Alzheimer’s disease is defined by two interconnected molecular cascades: the extracellular deposition of amyloid-beta peptides and the intraneuronal aggregation of hyperphosphorylated tau into neurofibrillary tangles. Neither pathway respects the blood-retina barrier. Aβ deposits have been identified in the retinas of AD patients and transgenic mice, including within the RGC layer and the optic nerve (Koronyo-Hamaoui et al., 2011, Neuroimage). Tau hyperphosphorylation impairs axonal transport in retinofugal axons, contributing to progressive optic nerve dysfunction in parallel with the broader synaptic and neuronal pathology in the brain. The key research challenge is translating these molecular observations into quantifiable functional endpoints that can be measured serially in the same animal, tracked against disease progression, and used to evaluate candidate therapeutics.

Conventional approaches – electroretinography (ERG), visual evoked potentials (VEPs), pattern ERG, and histological RGC quantification – are valuable but limited for this purpose. ERG requires anaesthesia and skilled preparation; VEPs typically require cortical electrode implantation; and histology provides only a single, terminal snapshot. These methods are poorly suited to the repeated longitudinal measurement that Aβ- or tau-tracking studies demand, particularly in aged animals where handling stress is a compounding variable. An automated, training-free, non-invasive paradigm that captures functional retinal output is therefore a practical necessity for this application area.

Researchers focused on non-visual aspects of Alzheimer’s disease should note that the optomotor reflex is mediated by the accessory optic system and the nucleus of the optic tract – a subcortical pathway that does not require and does not report on cortical visual processing. Studies documenting cortical visual plasticity changes in tau models (including visual cortex orientation selectivity and ocular dominance) are scientifically relevant context, but those measurements require separate methods such as VEPs or single-unit recordings. The OptoDrum specifically captures retinal and retino-brainstem pathway integrity. For cortical visual function assessment, AcuiSee – which uses an operant, reward-based discrimination paradigm requiring cortical visual processing – is the appropriate Striatech instrument.

For a broader discussion of aging-related visual decline in the context of neurodegeneration, see also the Systemic Aging & CNS Decline application page.

One of the central unresolved problems in Alzheimer’s disease research – and in neurodegenerative disease more broadly – is identifying biomarkers that report on disease status before significant neuronal loss has occurred. In the AD visual pathway, optic nerve beta-amyloid deposition and axonal transport failure are among the earliest detectable pathological events, potentially preceding cortical amyloid accumulation and cognitive symptoms (Hinton et al., 1986, N. Engl. J. Med.). Similarly, the differential vulnerability of RGC subtypes – with intrinsically photosensitive RGCs (ipRGCs) showing relative preservation relative to other RGC populations in some AD models (La Morgia et al., 2016, Ann. Neurol.) – means that visual function measurements may provide a functional stratification signal with diagnostic or staging utility.

The challenge for researchers is translating structural imaging findings (OCT, confocal retinal imaging, immunohistochemistry) into quantitative functional endpoints that can be obtained longitudinally in the same animal without sacrifice. Structural approaches confirm that the retina changes in AD but do not indicate whether those changes translate into a physiologically meaningful circuit-level deficit. Functional visual assessment closes this gap. Furthermore, understanding which RGC subtypes drive or sustain optomotor responses helps interpret why functional decline may lag behind or precede structural measures in AD models.

For a broader exploration of retinal degeneration mechanisms relevant to this question, see https://stria.tech/application/retinal-degeneration. For a detailed discussion of optic nerve axon damage across disease contexts, see https://stria.tech/application/optic-nerve-damage. For RGC-specific dysfunction, see https://stria.tech/application/retinal-ganglion-cell-dysfunction. For changes shared between AD and age-related macular degeneration, see https://stria.tech/application/age-related-macular-degeneration. (Note: some of these cluster pages may be coming soon – check availability at the linked URL.)

Parkinson’s disease is associated with measurable visual dysfunction in patients – including reduced contrast sensitivity, colour discrimination deficits, and electroretinographic changes – that predate or accompany motor symptoms (Archibald et al., 2009, Brain). The genetic basis of PD has expanded substantially over the past two decades. VPS35, encoding a core component of the retromer complex responsible for endosomal protein sorting and recycling, was identified as a rare but penetrant PD risk gene (Vilariño-Güell et al., 2011, Am. J. Hum. Genet.). Retromer dysfunction impairs lysosomal and autophagic clearance pathways central to α-synuclein turnover. When VPS35 is deleted specifically in rod photoreceptors, retinal degeneration ensues – and critically, this degeneration has a measurable functional correlate that is only detectable if both photopic and scotopic visual function are assessed.

Standard optomotor testing conducted under photopic conditions alone would miss rod-specific visual dysfunction that is the primary consequence of rod-targeted gene mutations. Scotopic testing with the ScotopicKit, following complete dark adaptation in the DarkAdapt box, isolates rod photoreceptor-mediated visual function and provides a distinct endpoint from photopic acuity. This dual-modality approach is essential not only for VPS35 models but for any PD study in which retinal dopamine depletion or photoreceptor-specific pathology may differentially affect rod and cone pathways. Researchers studying MPTP, rotenone, or α-synuclein overexpression models should similarly consider whether scotopic and photopic endpoints diverge, as this divergence itself carries mechanistic information.

For researchers studying neuroinflammatory contributions to PD-related retinal degeneration (the Fu et al. (2024) model also showed neuroinflammatory infiltration), see the related discussion of neuroinflammation on this page (below) and the Neuroinflammation & Autoimmune CNS Disease application page. For rod-specific night vision disorders more broadly, see https://stria.tech/application/night-vision (coming soon).

