Abstract
The determination of the figure-ground is essential for visual perception. Computational and psychophysical studies have reported that spatial attention in early vision facilitates the perception of border ownership (BO) that indicates the direction of figure (DOF) with respect to the border so that attended location appears as figures (Wagatsuma, et al. 2008). A recent physiological study has shown that the time course of BO-selective cells in V2 was affected by the ambiguity of the DOF on the previous display (P. O' Herron and R. von der Heydt, 2009). We investigated the mechanism behind this dynamics of the BO-selective cells through a computational model in which the early visual areas play critical roles for the determination of the activities of model BO-selective cells. Our model consists of V1, V2 and Posterior Parietal (PP) modules. The PP module is designed to represent spatial attention that could be considered as a saliency map based on luminance contrast. In the model, spatial attention alters contrast gain in the V1 module so that it enhances local contrast. The change in contrast signal then modifies the activity of model BO-selective cells in V2 because BO is determined solely from surrounding contrast (Sakai and Nishimura, 2006). The model was tested with the stimuli corresponding to Herron's physiological experiment. When new information regarding the DOF is presented, the activities of model BO-selective cells were rapidly modified. In contrast, if instead a new stimulus is presented with an ambiguous DOF, then the responses of model BO-selective cells decayed slowly. This model dynamics appears to depend on the ambiguity of the BO signal on the previous display and reproduce the same tendency found in the physiological study. These results suggest that network among PP and early visual areas could have crucial role for the time course of BO-selective cells.
This work was supported by Grant-in-Aid for JSPS Fellows (KAKENHI 09J02583).