Concept Maps - What they are and how to use them
Dr Steve Walsh - Mar 2010
Research on concept mapping has been ongoing for at least 35 years. Concept maps were developed in 1972 by Joseph Novak and colleagues at Cornell University.
- Concept maps are graphical representations of knowledge that are comprised of concepts and the relationships between them
- Concept maps contain content in context
How Are Concept Maps Structured?
- A concept is unit of meaning and is described by a word or phrase. Concepts consist of labels (enclosed in circles or boxes) and relationships (a connecting line) linking two concepts.
- The text on the line (linking words or phrases) describes the relationship between the two concepts.
- A proposition consists of two or more concepts connected using linking words or phrases to form a meaningful statement (semantic unit).
Figure 1 – A proposition, the basic unit of memory
Concept maps were developed in 1972 in the course of Joseph Novak's research program at Cornell University.
Concept maps have a strong psychological and theoretical foundations, based on Ausubel's Assimilatioin Theory (Ausubel, 1968, 2000) and Novak's Theory of Learning. We learn new things by using our current knowledge and, to a greater or lesser degree, integrating new knowledge with what we already know. A concept map is a graphical representation of these relationships between concepts in our cognitive structure.
Research showed that constructing concept maps is useful to represent knowledge and promote understanding.
Characteristics of Concept Maps
- Focus question - clearly specifies the problem or issue the concept map should help to resolve.
- Propositions - a concept map consists of a graphical representation of a set of propositions about a topic.
- Hierarchical structure - the most general concepts are at the top of the map and the more specific, less general concepts are arranged hierarchically below. Because of this, concept maps tend to be read from the top, progressing down towards the bottom. This is not a rule, it could be cyclic as long as there is logical order (e.g. using arrows or numbers). Also there can be more than one root.
- Cross-Links - relationships or links between concepts in different segments or domains of the concept map. Cross-links often represent new insights on the part of the knowledge producer.
Figure 2 – Overview of concept maps
What is a Concept?
A concept is defined as a perceived regularity or pattern in events or objects, or records of events or objects, designated by a label. Novak (1984), based on Ausubel's (1968; 2000) and Toulmin's (1972) work.
Concepts can be object as well as events. Words are use to describe concepts, e.g. "pneumonia", "electrolytes" and "collagen diseases" are examples labels assigned to objects. The regularities in the object determine its category - the way in which all the entities are alike. The label for most concepts is one or more words (and sometimes a symbols such as + or %).
Events such as a "depolarization" or "enzyme activation" can also be concepts. Events include changes in status like occurrences or improvements. For example, "improved quality of nutrition", "lowering blood pressure" and "formation of bone".
It is impossible to characterize any concept without its relation to other concepts.
What Are Linking Words?
The linking words or linking phrases are the set of words used to join concepts to express the relationships between the (usually) two concepts. Depending on the linking words selected, the relationship expressed will be static or dynamic. Picking the appropriate linking words to clearly express the relationship between two concepts is possibly the most difficult task during the construction of concept maps. This is what differentiates concept maps from mind maps.
Linking words are used to join two or more concepts thereby forming propositions. (Italics in the paragraphs below.)
The relations between concepts can be static or dynamic.
Static relationships between concepts help to describe, define, and organize knowledge for a given domain. There are many types of static relationships. They include, among others, relationships of inclusion (e.g., the head is part of the human body), common membership (e.g., squares and triangles are related to each other because they both belong to geometric shapes), intersection (e.g., geometric shapes may be symmetrical), and similarity (e.g., the soldier fought like a lion).
A dynamic relationship describes how the change in one concept affects the other concept. Two types of dynamic relationships are possible (Thagard, 1992); those based on causality (e.g. travel time is an inverse function of speed for a given distance) and those based on correlation/probability (e.g. academic performance in high school is a good predictor of academic performance in university).
What Are Propositions?
The smallest unit of knowledge according to Ausubel's cognitive theory (1963, 1968) are "concepts" and "propositions". They are the building blocks for knowledge in any domain. We can use the analogy that concepts are like the atoms of matter and propositions are like the molecules of matter. Understanding concepts and propositions is therefore a key step in learning about concept maps and how to construct good maps.
