Related Experiment Video
Updated: Jun 7, 2026

Defining the Role Of Language in Infants' Object Categorization with Eye-tracking Paradigms
Published on: February 8, 2019
Restructuring partitioned knowledge: the role of recoordination in category learning
David K Sewell1, Stephan Lewandowsky
1Psychological Sciences, The University of Melbourne, Australia. dsewell@unimelb.edu.au
This study provides the first empirical evidence for recoordination, a type of knowledge restructuring where people switch and reuse categorization strategies. A multiple-module model (ATRIUM) explains this phenomenon better than single-module models.
Area of Science:
- Cognitive Psychology
- Computational Neuroscience
- Machine Learning
Background:
- Knowledge restructuring involves changes in problem-solving strategies.
- Existing models differentiate between relearning (incremental updating) and recoordination (novel application of knowledge).
- Recoordination requires multiple-module models, but has lacked empirical support.
Purpose of the Study:
- To provide direct empirical evidence for recoordination in category learning.
- To compare the explanatory power of single-module (ALCOVE) and multiple-module (ATRIUM) models for knowledge restructuring.
- To investigate the role of dimensional attention in strategy shifts.
Main Methods:
- Two category learning experiments were conducted.
- Participants' ability to alternate and reinstate categorization strategies was assessed.
- Computational models (ALCOVE and ATRIUM) were used to fit the experimental data.
Main Results:
- Direct empirical evidence for recoordination was found, demonstrating fluid alternation and reinstatement of strategies.
- Single-module models (ALCOVE) did not adequately fit the observed knowledge restructuring data.
- A multiple-module model (ATRIUM) quantitatively accounted for recoordination, linking it to changes in dimensional attention.
Conclusions:
- Recoordination is a key aspect of knowledge restructuring, allowing flexible strategy use.
- Multiple-module models like ATRIUM are necessary to explain complex category learning and strategy shifts.
- Learning in complex tasks involves the strategic coordination of partial knowledge elements.
More Related Videos
Related Concept Videos
Associative Learning
Classical conditioning, also known...
Coordination Number and Geometry
Observational Learning
The Anchoring-and-Adjustment Heuristic
Purposive Learning
Chunking and Rehearsal in Sensory Memory

