Continuous Concurrent Feedback Degrades Skill Learning: Implications for Training and Simulation
Journal article by Richard A. Schmidt, Gabriele Wulf; Human Factors, Vol. 39, 1997
Journal Article Excerpt
Continuous concurrent feedback degrades skill learning: implications for training and simulation. by Richard A. Schmidt , Gabriele Wulf INTRODUCTION Over the past several decades, researchers have examined the role of numerous types of augmented (extrinsic, or supplemental) information feedback for the learning of movement skills. (For reviews, see Magill, 1993; Salmoni, Schmidt, & Walter, 1984; Schmidt, 1991.) One type of augmented feedback that has received both practical and theoretical attention in training and simulation settings is termed concurrent feedback: supplementary information presented to the learner during the actual action. This feedback can be presented discontinuously to signal that the performer is on target at that moment or that a certain level of performance is being achieved. This feedback can also be presented continuously (on-line, with essentially no delay) to indicate the level of momentary performance error (Karlin & Mortimer, 1963; Kohl & Shea, 1995; Phillips & Berkhout, 1978), deviations from a goal movement pattern (Van der Linden, Cauraugh, & Greene, 1993), the pattern of ongoing electromyographic activity (Mulder & Hulstijn, 1985), or other sources of biofeedback. Continuous concurrent augmented feedback using vision is the focus of the present article. Variants of this type of feedback are used in many practical situations of interest to the human factors/ergonomics field, such as simulators, in which a feature to be controlled (e.g., the location of an aircraft with respect to a glide path during landing; Lintern, 1980) is fed back to the learner concurrently and continuously with the action. In physical therapy, continuous feedback has been used for decades as a way to train weight-bearing in injured limbs (Winstein, 1991). On the surface, continuous feedback appears to be effective for learning because it guides the learner powerfully to the correct response, minimizes errors, and holds behavior on target. The problem is that the performance gains during practice are seldom carried over to retention or transfer tests in which the augmented feedback is withdrawn. The usual finding is that people who have practiced with concurrent continuous feedback often perform worse on no-feedback retention tests than do people who have practiced without such feedback. In other words, continuous concurrent feedback appears to enhance performance during practice when the feedback is operating, but it does not contribute to learning and may even degrade learning, as measured on retention and transfer tests (see Annett, 1959, 1969; Karlin & Mortimer, 1963; Kohl & Shea, 1995; Patrick & Mutlusoy, 1982; Van der Linden et al., 1993). If augmented continuous feedback generally degrades learning, it should be avoided in the design of learning settings. Nevertheless, augmented continuous feedback has been deliberately included in both simulators and training programs, often with impressive realism and fidelity but usually at considerable expense. Its inclusion is understandable in that it does generally improve performance when it is present, and it seems obvious at first glance that such information must contribute to learning as well. However, recent views of training (e.g., Schmidt & Bjork, 1992) emphasize that measures of performance during training are not generally good predictors of longer-term learning, which is usually best evaluated on retention or transfer tests that are separated from the training setting. Even more critical, there is considerable evidence that, relative to some so-called standard practice condition, several factors that facilitate performance during practice are detrimental to retention and/ or transfer performance, which is widely considered important in the success of training and simulation programs. Hypotheses for Feedback Processing The puzzling and counter-intuitive failure of continuous concurrent feedback to contribute to learning (even though it contributes strongly to performance during practice) has been explained mainly by the notion that concurrent feedback is overly guiding during practice. Originally suggested by Annett (1959, 1969), this idea has been formalized recently to form what is termed the guidance hypothesis for feedback (Salmoni et al., 1984; Schmidt, 1991; Schmidt, Young, Swinnen, & Shapiro, 1989). In this view the learner uses the guidance of concurrent feedback as a crutch to produce the correct action during practice. However, this reliance on feedback prevents the learner from exercising the capabilities that are needed on the retention or transfer test, on which augmented feedback is typically not available, so performance falls below even that of learners who did not have continuous feedback during practice. These overly guiding properties of concurrent feedback can be thought of in two ways.Guidance properties. Concurrent feedback seems to be most commonly used to guide behavior. This process is probably most apparent in tasks for which the movement duration is relatively long (greater ... |
End of free preview...
To continue reading this publication, you must have a Questia Subscription.
Try Us Today! Click Here
Questia provides the world's largest online library of scholarly books and journal articles, with integrated footnote and bibliography tools, highlighting, note taking and book marking. With a Questia subscription, you'll have access to the full text of more than 67,000 books and 1.5 million articles.
Already a subscriber? Login:
Sponsored Links
Free Newsletter
Get helpful how-to's, writing tips, search strategies, quizzes & more!
Search the Library
