Exploring Properties of Bioplausible Working Memory

Authors

• Willian Soares Girão - PhD student, University of Groningen, Faculty of Science and Engineering, Bio-inspired Circuits & Systems
• Thomas Tiotto - PhD student, University of Groningen, Faculty of Science and Engineering, Cognitive Modelling

References

• Mongillo, G., Barak, O. & Tsodyks, M. Synaptic Theory of Working Memory. Science 319, 1543–1546 (2008)
• Jug & Florian. On Competition and Learning in Cortical Structures (2012)
• Lundqvist, M., Rehn, M., Djurfeldt, M. & Lansner, A. Attractor dynamics in a modular network model of neocortex. Netw Comput Neural Syst 17, 253–276 (2009)
• Pals, et al. A functional spiking-neuron model of activity-silent working memory in humans based on calcium-mediated short-term synaptic plasticity. Plos Comput Biol 16, e1007936 (2020)
• Adaptations of original work from Ankatherin Sonntag's Master's thesis on the Chicca synaptic learning rule

Usage

• network_dynamics > RCN > rcn.py: Runs a recurrent competitive network (RCN) learning to sustain attractor activity for two stimuli. Output plots and data are saved to the network_results folder.
• graph_analysis > network_visualisation.py: Runs a recurrent competitive network (RCN) learning to sustain attractor activity for two stimuli. Builds and displays a graph from the RCN in order to characterise the network topology. Output files can be saved by calling the save_graph_results() function
• graph_analysis > network_visualisation.py: Loads and displays a pre-existing RCN graph in order to characterise the network topology.