metrix analyzes texts on multiple measures of language and discourse that are aligned with multilevel theoretical frameworks of comprehension. The authors discuss five major factors that account for most of the variance in texts across grade levels and text categories: word concreteness, syntactic simplicity, referential cohesion, causal cohesion, and narrativity. They consider the importance of both quantitative and qualitative characteristics of texts for assigning the right text to the right student at the right time.

Source Evaluation, Comprehension, and Learning in Internet Science Inquiry Tasks

In two experiments, undergraduates’ evaluation and use of multiple Internet sources during a science inquiry task were examined. In Experiment 1, undergraduates had the task of explaining what caused the eruption of Mt. St. Helens using the results of an Internet search. Multiple regression analyses indicated that source evaluation significantly predicted learning outcomes, with more successful learners better able to discriminate scientifically reliable from unreliable information. In Experiment 2, an instructional unit (SEEK) taught undergraduates how to evaluate the reliability of information sources. Undergraduates who used SEEK while working on an inquiry task about the Atkins low-carbohydrate diet displayed greater differentiation in their reliability judgments of information sources than a comparison group. Both groups then participated in the Mt. St. Helens task. Undergraduates in the SEEK conditions demonstrated better learning from the volcano task. The current studies indicate that the evaluation of information sources is critical to successful learning from Internet-based inquiry and amenable to improvement through instruction.