Abstract
In this commentary, the authors elaborate on the issue of going beyond curve fitting to the drawing of inferences about motor learning. They argue that the agenda of A. Heathcote and S. Brown (2004) is largely a theory-free, curve-fitting enterprise that can be useful for certain aspects of describing behavior change, but that its gold standard of percentage of variance accounted for can also be misleading in its relevance to the theory of learning. Clearly, analysis methods are necessary and some are better than others, but the researcher can more fully exploit the relevance of methods to the construct with a priori theorizing than with a data-driven strategy of maximizing percentage of variance accounted for in curve fitting.
| Original language | English |
|---|---|
| Pages (from-to) | 233-238 |
| Number of pages | 6 |
| Journal | Journal of Motor Behavior |
| Volume | 36 |
| Issue number | 2 |
| DOIs | |
| Publication status | Published - 2004 Jun 1 |
Fingerprint
Keywords
- Dynamic systems
- Learning curve
- Motor learning
ASJC Scopus subject areas
- Biophysics
- Orthopedics and Sports Medicine
- Experimental and Cognitive Psychology
- Cognitive Neuroscience
Cite this
Beyond Curve Fitting to Inferences about Learning. / Liu, Yeou-Teh; Mayer-Kress, Gottfried; Newell, Karl M.
In: Journal of Motor Behavior, Vol. 36, No. 2, 01.06.2004, p. 233-238.Research output: Contribution to journal › Review article
}
TY - JOUR
T1 - Beyond Curve Fitting to Inferences about Learning
AU - Liu, Yeou-Teh
AU - Mayer-Kress, Gottfried
AU - Newell, Karl M.
PY - 2004/6/1
Y1 - 2004/6/1
N2 - In this commentary, the authors elaborate on the issue of going beyond curve fitting to the drawing of inferences about motor learning. They argue that the agenda of A. Heathcote and S. Brown (2004) is largely a theory-free, curve-fitting enterprise that can be useful for certain aspects of describing behavior change, but that its gold standard of percentage of variance accounted for can also be misleading in its relevance to the theory of learning. Clearly, analysis methods are necessary and some are better than others, but the researcher can more fully exploit the relevance of methods to the construct with a priori theorizing than with a data-driven strategy of maximizing percentage of variance accounted for in curve fitting.
AB - In this commentary, the authors elaborate on the issue of going beyond curve fitting to the drawing of inferences about motor learning. They argue that the agenda of A. Heathcote and S. Brown (2004) is largely a theory-free, curve-fitting enterprise that can be useful for certain aspects of describing behavior change, but that its gold standard of percentage of variance accounted for can also be misleading in its relevance to the theory of learning. Clearly, analysis methods are necessary and some are better than others, but the researcher can more fully exploit the relevance of methods to the construct with a priori theorizing than with a data-driven strategy of maximizing percentage of variance accounted for in curve fitting.
KW - Dynamic systems
KW - Learning curve
KW - Motor learning
UR - http://www.scopus.com/inward/record.url?scp=2342446601&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=2342446601&partnerID=8YFLogxK
U2 - 10.3200/JMBR.36.2.233-238
DO - 10.3200/JMBR.36.2.233-238
M3 - Review article
AN - SCOPUS:2342446601
VL - 36
SP - 233
EP - 238
JO - Journal of Motor Behavior
JF - Journal of Motor Behavior
SN - 0022-2895
IS - 2
ER -