Lit review section on expertise

This will definitely help with the analysis of data 1!

This section on expertise attempts to summarize some of the general findings from the cognitive psychology field of expertise, which covers diverse fields such as chess, memory performance, typing, and medical analysis. The studies are almost always positivist and quantitative, and I believe they are limited in that they are studying an arguably subjective field, and trying to fit linear and statistical models to the ways experts go about making decisions in comparison to novices. In addition, there seems to be little stated about how the experts are determined, and if so, whether in fact the way experts were determined were actually valid and reliable.

In contrast to the psychological approaches to expertise, Bereiter and Scardamalia (1993) book is drawn upon as an authoritative and relatively recent text about the attainment of expertise, from the field of educational psychology. This text has been referred to by people such as James Gee (2000) and Colin Lankshear et al (1997), so it is arguably a seminal work in the area of expertise. The limitations of both of these approaches are discussed, and the reason for my study from the field of education and sociology is aptly justified.

The texts that are referred to below describe how expertise is acquired, and some have attempted to design models from which they claim how expertise is performed through linear models, which describe decisions that have been made. This, of course, is wrought with limitations, as humans are not linear. However, by presenting the information below, current ideas of the trajectory towards expertise, i.e., stages, time commitment, natural ability, intelligence levels, etc., all help to inform the base of literature of how expertise is gained.

Ericsson and Smith (1991) stated, “the study of expertise seeks to understand and account for what distinguishes outstanding individuals in a domain from less outstanding individuals in that domain, as well as from people in general” (p. 2), and that experts do not only have an extensive knowledge base in comparison to non-experts, but that they are able to more easily and more quickly access that knowledge. They claimed, that depending on the study, attention would be given to where expertise was inherited or acquired, and if it was general expertise or specific expertise, i.e., domain specific.

Expertise can refer to both nature and nurture, i.e., the result of both genes and/or environments (Ceci, Barnett, & Kanaya, 2003). Ceci et al (2003) argued that while some people may have genetic advantages towards expertise in certain fields, they also must have a high level of motivation to take advantage of the “relevant environmental factors” (p. 81) associated with developing the skill in their field of competence. Notwithstanding, Ceci et al (2003) also mentioned that some people with limited natural ability have reached expertise through sheer hard work, therein reinforcing the idea that motivation is a necessary component of expertise. This somewhat relates to Bourdieu’s notion that an agent makes the most of the opportunities before her, and is not just limited to reproduce what she knows (a critique of Bourdieu’s theory which he disputed).

In a book chapter on Eric Ericsson, Ackerman & Beier (2003) discussed Ericsson’s model of ‘deliberate practice’ (c. 1993) where Ericsson claimed that skill was obtained by the accumulation of deliberate practice over a period of time. Through the accumulation of training and experience, skills, performance, and the ability to select correct actions comprise expertise. Ericsson claimed that in this model, “The principal challenge for attaining expert performance is that further improvements require continuously increased challenges that raise the performance beyond its current level” (2003, p. 116). In reference to this model, Ceci et al (2003), stated, “accumulated deliberate practice causes acquired skill and characteristics, which in turn cause performance, and some of these characteristics increase the maximal amounts of possible practice” (p. 83).

This model of deliberate practice is similar to Dreyfus and Dreyfus’ (1986) model of five stages of progress towards expertise described by Bereiter & Scardamalia (1993, p. 17), presented below in Table 1.

Stage Description
1 Novice - where one rather blindly follows limited rules
2 More flexible and situationally sensitive rule use
3 Competence - one applies goal-directed plans and strategies
4 One has accumulated enough experience that one can often recognize what needs to be done and so have less need of planning and problem solving
5 Expertise - decision making becomes unnecessary and one just naturally does the right thing without having to think about it
Table 1 – Dreyfus and Dreyfus (1986): Five Stages towards Expertise

Glaser & Chi (1988, p. xvii - xx) listed some key components of experts and their expertise, which I present in Table 2, to use as a tool for analysis later in the thesis, to discuss and dispute how the experts in this study line up with these descriptors:

FORMATTING HERE IS AWOL
Characteristic Description
Experts Excel Mainly in Their Own Domains (p. xvii) Knowledge in one domain is difficult to transfer to another domain
Experts Perceive Large Meaningful Patterns in Their Domain (p. xvii) Reflects an organization of the knowledge base
Experts are Fast; They Are Faster than Novices at Performing the Skills of Their Domain, and They Quickly Solve Problems with Little Error (p. xviii) Resulting from either many hours of practice, or being able to arrive at a solution without conducting extensive research
Experts Have Superior Short-Term and Long-Term Memory (p. xviii) Experts’ recall seems to exceed limits of short-term memory, because automaticity of many portions of their skills frees up resources for greater storage
Experts See and Represent a Problem in Their Domain at a Deeper (More Principled) Level than Novices; Novices Tend to Represent a Problem at a Superficial Level (p. xviii) Demonstrated by asking experts and novices to sort problems and analyze nature of their groups; both experts and novices have conceptual categories, but experts’ categories are semantically or principle-based, where categories of novices are syntactically or surface-feature oriented
Experts Spend a Great Deal of Time Analyzing a Problem Qualitatively (p. xix) Experts typically build a mental representation from which they can infer relations that can define the situation, and they add constraints to the problem; novices plunge into applying equations and solving for an unknown
Experts Have Strong Self-Monitoring Skills (p. xx) Experts seem to be more aware than novices of when they make errors, why the fail to comprehend, and when they need to check their solutions
Table 2 – Key Characteristics of Experts’ Performances

