Illustration of the Hypothesis: If multisensory training effects specific multisensory regions (a) the training should affect differently the two groups, hence this region is trained by receiving input from 3 different modalities in the Auditory-Visual-Somatosensory group and 2 in the Audio-Visual one.

If training effects the interconnection of the structures (b) the effect should not differ between the groups, hence the trained structure is the same. MEG recordings of an audio-visual an auditory and a visual MMN were conducted pre- and post- training. The bold black lines mark the structures that are trained in each hypothesis. The grey and black shapes mark the structure that contributes to the results of the audio-visual MMN in each hypothesis.