Depression alters "top-down" visual attention: a dynamic causal modeling comparison between depressed and healthy subjects

Martin Desseilles, Sophie Schwartz, Thien Thanh Dang-Vu, Virginie Sterpenich, Marc Ansseau, Pierre Maquet, Christophe Phillips

Research output: Contribution to journalArticlepeer-review


Using functional magnetic resonance imaging (fMRI), we recently demonstrated that nonmedicated patients with a first episode of unipolar major depression (MDD) compared to matched controls exhibited an abnormal neural filtering of irrelevant visual information (Desseilles et al., 2009). During scanning, subjects performed a visual attention task imposing two different levels of attentional load at fixation (low or high), while task-irrelevant colored stimuli were presented in the periphery. In the present study, we focused on the visuo-attentional system and used "Dynamic Causal Modeling" (DCM) on the same dataset to assess how attention influences a network of three dynamically-interconnected brain regions (visual areas V1 and V4, and intraparietal sulcus (P), differentially in MDD patients and healthy controls. Bayesian model selection (BMS) and model space partitioning (MSP) were used to determine the best model in each population. The best model for the controls revealed that the increase of parietal activity by high attention load was selectively associated with a negative modulation of P on V4, consistent with high attention reducing the processing of irrelevant colored peripheral stimuli. The best model accounting for the data from the MDD patients showed that both low and high attention levels exerted modulatory effects on P. The present results document abnormal effective connectivity across visuo-attentional networks in MDD, which likely contributes to deficient attentional filtering of information.
Original languageEnglish
Pages (from-to)1662-8
Number of pages7
Issue number2
Publication statusPublished - 15 Jan 2011


  • Attention
  • Brain
  • Depressive Disorder, Major
  • Humans
  • Magnetic Resonance Imaging
  • Models, Neurological
  • Photic Stimulation
  • Visual Perception

Fingerprint Dive into the research topics of 'Depression alters "top-down" visual attention: a dynamic causal modeling comparison between depressed and healthy subjects'. Together they form a unique fingerprint.

Cite this