Contact Us
Precision OS icon

Precision OS News


August 8, 2019

The Concept of Deliberate Practice

There is a wide-held belief in our society that experts and masters are the result of innate talent, rather than the result of extended practice and training. When asked to describe what Innate talent is and how it can be measured people have great difficulty and typically refer to a vague combination of intelligence, creativity, and natural physical abilities, and creativity. Everyone agrees that individuals (including adults) can learn and dramatically improve their performance when they start as beginners in a domain of activity. With time in the domain most people reach an acceptable level and improvements eventually stop, but there are large individual differences in the attained level of performance.  Everyone knows golfer and tennis players, who play weekly, yet never get better even after decades of playing and remain at a modest level of achievement.

The belief that the individual differences in attained performance of experienced individuals reflect some upper-bound set by someone’s innate talent is discouraging because it suggests that it cannot be changed by more experience and practice. 

That’s when psychologist Dr. K. Anders Ericsson and his students examined several examples of these differences in achievement in everyday life, from typists to professional golf players and musicians. They specified how one could measure a superior performance in these domains, and then how the superior performance could be reproduced and studied in the laboratory so one could find out how the processes mediating their performance differed from average and low levels of performance. They then interviewed the superior performers and had them keep diaries so they could identify differences in how much time and what type of activities that the superior performers had engaged in order to develop their superior mechanisms. They found that the superior performers had engaged more in particular practice activities that were designed to attain particular goals and with immediate feedback and opportunities for gradual improvement with repetition. They named these activities Purposeful and Deliberate Practice. Specifically, Dr. Ericsson has proposed how this concept and the associated type of training can be applied to the field of medicine and healthcare, arguing that the understanding of the science behind acquisition of superior-performance would have a major impact on medical training and professional development. 

It was the concept of Deliberate Practice that inspired the founder of Precision OS, Dr. Danny Goel to explore surgical training using virtual reality technology, and this inspiration was furthered when Goel and Ericcson met in 2017. When Dr. Goel had the chance to interview the mind behind Deliberate Practice, he couldn’t help but jump on it—nor could he resist sharing it. 

A conversation between Precision OS founder, Dr. Danny Goel and the esteemed researcher Dr. Ericsson.

DG:  Can you give us a little bit of background, Dr. Ericcson, about yourself and how and when you started this path on deliberate practice?


DE:  I was always encouraged by my parents to do something that would make a difference. As a teenager, I wanted to become a nuclear engineer, but after my high school graduation when I was just about to start my studies at the Royal Institute of Technology, I got interested in psychology. So, I decided I was going to try to be a double student, and I would continue my training to be a nuclear physicist but in addition study psychology at the University of Stockholm.

I got interested to study the psychology of thinking, because I thought that the more I would be able to learn about superior thinking, selfishly I would be able to draw on some of that information myself. 


After a couple of years at the university I got completely absorbed  in research on thinking and memory and I ended up doing my dissertation on problem solving where I was asked people to think out loud while they were solving puzzles.

My main question concerned: what can we learn from the verbal reports of thinking that isn’t available from observing their behavior?


DG:  Was this in the US that you started down this path Dr. Ericcson or did you do this somewhere else?


DE:  No, this was all happening while I was still in Stockholm, Sweden. But, a the time of the defense of my dissertation, the University of Stockholm could not find anybody in Scandinavia that had done research on thinking aloud so they invited a professor of psychology at Carnegie-Mellon University, Herbert Simon, who later got the Nobel Prize, to be the officially appointed reviewer. When he came to Stockholm to be at my defense, he asked me if I wanted to come over to Pittsburgh, USA, and work with him for a couple of years as a postdoctoral fellow.


DG:  Outstanding - and what did your research for your dissertation show, Dr. Ericcson, with respect to thinking out loud?


DE:  By analyzing the thoughts verbalized by the participants I was able describe how each person tried to solve the puzzles by breaking the problem into parts (sub-goals) and showed individual differences in how people learn with practice to solve this particular types of puzzles faster and faster and do it in fewer moves.


DG:  And can you describe that in a little bit more detail as far as from a layperson’s understanding to see if people were actually solving the problem?


DE:  That is a very good question, because some researchers argued at the time that what participants say has no relation to their actual problem-solving processes. I addressed this challenge by two approaches. The first one involved showing that the sequence of verbalized sub-goals provided a valid solution to the puzzle presented to that participant.  I was able to translate the sequence of subgoals into instructions to my computer program, which was then able to reproduce the corresponding solution.  If a participant is verbally expressing thoughts of one valid solution and then at the same time is actually solving the problem by another set of processes then participants would need to generate two different solutions (one verbalized and the “real” solution.  To assess that possibility we compared participants’ performance when they thought aloud to participants in a control group who solved the same puzzle silently, we found no difference in the accuracy of performance between the two groups, so the verbalizations while thinking aloud most likely reveal the actual thinking processes and thus can provide insights on how we can improve learning and performance in every-day life..


DG:  So, can you give the audience an example of that Dr. Ericcson?


Thinking aloud during usability testing ended up being a very effective method. When [product designers] are testing out a program or a product, normally they would have [the testers] work with it and then interview them afterwards. What people have found much more informative is to have participants think out loud when they are trying to use the product, and hear their thoughts about what is confusing and frustrating. The designer of the product can then observe the development and resolution of problems that the participants are having - generating much richer information about what needs to be improved and actionable information about how it can best be improved. 


