Skip to content


  • Poster presentation
  • Open Access

Directed information structure in inter-regional cortical interactions in a visuomotor tracking task

  • 1, 4Email author,
  • 2, 3,
  • 2 and
  • 1
BMC Neuroscience200910 (Suppl 1) :P117

  • Published:


  • Interaction Structure
  • Task Difficulty
  • Functional Magnetic Resonance Image
  • Tracking Task
  • Transfer Entropy

Distributed computation in the brain is a complex process, involving interactions between many regions in order to achieve a particular task. We present an analysis of functional Magnetic Resonance Imaging (fMRI) measurements of brain activity in 16 localized regions recorded while eight subjects (who gave informed written consent) control a mouse with their right hand to track a moving target on a computer screen. We examine the underlying interaction structure between region pairs and the changes in that structure as a function of the difficulty of the tracking task. Our approach is distinguished in using asymmetric, multivariate, information-theoretical analysis, which captures not only non-linear relationships, but also collective interactions arising from groups of up to seven voxels in each region, and the direction of these relationships.

The statistical significance of the average transfer entropy [1, 2] (a directed measure of information transfer) yields a distinct 3-tier directed interaction structure that underlies all task difficulties (see Figure 1). Importantly, this structure connects movement planning to visual and motor control regions. Random-effects statistical parametric maps (SPM) analysis is then used to determine which pairs of regions have a) more in common (using the mutual information), or b) a more pronounced directional relationship (using the transfer entropy) as the task difficulty increases. Most significantly (see Figure 2), we identify an increased coupling between regions involved in movement planning (left SMA and left PMd) and execution (right cerebellum for right hand and right SC for eye movements) with task difficulty. It is likely these methods and extensions to them can be used to identify inter-regional structural changes due to other tasks and coherent information transfer structures in the cortex [3].
Figure 1
Figure 1

Directed relationships between selected regions.

Figure 2
Figure 2

The changes in the relationships (dashed green line = more in common, directed red line = a less directional relationship) as task difficulty increases.



JL thanks CSIRO and the Australian Research Council Complex Open Systems Research Network (COSNet) for their partial support of this work.

Authors’ Affiliations

School of Information Technologies, The University of Sydney, NSW, 2006, Australia
Bernstein Center for Computational Neuroscience, Charité-Universitätsmedizin Berlin, Philippstrasse 13, Haus 6, 10115, Berlin, Germany
Max Planck Institute for Human Cognitive and Brain Sciences, Stephanstrasse 1A, 04103 Leipzig, Germany
School of Computer Science and Engineering, The University of New South Wales, NSW, 2052, Australia


  1. Schreiber T: Measuring information transfer. Phys Rev Lett. 2000, 85: 461-464. 10.1103/PhysRevLett.85.461.PubMedView ArticleGoogle Scholar
  2. Chávez M, Jacques M, Le Van Quyen M: Statistical assessment of nonlinear causality: application to epileptic EEG signals. J Neuroscience Methods. 2003, 124: 113-128. 10.1016/S0165-0270(02)00367-9.View ArticleGoogle Scholar
  3. Lizier JT, Prokopenko M, Zomaya AY: Local transfer entropy as a spatiotemporal filter for complex systems. Phys Rev E. 2008, 77: 0261101-0261104. 10.1103/PhysRevE.77.026110.View ArticleGoogle Scholar


© Lizier et al; licensee BioMed Central Ltd. 2009

This article is published under license to BioMed Central Ltd.