Javascript must be enabled for the correct page display

Short-term Synaptic Plasticity Underlies Activity-Silent Working Memory: A Functional Spiking Neuron Model

Pals, Matthijs (2019) Short-term Synaptic Plasticity Underlies Activity-Silent Working Memory: A Functional Spiking Neuron Model. Bachelor's Thesis, Artificial Intelligence.

[img]
Preview
Text
Thesis_v2_Matthijs_Pals.pdf

Download (1MB) | Preview
[img] Text
Toestemming.pdf
Restricted to Registered users only

Download (121kB)

Abstract

We developed an end-to-end model of working memory (WM) based on the latest theories regarding the maintenance of information in the human brain. While it used to be widely assumed that storage in WM is maintained via persistent recurrent activity, recent studies showed that information can be maintained without persistent firing; there is storage in so called activity-silent states. A candidate mechanism underlying this type of storage is short-term synaptic plasticity (STSP), where the strength of connections between neurons rapidly changes to reflect new information being loaded in WM. In order to demonstrate that STSP by means of calcium-mediated synaptic facilitation can lead to behaviour similar to humans, STSP was integrated in a large-scale functioning spiking neuron model. This model was able to execute a delayed-response task in which a randomly oriented grating had to be maintained in WM. An earlier study measured the neural activity of human participants during this task. It was demonstrated that displaying a task-irrelevant stimulus during the maintenance period can reveal what is stored in activity-silent states. In support of our model’s plausibility, we showed that both its performance and neural activity during the task correspond to the human data of that study. We conclude that information in WM can be effectively maintained in activity-silent states by means of calcium-mediated STSP.

Item Type: Thesis (Bachelor's Thesis)
Supervisor:
Supervisor nameSupervisor E mail
Borst, J.P.J.P.Borst@rug.nl
Degree programme: Artificial Intelligence
Thesis type: Bachelor's Thesis
Language: English
Date Deposited: 12 Jul 2019
Last Modified: 12 Jul 2019 12:14
URI: http://fse.studenttheses.ub.rug.nl/id/eprint/20161

Actions (login required)

View Item View Item