gms | German Medical Science

Jahrestagung der Gesellschaft für Medizinische Ausbildung (GMA)

05.10. - 08.10.2011, München

Effective, ineffective and misleading ways of teaching non-analytical anatomy


  • corresponding author presenting/speaker Ralph Nawrotzki - Universität Heidelberg, Institut für Anatomie und Zellbiologie, Heidelberg, Deutschland
  • author Andreas Möltner - Universität Heidelberg, Medizinische Fakultät, Kompetenzzentrum Prüfungen, Heidelberg, Deutschland
  • Joachim Kirsch - Universität Heidelberg, Institut für Anatomie und Zellbiologie, Heidelberg, Deutschland
  • author Jana Jünger - Universität Heidelberg, Medizinische Klinik, Psychosomatische und Allgemeine Klinische Medizin, Heidelberg, Deutschland

Jahrestagung der Gesellschaft für Medizinische Ausbildung (GMA). München, 05.-08.10.2011. Düsseldorf: German Medical Science GMS Publishing House; 2011. Doc11gma193

doi: 10.3205/11gma193, urn:nbn:de:0183-11gma1937

Published: September 26, 2011

© 2011 Nawrotzki et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( You are free: to Share – to copy, distribute and transmit the work, provided the original author and source are credited.



Background: Despite over a century of debates about how anatomy should be learned and taught, data on the efficacy of ways of instruction are in scarce supply. Research clearly indicates that learners’ metacognitive abilities vary and that medical students apply superficial or deep learning as well as individual blends of memorization, understanding and visualization. Moreover, it is undisputed that, as teachers, we must guide our students to learn effectively so that they become experts in given times. But can we really warrant guidance if the efficacies of our teachings are not properly measured?

Research question: Experts often use non-analytical strategies – e.g., through pattern recognition – to solve problems. Here, we asked whether some knowledge in anatomy required non-analytical expertise, which kind of teaching supported its gain and whether this process can be measured.

Design / method: We performed two experiments on first-year medical students who volunteered to take part in a designed MC test, either at the end of the intervention (randomized posttest-only control group design; experiment 1) or in a cross-over design (randomized treatment with wait-list control group; experiment 2). In experiment 1, all participants completed their dissection module of the abdomen, but groups A and B received extra study materials – annotated ultrasound images (68 students) or sketches of ultrasound images (64 students), group C (88 students) received none. In experiment 2, students had either completed a five-day ultrasound class (posttest group, 116 students) or they had not (pretest group, 105 students). The MC test was timed (projections for 20 sec per question), it contained text or image questions (ultrasound, CT/MRT, angiograms, spatial rotations) on the anatomy of the abdomen, it had three possible answers and the option to choose “don’t know”. Participants were also asked to state whether they knew the answer to each MC question directly or after reflection.

Results: In experiment 1, all participants answered all types of questions equally well or equally poorly. Surprisingly, group B students chose “don’t know” significantly less often (p<.0001, ES d=1.26) and claimed to “know the answer directly” more frequently (p .0002, ES d=.66) than the control group. In experiment 2, the posttest group chose correct answers to text, ultrasound and CT/MRT questions more frequently than the pretest group (in all cases p<.0001; ultrasound ES d=1.59; text ES d=.74; CT/MRT ES d=.57). The posttest group also chose “don’t know” far less often and stated more frequently that they “knew the answer directly” (values were similar to those shown for correct answers). Pre- and posttest groups showed no difference when it came to answering angiogram or spatial rotation questions.

Conclusions: The data presented here indicate that (i) a five-day ultrasound class was effective in increasing the students’ knowledge in a specific context (answering questions about the anatomy shown in ultrasound images), and also through transfer to some related areas (text and CT/MRT) but not to others (angiograms). In fact, (ii) both experiments (extra study material and ultrasound class) were ineffective when it came to answering MC questions showing angiograms. (iii) simply distributing extra study material – without giving feedback – mislead students, as indicated by an apparent illusion of knowing (fewer choices of “don’t know” and more frequent choices of “knew the answer directly”). By contrast, (iv) completion of the ultrasound class enabled students to answer MC questions about abdominal anatomy non-analytically rather than by reflection – a sign that they acquired expertise in this particular context. In our view, more studies should measure the efficacy of teaching anatomy, especially of its non-analytical aspects.