Related Experiment Videos
Distributed memory and the representation of general and specific information
Journal of Experimental Psychology. General
|June 1, 1985
Summary
This study presents a distributed model of information processing and memory, demonstrating how abstract representations can emerge from the superposition of specific experiences. The model simulates experimental findings, offering insights into memory storage and retrieval.
Area of Science:
- Cognitive Science
- Computational Neuroscience
- Artificial Intelligence
Background:
- The debate on whether memory stores abstract representations or specific experiences is ongoing.
- Existing models struggle to reconcile both abstract and specific information storage.
Purpose of the Study:
- To propose and validate a distributed model of information processing and memory.
- To investigate the emergence of abstract representations from specific experiences within this model.
Main Methods:
- Developed a computational model with numerous simple processing units and modifiable connections.
- Simulated information processing as emergent activation patterns through excitatory and inhibitory interactions.
- Modeled memory traces as changes in connection strengths, with superimposed traces from multiple events.
Main Results:
- The model successfully simulated experimental findings previously interpreted as evidence for specific experience enumeration.
- Demonstrated the emergence of functional equivalents of abstract representations (prototypes, logogens, rules) from superimposed specific traces.
- Showcased how the model captures environmental structure, behaving as if it learned abstractions.
Conclusions:
- Distributed models can account for both abstract and specific information representation in memory.
- Abstract concepts can emerge organically from the accumulation and interaction of specific memories.
- This framework offers a unified approach to understanding memory's dual capacity for generalization and specificity.