CIRCSIM-Tutor: An Intelligent Tutoring System 
Using Natural Language Dialogue* 
Martha W. Evens 
Ru-Charn Chang S 
Yoon Hee Lee S 
Leem Seop Shim S 
Chong Woo Woo S 
Yuemei Zbang 
Department of CSAM 
Illinois Institute of Technology 
10 W. 31st Street 236-SB 
Chicago, IL 60616 
csevens @minna. iit. edu 
Joel A. Michael 
Allen A. Rovick 
Department of Physiology 
Rush Medical College 
1750 West Harrison 
Chicago, IL 60612 
jmichael@steve.iit.edu 
aar@rpslmc, edu 
Abstract 
CIRCSlM-Tutor version 2, a dialogne-based intelligent 
tutoring system (ITS), is nearly five years old. It 
conducts a conversation with a student to help the 
student learn to solve a class of problems in cardio- 
vascular physiology dealing with the regulation of 
blood pressure. It uses natural language for both input 
and output, and can handle a variety of syntactic 
constructions and lexical items, including sentence 
fragments and misspelled words. 
Using CIRCSIM-Tutor 
One of the important topics which beginning medical 
students must learn is how blood pressure is regulated 
in the human body. When something happens to 
change the blood pressure, such as a change in the 
volume of blood in the body, the body must 
compensate. The negative feedback loop which 
controls this process, known as the baroreeeptor 
reflex, can be a difficult topic for students. 
CIaCSlM-Tutor is based on a qualitative model 
involving seven core physiological parameters. The 
core parameters and the causal relationships between 
them are shown in the concept map in Figure 1. 
Figure 1 also shows the influence of the nervous 
system, which plays an essential role in blood pressure 
*This work was supported by the Cognitive Science 
Program, Office of Naval Research under Grant 
No. N00014-94-1-0338, to Illinois Institute of Technology. 
The content does not reflect the position or policy of the 
government and no official endorsement should be inferred. 
~:Ru-Charn Chang is now at Baxter Laboratories, North 
Chicago, IL. Yoon Hee Lee is currently Director of Training 
and Education, Institute of Defense Analysis, Seoul, Korea. 
Leem Seop Shim is currently at the Department of 
Information Science and Telecommunications, Hanshin 
University, Osan, Korea. Chong Woo Woo is currently Chair 
of the Department of Computer Science, Kookmin 
University, Seoul, Korea. 
regulation. (In the diagram, Baro=baroreceptor 
pressure and NS = nervous system response.) 
Students use CIRCSIM-Tutor to learn to solve 
problems like those taught in their physiology course. 
Students are asked to predict the value of the seven 
parameters at three points in time: the DR or direct 
response stage immediately atter the precipitating 
event, the RR or reflex response stage after the nervous 
system responds, and the SS or new steady state stage. 
When students start CIRCSIM-Tutor, they see the 
main user interface screen illustrated in Figure 2. The 
precipitating event (a broken pacemaker in this case) is 
shown at the top of the screen. One side of the screen 
contains room for students to enter their predictions. 
After the predictions are entered, the dialogue will 
unfold on the other side of the screen. 
Instructional and Discourse Planning 
The main components of CIRCSIM-Tutor v.2 are the 
input understander, the student modeler, the instruc- 
tional planner, the discourse processor, the text 
generator, and the knowledge base/problem solver. 
F 
F 
+ B 
' -\[- • + 
q 
'+ 
Figure 1 
13 
CIRCS\]M-Tutor picks a problem for the student to 
solve and obtains the correct answers from the problem 
solver. It collects the student's predictions and calls the 
instructional planner to conduct a conversation. 
The planner is given the student's predictions, plus a 
student model showing student errors and possible 
misconceptions. CIRCSIM-Tutor v. 2 uses an overlay 
model. For each causal relation in the knowledge base, 
the student model keeps a record of whether the 
student is correct or mistaken about that relationship. 
This record is built when the predictions are first 
entered and updated during the dialogue. 
The planning cycle is implemented with an 
augmented finite state machine. The tutoring plans are 
kept on a stack. State transitions are chosen based on 
the student's current input, whether the student has 
attempted the question before, and domain knowledge. 
For example, the system only generates a hint when the 
student makes a mistake on the first try at a question. 
Otherwise it gives the student the right answer. 
The planner calls the discourse processor with a list 
of discourse actions such as the following: 
1. Elicit the determinants of the erroneous variable 
2. Elicit the currently active determinant 
3. Elicit the relationship between the active deter- 
minant and the erroneous variable 
4. Elicit the correct value 
Alternatives to elicit are to give a declarative expla- 
nation or a hint ("Remember that ..."). 
Parsing, Interpretation and and Spelling Correction 
The input understanding component of CIRCSIM-Tutor 
v. 2 contains a bottom-up chart parser, producing first 
a phrase-structure parse and then a Lexical Functional 
Grammar f-structure. Matching the f-structure 
functional representation of the student's utterance to 
Problem: Pacemaker malfunctions 
DR RR SS Dialogue 
RAP 
CC 
SV 
HR 
CO 
TPR 
MAP 
the logic form of the tutor's question is largely 
performed by ad hoc code. This task is simplified by 
the fact that ten predicates cover most student answers. 
The CIRCSIM-Tutor v. 2 input lexicon is comprised 
of approximately 750 lemmata. Also in the lexicon are 
basic Lexical-Functional Grammar annotations to be 
used by the parser. Spelling correction is an important 
aspect of the input understander, as students frequently 
misspell words, abbreviate creatively, and make word 
boundary errors (two words joined together or a single 
word split in two). Spelling correction is based on a 
three-way match algorithm which slides a small 
window simultaneously across both the unknown input 
word and a candidate word from the lexicon. Trans- 
positions, elisions, substitutions, and similar errrors are 
counted and the most likely candidate is picked. 
Text Generation 
The text generator produces sentences from logic forms 
generated by the planner. A single tutor turn may 
involve several logic forms, such as an acknow- 
ledgment, an explanation expressed as a declarative 
statement, and a question. The logic forms are used to 
fill in abstract templates. Transformations such as 
passivization, insertion of auxiliary verbs and wh- 
movement are performed, and the final sentence is 
linearized with the help of an LFG grammar. 
Example of Generated Dialogue 
Here is an excerpt from a conversation generated from 
the tutoring tutoring tactic illustrated above: 
T: Remember that the Direct Response occurs imme- 
diately and produces changes in the system before 
any reflex is activated. What controls HR? 
S: Nervous system. 
T: Correct. Heart rate is neurally controlled. What is 
the correct value of HR? 
S: Unchanged. 
T: Correct, the value of HR is unchanged ... 
Acknowledgments 
We thank Stefan Brandle, Reva Freedman and Michael 
Glass for continued enhancements to v. 2 as part ol 
their research on v. 3, and for writing this document. 
References 
The CmcslM-Tutor project has published over 5( 
papers. For a complete bibliography, please see ouJ 
Web page http: //www. csam. iit. edu/~ci rcs im. 
Figure 2
