Search

How communication between neurons gives rise to natural vision remains a matter of intense investigation. The mid-level visual areas along the ventral stream, as studies in primates have shown, are selective to a common class of natural images—textures—but a circuit-level understanding of this selectivity and its link to perception remain unclear. We addressed these questions in mice, first showing that they can perceptually discriminate between texture types and statistically simpler spectrally matched stimuli. Then, at the neural level, we found that the secondary visual area (LM), more than the primary one (V1), was selective for the higher-order statistics of textures, both at the mesoscopic and single-cell levels. At the circuit level, textures were encoded in neural activity subspaces whose relative distances correlated with the statistical complexity of the images and with the mice’s ability to discriminate between them. These dependencies were more significant in LM, in which the texture-related subspaces were smaller and closer to each other, enabling better stimulus decoding in this area. Together, our results demonstrate texture vision in mice, finding a linking framework between stimulus statistics, neural representations, and perceptual sensitivity—a distinct hallmark of efficient coding computations.

The human visual ability to recognize objects and scenes is widely thought to rely on representations in category-selective regions of visual cortex. These representations could support object vision by specifically representing objects, or, more simply, by representing complex visual features regardless of the particular spatial arrangement needed to constitute real world objects. That is, by representing visual textures. To discriminate between these hypotheses, we leveraged an image synthesis approach that, unlike previous methods, provides independent control over the complexity and spatial arrangement of visual features. We found that human observers could easily detect a natural object among synthetic images with similar complex features that were spatially scrambled. However, observer models built from BOLD responses from category-selective regions, as well as a model of macaque inferotemporal cortex and Imagenet-trained deep convolutional neural networks, were all unable to identify the real object. This inability was not due to a lack of signal-to-noise, as all of these observer models could predict human performance in image categorization tasks. How then might these texture-like representations in category-selective regions support object perception? An image-specific readout from category-selective cortex yielded a representation that was more selective for natural feature arrangement, showing that the information necessary for object discrimination is available. Thus, our results suggest that the role of human category-selective visual cortex is not to explicitly encode objects but rather to provide a basis set of texture-like features that can be infinitely reconfigured to flexibly learn and identify new object categories.Significance StatementVirtually indistinguishable metamers of visual textures, such as wood grain, can be synthesized by matching complex features regardless of their spatial arrangement (1–3). However, humans are not fooled by such synthetic images of scrambled objects. Thus, category-selective regions of human visual cortex might be expected to exhibit representational geometry preferentially sensitive to natural objects. Contrarily, we demonstrate that observer models based on category-selective regions, models of macaque inferotemporal cortex and Imagenet-trained deep convolutional neural networks do not preferentially represent natural images, even while they are able to discriminate image categories. This suggests the need to reconceptualize the role of category-selective cortex as representing a basis set of complex texture-like features, useful for a myriad of visual behaviors.

Many everyday tasks require prioritizing some visual features over competing ones, both during the selection from the rich sensory input and while maintaining information in visual short-term memory (VSTM). Here, we show that observers can change priorities in VSTM when, initially, they attended to a different feature. Observers reported from memory the orientation of one of two spatially interspersed groups of black and white gratings. Using colored pre-cues (presented before stimulus onset) and retro-cues (presented after stimulus offset) predicting the to-be-reported group, we manipulated observers' feature priorities independently during stimulus encoding and maintenance, respectively. Valid pre-cues reliably increased observers' performance (reduced guessing, increased report precision) as compared to neutral ones; invalid pre-cues had the opposite effect. Valid retro-cues also consistently improved performance (by reducing random guesses), even if the unexpected group suddenly became relevant (invalid-valid condition). Thus, feature-based attention can reshape priorities in VSTM protecting information that would otherwise be forgotten.