Advanced search
Add to list

The challenge of successive dynamic causal models

Hannes Almgren (UGent) , Frederik Van de Steen (UGent) , Roma Siugzdaite (UGent) and Daniele Marinazzo (UGent)
Author
Organization
Abstract
Effective connectivity during both resting state and cognitive tasks has been the topic of many studies. A widely used technique to infer effective connectivity, and context-related changes thereof, is dynamic causal modelling (DCM; Friston, Harrison, & Penny, 2003). For task-related fMRI, most studies investigate context-related perturbations of effective connections between brain regions. This is most commonly done with basic deterministic DCM (Friston et al., 2003). For resting-state fMRI, two different DCM methods have been applied. The first is stochastic DCM (Li et al., 2011), which is based on estimation of random neural fluctuations, the second is spectral DCM (Friston, Kahan, Biswal, & Razi, 2014), which is based on observed cross spectra between regions. The purpose of the present study is to apply DCM on successive working memory and resting-state fMRI sessions (n=15 and n=104, respectively), acquired from a single subject. For resting-state sessions, both DCM methods will be applied and compared. The second objective is to explore whether physiological, behavioral, and psychological parameters (e.g., blood pressure, quality of sleep, positive and negative emotions) are related to session-specific changes in effective connectivity. Preliminary results revealed changes in communication between primary visual areas over sessions during a face memory task. Specifically, forward connections from the primary visual cortex (V1) to fusiform face area (FFA) increased, while backward connections decreased over sessions. This change in dynamics most likely indicates more efficient face processing.

Citation

Please use this url to cite or link to this publication:

MLA
Almgren, Hannes, et al. “The Challenge of Successive Dynamic Causal Models.” Front. Neuroinform. Conference Abstract: Second Belgian Neuroinformatics Congress, vol. 9, Frontiers Media, 2015.
APA
Almgren, H., Van de Steen, F., Siugzdaite, R., & Marinazzo, D. (2015). The challenge of successive dynamic causal models. In Front. Neuroinform. Conference Abstract: Second Belgian Neuroinformatics Congress (Vol. 9). Leuven, Belgium: Frontiers Media.
Chicago author-date
Almgren, Hannes, Frederik Van de Steen, Roma Siugzdaite, and Daniele Marinazzo. 2015. “The Challenge of Successive Dynamic Causal Models.” In Front. Neuroinform. Conference Abstract: Second Belgian Neuroinformatics Congress. Vol. 9. Frontiers Media.
Chicago author-date (all authors)
Almgren, Hannes, Frederik Van de Steen, Roma Siugzdaite, and Daniele Marinazzo. 2015. “The Challenge of Successive Dynamic Causal Models.” In Front. Neuroinform. Conference Abstract: Second Belgian Neuroinformatics Congress. Vol. 9. Frontiers Media.
Vancouver
1.
Almgren H, Van de Steen F, Siugzdaite R, Marinazzo D. The challenge of successive dynamic causal models. In: Front Neuroinform Conference Abstract: Second Belgian Neuroinformatics Congress. Frontiers Media; 2015.
IEEE
[1]
H. Almgren, F. Van de Steen, R. Siugzdaite, and D. Marinazzo, “The challenge of successive dynamic causal models,” in Front. Neuroinform. Conference Abstract: Second Belgian Neuroinformatics Congress, Leuven, Belgium, 2015, vol. 9.
@inproceedings{8560566,
  abstract     = {Effective connectivity during both resting state and cognitive tasks has been the topic of many studies. A widely used technique to infer effective connectivity, and context-related changes thereof, is dynamic causal modelling (DCM; Friston, Harrison, & Penny, 2003). For task-related fMRI, most studies investigate context-related perturbations of effective connections between brain regions. This is most commonly done with basic deterministic DCM (Friston et al., 2003). For resting-state fMRI, two different DCM methods have been applied. The first is stochastic DCM (Li et al., 2011), which is based on estimation of random neural fluctuations, the second is spectral DCM (Friston, Kahan, Biswal, & Razi, 2014), which is based on observed cross spectra between regions.

The purpose of the present study is to apply DCM on successive working memory and resting-state fMRI sessions (n=15 and n=104, respectively), acquired from a single subject. For resting-state sessions, both DCM methods will be applied and compared. The second objective is to explore whether physiological, behavioral, and psychological parameters (e.g., blood pressure, quality of sleep, positive and negative emotions) are related to session-specific changes in effective connectivity.

Preliminary results revealed changes in communication between primary visual areas over sessions during a face memory task. Specifically, forward connections from the primary visual cortex (V1) to fusiform face area (FFA) increased, while backward connections decreased over sessions. This change in dynamics most likely indicates more efficient face processing.},
  author       = {Almgren, Hannes and Van de Steen, Frederik and Siugzdaite, Roma and Marinazzo, Daniele},
  booktitle    = {Front. Neuroinform. Conference Abstract: Second Belgian Neuroinformatics Congress},
  issn         = {1662-5196},
  language     = {eng},
  location     = {Leuven, Belgium},
  publisher    = {Frontiers Media},
  title        = {The challenge of successive dynamic causal models},
  url          = {http://dx.doi.org/10.3389/conf.fninf.2015.19.00034},
  volume       = {9},
  year         = {2015},
}

Altmetric
View in Altmetric