Developing Expertise Notes

Contents
Role of Deliberate Practice
The Acquisition of Expert Performance
The Acquisition of Medical Expertise
Reasoning and Instruction in Medical Curricula
Changing the Agency for Learning
Field Study in SW Design
Brooks on Great Designers
Conceptualizations of Practice

Role of Deliberate Practice

Prediction
Ability tests are shown to be fairly good predictors of early job performance, but predicts final performance poorly.

Domain specificity
Expert performance tends to be domain specific. Experts performance in their domain of expertise exceeds that of the uninitiated, yet in neutral domains the expert cannot be differentiated from the novice. (Lack of transfer to other domains for both speed and memory)
"The domain-specific nature of experts' superior performance implies that acquired knowledge and skill are important to attainment of expert performance.(p. 365)"

Deliberate Effort
Performance is arrested sometimes at less than optimal levels. To overcome these plateaus of performance requires "effortful reorganization of the skill".

Preparation Time
Data indicates it takes on the order of ten years to become expert in a domain. Yet experience alone is a weak predictor in that it is the nature of the experience (referred to as deliberate practice) that makes the difference. 10-year rule is supported by data from music, mathematics, tennis, swimming, and long-distance running. For music composition, Hayes found an average of about 20 years from the time an individual started to study music until they first composed an outstanding piece.

Improvement
"the maximal level of performance for individuals in a given domain is not attained automatically as a function of extended experience, but the level of performance can be increased even by highly experienced individuals as a result of deliberate efforts to improve.(p. 366)"
Mere repetition does not automatically lead to improvement.

Deliberate Practice

Includes activities that are especially designed to improve the current level of performance. Would allow for repeated experiences in which the individual can attend to the critical aspects of the situation and incrementally improve her performance in response to knowledge of results, feedback or both. Explicit goal is to improve performance.

Framework

Resource Constraint

Access to teachers/tutors, training materials, etc.

Effort Constraint

Practice sessions should be limited. The reasoning for this is to insure that the student has the mental stamina to concentrate on practice. Recall that practice for practice sake does not lead to the types of performance we seek, thus practice sessions should occur when the mental faculties of the student permit the levels of concentration required to focus on the task.

Motivation Constraint

The Acquisition of Expert Performance

Experience is a weak predictor of performance in a domain.
Representation

Experts acquire and preserve highly adapted representations that aid in planning, prediction, and evaluation.

Expert performance show to be acquired over an extended period and its mediating mechanisms were found to reflect major adaptions to domain-specific constraints acquired through extended intense training activities.

Effective Learning

"the most effective learning requires a well-defined task with an appropriate difficulty level for the particular individual, informative feedback, and opportunities for repetition and corrections of errors. (p. 20-21)" When all these items are present they use the phrase deliberate practice.

"If improved performance was a mechanical consequence of engaging in the assigned training tasks, the problem of skill acquisition would be essentially solved. However, to attain changes in behavior, attention is necessary to generate the correct desired action and to thus override the activation of th old habitual response. For effective learning the subjects also need to monitor their processes and performance to determine necessary adjustments and corrections."(p. 33) This leads to issues of self-regulation as the agency for learning discussed in Glaser.

"Effective learning requires attention and monitoring of goals, processes, and performance." (p. 34) "for effective learning the subject must acquire mechanisms supporting reasoning, planning, preduction, and expectation in order to generate feedback and effective error diagnosis with appropriate correction." (p. 39)

The Acquisition of Medical Expertise

Forward Reasoning

A chain of inferences from data toward an incremental refinement of hypotheses resulting in a solution. This ia claimed as one of the hallmarks of expert performance in several domains. Novices and intermediates tend to use backward reasoning (hypothesis to confirmation). Forward reasoning is a function of a highly structured knowlege based and well-developed pattern recognition capabilities. When an expert works outside her domain of expertise or on more complex problems then we notice the use of less efficient strategies. This may support the notion of opportunistic reasoning noted by Guindon in design studies.

Looks at issues related to problem solving teams such as those found in ICU.

As repertory of expertise increases there is a corresponding decrease in the use of "basic" science in decision making.

Definitions:
Cognitive Complexity - Amount of cognitive activity /unit of time.
Cognitive Effort - subjective measure of mental workload

Reasoning and Instruction in Medical Curricula

Interestingly the paper explores role instruction plays in the acquisition of diagnostic behavior. The findings concern the use of basic science knowledge in diagnosis. Subjects taught under the two instructional paradigms show very strong differences in their ability to utilize scientific knowledge in diagnosis. The research suggests that if basic knowledge is needed for a particular application then treating that knowledge base separately prior to its use may be a valid, necessary approach to insure the development of an adequate base. Attempting to integrate the learning of basic science and diagnostic skill simulatenously tended to degrade performance in basic science knowledge. There was evidence of increased use of deductive reasoning in subjects where deductive reasoning was explicitly taught as part of the training paradigm.

Changing the Agency for Learning

The claim is that a fundamental principle underlying the acquisition of competence is the change in agency for learning as expertise develops. Initially learning involves enormous external support (on the part of teachers or coaches). The role these individuals plays is importants to the development of expertise. As expertise and competence increases the individual increasingly internalizes control over the learning situation and the honing of performance. Note that part of the learning must be to improve perforamance/skill and self-regulation.

Glaser's remarks encourage us to pay explicit attention to issues of self-regulation and developing the skills (metacognitive) necessary to support self-regulation in designing instructional activities. Empirical support for training in such skills is available, and encouraging.

