Trial modeling of the hemodynamic response using event-related fMRI

Investigates the modeling of the intra-run trials within auditory event-related fMRI experiments. Building upon previously accepted techniques and models, a new, less rigid model is presented that allows for the estimation of the hemodynamic response across a much larger set of timepoints. This is accomplished by avoiding the reduction of timepoints that occurs when using trial averaging techniques. Instead, variations between trials are incorporated into a new model while still preserving the general characteristics of the response. The addition of a trial-dependent proportionality factor in the model significantly increases the accuracy of the coefficients estimated through a nonlinear least squares regression.