Neurodegenerative disease models are inherently longitudinal: disease develops over weeks to months, treatments must be evaluated across extended time courses, and identifying the optimal treatment window requires dense, early-stage data. Yet most standard preclinical endpoints are either terminal (histology, biochemistry from brain tissue) or technically demanding to repeat (ERG, VEP, MRI, cognitive behavioural tests). Morris water maze, for example, introduces substantial handling stress and confounds motor impairment with cognitive readout; it cannot be conducted daily and provides poor temporal resolution of the slow functional decline typical of most AD or PD models. Pattern ERG, which provides retinal functional information, requires anaesthesia, pupil dilation, and corneal electrodes – procedural demands that compound across repeated sessions and produce variability that obscures subtle early changes.

Automated optomotor testing with the OptoDrum addresses this problem directly. Animals are placed on a central platform in an enclosed arena; no restraint, electrode placement, injection, or training is involved. The automated threshold determination algorithm requires approximately four minutes per animal and produces consistent, operator-independent results (Prusky et al., 2004, Invest. Ophthalmol. Vis. Sci.). This makes it feasible to establish a high-resolution temporal profile of visual function across the full disease course in every individual animal, not only at selected group-average time points. Dense longitudinal profiles enable precise identification of the onset of functional decline, the plateau of degeneration, and – when interventions are applied – the time points at which therapeutic effects emerge or fade.

In aged animals or animals with systemic disease, handling stress is a significant source of measurement variability and animal welfare concern. The Non-aversive Animal Platform, introduced at ARVO 2025, minimises this by allowing animals to enter the testing platform voluntarily from their home cage via an innovative tunnel-lid design, eliminating the need for forced handling. This is particularly relevant for neurodegenerative disease models that are tested repeatedly from young adulthood through advanced age. For studies considering aging as a co-variable, see also the Systemic Aging & CNS Decline application page and the aging cluster page (coming soon).

Neuroinflammation is not a secondary consequence of neurodegeneration – it is a co-driver of disease progression. In Alzheimer’s disease, activated microglia cluster around amyloid plaques, release pro-inflammatory cytokines including TNF-α, IL-1β, and IL-6, and activate the complement cascade; this inflammatory milieu accelerates RGC dysfunction and retinal degeneration independently of direct amyloid toxicity (Heneka et al., 2015, Lancet Neurol.). In Parkinson’s disease, microglial activation in the substantia nigra has long been recognised, but neuroinflammation also occurs in the retina: microglial cells adopt activated morphologies in dopaminergic amacrine cell-depleted retinas, and the Fu et al. (2024) VPS35 model demonstrated that rod-specific gene deletion triggers a neuroinflammatory response alongside retinal degeneration.

The mechanistic question for researchers is whether inflammation is driving, mediating, or simply coinciding with retinal visual pathway damage in their model. OptoDrum-based functional testing enables researchers to ask precisely this question: if suppressing a specific inflammatory target (for example, microglial activation with PLX5622, complement inhibition, or cytokine blockade) preserves or rescues optomotor visual acuity, the functional link between that inflammatory pathway and visual circuit damage is established. This approach has been productively applied across the neuroinflammation literature (see the Neuroinflammation & Autoimmune CNS Disease application page, which catalogues 21 Striatech publications using OptoDrum to assess neuroinflammatory visual pathway damage across multiple disease contexts).

Microglia-mediated neurotoxicity operates through both direct synaptic stripping and indirect oxidative and excitotoxic mechanisms (Block et al., 2007, Nat. Rev. Neurosci.). In the retina, these mechanisms converge on RGC survival and on the integrity of the inner plexiform layer synapses through which RGCs receive input from bipolar cells. Functional visual testing therefore captures an integrated downstream output of the entire inflammatory cascade, rather than any single molecular intermediate. For a focused cluster-level treatment of neuroinflammation as a research topic across disease contexts, see https://stria.tech/application/neuroinflammation (coming soon).

Therapeutic development for Alzheimer’s and Parkinson’s disease requires efficacy endpoints that are sensitive, reproducible, and measurable in the same animal at multiple time points. Cognitive behavioural tests (Morris water maze, novel object recognition) are confounded by motor impairment, stress, and motivational variation; they are poorly suited to aged animals or animals with systemic burden of disease. Brain tissue endpoints (amyloid ELISA, tau western blot, RNAseq from hippocampal extracts) are terminal and provide no information on functional circuit-level effects of the treatment. Visual function testing bridges this gap: it reports on the functional integrity of a defined CNS circuit (the retinofugal pathway), can be repeated indefinitely in the same animal, and produces a continuous quantitative scale (cycles per degree; contrast sensitivity threshold) rather than a categorical or subjective score.

For neurodegenerative disease studies, the visual endpoint carries additional translational relevance: retinal pathology in AD and PD is increasingly viewed not only as an epiphenomenon of brain disease but as a primary target for neuroprotective interventions in its own right. Compounds that rescue RGC survival or optic nerve integrity in AD or PD models may have translational value for the co-occurring visual impairment experienced by patients. Integrating OptoDrum-based visual endpoints into drug development pipelines therefore adds scientific value at two levels: as a functional biomarker for overall CNS treatment response and as a direct efficacy measure for the visual pathway.

For a comprehensive catalogue of therapeutic rescue approaches evaluated with Striatech visual function endpoints – including gene therapy, optogenetics, neuroprotection, and stem cell strategies – see the Maintaining & Restoring Vision application page, which covers 13 publications across these modalities.