Daniel BuenoDiagrams to summarize a text, summary tables, and development of outlines before writing a thesis, organizational charts, figures illustrating a news article—graphical organization methods are everywhere, demonstrating the usefulness of visual representations for understanding the processes that they link. These examples, however, have their limits, and one sure way to overcome them is by using concept maps, which are being studied as a form of knowledge management by Professor Paulo Correia, of the School of Arts, Sciences and Humanities (EACH), University of São Paulo (USP). Since 2005, he has led the Concept Map Research Group, which brings together undergraduate and graduate students on the Eastern USP campus and from other university departments.
What makes concept maps different from intuitive alternatives to showing connections between different pieces of information is the organization of concepts in a “propositional network.” This means that two concepts represented separately in a diagram need to have their relationship expressed semantically, and not just visually (an arrow or a link). A beautiful song by Dorival Caymmi, O Vento (The Wind), provides a good example. In the verses the composer unites concept words: wind and sail, sail and boat, boat and people, people and fish, fish and money. Seen in this way, the most we can do is make assumptions about the relationships between them. The actual verses, however, reveal the propositions: “Wind on the sail / sail that moves the boat / boat that carries people / people that carry fish / fish that brings in money.”
With these connections, the lyrics are expressed “clearly and precisely” like in concept maps. According to Correia, any information or conceptual knowledge can be represented in the form of concept maps. This type of knowledge is useful in education, research and the corporate environment. As a teaching tool for representing knowledge, one of the virtues of concept maps is that they work in two ways: both to explain course content and to assess what the student has learned. “When we ask a student, someone who is not an expert in the subject, to draw a concept map, we can detect conceptual questions and errors, which makes it easier for the professor to help the student understand the concept correctly,” says Correia.
Concept maps were created in response to the idea of ”meaningful learning,” formulated by the American psychologist David Ausubel as a process by which new information is related to the student’s prior knowledge. It was another American educator, Joseph Novak, at Cornell University, who developed the idea of concept maps in the 1970s. “Since Novak was a biologist, the use of concept maps has, by tradition, developed more in the natural sciences, but they can be used in any field of knowledge,” says Correia, who majored in chemistry, and studied chemistry through his post-doctoral research before devoting himself to concept maps due to his interest in science education. The use of maps fit perfectly into his work at EACH, whose pedagogical proposal favors an interdisciplinary approach to teaching and research.
Two years ago, searching for “new pedagogical architectures,” undergraduate and graduate students in Correia’s group developed a study that, using concept maps, unified apparently distant realms of knowledge, revealing propositional interfaces between the syllabuses of two of the introductory courses at EACH: natural sciences and psychology. The discovery of latent content present in the syllabuses of the two courses led the professors to work together to develop teaching techniques and common, interdisciplinary study material.
In another study, 55 EACH students were asked to develop collaborative maps in various fields of study that have become a “knowledge model.” Each student contributed information on a specific subject, with hypertext links on the Internet, and they then created combinations of them in work group meetings and, later, through concept maps. Once again, the experience revealed a series of topics and connections that were not clear in the individual bodies of knowledge.
Daniel BuenoHistorically, the use of concept maps expanded between the 1970s and the early 1990s, when they were still being drawn on paper, because of their recognized educational potential. Their possibilities were maximized in the last decade of the twentieth century with the use of the Internet to develop concept maps collaboratively. This was also when the program CmapTools appeared, and is still being developed by the Institute for Human and Machine Cognition, of the University of West Florida. This program allows construction, navigation, sharing and commenting on knowledge models represented by concept maps. CmapTools is free and available at cmap.ihmc.us/download. A new impetus came in 2004, when the 1st International Conference on Concept Mapping was held, an event that is now biennial. “Until then there were no regular meetings of the scientific community to disseminate this technique for representation and modeling of knowledge,” says Correia.
The introduction of the Internet has enabled concept maps to include hypertext, with links to information available in other virtual spaces, outside CmapTools. “All this has opened up a new world for the organization of knowledge, with the possibility of cooperative, synchronous and remote sharing,” says Correia. Digital media also makes refinement of the maps easier. “Redoing maps is important, because we never stop learning. There is no definitive map. Indeed, before the Internet, we used Post-it Notes for the most complex cases, to be able to make changes more quickly,” says Correia. The problem is that it became a bit too easy to create what looked like concept maps. “CmapTools is overly user-friendly, and sometimes leads to the construction of naive and purely recreational maps. The tool itself is not synonymous with a good map, just like Word is not synonymous with a good text.”
To draw up good maps, the Concept Map Research Group developed a list of four conditions. The first is the proposition, discussed above: two concepts are not enough. What binds them, their conceptual relationship, must be included. The second is the definition of a focal question. “You cannot map everything. If a theme is not defined, there will be dispersion into areas adjacent to the main focus. You need to decompose a broad subject into several questions that will correspond to various maps,” says Correia. The third item is the hierarchy of knowledge: starting with general concepts and moving to progressively more detailed ones. “It’s the way we articulate information, according to the theory of meaningful learning.” Finally, we must maintain the principle of recursive revision—knowledge changes all the time, and the maps need to be updated constantly. “This ruptures a paradigm typically found in a classroom, that of definitive knowledge, something that does not exist.”
Daniel BuenoOne of the qualities of concept maps is that they can be read according to the reader’s interests, enriching the connection potential. “Despite this advantage, there is a risk of generating an additional cognitive load: the reader is forced to decide the sequence in which to read the concept map, something that is unnecessary when we read a text,” says Correia. Concept maps with lots of information can generate what they call map shock, a confused or erroneous reading of a map, which has been the subject of studies by the group led by Correia. “If you have mastered the subject, the reading order is unimportant. But if not, the decisions made when ordering sequences on the map could cause difficulty and misunderstanding.
Correia’s interest in map shock led him to purchase, with FAPESP resources, equipment to track the eye, in order to identify how a reader interacts with a concept map while reading it. Heat marks identify the direction of the reader’s gaze and the time he spends on each point on the map—the warmer colors indicate greater time was spent. The device works quickly, and records the movement of the eye much more precisely than a digital camera. “The technique allows you to capture the nuances of what we want to communicate in a concept map,” says Correia.
The map makes sense to the person who developed it, but not always to the person reading it. This is the problem that map shock reveals. There are very elaborate maps, covering a large amount of knowledge, but for most readers “they exceed their capacity to understand so much content at once,” making them useless. The study of the reading measurements allows us to understand that, in some cases, the relationships seen by a specialist may confuse a beginner, showing the expert which items should be reviewed or improved. This is one of the main issues that is likely to attract attention at the 6th International Conference on Concept Mapping, to be held September 23–25, 2014 in Santos (SP), led by Correia, and whose honorary president will be Joseph Novak himself.
Evaluation of the disorientation caused by concept maps used as organizers of instructional materials (nº 2012/22693-5); Grant Mechanism Regular research project; Investigator Paulo Rogério Miranda Correia; Investment R$ 86,658.64 (FAPESP).