Training Design and Training Delivery
Research on training design and delivery can be categorized into two general themes: research on new approaches to engage learners in mean- ingful learning processes and research on spe- cific training methods. Several studies in each of these two domains provide information on how to maximize the benefits of training.
Linou & Kontogiannis (2004) compared im- mediate recall and follow-up retention levels (after six weeks) in four groups. Trainees were production engineering undergraduates. The primary objective of training was to help par- ticipants develop diagnostic strategies to iden- tify symptoms and problems given a set of fault scenarios. One group received systemic training (focusing on structural, functional, and physical relationships among subsystems), two groups received either low-level or high-level diagnos- tic information, and one group received general training on theories related to manufacturing plants. The theory group and both diagnostic groups performed better on the immediate re- call measures, whereas the systemic group per- formed better on the retention measure, pre- sumably because group members built a more stable organization (mental model) of the train- ing content. Similarly, Holladay & Quin ̃ones (2003) showed that adding variability to prac- tice trials resulted in better long-term reten- tion, presumably because trainees had to exert greater effort during skill acquisition.
Researchers continued to explore error training as a strategy for increasing perfor- mance and maintaining performance under changing environmental demands. In contrast to traditional training design approaches that focus on teaching correct methods (and avoid- ing errors), error management training encour- ages trainees to make errors and engage in reflection to understand the causes of errors and strategies to avoid making them in the fu- ture. Heimbeck et al. (2003) implemented error training using a sample of undergraduate stu- dents. The task consisted of learning how to use spreadsheet software (i.e., Excel). Performance was assessed by raters who reviewed videotaped
sessions and rated whether discrete tasks such as entering data correctly or formatting a table were performed correctly. Trainees who were provided the opportunity to make errors (to- gether with explicit instructions encouraging them to learn from these errors) performed sig- nificantly higher than those in error-avoidant conditions. In a follow-up experiment, partic- ipants learning how to use presentations soft- ware (i.e., PowerPoint) performed better in the error training with metacognition prompting (i.e., instructions encouraging trainees to think explicitly about what the problem is, what they are trying to achieve, and so forth) compared to the error-avoidant condition (Keith & Frese 2005). A recent meta-analysis by Keith & Frese (2008) reported that overall, error management training was superior to either proceduralized error-avoidant training or exploratory training without error encouragement (d = 0.44). Ef- fect sizes were moderated by two important fac- tors: Effect sizes were greater for posttransfer measures compared to within-training perfor- mance, and for adaptive transfer tasks (as op- posed to tasks structurally similar to training). Thus, error training may be appropriate for de- veloping a deeper task understanding that facil- itates transfer to novel tasks.
Research on error training highlights the importance of understanding and affecting learner states and answers long-standing calls to engage in research on how individuals learn, not in just the latest training fads (e.g., Campbell 1971, Kraiger et al. 1993). For ex- ample, Schmidt & Ford (2003) reported that levels of meta-cognitive activity mediated the effects of a computer-based training program on declarative knowledge, task performance, and participants’ self-efficacy. An increasing amount of evidence suggests that trainees’ self-regulatory processes mediate the training– learning relationship. Self-regulation refers to the extent to which executive-level cognitive systems in the learner monitor and exert con- trol on the learner’s attention and active en- gagement of training content (Vancouver & Day 2005). Chen et al. (2005) trained 156