Related Experiment Videos
Beyond bistability: biophysics and temporal dynamics of working memory
1Centre for Theoretical and Computational Neuroscience, University of Plymouth, A 220 Portland Square, Drake Circus, Plymouth PL4 8AA, UK. daniel.durstewitz@plymouth.ac.uk
Neuroscience
|December 6, 2005
Summary
Working memory is a dynamic neural process, not a simple switch. Recent research reveals complex temporal activity profiles and network dynamics, moving beyond bistable models.
Area of Science:
- Neuroscience
- Computational Neuroscience
- Cognitive Science
Background:
- Working memory is traditionally modeled as a bistable switch with persistent neural activity.
- Related concepts like up- and down-states have been studied for their synaptic and ionic basis.
- Single-cell mechanisms for bistability have been explored computationally and in vitro.
Purpose of the Study:
- To review empirical findings on dynamic activity profiles in working memory.
- To discuss neural models explaining temporal structure and computational significance.
- To explore new experimental preparations for studying working memory biophysics.
Main Methods:
- Review of empirical findings on ramping activity and temporal structure.
- Analysis of neural models for working memory dynamics.
- Discussion of recent in vivo, neural culture, and in vitro preparations.
Main Results:
- Clear bistable activity is rare in working memory tasks; neuronal activity is highly dynamic.
- Neurons exhibit diverse temporal profiles and spiking dynamics.
- Evidence for temporal structure and spatio-temporally organized attractor states in cortical networks.
Conclusions:
- Working memory is a highly dynamical neural process, challenging traditional bistable models.
- Temporal structure and network dynamics are crucial for understanding working memory.
- Biophysical principles of dynamic networks may form the basis for future working memory theories.