Ownsworth, T., Fleming, J., Tate, R., Beadle, E., Griffin, J., Kendall, M., Schmidt, J., Lane-Brown, A., Chevignard, M., & Shum, D. H. K.
Do people with severe traumatic brain injury benefit from making errors? A randomized controlled trial of error-based and errorless learning.
What the study is about
Individuals who have sustained a traumatic brain injury (TBI) often need to relearn a lot of skills they had previously mastered. There is some disagreement as to the best strategy to relearn these skills, however, one method that has been successful is ‘errorless learning’. Errorless learning refers to teaching the skill without ever allowing for errors to occur. The training session is designed in a way that requires the trainer to prompt with the correct response, rather than allowing the trainee to guess, and therefore risk laying down the memory of the error response. Learning in this method is often very specific and skills do not often generalise to other situations where the skill is needed outside of the training session. Error-based learning on the other hand involves structured feedback on performance (e.g., use of videos), graded prompts, and post-task reflection to teach how to anticipate errors, check for and correct errors, and generate strategies for overcoming those errors. This ‘metacognitive’ method has previously been shown to reduce errors on trained tasks, increase self-regulation and self-awareness, however, it was previously not known if this approach would promote greater generalisation of skills than errorless learning. This study aimed to determine this.
What we did
Fifty-four individuals with a severe traumatic brain injury were randomised either to the errorless learning group OR to the error-based learning group. They received 8 x 1.5 hour individual training sessions focused on meal preparation. The success of training was measured by total errors made during the Cooking Task (a standardised test of error self-regulation), as well as measure of broader generalisation (Zoo map test) and a number of other secondary outcome measures.
What we have found
After accounting for initial performance and level of pre-injury education, individuals in the error-based learning group demonstrated significantly fewer errors on the Cooking Task following training when compared to those in the errorless learning group. Those who received the error-based training strategies also demonstrated greater levels of self-awareness and behavioural competency at completion. Forty-one participants were re-tested 6 months following completion of training. There were no significant differences in social and vocational outcomes at this point. Results from this study demonstrate that having structured opportunities to make errors and learn to correct these plays an important role in the learning of skills during rehabilitation after severe traumatic brain injury.
Acknowledgements
The authors would like to thank the willing volunteers who participated in this research. The research would also not have been possible without funding from the National Health and Medical Research Council.