Competition for Attention Among Spatial Modulations of Brightness Gradients

Introduction. The pre-attentive stage of visual processing includes the stage of selection and coding of brightness gradients, and the stage of spatial integration of this information. The first operation is realized by the first-order visual filters (striatal neurons), the second one – by the second-order visual filters (extrastriatal neurons). The second-order filters transmit image areas with spatial modulations of contrast, orientation, and spatial frequency (modulation dimensions) and can function as attention gates. This investigation is aimed at establishing priorities among the modulations in their competition for attention. Methods. All the stimuli in the study represented three images of the same object. These images were constructed from the areas of the original object and contained modulations of a) contrast (first image), b) orientation (second image), and c) spatial frequency (third image). These areas were in advance isolated from the object by means of the model of the second-order filters developed by the authors. In the first experiment, images composed of different in dimension modulations competed for participants' attention. In the second experiment, the authors used images of the same dimension, but different in their spatial frequency. Attention focus was determined by recording eye movements. Results. The findings suggest that the modulations of contrast and orientation had an advantage in the competition for attention among dimensions. When the choice was made between the images of the same dimension, the images formed from the middle-range spatial frequencies had the advantage in the competition for attention. Discussion. The present study is the first to address the priorities among the modulations in competition for attention. Higher average rates of output signals in the second-order filters that form the test image increase the probability that this image will attract participants' attention. Conclusion. The second-order filters can play the role of attention gates. In addition, the filters with higher rates of their output signals have an advantage in competing for attention in visual search. The Appendix describes the models of the second-order visual mechanisms, which the authors used for preparing the study stimuli.