Related Experiment Videos
Using segmented linear regression models with unknown change points to analyze strategy shifts in cognitive tasks
1Center for Instructional Psychology and Technology, University of Leuven, Vesaliusstraat 2, B-3000 Leuven, Belgium. koen.luwel@ped.kuleuven.ac.be
Behavior Research Methods, Instruments, & Computers : a Journal of the Psychonomic Society, Inc
|January 31, 2002
Summary
This study introduces SegcurvN, a new method for analyzing cognitive strategy shifts using segmented linear regression. The technique effectively identifies strategy changes in tasks like numerosity judgment.
Area of Science:
- Cognitive Psychology
- Statistical Modeling
Background:
- Segmented regression analysis is valuable for detecting strategy shifts in cognitive tasks.
- Previous work introduced Segcurve for two regression lines with one change point.
- SegcurvN extends this to n regression lines with n-1 change points.
Purpose of the Study:
- To present the SegcurvN technique for fitting segmented regression models.
- To demonstrate the utility of three-phase segmented linear regression for identifying cognitive strategies and shifts.
- To apply the method to data from a numerosity judgment experiment.
Main Methods:
- Utilized the SegcurvN technique for segmented linear regression analysis.
- Applied a three-phase segmented linear regression model.
- Analyzed data from a numerosity judgment experiment.
Main Results:
- The three-phase segmented linear regression model proved useful for identifying strategies and strategy shifts.
- The study demonstrated the application of SegcurvN to cognitive task data.
Conclusions:
- SegcurvN is a valuable tool for analyzing cognitive strategy shifts.
- The technique offers insights into strategy use and changes in cognitive tasks.
- Advantages and limitations of the method were evaluated.