Rebecca Willis¹ (rebecca.willis@ndcn.ox.ac.uk), Megan Groombridge¹, Aaron T. Hess¹, William T. Clarke¹, Juliet Semple², Jon Campbell¹, Pete R. Jones^3,4, Tessa M. Dekker^4,5, Dinesh K. Deelchand⁶, Pierre-Gilles Henry⁶, Ravi Purohit⁷, Holly Bridge¹, I. Betina Ip¹; ¹The University of Oxford Centre for Integrative Neuroimaging, FMRIB, Nuffield Department of Clinical Neurosciences, University of Oxford, UK, ²Oxford Centre for Human Brain Activity, Department of Psychiatry, University of Oxford, UK, ³Department of Optometry and Visual Sciences; School of Health and Medical Sciences; City St George’s, University of London, UK, ⁴Institute of Ophthalmology, University College London, UK, ⁵Experimental Psychology, Division of Psychology and Language Sciences, University College London, UK, ⁶Center for Magnetic Resonance Research, Department of Radiology, University of Minnesota, Minneapolis, MN, USA, ⁷Oxford Eye Hospital, Oxford University Hospitals NHS Trust, UK

Amblyopia is a common neurodevelopmental condition, marked by extreme eye dominance. While amblyopia in adults has been associated with GABAergic inhibition in the visual cortex (Mukerji et al. 2022; Ip et al. 2024), the role of GABA in childhood amblyopia is unexplored. We created a child-friendly protocol to measure GABAergic inhibition in children. Here we present the protocol and preliminary data from controls. 5-8 year old participants have two visits six months apart, between which amblyopes have occlusion therapy and controls have no intervention. Each visit includes an MRI scan and vision tests. The MRI comprises motion corrected T1-structural and magnetic resonance spectroscopy (MRS) sequences, and resting state fMRI. MRS is conducted in early visual cortex (EVC) and sensorimotor cortex (M1 control) voxels. Child-friendly vision tests are used to measure visual acuity (Keeler Crowded LogMAR), contrast sensitivity (PopCSF, Elfadaly et al. 2020 and FrACT, Bach 1996) and stereopsis (Frisby; Asteroid, Vancleef et al 2019). In controls (N=14, mean age=6.96 years), 91.3% of attempted MRS scans resulted in data acquisition. Of those data, 85.7% of scans could be analysed without further processing and are reported here. When analysed using a standard basis set, GABA+ concentration was comparable between voxels (F(1,3)=4.32, p=0.13) and timepoints (F(1,3)=3.12, p=0.09). Absolute CRLB indicated that MRS data quality was good (mean= 0.53, SD=0.70) and did not differ between voxels (F(1,3)= 0.79, p= 0.44) or timepoints (F(1,3)= 1.32, p= 0.33). Visual acuity (logMAR mean=-0.05, SD=0.08), contrast sensitivity (popCSF AUCSF, mean=29.4, SD=12.8), and stereopsis (Asteroid arcsec=77.8, SD=90.6) were within the range for normal vision. These preliminary results demonstrate that our protocol can reliably obtain high quality non-invasive measures of GABA+ concentrations from 5-8 year old children. This feasibility is an important step for improved understanding of neuroplasticity during the critical period in humans.

Acknowledgements: This research is funded by The Royal Society (studentship to R.W., Dorothy Hodgkin Research Fellowship to I.B.I.) and the Medical Research Council (MR/V034723/1).