Consciously detecting and recognizing a stimulus without knowing what it looks like

Does conscious perception occur during initial sensory processing, or does it arise later in a supra modal fashion? If conscious access truly depends on supra modal processes, we may be able to induce “asensory perception”, where only the semantic features of a meaningful stimulus are accessed, untied to its sensory attributes. Here we tested this prediction by degrading the low-level sensory representations of visual words in the brain using pattern masking, and subsequently presenting audio words that were either semantically related to the masked word or not. We hypothesized these retrospective semantic cues would reactivate the remaining traces of the masked word in the brain, and induce awareness of any information that was not disrupted by masking. In three separate experiments we show that, when presented with retrospective cues that are semantically related to the masked word, participants are better at detecting the presence of the preceding masked word and naming it, while at the same time being unable to report its visual features. In other words, participants could consciously detect and recognize the preceding masked word, without knowing what it looked like. These findings suggest that non-sensory information can be consciously accessed in relative independence from the build-up of sensory representations. ### Competing Interest Statement The authors have declared no competing interest.