Next: Average Group Activation
Up: tr01cb1
Previous: Limitations
Examples
In this section we show how various group level
parameters of interest can easily be calculated within the GLM
framework. This amounts to specifying a suitable group design matrix
, a covariance structure
and possibly a contrast vector
. Note that unlike the case of first-level designs, the mean
parameter value is often of interest and hence the design matrix,
, must always model the group mean activation, that is, the unit
vector must always be included in the span of
.
For several of the examples it is easy to show the added benefit of
the proposed framework using numerical simulations. These principally
contrast the heteroscedastic model (allowing for different individual
first-level variances) with the homoscedastic model (where first-level variances are assumed to be identical). These comparisons
show substantial increases in -statistics over a wide range
of realistic scenarios.
Subsections
Christian Beckmann
2003-07-16