ИСТИНА

Illusory contours comprise a convenient model for studying neurophysiological properties of the visual system allowing it to support realistic visual perception. This basic feature of visual system functioning seems to be essential for vision, since it allows perception of complete Gestalt objects derived from fragmented physical images. A large body of evidence hints to higher visual areas as critical nodes responsible for the illusory contour detection, with lateral occipital complex most often cited (Murray & Herrmann, 2013; Shpaner et al., 2013). Yet presently there is a significant controversy concerning the role of the early visual cortices in processing illusory contours, and contradictory data were reported concerning even the sign of the effect. Effects in the V1/V2 related to IC perception mostly remain evasive for EEG and MEG recording with few exceptions, hinting at rather late activations (Ohtani et al., 2002; Proverbio & Zani, 2002; Halgren et al., 2003; Khoe et al., 2004; Knebel & Murray, 2012; Mijović et al., 2014). Animal studies reported both excitation and inhibition of V1 cells in response to illusory contours (Peterhans & Von der Heydt, 1989; Von der Heydt & Peterhans, 1989; Grosof et al., 1993; Ramsden et al., 2001; Lee & Nguyen, 2001). Relying on the EEG findings of short-latency suppression of early visual cortices in response to Kanizsa figure compared to control stimuli in children (Stroganova et al., 2012), we attempted to find direct evidence, as well as localization and timing of this effect. The stimuli sized 4.5° and 9.0° were presented centrally to the participants in passive conditions. By applying strict spatial and temporal confinements as well as using threshold-free cluster enhancement technique and permutation statistics, we were able to detect the statistically significant inverted illusory contour effect – relative suppression of the response to the Kanizsa figure compared with the control stimulus within the 40-120 ms time window after the stimulus onset. In view of current literature data this effect is most likely explained by iso-orientation suppression (Sillito & Jones, 1996; Jones et al., 2001; Ishikawa et al., 2010; Hashemi-Nezhad & Lyon, 2012; Henry et al., 2013; Vanni & Casanova, 2013) – an intrinsic feature of V1 cells response to collateral borders, which are present in Kanizsa figures and absent in relevant control stimuli. The effect may relate to the principle of "sparse" coding, or information maximization principle (Series et al., 2003; Zhu & Rozell, 2013), according to which the V1 suppresses representations of inner parts of collinear assemblies as being informationally redundant.