Bruett, Heather
(2018)
Connectivity Based on Multi-Voxel Patterns Can Selectively Identify Brain Networks Where Condition-based Functional Connectivity Does Not: Evidence from the Scene Network.
Master's Thesis, University of Pittsburgh.
(Unpublished)
Abstract
With a major focus of neuroimaging research on mapping brain network connectivity, it is essential that researchers use the most effective methods for determining which regions comprise functional networks. Here, we compare the functional magnetic resonance imaging (fMRI) brain networks that can be identified through shared fluctuations in regions’ univariate responses to conditions (i.e., condition-based functional connectivity), with those identified by shared fluctuations in multivariate information. To do this, we compare brain networks generated by two approaches for measuring connectivity: psychophysiological interaction (PPI), which measures the effect of conditions on shared univariate responses, and informational connectivity (IC), which measures shared fluctuations in the discriminability of multi-voxel patterns. We compare the findings generated by applying these methods to data collected while people perceptually process scenes and control (pseudo) scenes. Prior work establishing the regions involved in scene processing give us an opportunity to compare the sensitivity and selectivity of these approaches to detect a stimulus-relevant network. We find that, while each measure produces useful information, the PPI method was less selective than IC in detecting scene-related regions. Using PPI led to identifying networks containing both scene and object regions, with little specificity in connections between scene regions. In contrast, the network identified by IC was more consistent with prior literature examining the brain’s scene network. We recommend that – for conditions known to be represented in multi-voxel patterns – researchers wishing to prioritize specificity in mapping networks should examine informational connectivity over univariate connectivity approaches such as PPI.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
29 August 2018 |
| Defense Date: |
16 July 2018 |
| Approval Date: |
21 June 2019 |
| Submission Date: |
23 January 2019 |
| Access Restriction: |
No restriction; Release the ETD for access worldwide immediately. |
| Number of Pages: |
33 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
Dietrich School of Arts and Sciences > Psychology |
| Degree: |
MS - Master of Science |
| Thesis Type: |
Master's Thesis |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
informational connectivity, psychophysiological interaction, connectivity, scenes, decoding, mvpa |
| Date Deposited: |
21 Jun 2019 15:51 |
| Last Modified: |
21 Jun 2019 15:51 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/35926 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](/images/feed-icon-32x32.png){kind=icon}