Glaser and Chi (1988) claimed that research examined knowledge-rich tasks, namely ones that require thousands of hours spent on learning and experiencing within the field. Many of the studies in Chi et al’s (1988) edited book compared the performances of novices and experts. They claimed the results show a strong correlation between the acquisition of knowledge, and the ability to access that knowledge stored in one’s memory quickly (Posner, 1988; E. Johnson, 1988). Posner (1988) also referred to the motivation needed for a person to become an expert.

Eric Johnson (1988) discussed the models of cognitive science designed to emulate the decisions made by experts in relation to two earlier studies he conducted: the first on physicians’ selection of house officers (internships and residences) for teaching hospitals, and the second on the prediction of stock prices by expert security analysts and by novice MBA students. He suggested there were three possible hypotheses as to why the application of cognitive models had not proved to be effective in the contexts he presented:
“Experts are fallible linear models” (p. 212).
“Experts use nonlinear rules” (p. 213).
“Experts attend to different variables than do models” (p. 213).
This demonstrates to me that trying to use/design/apply models to how experts make decisions is impossible and useless, as humans are non-linear, and the subjective nature of what expertise is, does not lend itself to the quantitative analysis of the acquisition and performance of expertise!

2.3.1 Bereiter and Scardamalia

Bereiter and Scardamalia (1993) discussed some of the history of research in expertise and presented basic premises about the nature of expertise including that many cognitive psychologists believe expertise comprises knowledge, and that skill is viewed as a type of knowledge, that it may take 10,000 hours or an arbitrary figure of ten years for a person to become an expert in a field. They argued for a method of expertise named progressive problem solving, which they believed constituted a process of expertise, rather than viewing expertise as a product (Gee, 2000). Whilst they acknowledged that some inherent abilities lack in those wishing to become experts (though natural ability or intelligence is not always an indicator of possible expertise), they argued that in general, people need to become expert in becoming experts (p. 2).

In their research on expertise, Bereiter and Scardamalia (1993) found that in comparing novices and experts, experts would “work harder and do a great deal more thinking” (p. x) than novices. This is in contrast to previous studies expounded by others (e.g. Glaser & Chi, 1988; Posner, 1988; E. Johnson, 1988), which found that experts were able to do more with less effort.

Bereiter and Scardamalia (1993) explained how a person can become an expert through the process of reinvesting in learning through seeking out more difficult problems, problem reduction, which they term progressive problem solving. They argue that through this process, fewer mental resources are needed to accomplish the same results because problems are reduced, so that agents can focus on bigger problems. This process of expertise is notable because it represents what is done above and beyond the normal course of learning, i.e. what novices do. By reducing problems, more energy and focus is set aside for analysing and focusing on bigger problems.
2.3.1.1 Flow

They presented the term flow (Csikszentmihalyi & Csikszentmihalyi, 1988) to “refer to an experience of sustained pleasure that he found to be reported by artists of all kinds, athletes, scientists, mountain climbers, and many others, when they were absorbed in an activity that sounds to us very much like the process of expertise” (Bereiter & Scardamalia, 1993, p. 102).

Bereiter and Scardamalia (1993) described common characteristics of the flow experience, which are listed below:
• Total absorption in the activity
• A feeling of being in control
• A loss of self-consciousness (“which Csikszentmihalyi attributes to all mental resources being invested in the activity, so that none are available for self-reflection” p. 102)
• A loss of the sense of time
• A balance between ability and challenge (so that there is neither too much anxiety and frustration, nor too much boredom)
• Too much repetition means it gets too easy resulting in boredom, so the nature of flow means that the task will be progressive, i.e. increasingly more difficult
• Sustained pleasure, which becomes a motivator for more flow, which could result in the feeling of enjoyable addiction (p. 102 – 103, emphasis mine)
All of the above characteristics will be referred to in the data analysis chapters on the attainment and performance of expertise by the teenagers in this study.

2.3.2 Critique

Does it really take 10 years, or 10,000 hours to become an expert in a field? With the changes in popular culture, and the changes in knowledge (??), is this traditional notion still applicable?

Gee (2000) claimed that at the end of their book, Bereiter and Scardamalia (1993) connect their ideas directly to the new capitalism.

The idea of progressive problem solving sounds very similar to Ericsson’s model of deliberate practice, mentioned previously, where in order to attain expert performance, challenges must be presented which continually raise the expert’s performance above its current level. Lankshear et al (1997) also suggested that the idea of progressive problem solving simply resembled quality improvement.

It's not finished, but it's a start.