Millington-Truby, R. | Search bioRxiv

bioRxiv2026

Quantifying visual pathway overlap to identify training-related microstructural change in hemianopia

BackgroundVisual training can improve residual vision in hemianopia, but there is significant variability in patient outcomes which may reflect differences in neural structures preserved after injury. Diffusion MRI tractography offers a means of assessing white-matter microstructure in vivo, however substantial overlap between neighbouring visual pathways complicates attribution of differences to specific tracts. We quantified tract overlap to identify relatively unique pathway segments to assess training-related microstructural change. MethodsDiffusion MRI data was acquired in six participants with homonymous hemianopia before and after 3-6 months of visual training. Tractography was used to reconstruct three pathways implicated in residual vision: lateral geniculate nucleus (LGN)-V1, LGN-V5 and pulvinar-V5, together with secondary superior colliculus-LGN and superior colliculus-pulvinar pathways. Streamline overlap was quantified along neighbouring tracts, and comparison of pre- to post-training fractional anisotropy was restricted to unique or pair-specific non-overlapping segments. Exploratory relationships with visual outcomes were also examined. ResultsFractional anisotropy increased after training in both early and late pair-specific segments of the LGN-V5 pathway in the lesioned hemisphere. In the late segment, this increase was significantly greater than in the corresponding unique LGN-V1 segment, whereas change in the early LGN-V5 segment did not differ significantly from the unique pulvinar-V5 segment. No significant training-related changes were observed in collicular pathways. Brain-behaviour analyses also preferentially implicated LGN-V5: baseline fractional anisotropy was strongly associated with post-training Gabor detection, and training-related fractional anisotropy change showed a positive association with improvement in Gabor detection, although the latter did not survive multiple comparison correction. ConclusionThese findings support a role for the LGN-V5 pathway in visual plasticity after hemianopia and demonstrate the importance of explicitly quantifying tract overlap when attributing microstructural change to small, neighbouring white-matter pathways.

neuroscience↗

bioRxiv2026

Generalisation of training-induced recovery in occipital stroke: neurochemical and fMRI correlates

BACKGROUNDDamage to the early visual cortex after an occipital stroke typically results in the loss of conscious vision in the contralateral hemifield. Nonetheless, extensive perceptual training can restore visual motion discrimination in the blind-field. Here, we assessed, in a cohort study, whether improvements transferred to an untrained Gabor detection task and whether awareness within the blind field increased. We then explored the neural underpinnings of these changes. METHODSEighteen participants (6 female; aged 24-74 years; >6 months post-stroke) completed at least six months of visual rehabilitation in their blind field. Rehabilitation consisted of participants practicing a two-alternative, forced-choice, motion discrimination task using random dot stimuli, five days/week, at home, at one or two non-overlapping, locations in their blind-field. Each participant also completed two in-lab visits: one pre- and one post-training. A subset returned to the lab for a follow-up visit three months later to assess persistence of recovery. In addition to the trained task, an untrained, drifting-Gabor detection task was used to measure transfer of learning and changes in visual awareness at the trained locations. To investigate neural mechanisms underlying generalisation of improvements, participants completed MRI scanning at each lab visit. Magnetic resonance spectroscopy (MRS) was used to quantify GABA and glutamate concentrations in the ipsilesional motion sensitive area, hMT+, and a control voxel in the sensorimotor cortex. Functional MRI was conducted to assess BOLD signal changes in hMT+ and across the rest of the brain during passive viewing of high contrast Gabor stimuli in the blind field. RESULTSParticipants showed significant improvements in motion direction discrimination (trained task) between pre- and post-training in-lab visits, which generalised to improvements in Gabor detection and awareness (untrained task). Reduced GABA and glutamate in ipsilesional hMT+ was linked to improved Gabor detection, but not awareness. Increased BOLD signals in hMT+ and dorsolateral prefrontal cortex also correlated with improved Gabor detection, while awareness changes were linked to higher-level areas associated with visual attention in the contralesional prefrontal cortex (area 46) and inferior parietal lobule. CONCLUSIONSLong-term visual rehabilitation using a global motion discrimination task generalised to enhance both detection and awareness of moving Gabors within the blind field of occipital stroke survivors. Improvements were supported by selective changes in brain regions known to be involved in motion perception and attention respectively, suggesting that a broad network supports recovery, which could be targeted to enhance outcomes.

neuroscience↗