Roni Maimon-Mor^1,2 (), Mahtab Farahbakhsh^1,2, Andrew Rider¹, Mohamed Katta^1,3, Andrew Stockman¹, Michel Michaelides^1,3, Tessa Dekker^1,2; ¹UCL Institute of Ophthalmology, University College London, London EC1V 9EL, UK, ²Experimental Psychology, University College London, London WC1H 0AP, UK, ³Moorfields Eye Hospital, London EC1V 2PD, UK

Recent advances in cell- and gene-therapies make it increasingly possible to halt or reverse vision loss for many eye conditions. To develop effective treatments, we need to understand how rescued retinal signals integrate into developing brain function. One promising gene therapy treatment is for the congenital genetic condition achromatopsia, which causes retinal cone dysfunction. In a typically developing brain, cones are responsible for providing fast, chromatic information from across the retina, and all information from the fovea. In research alongside pioneering gene therapy trials, we measured cone recovery in 7 patients with achromatopsia treated between the ages of 8 and 16 years of age. This scenario offers an unprecedented opportunity to untangle the role of early experience in shaping neural specialisation for rod- and cone-mediated functions. Using photoreceptor-specific pRF mapping (fMRI), we found robust and consistent rescue of retinal cone-mediated signals in visual cortex in the majority of children, which was not present before treatment. Crucially, each of these patients could use these rescued signals to detect cone-selective stimuli previously invisible to them. Next, in two patients, we used high-precision psychophysics to determine which additional new visual functions the rescued retinal signals could support, focusing on cone-specific functions not available through rod-only vision: high-speed, foveal, and chromatic (colour) vision. After treatment, visual processing speed was substantially higher. Foveal cortex showed limited cone-mediated function, in line with previous reports of inconsistent benefits for acuity. Several complementary tests for colour vision (lvvCCT, Ishihara, Rayleigh, Stockman) revealed different recovery patterns and perceptual experiences across the two patients. Together this work provides novel insights into how a brain that has developed and organised with one set of retinal signals (rods) can incorporate a whole new parallel set of signals (cones) into its existing structures, and which novel functions can be restored.

Acknowledgements: NIHR Biomedical Research Centre at Moorfields Eye Hospital NHS Foundation Trust and UCL Institute of Ophthalmology, the Economic and Social Research Council (ESRC) of the UKRI; MeiraGTx, Retina UK, Moorfields Eye Charity, Foundation Fighting Blindness (USA) and The Wellcome Trust