Learning expert performance

In general, experts possess vast amount of knowledge and long experience in a particular domain. Once the concept of expertise is teased out the nuances become more problematic as experience and knowledge alone are not enough to demarcate who is an expert and who is not. It goes without question that experts possess more knowledge and perform better than novices, but that is also true of anyone who has spent enough time in any field but are not yet experts. The expert performance approach to expertise according to Anders Ericsson "avoids the problem of using questionable criteria that is based on professional experience and peer nomination to identify reproducibly superior performance."¹ In a recent book authored by Ericsson and Robert Pool advise that when "trying to identify the best performers in an area that lacks rules-based, head-to-head competition or clear, objective measures of performance (such as scores or times), keep this one thing at the front of your mind: subjective judgments are inherently vulnerable to all sorts of biases."² This in turn has major implications for criteria selection in different schools of expertise in decision making. While heuristics and bias decision making rely on objective models and quantitative performance, most naturalistic decision making rely on peer judgment when deciding who is an expert.³

What separates experts from novices is "the quality and quantity of their mental representations" which "allow them to make faster, more accurate decisions and respond more quickly and effectively in a given situation."² Mental representations are acquired via deliberate practice since experience, daily work, and generic practice are insufficient for expertise development.² What mental representations are, their content, and how they function continue to be debated among many professions,⁴ but the fact they are not directly observable is indisputable. The search of an objective criteria is further complicated by the fact that learning, like mental representations, is not directly observable and must also be inferred from performance.⁵

Elizabeth and Robert Bjork have explored the unreliability of performance quality as a surrogate of learning and how easily performance can mislead not only learners, but also instructors.^1,6 They call desirable difficulties those activities which induce forgetting while simultaneously increase learning and reduce performance. Variability in performance, therefore, is present not only between practitioners, but also in the same practitioner over time. Daniel Kahneman reported the difficulty in detecting variability when instructors believed their feedback influenced the students' performance quality.⁷ This statistical concept, called regression to the mean, and Kahneman's example are explained by Derek Muller in the following video.

The concept of mental representation is essential in the understanding and application of Ericsson's expert performance approach. This rules out any framework which do not support a cognitive science approach to expertise development, for example Taylorism and behaviorism. Understanding error in performance then, becomes an important factor in expertise development and maintenance if deliberate practice and desirable difficulties are taken seriously. Performance under deliberate practice and desirable difficulties is so difficult that errors become inevitable. The notion that experts do not commit errors is a myth according to Vimla Patel, instead what happens is that experts detect and recover more quickly than non-experts from errors.⁸

Understanding that not everyone with experience in a particular domain is an expert may be difficult for a field that relies heavily on time as a marker of expertise. How counterintuitive and the cognitive effort it takes to not only become an expert, but to maintain a high level of performance throughout a life time is quite a challenge to say the least. It is a daily, concentration hungry, lonely, highly specific, non-enjoyable endeavor. We must rely on imperfect inferences to assess the underlying process of our performance, even more so when observing others. In order to diminish errors the best we can hope for is constructive criticism that takes into account the faulty tools we are destined to use.

Tweet to @jvrbntz

References:

1. Ericsson, A., Development of professional expertise: toward measurement of expert performance and design of optimal learning environments., 2009
2. Ericsson, A., and Pool, R., Peak: Secrets from the New Science of Expertise., 2016
3. Kahneman, D., & Klein, G. (2009). Conditions for intuitive expertise: A failure to disagree. American Psychologist, 64(6), 515-526.
4. Pitt, David, "Mental Representation", The Stanford Encyclopedia of Philosophy (Spring 2017 Edition), Edward N. Zalta (ed.), URL = .
5. Soderstrom, N., Bjork, R., Learning Versus Performance: An Integrative Review., Perspect Psychol Sci., 2015 Mar;10(2):176-99.
6. Applying Cognitive Psychology to Enhance Educational Practice., Bjork learning and forgetting lab, UCLA., Accessed October 30, 2017
7. Kahneman, D., Thinking, fast and slow., 2011
8. Patel, V.L., Cohen, T., Murarka, T., Olsen, J., Kagita, S., Myneni, S., Ghaemmaghami, V. (2011). Recovery at the Edge of Error: Debunking the Myth of the Infallible Expert. Journal of Biomedical Informatics . Jun 44(3):413-24.