Home WakeSpace Scholarship › Electronic Theses and Dissertations

A Novel Model of Task-Switching and Decision Making

Electronic Theses and Dissertations

Item Files

Item Details

A Novel Model of Task-Switching and Decision Making
Bentley, Nick
To survive, an organism must pursue multiple goals and switch between them at appropriate times, yet little is known about how a network of neurons can underlie such behavior. A network’s behavior can only change with an organism’s goals if it receives information about those goals as input. Thus, network responses must depend jointly on both current stimuli and current goals. How are these sources of information combined to generate behavior? In Chapter II we present a plausible network model that can alter it’s goals on the fly. We illustrate the model’s performance in visual search tasks for which human behavioral data are available. We show that a single fixed network can perform all of these tasks correctly, reproducing a number of experimental observations, as well as making a novel prediction. Our model demonstrates that three key properties of cortical neurons — gain-modulation, recurrent connectivity, and race-to-threshold dynamics — can be combined to generate a powerful and flexible decision-making model. A similar mechanism may be at work in real brains. One network dynamic required for our model, but which we did not simulate, is multistability, which refers to a neural population’s ability to exist in multiple stable states. In chapter III we describe a study of the effects of known features of brain structure and activity on multistability in a simple neural network model. We found that multistability can coexist with all the tested factors, suggesting that multistability is a robust phenomenon which can operate under realistic conditions. We still lack adequate quantitative ways to describe the capacity of neural populations to transform stimuli into behavior. Although measures exist which describe how well neural populations can represent the environment, there are none which describe how well they can use such representations to modify behavior. In Chapter IV we introduce just such a measure, called the Basis Set Error, which describes how well a neural population represents stimulus information in a way that can be used by the mechanisms which control behavior. We illustrate the measure’s behavior on artificial data and compare it to preexisting measures of representational capacity.
computational model
race model
visual search
Pete Santago (committee chair)
Christos Constantinidis (committee member)
Terrence Stanford (committee member)
Paul Tiesinga (committee member)
Emilio Salinas (committee member)
Bentley, Nick
2008-09-28T10:52:08Z (accessioned)
2010-06-18T18:57:07Z (accessioned)
null (available)
2008-09-28T10:52:08Z (available)
2010-06-18T18:57:07Z (available)
2008 (issued)
null (defenseDate)
Neurobiology & Anatomy (discipline)
Wake Forest University (grantor)
PHD (level)
http://hdl.handle.net/10339/14667 (uri)
etd-08202008-163145 (oldETDId)
Release the entire work immediately for access worldwide. (accessRights)
I hereby certify that, if appropriate, I have obtained and attached hereto a written permission statement from the owner(s) of each third party copyrighted matter to be included in my thesis, dissertation, or project report, allowing distribution as specified below. I certify that the version I submitted is the same as that approved by my advisory committee. I hereby grant to Wake Forest University or its agents the non-exclusive license to archive and make accessible, under the conditions specified below, my thesis, dissertation, or project report in whole or in part in all forms of media, now or hereafter known. I retain all other ownership rights to the copyright of the thesis, dissertation or project report. I also retain the right to use in future works (such as articles or books) all or part of this thesis, dissertation, or project report. (license)

Usage Statistics