ePoster

ON–OFF CODING IS LATENT IN VERTEBRATE VISUAL CIRCUITS

Xinwei Wangand 3 co-authors

University of Sussex

FENS Forum 2026 (2026)
Barcelona, Spain
Board PS02-07PM-641

Presentation

Date TBA

Board: PS02-07PM-641

Poster preview

ON–OFF CODING IS LATENT IN VERTEBRATE VISUAL CIRCUITS poster preview

Event Information

Poster Board

PS02-07PM-641

Abstract

Neural circuits across sensory systems represent stimulus changes of opposite sign as On and Off signals. In vertebrate vision, this polarity split is classically attributed to the photoreceptor–bipolar cell synapse via a fixed molecular dichotomy: Off bipolar cells use sign-preserving ionotropic glutamate receptors, whereas On bipolar cells use the sign-inverting receptor mGluR6. This textbook model implies that downstream circuits inherit pre-segregated polarity channels. However, many non-mammalian vertebrates exhibit predominantly mixed On–Off responses at the retinal output, challenging this view.
Here we combine comparative transcriptomics, pharmacology, genetics, and in vivo imaging in the diurnal zebrafish retina to test whether polarity splitting is actively maintained. We find that bipolar cells frequently co-express receptors of opposing polarity and many are intrinsically On–Off. Here, Off signalling depends on kainate receptors, whereas On signalling is mediated primarily by the glutamate transporter EAAT5b. Blocking either pathway or removing inhibition unmasks strong latent responses of the opposite polarity. Strikingly, additional sign inversions re-emerge downstream via Group III mGluR, and removing inhibition at the retinal output and in central visual circuits reveals widespread mixed-polarity signalling.
Together, our results show that On–Off coding is latent, whereas clean On or Off responses emerge through active suppression of one polarity. We suggest that the On/Off segregation prominent in mammals represents a derived specialization shaped by a nocturnal bottleneck where low-light constraints favoured early, robust polarity splitting. Polarity segregation is therefore repeatedly reconstructed across processing stages, challenging the canonical model of a single early pathway split.

On–Off polarity in zebrafish vision is not fixed at the photoreceptor–bipolar synapse but arises dynamically through multiple sign inversions and context-dependent inhibition. Panel 1: Conceptual model The left schematic illustrates the classical view where photoreceptors drive distinct On or Off bipolar cells, which in turn feed ganglion and central neurons that inherit this segregated polarity. The right schematic proposes a dynamic polarity model in which many bipolar cells are intrinsically On–Off, additional inversions occur downstream, and inhibitory circuits selectively suppress one polarity to generate clean On or Off outputs when needed. Panel 2: Bipolar cell imaging The middle panels show population calcium signals in zebrafish bipolar cell terminals during light onset (red) and offset (green) under different manipulations. In control conditions, activity segregates into discrete On and Off layers, but blocking kainate receptors (ACET), removing EAAT5b, or pharmacologically blocking amacrine-cell inhibition unmasks widespread mixed On–Off responses within individual bipolar channels. Panel 3: Brain neuron imaging The lower panels display light-evoked activity in downstream brain regions, again separated into light-on (red) and light-off (green) components and their merge. When inhibition is blocked at retinal output and in central circuits, many brain neurons also become On–Off, revealing latent mixed-polarity signalling beyond the retina. Overall, the figure illustrates that On–Off coding is broadly latent across the visual pathway, and that apparently pure On or Off responses emerge through active inhibitory suppression of the opposite polarity at multiple processing stages.

Recommended posters

Cookies

We use essential cookies to run the site. Analytics cookies are optional and help us improve World Wide. Learn more.