LAUSR

Natural images contain information at multiple spatial scales. Although we understand how early visual mechanisms split multi-scale images into distinct spatial frequency channels, we do not know how the outputs of these channels are processed further by mid-level visual mechanisms. We have recently developed a naturalness discrimination task that uses synthesized, multi-scale textures to isolate these mid-level mechanisms (Freeman et. al. 2013). Here, we use three experimental manipulations (image blur, image rescaling, and eccentric viewing) to show that naturalness sensitivity is strongly dependent on image features at high object spatial frequencies (measured in cycles/image). As a result, sensitivity depends on a texture acuity limit, a property of the visual system that sets the highest retinal spatial frequency (measured in cycles/degree) that can be used to solve the task. A model observer analysis shows that high object spatial frequencies carry more task-relevant information than low object spatial frequencies. Comparing the outcome of this analysis with human performance reveals that human observers’ efficiency is similar for all object spatial frequencies. We conclude that the mid-level mechanisms that underlie naturalness sensitivity effectively extract information from all image features below the texture acuity limit, regardless of their retinal and object spatial frequency.