Characterizing across-trial variation in the hemodynamic responses using ultra-fast fMRI data at 7T

Characterizing hemodynamic responses with great temporal precision is crucial for the interpretation of fMRI data with respect to the underlying neuronal activity. Recent literature highlights that analyses based on a canonical hemodynamic response function (HRF) may fail to capture relevant aspects of neuronal activity in certain conditions1. Ultra-fast fMRI enables precise temporally precise characterizations of hemodynamic responses and their trial-by-trial variation. Here, we analyzed fMRI timeseries sampled at which allowed us to model hemodynamic responses for individual trials during a visuo-motor task. Trial-by-trial variability of hemodynamic responses are analyzed by principal components analysis (PCA) and related to trial-specific conditions.