#KeyLIMEpodcast 137: Can many wrongs eventually make a right?

As the old adage goes- does ‘practice make perfect’?  Or the not so old saying coming from mastery learning – does ‘perfect practice makes perfect’? The authors of Linda’s selected paper – a report on a randomized education trial – want to convince us that ‘imperfect practice makes perfect’.

What are your thoughts? And do they succeed in convincing the KeyLIME hosts? Check out the podcast here (or on iTunes!) to find out!


KeyLIME Session 137 – Article under review:

Listen to the podcast

View/download the abstract here.


Dyre L , Tabor A, Ringstead C, Toldsgaard, MG. Imperfect practice makes perfect: error management training improves transfer of learning.  Medical Education. 2017 Feb;51(2):196-206.

Reviewer: Linda Snell (@LindaSMedEd)



In simulation training we see that ‘that error is often perceived as a negative event that we train our learners to avoid’. However, in studies outside medicine, in psychology, there is evidence that errors are necessary in early stages of learning and that it may be essential to elicit errors in order to avoid them during subsequent practice. A meta-analysis from psychology showed that error management training (EMT) resulted in better learning outcomes than traditional error avoidance training (EAT). This is particularly important in for transfer of learning to new tasks and contexts.

EMT focuses on using errors for learning rather than on learning to avoid errors as is practised in EAT.

Characteristics of EMT:

  1. provision of minimal guidance to support learners’ own reflection on action and on the errors, which forces learners to be more actively engaged, explore solutions on their own but may cause them to make more errors than they would if given more detailed, intensive supervision.
  2. error management instructions, encouraging making errors and framing these as positive events, used for feedback; training is focused on the recognition of and reflection on errors in order to prevent them from reoccurring.

Concern: instructing learners to deliberately commit errors may lead to repeat mistakes if they are not aware of the occurrence of an error, however authors say ‘successful learning from errors occurs when learners are instructed to anticipate and appreciate errors in order to ensure that these are framed as desirable rather than as undesirable difficulties, and that sufficient support is provided when errors become obstructive rather than constructive for learning.’


Goal of study: ‘to explore the effects of EMT instructions compared with EAT instructions during simulation-based ultrasound training on the transfer of skills to real patients in the clinical setting.’

Type of Paper

Research – randomized education trial

Key Points on Methods

60 5&6th year med students randomized to EAT vs EMT 3 hour simulation training in ultrasound estimation of fetal weight. Only difference was instructions quoted in text.

Pre-and post-intervention OSAUS – scale for assessing ultrasound performance with trained, blinded raters [secondary outcomes: performance scores and diagnostic accuracy]; and a transfer test on real patients [primary outcomes: performance scores and diagnostic accuracy.]

T-test, ANOVA and sample size calculation done.

Key Outcomes

Similar performances immediately after training, but significantly improved performances of EMT group during the transfer test in the clinical setting, and modest effect on diagnostic accuracy.

Key Conclusions

The authors conclude ‘active exploration and positive framing of errors, which yield a deeper understanding of the task at hand.10 These two components may have served as preparation for future learning.’ Learning from errors can occur as long as errors are expected, noticed and handled correctly. Further exploration needed.

Spare Keys – other take home points for clinician educator

Paper titles are important – make them ‘catchy’ as well as accurate

RCTs in med ed can be done!

Access KeyLIME podcast archives here

Check us out on iTunes