Back

Brain regions with positive and negative task-evoked responses engage in cooperative rather than competitive interactions

Csumitta, K. D.; Ossowski, A.; Gilmore, A. W.; Gotts, S. J.; Martin, A.

2019-07-02 neuroscience
10.1101/687897 bioRxiv
Show abstract

Brain regions displaying task-induced decreases in activity, sometimes referred to as \"Task-Negative\", have been proposed to reflect cortico-cortical competition from regions showing task-induced increases in activity, sometimes referred to as \"Task-Positive\". If these functional networks are competitive, trial-level fMRI BOLD responses across voxels exhibiting Positive and Negative responses should be anti-correlated. Additionally, the correlation between the BOLD response and behavior in Positive and Negative regions should have opposite sign. Here, we test these predictions using trial-to-trial variability in an object naming task. In contrast to the negative coupling proposed by a competition model, we find positive coupling between Positive and Negative regions and show that this coupling aligns with trial-level behavioral correlations. We provide evidence that these cooperative interactions are not due to a non-specific global factor and discuss alternative proposals, such as thalamo-cortical gating.\n\nHighlightsO_LITested competitive interactions among positive and negative task-evoked regions\nC_LIO_LIPositive and negative regions exhibited positive coupling at trial and run levels\nC_LIO_LISmaller competitive effects seen in BOLD-behavior correlations\nC_LIO_LIStrong cooperative effects not due to a non-specific common global factor\nC_LI\n\neTOCCsumitta et al. show that positive and negative BOLD responses do not exhibit cortical competition among corresponding brain regions. Rather, these networks display positive coupling. The spatial distribution of positive coupling aligned with behavioral correlations, suggesting normal, cooperative network interactions.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.