Residual Analysis for Detecting Mis-modeling in fMRI.

Citation: Loh, J. M., Lindquist, M. A., Wager, T. D. (2008). Residual Analysis for Detecting Mis-modeling in fMRI. Statistica Sinica, 18, 1421-1448.

Full Text


The voxel-wise general linear model (GLM) approach has arguably become the dominant way to analyze
functional magnetic resonance imaging (fMRI) data. The approach relies on specifying predicted patterns of
signal change a priori. In this work we develop methods for detecting mis-modeling in the GLM framework,
and derive mathematical expressions for quantifying the e?ects this has on bias and power. We show that
even a relatively small amount of mis-modeling can result in severe power loss, and can in?ate the false
positive rate beyond the nominal value. Due to the massive amount of data, examining the appropriateness
of the model is challenging in fMRI. We propose a simple procedure involving the residuals that can be used
to identify possible voxels or regions of the brain where model mis?t may be present. The key idea is that
if there is model mis?t ? such as a mis-speci?cation of onset, duration, or response shape ? residuals will
be systematically larger in mis-modeled segments of the time series. By looking at the weighted sum of
consecutive residuals using a moving window, our method can pick out regions of a residual time series in
which the residuals are consistently larger than expected by chance, while ignoring spurious large residuals
that are expected based on the noise distribution. It may also be used more generally for identifying arti-
facts in fMRI time courses. We investigate the e?ectiveness of this method using a simulation study, and by
applying it to an fMRI dataset. We develop a method and accompanying software for creating whole-brain
maps showing power loss and bias due to mis-modeling. Such maps could be a valuable tool in assessing
violations of statistical assumptions and informing about di?erences in the shape and timing of the hemo-
dynamic response function (HRF) across the brain.

Key words: fMRI, GLM, model diagnosis, model mis?t, power, residual analysis