Page 272 - Psychological Management of Individual Performance

256 enhancing performance through training
(1990) showed that training involving general instructions for a word processor was
less effective than when instructions were tailored to the specific system. However, the
general instructions were transferred more easily to another word-processing system
than the specific instructions. The gap between abstract theories and specific instances
can be bridged to some extent by accompanying rules and principles with examples il-
lustrating their application (Gick & Holyoak, 1983). Multiple and varied examples, and
instructions to compare examples, appear to enhance schema abstraction (Catrambone &
Holyoak, 1989; Elio & Anderson, 1984; Gick & Holyoak, 1987), presumably because
these processes encourage learners to distinguish common structural features that should
be included in the schema from irrelevant surface features that are particular to a specific
example (Holyoak, 1984).
Despite the advantages of training in rules and principles, there may be times when
trainingwithexamplesaloneismorebeneficial.Wheretheunderlyingprinciplesarecom-
plex or non-linear, trainees may find it easier to reason from similar training examples
than on the basis of abstract principles (Neal, Hesketh, & Bamford, 1999). However, care
needs to be taken that the training examples are representative of the range of likely prob-
lems and that an optimum number of examples are presented, as specific instances may
be difficult to recall if too many examples are used in training (DeLosh, Busemeyer, &
McDaniel,1997).Theshortcomingsofrelyingonexemplar-basedtrainingstrategiesmay
be overcome to some extent by following examples with reflective questions designed
to encourage elaboration (Lee & Hutchinson, 1998).
As with other design issues already discussed, the appropriate level of specificity
or abstraction that needs to be acquired should be identified during the training needs
analysis. In most cases, a “moderately abstract” level of representation may represent a
workablecompromisebetweenreasoningatthehighestlevelofabstractionandreasoning
from a specific example (Zeitz, 1997).

INSTRUCTION VERSUS PRACTICE

Another issue to consider when choosing a training method is the degree to which key
concepts are acquired through direct instruction versus active involvement on the part of
the learner. Substantial direct instruction may be required during training if the subject
matter is complex or if trainees have little background knowledge they can bring to bear
on the current task. Direct instruction using rules and examples is one way of giving
trainees a good grasp of the major theoretical issues involved in a particular domain.
However, the importance of incorporating activities that promote “active” learning is
increasingly recognised (Frese, 1995; Ivancic & Hesketh, 1995/1996), particularly in
light of the capacity of these training methods to provide opportunities for transfer
appropriate processing. The ways in which active forms of learning such as behavior
modeling and role playing, computer simulation, and error training can be used in training
to enhance expertise are discussed next.

BEHAVIOR MODELING AND ROLE PLAYING

The popularity of behavior modeling as a training method has been enhanced by recent
research demonstrating its effectiveness in training technical computer skills (Simon &
Werner, 1996), as well as more interpersonal skills such as active listening (May &