Near-infrared spectroscopy for functional studies of brain: a challenge for BOLD model

The nonlinearity of blood oxygenation level-dependent (BOLD) response to stimuli of different duration (in a special way those of short duration) has been studied by functional magnetic resonance imaging (fMRI). This nonlinearity is assumed to be due to neural adaptation and the nonlinearity of the response in the oxygen extraction fraction (OEF). Simultaneous measurements made with fMRI and near-infrared spectroscopy (NIRS) give the possibility to evaluate OEF response contributes for a large time scale. The hemodynamic response nonlinearity can be quantified using an impulse response function model with saturation nonlinearity scaling in the response amplitude, assuming that the unknown neural adaptation parameters vary with a physiologically feasible range. In this paper, we adopt an inferential method for clarify the role of a set of unknown parameters on the nonlinearity BOLD model as measured by fMRI and NIRS.