This title appears in the Scientific Report :
2014
Temporal sequence learning via adaptation in biologically plausible spiking neural networks
Temporal sequence learning via adaptation in biologically plausible spiking neural networks
The ability to acquire and maintain appropriate representations of time-varying, sequentialstimulus events is a fundamental feature of neocortical circuits and a necessary first steptowards more specialized information processing. The dynamical properties of such representationsdepend on the current...
Saved in:
Personal Name(s): | Duarte, Renato (Corresponding Author) |
---|---|
Morrison, Abigail | |
Contributing Institute: |
Computational and Systems Neuroscience; IAS-6 Computational and Systems Neuroscience; INM-6 |
Published in: | 2014 |
Imprint: |
2014
|
Conference: | Donders Discussions 2014, Nijmegen (Netherlands), 2014-10-30 - 2014-11-01 |
Document Type: |
Poster |
Research Program: |
W2/W3 Professorinnen Programm der Helmholtzgemeinschaft (Dys-)function and Plasticity ohne Topic |
Publikationsportal JuSER |
The ability to acquire and maintain appropriate representations of time-varying, sequentialstimulus events is a fundamental feature of neocortical circuits and a necessary first steptowards more specialized information processing. The dynamical properties of such representationsdepend on the current state of the circuit, which is determined primarily by theongoing, internally generated activity, setting the ground state from which input-specifictransformations emerge. Here, we begin by demonstrating that timing-dependent synapticplasticity mechanisms have an important role to play in the active maintenance of an ongoingdynamics characterized by asynchronous and irregular firing, closely resembling corticalactivity in vivo. Incoming stimuli, acting as perturbations of the local balance of excitationand inhibition, require fast adaptive responses to prevent the development of unstable activityregimes, such as those characterized by a high degree of population-wide synchrony. Weestablish a link between such pathological network activity, which is circumvented by theaction of plasticity, and a reduced computational capacity. Additionally, we demonstratethat the action of plasticity shapes and stabilizes the transient network states exhibited inthe presence of sequentially presented stimulus events, allowing the development of adequateand discernible stimulus representations. The main feature responsible for the increased discriminabilityof stimulus-driven population responses in plastic networks is shown to be thedecorrelating action of inhibitory plasticity and the consequent maintenance of the asynchronousirregular dynamic regime both for ongoing activity and stimulus-driven responses,whereas excitatory plasticity is shown to play only a marginal role. |