TY - JOUR T1 - Task activations produce spurious but systematic inflation of task functional connectivity estimates JF - bioRxiv DO - 10.1101/292045 SP - 292045 AU - Michael W. Cole AU - Takuya Ito AU - Douglas Schultz AU - Ravi Mill AU - Richard Chen AU - Carrisa Cocuzza Y1 - 2018/01/01 UR - http://biorxiv.org/content/early/2018/08/01/292045.abstract N2 - Task-state functional connections are thought to reflect distributed cognitive processes throughout the brain. Utilizing a neural mass computational model we found that task activations substantially and inappropriately inflate task-state functional connectivity (FC) estimates (temporal correlations), especially in functional MRI (fMRI) data. Such activation-induced inflation of task FC estimates was postulated previously, but this phenomenon has not been conclusively established either theoretically or empirically, nor have corrective methods been systematically evaluated. We found that activation-based task FC inflation was primarily driven by task-onset neural responses propagating forward in time into the task performance period (temporal autocorrelation), increasing the temporal similarity of task-activated brain regions regardless of whether they were actually functionally connected. This effect was enhanced in fMRI data, with task-evoked fMRI activations introducing a temporally-extended hemodynamic response shape to underlying neural time series. This demonstrates that isolating task-state network changes from task-evoked activations is essential for ensuring discovery of FC effects that are distinct from task activation effects. Standard approaches for fitting and removing task-evoked activations were unable to correct these inflated correlations. In contrast, methods that flexibly fit hemodynamic response shapes (especially finite impulse response-based regression) effectively corrected the inflated correlations, without introducing false negatives. Results with empirical fMRI data confirmed the model’s predictions, revealing activation-induced task FC inflation for both Pearson correlation and psychophysiological interaction (PPI) approaches. These results demonstrate that removal of task activations using an approach that flexibly models hemodynamic response shape is an important preprocessing step for valid estimation of task-state FC.Computational model shows task inflation of functional connectivity estimatesHemodynamic responses cause task activations to further inflate estimatesStandard approaches to remove task activations leave many false positivesMethods that flexibly fit hemodynamic response shape effectively correct inflationCorrection of functional connectivity inflation verified with empirical fMRI data ER -