Related Experiment Video
Updated: May 8, 2026
An Appetitive Spatial Working Memory Task for Mice in a Semi-Automated 8-Arm Radial Maze, Reducing Fearful Memory Association in the Maze
Published on: July 29, 2025
A note concerning analysis of error scores from motor-memory research
1a Department of Physical Education , Louisiana State University.
Abstract:
Exception is taken to the statistical analysis for motor memory research suggested by Roy (1976) in his paper on "Measuring change in motor memory. " The hazardous effects of multicollinearity which result from using AE(or E), CE, and VE as multiple dependent variables in MANOVA are explained and an alternate strategy for statistical analysis is suggested. I would like to express my appreciation to Ken Brewer, Program of Evaluation and Research Design at Florida State University, for his assistance with the major points of this paragraph.
Insights
This study critiques Roy's (1976) statistical approach for motor memory research. It highlights multicollinearity risks with multiple dependent variables in MANOVA and proposes an alternative analysis strategy.
Area of Science:
- Motor learning and memory research
- Statistical methodology in psychology
Background:
- Roy (1976) proposed a statistical analysis for motor memory research.
- The proposed method involves using multiple dependent variables in MANOVA.
Purpose of the Study:
- To critique the statistical analysis suggested by Roy (1976).
- To explain the hazardous effects of multicollinearity.
- To suggest an alternate statistical analysis strategy for motor memory research.
Main Methods:
- Critique of existing statistical methods.
- Explanation of multicollinearity in MANOVA.
- Proposal of an alternative statistical approach.
Main Results:
- The use of Absolute Error (AE), Constant Error (CE), and Variable Error (VE) as multiple dependent variables in MANOVA can lead to multicollinearity.
- Multicollinearity poses risks to the validity of statistical findings in motor memory research.
Conclusions:
- Roy's (1976) statistical analysis for motor memory research is problematic due to multicollinearity.
- An alternative statistical strategy is recommended to mitigate these risks.
