|
| Variables Affecting the Shaping of Human Behavior |
| Sunday, May 27, 2007 |
| 3:00 PM–3:50 PM |
| Cunningham AB |
| Area: TPC |
| Chair: Parsla Vintere (The Graduate Center, City University of New York) |
| |
| Error-Correction Procedures in Movement Training. |
| Domain: Theory |
| PARSLA VINTERE (The Graduate Center, City University of New York), Claire L. Poulson (Queens College, City University of New York) |
| |
| Abstract: Repeated errors in well-learned movement performance may render movement inefficient or lead to physical injuries. Thus, it is important to understand the reasons for the reoccurring errors and develop effective error-correction procedures. There have been a number of behavior analytic studies that have experimentally examined the effectiveness of different behavioral coaching procedures for athletic skill development. Behavioral procedures have been particularly effective in decreasing persistent errors in athletic skill training arising from lack of stimulus control and insufficient reinforcement for correct performance. Behavioral error-correction procedures address topography of movement often leaving the dynamic properties of movement not fully investigated. The purpose of this paper is (a) to review the behavioral literature on error-correction procedures in movement training; (b) to discuss methodological difficulties associated with movement analysis; and (c) to examine possible alternatives to conventional error-correction procedures. Findings are discussed in terms of possible extension of movement-training models that may be applied in sports, physical education, dance, fitness training, and physical therapy. |
| |
| Understanding Why Practice Should be Fast and Accurate. |
| Domain: Theory |
| PHILIP IRVIN PAVLIK (Carnegie Mellon University) |
| |
| Abstract: In this talk, I will introduce the ACT-R cognitive model and how it explains the learning of simple behaviors. Specifically, this modeling system proposes a series of equations that capture the effects of practice and forgetting on long-term performance. Interestingly these equations predict that learning should be optimal when it is fast and accurate. In contrast to this prediction, cognitive psychology has generally asserted that high error rates (particularly from wide temporal spacing of repetitions) increase the amount of learning for each presentation of the stimulus. While the model does agree with this conclusion, the model also specifies that increased error rates lead to slower learning per trial. Furthermore, laboratory experiments suggest that the cost an increased error rate tends to outweigh the advantage of an increased error rate except at fairly high levels of performance (e.g. greater than 90% correct). Of course, behavior analysis has long advocated this conclusion. To better understand why practice should be fast and accurate I will highlight how the model supports the claims of both Precision Teaching and Direct Instruction. The crux of the analysis is that Precision Teaching and Direct Instruction are more efficient than training methods that result in high error rates. |
| |
|
| |