Transient and linearly graded deactivation of the human default-mode network by a visual detection task.

In this fMRI study, we show that an extended network of brain areas, previously described as the default-mode network, is suppressed during the performance of a global visual motion discrimination task. For the first time, we demonstrate that this network is transiently suppressed in an event-related fashion, reflecting a true negative activation compared to baseline, and that this deactivation occurs in a strongly graded fashion depending on the strength of the global motion signal. Deactivation across the network varied in an inverse linear relationship with motion coherency, demonstrating that the strongest suppression occurs for the most error-prone tasks. Deactivations were absent for the easiest of the tasks (100% coherence). We also show that the magnitude of task-related activation of the individual sub-components of the default-mode network are strongly correlated, indicating a highly integrated system. The results offer a striking indication of a rapid, highly reactive and tunable system within the brain for active suppression of this network of brain areas.