Neural representation of the bottom-up saliency map of natural scenes in human primary visual cortex

摘要

Zhang and colleagues recently showed that neural activities in V1 could create a bottom-up saliency map (Neuron, 73, 183-192, 2012). In that study, they used simple bar textures and a salient region was created by the orientation contrast between foreground and background bars. Here, we tested if their conclusion can generalize to complex natural scenes. Fifty natural images were selected from the internet based on the output of a prominent saliency model proposed by Itti and Koch (1998). The model predicted that all the natural images had a focal, lateral salient region, which was confirmed by a psychophysical experiment. In the experiment, to avoid top-down influences, each image was presented with a low contrast for only 50 ms and was followed by a high-contrast mask, which rendered the whole image invisible to subjects (confirmed by a forced-choice test). The Posner cueing paradigm was adopted to measure the spatial cueing effect (i.e. saliency) of the predicted salient region on an orientation discrimination task. A positive cueing effect was found and the magnitude of the cueing effect was consistent with the saliency prediction of the model. In a following fMRI experiment, we also used the masked natural scenes and measured BOLD signals responding to the predicted salient region (relative to the background). We found that the BOLD signal in V1, but not in other cortical areas, could well predict the cueing effect. These findings suggest that the bottom-up saliency map of natural scenes could be constructed in V1, providing further compelling evidence for the V1 saliency theory (Li, 2001).