Field Study in SW Design

Suggests a layered model to explore factors that affect the psychological, social, and organizational processes involved in software development. This model views the individual, the team, the project, the company, and the busines milieu. At the core of the software development activity is the individual software designer. To understand software development is explore the individual's cognition and motivation. A focus on the team or project requires a look at group dynamics. At the top layers are the company and business with isses of organizational behvarior.

The authors point out three issues:

  1. the effects of tools and methods are small
  2. variability across designers is high
  3. success at design requires more than technical expertise
The first indicates the continuing search for the technological/methodological silver bullet (Brooks, 1987) is not likely to achieve the goal. The second speaks to issues of expertise. Surely the initial backgrounds, knowledge and grades for instance, of the designers would not have predicted such variability (this outcome is as suggested by Ericsson et. al. (1993) that aptitude is not a good predictor of future performance). The third item implies a variety of factors outside of computer science influence competent designs.

Exceptional Designers

Application Domain Knowledge
Those identified as exceptional designers by either management or project staff were more likely to possess more application domain knowledge than others working on the project. Thus, expert designers are not typically narrow technophiles with only high levels of computer related knowledge. They were better able to map between the behavior of the application and the computational structures that implement this behavior.

Roles
Exceptional designers assumed roles other than designer. These individuals were identified as the intellectual "core" of the project (visionary). They provided more insight into the design process. Their understanding of the developers and customers allowed them to integrate perspectives.

Communication
Exceptional designers were skills at communicating their technical vision to other project members.

Project Performance
Exceptional designers became consumed with the performance of the project. In doing so are the source of coordination activities, not formal. This implies that exceptional designers have a better sense of process and the roles the various players have in the project. Furthermore, it implies sufficient understanding in process to support accomplishing project goals.

Brooks on Designers

"The central question in how to improve the software art centers, as it always has, on people."(p. 18)
"Sound methodology can empower and liberate the creative mind; it cannot inflame or inspire the drudge."

How to grow great designers?

  1. Systematically identify top designers as early as possible
  2. Assign a career mentor to be responsible for the development of the prospect and carefully keep a career file
  3. Provide oportunities for growing designers to interact and stimulate each other

New Conceptualizations of Practice

Training has been studied by looking at the nature of manipulations during the skill acquisition period, and accessing how these work in terms of the learner's acquisition of the skill being taught in a number of trials or a fixed period of time. Unfortunately, the measures of learning tend to be short-sighted and look only at skills gains during the acquisition period, not at retention or transfer. The authors argue convincingly that the training regimen should be designed to have a greater impact on posttraining attributes.

Examples

Scheduling of Tasks During Practice

Blocked practice demonstrates a clear advantage over random presentation during acquisition. Random conditions however produced better retention. It could be argued that what is learned from the random practice sessions is very different from the blocked presentation. What the subject may have to develop is the skill in determining how best to approach the task when presented since it is not known a priori.

Similar findings result from experiments that compare fixed versus expanded presentation/test intervals. The fixed interval subjects showed better acquisition performance, but the expanded interval subjects had better posttest retention.

"The condition that produced the best retention performance seemed toohave the characteristics that it provided added "difficulty" for the learner during the acquisition phase." p. 212

Feedback During Skill Acquisition

  1. Frequent feedback comes to be a part of the task, so that performance is disrupted in retention when the feedback is removed or altered.

  2. Frequent feedback blocks the processing of response-produced feedback, leading to less effective error-detection capabilities for use in retention.

  3. Freqent feedback makes performance too variable during pratice, preventing the learning of a stabilized representations of the kind necessary to sustain performance in later retention tests.


See also

References

Brooks, F. P. (April 1987) No Silver Bullet. IEEE Computer, 20. p. 10-19.

Curtis, B., H.Krasner, and N. Iscoe (1988) A Field Study of the Software Design Process for Large Systems. CACM, 31. p. 1269-1287.

Ericsson, K. A., R. Krampe, and C. Tesch-Römer (1993) The Role of Deliberate Practice in the Acquisition of Expert Performance. Psychological Review, 3. p. 363-406.

Ericsson, K. A. (1996) The Acquisition of Expert Performance: An Introduction to Some of the Issues. In K. A. Ericsson (ed.), The Road to Excellence. Mahwah, NJ: Lawrence Erlbaum. p. 1-50.

Glaser, R. (1996) Changing the Agency for Learning: Acquiring Expert Performance. In K. A. Ericsson (ed.), The Road to Excellence. Mahwah, NJ: Lawrence Erlbaum. p. 127-165.

Patel, V. L., G. J. Groen, and G. R. Norman (1993) Reasoning and Instruction in Medical Curricula. Cognition and Instruction, 10. p. 335-378.

Patel, V. L., D. R. Kaufman, S. A. Magder (1996) The Acquisition of Medical Expertise in Complex Dynamic Environments. In K. A. Ericsson (ed.), The Road to Excellence. Mahwah, NJ: Lawrence Erlbaum. p.303-311.

Schmidt, R. A. & R. A. Bjork (1992) New Conceptualizations of Practice: Common Principles in Three Paradigms Suggest New Concepts for Training. Psychological Science, 3(4), pp. 207-217.


Comments should be sent to
Richard Upchurch (rupchurch@umassd.edu)
or
ciswebster@wwwmail.umassd.edu
Computer and Information Science Department
University of Massachusetts Dartmouth
285 Old Westport Rd.
N. Dartmouth, MA 02747-2300
RUpchurch@umassd.edu

This document
Created: December 18, 1996
by RLU
Modified: June 6, 1998