DG:  Why is there an issue with them providing feedback afterwards?


DE:  If you are asking participants about what they recall about their thinking after engaging in testing a product for five, 10, 15, 20 minutes, what you find is that people do not really remember what they were doing and thinking during the entire time period – their subsequent thinking creates interference with their memories of their thoughts during the earlier parts of the activity.  This is particularly problematic when the participants are trying to make something work and fail several times before being successful and in those cases they seem to recall primarily their final attempt, which often led to an acceptable outcome. 


DG:  It is fascinating - if you could tell us a little bit about how this fits into the concept of deliberate practice - a term that you coined - how did that term come about and how does thinking that you are describing out loud feed into the issue or the concept of deliberate practice?


DE:  There is one step in between my dissertation work and the work on deliberate practice. When I worked as post-doc at Carnegie-Mellon University I worked with Professor Bill Chase, who was interested in whether you could increase your short-term memory (STM) capacity. A typical trial for testing the capacity of STM involves reading a series of random digits at one digit per second, such as 4, 7, 1, 0, 2, 3, 9.  After the last digit is presented the participant is instructed to repeat the digits back immediately in exactly correct order from memory. The average performance of adults allows them to reproduce sequences of around seven digits (a local phone number), after lengthy training our participant (a regular college student) was able to improve his memory performance of seven digits to recall perfectly digit sequences with over 80 digits. Based on the students’ verbal reports after each memory trial and designed experiments we were able to show that the right training can fundamentally change the working memory structure and expand working memory capacity for a range of materials relevant in different domains of expertise. 

After research in several domains of expertise we abstracted the critical variables of this type of effective training, and called it deliberate practice. For this training to work best  you have a teacher who is knowledgeable about the effective training methods in a domain. In the first step the teacher examines a particular student’s current level of performance and then identifies what that individual should try to improve next. Then the teacher describes to the student the identified goals for next steps and what kind of training has been successful previously for other students to attain those goals. The teacher describe the clear goal for improvement and check that the student can represent this goal mentally (even though the students cannot yet produce that performance) as well as the best practice methods for attaining the new goal. This information allows the students to engage in the practice activity by themselves. The students will generate an attempt to reach the new goal and will be able monitor their performance so they get  feedback about how close to that goal their generated attempt was so the students can iterate and gradually improve their performance and eventually achieve the goal. This type of practice is very different from just repeating the same behavior over and over so my colleagues and I came up with the term deliberate practice to distinguish it from other types of practice.  


DG:  Interesting. Now, I know that when you and I met a couple of years ago Dr. Ericsson you had a chance to try out virtual reality, so how do you think that deliberate practice could be implemented in virtual reality or a simulation so to speak?

DE: Since then I have been involved in a project studying how to train expert surgeons who will have attain superior patient outcomes. As part of this project I was invited to shadow half a dozen master surgeons and I had a chance to talk to them about their development of their superior surgical performance. I also reviewed the literature on surgical simulation and I found several groups, who are applying deliberate practice principles and have been able to demonstrate that deliberate practice in a simulator improves patient outcomes significantly more than traditional training when you measure outcomes by the number of days that patients have to remain in post-op after surgery.  Some of these groups even have shown the superiority of the deliberate practice with simulators using designed randomized controlled trials where they randomly assign their residents to different groups so one can compare outcomes for residents, who are traditionally trained versus deliberate practice approach with a simulator with mastery training.


DG:  This is very encouraging. You mentioned that several simulators discuss deliberate but they never actually end up doing it. What are the implications of this?

In my first job as a researcher I worked with airplane simulators for the Institute of Aviation Medicine in Sweden during the 1970ies. At that time many instructors viewed simulators as a safer and cheaper device for providing pilots with the experience of flying. However, many studies failed to show clear benefits from hours of this type of training in the simulator. Decades later I was able to find one study where they actually demonstrated real-world benefits from simulator training. This study analyzed the outcomes of emergency situations happening in the air for fighter pilots during actual missions. These researches were able to show that if the pilot had trained in the simulator for the particular emergency situation that actually happened during a mission, they were significantly more successful in dealing with that situation than if they had not trained in the simulator for that emergency situation. The cost of failure in these emergency situations is tremendous if pilots are injured and cost of replacement of an advanced military aircraft is so high that it provides compelling cases for return on investment of simulator training.


The situation is rather similar if you are working on improving a particular laparoscopic procedure on a simulator by making adjustments, then you will see a reduced occurrence of related problems during the subsequent surgeries when you are actually encountering your first patients. Being able to reduce or even eliminate certain types of negative outcomes during surgery by prior training with the simulator is similarly likely to result in the reduction of patient suffering and lengthy recovery with associated loss of income as well as the more direct costs associated with extended stay in hospitals. 

Models like deliberate practice are foundational and important learning techniques, especially for high-consequence skills like surgery. Cutting edge researchers like Dr. Ericcson will continue to be a guide for how Precision OS approaches training module planning and development. The verbalization technique is another potential way for us to explore how surgeons can maximally learn, retain and sharpen their skills.

Share this article
Link has been copied.
Interested in the PrecisionOS™ platform?
Would you like to know more about how we can craft custom experiences for your organization?
Please Contact Us
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.