Propositions are statements about some object or event (a concept), either naturally occurring or constructed. Propositions contain two or more concepts connected using linking words or phrases to form a meaningful statement. Propositions can be read independently and understood, for this reason, they are sometimes called semantic units, or units of meaning.
Why The Focus Question?
A focus question clearly specifies the problem or issue the concept map should help to resolve. Every concept map responds to a focus question, and a good focus question can lead to a much richer concept map. When learning to construct concept maps, learners tend to deviate from the focus question and build a concept map that may be (somewhat) related to the domain, but which does not answer the question. This is fine in the sense that the map built probably answers another focus question, and so the focus question of the map should be changed to reflect this.
Derbentseva, Safayeni, & Cañas (2004, 2006) report that a more dynamic focus question has an effect on the nature of the propositions generated, but adding a "quantifier" to the root of the concept map that has the greatest impact on encouraging thinking. Although the results of the experiments are preliminary, they report on three methods by which more dynamic thinking can be encouraged: cyclic maps, a dynamic focus question and a quantified root concept.
Unfortunately, we often see that concept mappers fail to define a focus question in advance of building a concept map, or they simply ignore the question as their map construction progresses.
Concept Maps Comapred To Mind Maps
Concept Maps differ from other types of mapping systems, such as Knowledge Maps, Conceptual Graphs, and Mind Maps because of: their grounding in Ausubel’s Assimilation theory of learning, their semantic and syntactical (structural) organization, the nature of concepts that comprise the nodes in a Concept Map, and the unconstrained nature of linking phrases. A standard procedure for Concept Map construction involves defining the topic or focus question, identifying and listing the most important or “general” concepts that are associated with that topic, ordering the concepts from top to bottom in the mapping field, and adding and labeling linking phrases. Once the preliminary Concept Map has been built, cross-links are identified and added, and a review of the map for completeness and correctness is performed.
Table 1 - Mind Maps vs. Concept Maps
How Are Concept Maps Used?
Concept maps have been used in loads of ways. Some of them are illustrated in the figure below. They are easy enough to produce that primary school children can create them, but rich enough functionally that NASA has used them to capture the knowledge of retiring rocket engineers so as not to lose this valuable expertise.
Figure 3 - Some uses of concept maps
Meaningful Or Deep Learning
One of the most fundamental goals in the use of concept maps is to foster meaningful learning. Ausubel made the very important distinction between rote learning and meaningful learning, and stated that meaningful learning requires three conditions:
- The material to be learned must be conceptually clear and presented with language and examples relatable to the learner’s prior knowledge. Concept maps can be helpful to meet this condition, both by identifying general concepts prior to instruction in more specific concepts, and by assisting in the sequencing of learning tasks though progressively more explicit knowledge that can be anchored into developing conceptual frameworks.
- The learner must possess relevant prior knowledge. This condition can be met after age 3 for virtually any domain of subject matter, but it is necessary to be careful and explicit in building concept frameworks if one hopes to present detailed specific knowledge in any field in subsequent lessons. We see, therefore, that conditions (1) and (2) are interrelated and both are important.
- The learner must choose to learn meaningfully. The one condition over which the teacher or mentor has only indirect control is the motivation of students to choose to learn by attempting to incorporate new meanings into their prior knowledge, rather than simply memorizing concept definitions or propositional statements or computational procedures. The creation of Concept Maps supports the incorporation of new meanings into prior knowledge.
When Concept Mapping is compared with other sorts of activities, such as outlining or defining concepts, that also can induce the learner to take a thoughtful, systematic approach to engaging subject matter, the positive benefit of Concept Mapping often diminishes (a finding noted also in the review by Horton, 1993). However, even in these situations, it appears that Concept Mapping is especially good, in comparison to other interventions, for the learning of relationships among concepts.
From several of the studies reviewed, there is indication that Concept Mapping may be particularly beneficial for lower ability learners, partly because it does induce the active, inquiring, orderly approach to learning that is likely a more natural part of the higher ability student’s approach to learning. On the other hand, when learners are not yet facile with constructing Concept Maps, there is some indication that the cognitive load of creating maps from scratch may hinder learning. When students are novice mappers, other “scaffolded” ways of interacting with Concept Maps, for example, filling in the blank content nodes of a Concept Map already containing the labeled relationships of a completed Concept Map, may be beneficial.
