Diagnostic Information
Clinical Applications

Features extraction and information processing in vision

Using EEG we investigate how stimulus features are extracted and how categories are processed by the brain. A typical example is a study in which we were interested in testing if stimulus phase information is processed and when. In a 1st paper (Rousselet et al., (2008) 1, 98) we showed that the visual system sensitivity to global phase information emerges progressively in a short time window between the P1 and the N170 components, roughly at about 120–130 ms after stimulus onset and continues for another 25–40 ms. During this period, single-trial activity is not only sensitive to phase information, but also to the kurtosis of the pixel distribution in each image, and phase by kurtosis interaction. These latter effects indeed correspond to the extraction of local phase information, especially the eyes for faces, as demonstrated in subsequent papers. Using a simlar approach, current work will test which features, based on low level vision model, are actually extracted and which model gives the best predictions about brain activity.

Modelled data and topographical maps from 1 subjects . The upper part of the figure shows 3D (left) and 2D (right) representations of single trials, sorted by phase coherence level. In the lower left corner, single-trial modelled data were averaged according to phase coherence level, and colour coded from blue (0%) to red (100%). The data are from the electrode at which the maximum variance explained was obtained. The topographic maps show the interpolated ERP signal (left, in μV) and explained variance (right, in %) at the latency of maximum R2 (148 ms). The electrode showing the best fit is at the centre of the lower left red cluster in the explained variance map.