Steffen Stoewer

I'm using 7T fMRI of the awake behaving macaque to investigate executive function in the non-human primate brain. More specifically, I am interested in the Multiple Demand Network (MD).

I'm also working on methodological issues of awake high-field fMRI. Algorithms and functions I have developed have been integrated into fMRI Sandbox (FSB), an interactive Matlab toolbox for the analysis of awake NHP fMRI data.

To download fMRI Sandbox (FSB), please click . Newer versions are available from Google code ()

Investigation of the Multiple Demand Network (MD) with high-field fMRI

Steffen Stoewer (MPI Dept. Logothetis), Natasha Sigala (University of Sussex, UK), John Duncan (MRC-CBU Cambridge, UK), Nikos Logothetis

Introduction

A network of prefrontal, parietal and subcortical areas has consistently been found to be active in human fMRI studies of higher cognitive functions such as conflict monitoring, working memory, novelty detection or the processing of perceptual difficulties, for instance recognition and analysis of degraded stimuli [1]. It is hypothesized that its function is behavioural control through selective representation of task-relevant information including stimuli, rules, working memory contents etc. A detailed investigation of this network named Multiple Demand (MD) Network is of importance for our insight into the neural basis of many human brain disorders.

Goals

The main goal of the project is the investigation of the MD Network  in non-human primates with high-field fMRI.

Methods

We tried to elicit MD activation with a passive viewing task adapted to the one used in a human fMRI study [2] where it could be shown that passive viewing of attended stimuli with no active task produced the full pattern of MD activation when stimulus changes occurred.

Secondly, we modified a saccade choice task by introducing perceptual difficulty or a response conflict to allow us to directly compare MD activity for different kinds of cognitive demand.

Results

We could show an activation of key components of the MD network in three primates for the passive viewing task [3]. Scans for the second experiment are ongoing.

Figure legend:

Statistical parametric maps (T-score maps) indicating voxels (colored red to yellow) significantly more active during the stimulation phase than when viewing a blank screen for a group analysis of three animals [3], thresholded at P<.05 (FWE-corrected).

45: Brodmann area 45; 46v: BA 46, ventral subdivision, F5: agranular frontal area F5; 8A: area 8A in arcuate sulcus; MT: middle temporal area; MST: middle superior temporal area; V1: Visual area 1; V2: Visual area 2; LIP: lateral intraparietal are; VIP: ventral intraparietal area; MIP: medial intraparietal area

Conclusion

Our results suggest that visual stimulation alone already activates key components of the MD network. Additional task demands activate additional cortical regions. Our data support the idea of a possible overlap between MD and the dorsal frontoparietal network.

References

1. Duncan, J. and Owen A.M.,  Trends Neurosci, 23(10), 475–483 (2000).
2. Hon, N., Epstein, R.A., Owen, A.M. and J. Duncan, Journal of Neuroscience, 26(38), 9805-9809 (2006)
3. Stoewer, S, Ku, S.-P., Goense, J, Steudel, T, Logothetis, N.K., Duncan, J, Sigala, N, Magnetic Resonance Imaging 28 1120-1128(2010)