Given that I have spent some time studying expertise from an academic perspective, I figure it would be good to summarize some of the things we know about experts and the process of developing expertise. Keep in mind that academic studies will rarely give you the simple answer – there is none. But what they can reveal is the chain of factors that, if done right, will lead to better results.
Experts don’t process faster, they process better
Classic studies such as de Groot (1965) and Chase & Simon (1973) have shown that experts in chess do not consider more moves as much as they consider better moves and have a better knowledge (memory) of real chessboard configurations.
The effect of domain-specific knowledge is shown in the observation that chess experts do not perform any better at remembering random chessboard configuration but perform significantly better at remembering chessboard configurations from real games. However, the distinction between knowledge and processing ability is not as simple as it seems initially. Chess experts for example remember features relevant to the game better – they have learned to identify the most significant patterns in a chessboard configuration.
That is, their higher performance is a result of a superior processing strategy. Thus, to be an expert, you should improve your processing strategy – it is not about having more raw brainpower but rather having developed better mental models (schema) to process problems. As Detterman & Spry (1988) point out:
“It is highly unlikely that graduate student performance would correlate with IQ, because graduate students have been so intensively selected on the basis of intelligence that other factors become increasingly important. [This] does not mean that intelligence is not important. [C]ompared to that of the general population, graduate students would have a substantially higher IQ. [I]f graduate students were selected randomly from applicants [...] a relation between IQ and graduate school performance would be obtained. Samples that have a high degree of selection must be examined closely when discussing validity.”
In other words, in a select group of people, non-IQ factors are emphasized more because simply being able to do a particular activity entails a level of selection. To be really good, you need to differentiate based on other factors. (This comes up interestingly in executive training, where rather subtle soft skills are emphasized over other things. I suspect this is because the high degree of selection does not leave space for incompetence in less subtle areas.)
The example of two teachers
How does such a superior processing strategy come about? Scardamalia & Bereiter (1993) give the example of two teachers who begin their careers fresh out of college.
In their example they demonstrate the development of the teachers in terms of the skill development strategy the two teachers take. Initially, both teachers experience similar problems and have essentially the same list of difficult problems to solve. For example, maintaining order in the classroom, organizing their teaching material and keeping track of homework.
As they learn while working they address the problems on the top of the list (such as maintaining discipline) and can remove the problem from their lists by developing an effective routine for dealing with (e.g. shout at the top of their lungs – I don’t know, I’m not a teacher) or preventing the issue.
However, the key difference between the two teachers – who both are doing good at this point in their careers – is that while one of the teachers is simply reducing the number of problems without revising the goals of her efforts, the other teacher is setting new goals as she is more able to cope with her initial problems.
Problem reduction versus progressive problem solving
In other words, simply knowing more does not make you an expert. If your learning goal is to get rid of all the problems in your day-to-day work – rather than to both solve those problems and add new, tougher problems on the other end – you are becoming a specialist in dealing with those particular kinds of problems.
The problem-reduction approach taken by the first teacher in the example is a typical example of normal adaptation to an environment. Initially, the problem-reducing teacher is challenged by the environment, but as she learns more she is less and less challenged by daily events in her environment. She is able to get the kids to shut up and sit down, and life gets easier each day.
The problem-reducing approach distinguishes experienced people from less experienced people, but does not result in expertise as much as adaptation. At some point, the first teacher can essentially handle a set of routine problems without a lot of challenge. However, they will never start tackling the heavier issues, such as how to get a student with a learning difficulty to learn better. The teacher who took the progressive problem solving approach by setting new goals continues to advance towards expertise in her profession.
Reinvesting in learning
As Scardamalia & Bereiter (1993) point out, the experts skill development strategy is characterized by reinvestment of effort into the work instead of simply reducing the amount of effort taken by the work. The authors argue that the general characteristics of expert skill development through progressive problem solving are:
- Reinvestment in learning. As the learner needs to spend less effort in achieving the same performance, the reduction in effort is used in learning, thus resulting in continued improvement.
- Seeking out more difficult problems. The learning effort is not simply characterized by a quantitative change in the amount of effort required to solve the problems – it represents a qualitative change in the type of problems and knowledge requited to solve them.
- Tackling more complex representations of recurring problems. Initially, many problems are represented in a simplified form in order to cope with the demands of the environment and task at hand. However, as expert learners develop, the aim to reformulate the problem more specifically, allowing them to deal with the problem with greater flexibility. (For example, rather than coercing compliance from difficult students, the teacher recognizes and begins to address the underlying learning problems.)
The skill development strategy leading to expertise differs from the one leading to non-expert performance in that it requires that the learner not only improves in performance but also that he or she continues to set new goals and new targets, and that the improvements in performance are used in order to address ongoing issues at a greater level of depth. This process is principally internally driven and motivational processes play a major part in enabling expert skill development.
Different kinds of learning goals
The different orientations of expert and nonexpert students can be expressed quite well in terms of goals. Scardamalia & Bereiter (1993) suggest three different types of goals which different kinds of learners use to guide their actions:
- Task accomplishment goals. Task goals are focused on performance on a particular task or activity; learning is incidental.
- Instructional goals. Instructional goals are based on explicit or inferred learning objectives which are based on an external standard but go beyond task accomplishment.
- Knowledge-building goals. Knowledge-building goals are set by the learner him/herself and may include other types of goals but are not limited to them. Knowledge-building goals are usually related to a personal aim to understand a problem or a field of knowledge.
Goals are important because they orient the learner and guide task performance and selection, which lead to action and learning. Ng & Bereiter (1991) studied participants in a programming language learning task and classified their statements based on the criteria above. Knowledge-building goals are progressive and useful, but results from the study show that other factors are important in addition to the goal setting.
Knowledge-building goals were not a direct predictor of performance in the Ng & Bereiter (1991) study. Instead, more important for performance was in the effectiveness in the direction of efforts towards self-set goals – for example by skipping a task which the learner assessed to have little new knowledge to offer.
Do our current educational and instructional systems support the development of better self-regulatory skills?
While some individuals are able to learn the self-regulatory skills and motivational skills needed by themselves, Scardamalia & Bereiter (1993) stress in their conclusion that in order to help learners develop into experts and not nonexpert specialists, our schools and institutions should place more emphasis on supporting knowledge-building – instead of the more traditional task accomplishment based goals exemplified by “educational activities”.
Expertlike learning consists of a specific set of skills which can either be fostered or suppressed in our formal education. In order to help more people to develop into experts, we ought to reconsider the idea of expertise as an end-state and focus more on the process of knowledge acquisition through “expertlike learning”.