Neurostatistical Model of Memory

From supermemo.guru
Jump to navigation Jump to search

Outline

Neurostatistical Model of Memory describes processes underlying the formation of long-term memory. In the model, memories are described by four core properties: stability, coherence, retrievability, and complexity.

Origins

The neurostatistical model of memory was based on statistical properties of long-term memory derived from the memory effects of varied spaced review schedules. It expands the two component model of memory beyond the molecular interpretation by involving structural and neural aspects of long term changes in properties of memory. The model provides basic functional ingredients needed for the brain to function as a concept network.

Neurostatistical Model of Memory resolves the plasticity-stability dilemma

New concepts

In addition to the skeletal ideas of the two component model of memory, the new contributions of the model are:

Learning

According to the model, learning would result in dendritic spine tagging, insertion of AMPA receptors, and elimination of untagged spines. Inverse tagging might be use to eliminate interfering connections. Learning would then result in a net loss of dendritic spines. In addition, spacing effect could arise from the need to clear dendritic filopodia from the site undergoing stabilization. The exploratory growth of filopodia would be a response to reduced postsynaptic activity.

Contributors

Piotr Wozniak, George Zonnios (spacing effect), Darek Murakowski, Janusz Murakowski (two component model of memory), Edward Gorzelanczyk (molecular memory).

This glossary entry is used to explain texts in SuperMemo Guru series on memory, learning, creativity, and problem solving