RESULTS

Figure 1 presents, at a subset of electrodes, the ERPs elicited by face (a), and by shape stimuli (b,c) within 100 ms post-stimulus in the one-class and the two-class conditions. ERPs to face stimuli differed significantly ( p ,0.05) between the two conditions at Pz, POz, O2 and P24 between about 50 and 65 ms (mean amplitude of the difference within 50±65 ms over the 4 electrodes, 0.48 μV; peak amplitude, 0.57 μV at P24 and 55 ms). Figure 2a depicts the detailed statistical significance of the differences at posterior electrodes, and the occipito-parietal distribution of the effect with the corresponding Student t-map at 55 ms latency.

Significant differences between the one-class and the two-class conditions were also observed in ERPs to shape stimuli, but only when the stimuli were separated by salient visual features (wide-hatched session; Fig. 1b,c; Fig. 2b,c). The differences reached statistical significance ( p ,0.01) at Iz, O2, IMb, Ma2, T6, and P24 between about 40 and 80 ms (Fig. 2b left; mean amplitude of the difference within 40±80 ms over the six electrodes, - 0.66 μV; peak amplitude, - 0.82 μV at T6 and 59 ms). No corresponding effect was found in the thin-hatched session (Fig. 2c, left).

Figure 2b,c (right) illustrate the topography of the differences at 55 ms latency in the two (wide- and thin-hatched) shape sessions, with the corresponding Student t-maps.

Fig. 1. Grand-average ERPs, at selective posterior electrodes, elicited by non-target faces (a) and geometrical shapes (b,c) according to whether, in each case, two categories of stimuli (male and female for faces; grey and hatched figures for shapes) are presented randomly mixed in the same run (two-class condition), or in separate runs (one-class condition; see Materials and Methods). The shaded rectangles indicate the first time period (between about 40 and 80 ms) when the responses differ between the two conditions.