Institute for Human and Machine Cognition


Using Concept Maps with Technology to Enhance Collaborative Learning in Latin America[1]


Alberto J. Cañas, Kenneth M. Ford, Joseph D. Novak, Patrick Hayes, Thomas R. Reichherzer, Niranjan Suri

Institute for Human and Machine Cognition
Pensacola Fl, 32502



In the early 1990’s, The University of West Florida and IBM Latin America initiated a joint partnership to investigate how to take advantage of telecommunication links between schools. Since the Internet was not readily available in most Latin-American countries at the time, the IBM telecommunications network was used initially to link students at schools throughout the region and with one another. Utilizing the existing international computer network facilities of IBM, local offices provided network access to students in local schools. Project Quorum (Cañas et al, 1995a) was funded by IBM and provided for development of software, teacher training programs, and curriculum materials to support collaborative learning across classrooms and countries.

Project Quorum embraced the idea that meaningful learning (Ausubel, Novak, & Hanesian, 1978) can be encouraged through the use of concept maps (Novak & Gowin, 1984; Novak, 1998) that serve as a vehicle for collaborative learning by students within classrooms, and between schools via the international IBM network facilities. Software was developed to allow elementary and secondary schools students to produce text and to build concept maps on the computer. These concept maps represented children’s ideas on some topic, problem or issue, and the texts and maps were shared through a phone call to the closest IBM office, usually accessible with a local phone call.

As part of the project, a collaborative software system was developed called the Knowledge Soup (Cañas et al 1996), which allows students from distant schools to share claims (propositions) derived from their concept maps regarding any domain of knowledge being studied. Sharing takes place through the Knowledge Soup, a repository of propositions submitted by the students. Propositions in the Soup that are found to be similar to those submitted by the student are displayed on the student’s screen. He or she can use these propositions from other students to enhance his or her concept map. In addition, the student can question or criticize propositions submitted by other students, leading to a peer-review type of environment, where students themselves are responsible for the validity of the propositions in the Soup. In addition, a software agent called the Giant reasons over the propositions in the Soup, coming up with his own suggestions that he presents to the students. Thus for any domain of knowledge, the students, with some help from their teachers, can create over time a large knowledge soup with almost unlimited related, valid propositions. The nature of the project is shown in a concept map in Figure 1.

(click on the image for a larger view)

Figure 1: Concept Map describing the Quorum Project


Enhancement of Meaningful Learning

Meaningful learning requires the learner to choose to relate new knowledge to relevant knowledge he or she already possess in some substantive, non-trivial way. This process involves all three memory systems of our brain, where information is first received in sensory memory, processed in working memory, or short term memory (STM), and finally stored in an organized fashion in long term memory (LTM). Information that is memorized, or learned by rote, is also stored in LTM, but it is not integrated with existing relevant knowledge and does not lead to construction of powerful knowledge structures in LTM. It is easier for most students simply to memorize facts and definitions rather than to integrate these into their knowledge structures. Therefore, students need encouragement to learn meaningfully. The use of concept maps can be a powerful aid to achieve meaningful learning (Novak, 1991; 1993a). Better evaluation practices can also be helpful (Novak, 1998, Chapter 9; Mintzes, Wandersee, & Novak, 2000).

Meaningful learning is also encouraged when students gain understanding of epistemology (Novak, 1993b), i.e., when they recognize that all knowledge is a product of human meaning making, and hence possesses both the power and limitations expressed in human learning. Human constructivism holds that all knowledge consists of concepts and relationships between concepts called statements, claims, or propositions. In the Quorum project, the activities associated with collaborative problem solving, and the recognition of the unique contributions made by other individuals, helps students to acquire an understanding of the way in which knowledge is constructed and its conceptual and propositional nature. The project activities can do much to help students become better at and to understand the process of knowledge creation and its basis in the meaningful learning process (Novak, 1993b).

Creating a Knowledge Soup

The first task is to train teachers and students in the use of the concept mapping and other software. A new version of the software which takes advantage of Internet’s connectivity, and freely available for teachers and students, makes it very easy to guide learners through the process of constructing a concept map. Figure 2 shows an example of a concept map dealing with plants. This map shows 12 concepts and 13 propositions or knowledge claims.