Another very powerful use of concept maps is as an evaluation tool, thus encouraging students to use meaningful-mode learning patterns (Novak & Gowin, 1984; Novak, 1998; Mintzes, Wandersee & Novak, 2000). Concept Maps are also effective in identifying both valid and invalid ideas held by students. They can be as effective as more time-consuming clinical interviews for identifying the relevant knowledge a learner possesses before or after instruction (Edwards & Fraser, 1983).
There are numerous methods of scoring concept maps. Scoring systems focused on concepts more closely correlate with traditional testing scores. Using all the elements of the concept map for scoring often produces a different outcome from traditional testing methods and promotes meaningful learning.
I am not going to go into details regarding the reliability and validity of the different scoring systems except to say that the inter-rater reliability of scoring concept maps depends on the scoring system used and that construct validity comparing concept maps to traditional assessments has a correlation of around r=0.5. Concept maps measure something related to traditional exams and should probably be used as a complimentary method of assessment.
Concept maps have been used to construct individual courses or entire curricula. Tradition curriculum design emphasizes content delivery with the teacher as the conduit and the student as recipient. Teaching using concept maps provides a scaffolding for learners, and getting students to construct their own concept maps produces far greater ownership of the course material. They advocate:
"While constructing one’s own knowledge structure is sine qua non for assimilation of new knowledge for a learner, as highlighted in this paper, many times, the beginners in any domain may start with a wrong concept. If the wrong concept is not addressed, all additional learning may have an incorrect foundation and the process of unlearning this misconception may become problematic for the learning facilitator. A way round this is to develop a set of expert skeletal concept maps on the basic issues in the domain, which could provide the right foundations for the learners and they can subsequently build the additional concepts and add these to the expert skeletal maps."
Efficacy studies reveal that when Concept Mapping is used in a course of instruction, it is better that it be an integral, on-going feature of the learning process, not just some isolated “add-on” at the beginning or end. In this regard, Concept Mapping appears to be particularly beneficial when it is used in an on-going way to onsolidate or crystallize educational experiences in the classroom, for example, a lecture, demonstration, or laboratory experience. In this mode, learners experience an educational event and then use Concept Mapping in a reflective way to enhance the learning from the event. There is also indication that learning effects are enhanced when in the course of Concept Mapping learners adopt an active, deep and questioning approach to the subject matter. Such active, self-engaging, transformational interaction with learning material has been suggested to enhance learning in general (e.g., Feltovich, Spiro, & Coulson, 1993) and this appears to carry over to learning with Concept Maps as a tool.
Novak (1998) suggests that using Concept Maps in planning a curriculum or instruction on a specific topic helps to make the instruction “conceptually transparent” to students. When Concept Maps are developed at the course or curriculum level, it is often desirable to organize them. This involves creating a global “macro map” which shows the main topics and their interrelationships, and more detailed “micro maps," which show more specific details for a particular portion of the instructional material. Concept Maps arranged in this way avoid some of the difficulties that are associated with processing large expert maps, or maps that attempt to cover too many focus questions or topics.
An example of an advance organizer for a college-level course has been implemented by Arguea & Cañas (1998) who developed a set of Concept Maps and associated resources that are used in a class on quantitative methods in business. The Concept Maps detail the use of applied statistics at the graduate level. Students use them as an adjunct rather than a replacement to in-class learning and assignments.
Building A Concept Map
There is no simple recipe or set of steps for constructing a concept map, everyone has their own style. Some start by listing a set of concepts, others go directly to placing a root concept and start linking other concepts from it. Here is a suggested sequence:
Figure 4 - Suggested sequence for constructing concept maps
Cañas A. A Summary of Literature Pertaining to the Use of Concept Mapping Techniques and Technologies for Education and Performance Support.
Chatterjea K. Constructivism in curriculum development - a case study at the Singapore maritime academy.
Kinchin I, Cabot L. Using Concept Mapping Principles in Powerpoint.
Moon B, Hoffman R, Shattuck L, et al. Rapid and accurate idea transfer - evaluating concept maps against other formats for the transfer of complex information.