Single-neuron correlates of visual consciousness in human lateral occipital complex
Michaël Vanhoyland, Peter Janssen, Tom Theys
Nature Communications December 1, 2025 DOI: 10.1038/s41467-025-67077-w (opens in new tab)
Study at a glance
AI-extracted from the abstract| Characteristics | Observational neurophysiology study using single- and multiunit recordings Peer reviewed |
|---|---|
| Population | Human participants undergoing intracranial recordings in the lateral occipital complex |
| Key findings | Conscious perception increased decoding accuracy and decoders assigned higher probabilities to the consciously perceived stimulus during dichoptic presentation. The authors report that most neurons in the human lateral occipital complex tracked perceptual awareness in a graded fashion, while a smaller subset encoded the visual input itself, suggesting LO responses predominantly align with subjective phenomenology. |
Abstract
Conscious perception, a critical aspect of human cognition, is assumed to emerge from a complex network of interacting brain regions that transmit information via feedforward and recurrent pathways. This study presents single- and multiunit recordings from the human lateral occipital complex (LO), a key region for shape and object recognition, during three distinct perceptual paradigms: backward masking, flash suppression and binocular rivalry. Stimulus awareness increased decoding accuracy and decoders assigned higher probabilities to the consciously perceived stimulus during periods of dichoptic stimulus presentation. These findings highlight the intricate neural mechanisms underlying visual awareness and show that LO responses predominantly align with subjective phenomenology, offering new insights into the neural correlates of visual consciousness. How individual neurons in the human lateral occipital complex respond to perceptually suppressed visual input remains unclear. Here, the authors demonstrate that while most neurons track perceptual awareness in a graded fashion, a smaller subset instead encodes the visual input itself.