Coherence and Structural Knowledge

L. E. Olson

That is where the danger lies, in the ever crowded curriculum of the four years course. The ???.. student of ?? finds himself ground between the upper and the nether millstone. The later is the laboratory which in increasing number and variety is constantly tending to consume more and more of the student?s energy and time. The former is the specialist who in the enthusiasm of rapidly accumulating data and the elaboration of technic is liable to distort somewhat the symmetrical balance of the curriculum and demand each year for his particular department more and more hours of the already overworked student.1

I can't recall how I came on this article, but it struck a chord. Certainly one of the challenges in any professional curriculum is coverage and coherence. Teaching in the basic science curriculum requires deciding what nuggets of knowledge should be distilled from an entire discipline and transmitted in a way that adds value to the professional curriculum. One of the take-home messages from my foray into the nature of clinical expertise is that clinical experts re-arrange their knowledge so that it no longer represents the form in which it was acquired. As I continued to delve into the literature of expert knowledge, I learned of several techniques that have been used to assess the manner in which experts structure knowledge in their domain of expertise quantitatively. The results are then displayed graphically as concept maps.

I reviewed a series of papers2,3,4,5 that used these methods to assess how faculty, medical students, and veterinary students structure knowledge in pulmonary physiology. The studies were undertaken based on the assumptions that expertise requires being able to organize, simplify and retain a large body of knowledge and that coherence increases as individuals go through the knowledge structuring process. Meaningful learning is then defined as the ability to structure knowledge in a way that facilitates recall and use, rather than the ability to accumulate volumes of information.

The authors started by identifying 12 or 13 concepts in pulmonary physiology that students should master, such as alveolar ventilation, compliance, dead space, diffusion, gas exchange, hydrostatic pressure, pleural pressure, lung perfusion etc. Respondents (internists, anesthesiologists, physiologists, students) were then presented with the 66 or 78 resulting pairs of concepts and asked to rate their degree of similarity or difference on a numerical scale ranging from "unrelated" to "highly related". Concept maps were created detailing how respondents related the concepts. Significant findings were that:

  • expert groups differed in how they structured the concepts;
  • student concept maps became more like faculty concept maps following classroom instruction in pulmonary physiology; and
  • unlike studies in other subject areas, there was no correlation between the similarity of student-faculty concept maps and performance on multiple choice exams, a finding that disappointed the authors and me.

Proponents of this type of analysis speculate that information gained by comparing how experts and novices structure knowledge could be used to design specific educational interventions to facilitate meaningful learning6. I was most intrigued to learn that experts structure knowledge differently (and by the way, the anesthesiologists were the most unique in this sample). I believe that this finding has significance for peer review of teaching, suggesting that the organization and presentation order of concepts is likely to be a bit idiosyncratic. Individuals reviewing the course materials as a component of peer review of teaching should be aware of this finding and adjust their responses accordingly.

And the opening quote? A response to Dr. William Osler?s proposal to incorporate "radical reform in the methods of teaching senior medical students" ? written in 1903. The more things change???..


  1. Thompson, W. G. The Instruction of Senior Students In Medicine. The Medical News. 82:53-55, 1903.
  2. McGaghie, W. C., R. Boerger, D. McCrimmon, and M. M. Ravitch. Agreement among Medical Experts about the Structure of Concepts in Pulmonary Physiology. Acad. Med. 69:S78-S80, 1004.
  3. McGaghie, W. C., R. L. Boerger, D. R. McCrimmon, and M. M. Ravitch. Learning Pulmonary Physiology: Comparison of Student and Faculty Knowledge Structures. Acad. Med. 71:S13-S15, 1996.
  4. McGaghie, W. C., D. R. McCrimmon, J. A. Thompson, M. M. Ravitch, and G. Mitchell. Medical and Veterinary Students? Structural Knowledge of Pulmonary Physiology Concepts. Acad. Med. 75:362-368, 2000.
  5. McGaghie, W. C., D. R. McCrimmon, G. Mitchell, J. A. Thompson, and M. M. Ravitch. Quantitative Concept Mapping in Pulmonary Physiology: Comparison of Student and Faculty Knowledge Structures. Advances in Physiology Education. 23:72-81, 2000.
  6. Goldsmith, T. E., P. J. Johnson, and W. H. Acton. Assessing Structural Knowledge. J. Educational Psych. 83:88-96, 1991.