next up previous
Next: Spatio-temporal Noise Modelling Up: tr03mw2 Previous: Introduction

Modelling Framework

It is assumed that the observed data, $ y_{it}$, is decomposed into the addition of the response to the stimulus, $ r_{it}$, and some noise process, $ e_{it}$:

$\displaystyle y_{it}=r_{it}+e_{it}$ (1)

where $ i$ indexes the voxel and $ t$ indexes the time (or volume). Generally, $ r_{it}$ would be a function of the stimulus $ s_t$. Next we need to turn our attention to the modelling of the separate components in equation 1, the response to the stimulus, $ r_{it}$, and the noise process, $ e_{it}$.