Computational Modelling Group

Seminar  10th October 2012 2 p.m.  Building 32, Room 3077

THE RISE AND FALL OF MEMORY IN A MODEL OF SYNAPTIC INTEGRATION

Konstantinos Lagogiannis
University of Southampton

Web page
http://www.ecs.soton.ac.uk/people/kml206
Categories
Complex Systems
Submitter
Luke Goater

How do we acquire memories without immediately overwriting old ones? Acquisition of new memories involves changing the coupling strength between neurons. These plastic changes occur in response to electrical and chemical neural stimulation at structures called synapses found between neurons. Synapses need to resolve a dilemma between remaining stable to retain old memories and changing to acquire new memories. Previous attempts addressing this issue use complex synaptic state models as abstractions of the numerous biochemical reactions and pathways underlying synaptic plasticity. These states exists to support the stability of plastic changes for multiple timescales effectively slowing down the degradation of old memories in the face of ongoing memory storage.

In this talk we will propose a simple synaptic mechanism that enables synapses to resolve the dilemma by reading the recent history of stimuli that produce plastic changes. These simpler synapses exhibit natural timescales for the augmentation and decline of memories. Based on these natural timescales new models for the transition to late phase long-term plasticity (l-LTP) can be built. For this purpose, we will discuss the importance of the intracellular signalling pathway known as the cAMP-dependent protein kinase (PKA) in certain forms of long-term potentiation and then discuss an abstract mechanism that uses the proposed synaptic mechanism to activate PKA and signal the transition to l-LTP.