Figure 2: Concept Map about Plants

Next, the student must join a Knowledge Soup, by connecting to a Soup server. The software allows the server to be anywhere on the Internet. A Knowledge Soup on a particular topic would usually be accessed by a group of students who are studying that knowledge domain. They can be in the same classroom, in different classrooms within their school, and/or in different schools anywhere on the Internet. By joining a Knowledge Soup, the learner is able to share his propositions with other students. When the student joins a Knowledge Soup, the system opens a Scratchpad window and displays a list of the propositions it is able to derive from the student’s concept map. These propositions have not yet been shared with other students, nor have they yet been stored in the Soup server. The student can delete propositions from this list, or add propositions if the software has failed to identify a sequence of concepts and links in the map that form a proposition. Figure 3 shows the result of joining a Knowledge Soup with the concept map from Figure 2.

(click on the image for a larger view)

Figure 3: Knowledge Soup Collaboration Software

The student can now decide which of the propositions he or she wants to share with other students. The student selects the proposition from the list in the Scratchpad window (top right in Figure 3), and selects the Share option under the Collaboration menu entry. We refer to this action as “publishing” the proposition. And since the student must be prepared to defend his proposition from criticism or questions from other students, we refer to it as a “claim”. The claim is sent to the Soup Server, added to the Knowledge Soup he joined, and an icon is displayed to the left of the claim on the Scratchpad window. The Soup immediately sends back claims from other students that are similar to the one the student sent, and these are displayed in the window Soup (lower right window in Figure 3). If the student does not share his propositions with other students, the Soup window will display only his or her own claims. The more he or she shares, the more likely that claims from other students will be displayed on his workspace. The student can then analyze other student’s claims, and decide whether he or she should add them to his concept map. As other students contribute similar claims to the Knowledge Soup, these will be added to the student’s Soup window.

Notice that all claims are anonymous. The student does not know who submitted any particular claim, nor will other students know which claims he or she submitted. At any point in time, the student can decide to “unshare” a particular claim, removing it from the Soup. Claims in the Knowledge Soup that were close in meaning to the unshared claim are removed from the Soup window (but not from the Knowledge Soup -- only the contributing student, or the teacher, can remove a claim from the Knowledge Soup).

Peer Review

A student can challenge, criticize, or add a comment to another student’s claim that he or she disagrees with or doesn’t understand, or to which the learner wants to add a personal opinion. When the student clicks on the Attach Discussion Thread option in the Collaboration menu after selecting a claim in the Soup window, he or she can create a Discussion Thread that will be attached to the claim. Then, by selecting the View Discussion Thread menu option, the student can write his comment, criticism or question. The message is sent to the Knowledge Soup Server and an icon will appear to the left of the claim, in the Soup window of the claim’s owner, and of all other students who are able to see this claim (those who have published a claim similar to the one being commented on) (see Figure 3). If the Anonymous option is selected when the Knowledge Soup is created, the message sent does not identify the questioning student or the originating student, thus avoiding the ego assault that comes when wrong answers are identified publicly. All students who are able to see the claim can respond to this message, creating a discussion thread on that particular claim. It is up to the students (with the aid of the teachers, if needed) to agree, if necessary, on the answer to the question. The system itself makes no attempt to identify claims as valid or invalid. Through this collaborative effort, the students work with each other to elaborate, refine, and improve their own knowledge structures. (The teacher can participate directly in the Knowledge Soup discussions without having to construct a concept map.)

The “Knowledge Soup” created by students in one classroom can be shared with the students in other schools, and even students in other countries, by using the Internet. Exchanges between students in different countries allow for opportunities to recognize cultural differences, or differences in fauna, flora, temperatures, etc. This multicultural shared learning serves to enhance further the quantity and quality of the knowledge structures the students build.

