Fig. 3.

Evolution and principal components of the evoked fields.

A: The evoked response fields, averaged across all subjects, are plotted here at various time points in the 0–200 ms time window. The Smooth condition responses are on the top row, and the Noisy condition responses on the bottom.

B: The first two principal components of the data matrix M – i.e. the first two columns of U – are plotted on two topoplots. The second principal component contains substantial eye artefact signal.

C: The evoked responses at source level, for the grand averaged data. The neuronal populations whose activity is plotted on each graph (the spiny stellate cells and superficial and deep pyramidal cells) are those contributing to the MEG signal (i.e. omitting the inhibitory interneurons — see also Fig. 4).