One of the problems in teaching and learning in any subject field is that students (and some teachers) possess invalid ideas or misconceptions. There has been much research on the problem of helping students overcome misconceptions as well as to build better understanding of science and mathematics concepts (for access to many research studies on this problem see What this research has shown is that misconceptions are not remediated simply by telling the student the correct ideas. Each student must actively and deliberately seek to reconstruct the relevant knowledge in their LTM in which the misconception is embedded (Mintzes, Wandersee & Novak, 1998 p.332-339). The kind of study made possible with use of concept maps and the Knowledge the Soup software is ideal for helping students remediate their misconceptions and build powerful, valid knowledge structures.

Enhancements to the original version of the software include the addition of the Giant, a software module that proposes claims to the students, based on logical combinations of propositions in the Soup (Cañas et al, 1995; Reichherzer et al, 1998). For example, the Giant might ask, “Do all green plants have flowers?”, and of course the answer is no; ferns, mosses, conifers, and many other green plants don’t have flowers. But it’s up to the student to teach the Giant.

The materials developed through the Quorum project have been utilized in other settings. For example, secondary school students have been using concept maps to plan fieldwork and to summarize their findings in field studies. Figure 4 shows a student on a field trip, working on a transect on the intertidal zone, making observations and collecting materials and data his group will study when they returns to the lab.

Figure 4: Brazilian student from CEDI school making observations, and collecting material and data in a field trip in the Bay of Angra du Rei

Although the Quorum project is now terminated, many of the schools that participated continue to use the system, and improved Internet access will encourage this process. Funds for new software development are no longer available under this project, but other projects are underway at the University of West Florida to develop further methods for using concept mapping as acknowledge representational tool to organize, store, and provide new and more powerful retrieval mechanisms to facilitate learning by students, professionals, and corporate personnel. New and improved versions of the software for concept mapping and collaboration using Knowledge Soups are available for free for not-for-profit educational use at


Ausubel, D. P., J. D. Novak, & H. Hanesian. 1978. Educational Psychology: A Cognitive View (2nd ed.). New York: Holt, Rinehart and Winston. Reprinted, New York: Werbel & Peck, 1986.

Cañas, A. J., K. M. Ford, G. Hill, J. Brennan, R. Carff, N. Suri, J. Coffey. 1995a. Quorum: A Collaborative Network for Latin American Schools, presented at the 12th International Conference on Technology and Education - ICTE '95, in Orlando, FL.

Cañas, A. J., K. M. Ford, J. Brennan, T. Reichherzer, P. Hayes. 1995b. Knowledge Construction and Sharing in Quorum, 7th World Conference on Artificial Intelligence in Education, Washington DC.

Cañas, A. J., K. M. Ford, J. Brennan, T. Reichherzer, P. Hayes, N. Suri. 1996. An Environment for the Construction and Sharing of Knowledge, Proceedings of the Ninth Florida Artificial Intelligence Research Symposium, Key West, FL.

Mintzes, J. J., J. H. Wandersee & J. D. Novak. 2000. Assessing Science Understanding. San Diego, CA: Academic Press.

Mintzes, J.J., J. H. Wandersee & J. D. Novak. 1998. Teaching Science for Understanding. San Diego, CA: Academic Press.

Novak, J D. & D. B Gowin. 1984. Learning How to Learn.New York and Cambridge, UK: Cambridge University Press. Also published in Spanish, Italian, Thai, Chinese, and Japanese; Arabic, Finnish, and Portuguese.

Novak, J. D. 1998. Learning, Creating, and Using Knowledge: Concept Maps as Facilitative tools in Schools and Corporations. Mawah, NJ: Lawrence Erlbaum and Associates.

__________. 1991. Clarify with concept maps. The Science Teacher 58(7):45-49.

__________. 1993a. How do we learn our lesson? The Science Teacher 60(3):51-55.

__________. 1993b. Human constructivism: A unification of psychological and epistemological phenomena in meaning making. International Journal of Personal Construct Psychology 6:167-193.

Reichherzer, T. R., A. J. Cañas, K. M. Ford & P. J. Hayes. 1998. The Giant: A Classroom Collaborator, ITS ’98 Workshop on Pedagogical Agents, San Antonio, TX.



[1] This article has been accepted for publication in Science Teacher.