The retina is regarded as a specialized sensory neural tissue capable of detecting optical defocus signals and generating molecular signals specific to the type of defocus (11). This capability allows the retina to determine whether images are accurately focused on it and to locally regulate growth-related optical changes in the eye (12, 13). Applying positive lenses to the eye induces myopic defocus, causing images to focus in front of the retina. This leads to the inhibition of axial elongation and the development of hyperopia. Conversely, negative lenses induce hyperopic defocus when applied to the eye, resulting in axial elongation and myopia (14). The involvement of the retina in the formation and development of myopia encompasses a variety of molecular, cellular, and physiological mechanisms.

The sclera is a highly resilient and structurally complex connective tissue whose primary function is to provide a firm and stable environment for the retina. When there are defocus factors or other visual stimuli such as FD, light intensity, contrast sensitivity, and other factors, they trigger a retinal cascade response that ultimately transmits to the sclera (60, 61). This leads to the expression of related scleral genes, such as HIF-1α, MMPs, TIMPs, and TGF-β, resulting in corresponding changes in scleral structure, biomechanics, and composition, a process known as scleral remodeling. Wu discovered that signaling of HIF-1α promotes myopia by causing fibroblasts to transform into myofibroblasts, and that treatment targeting hypoxia prevented the molecular changes associated with HIF-1α, thus stopping the progression of myopia (62).

The intraocular mechanisms are also reflected in the localized response within the eye. Even when the optic nerve is transected or the RGC action potential is chemically blocked, preventing signal generation and transmission to the brain, FDM still occurs (8, 63). Furthermore, if diffusers or negative lenses are applied to cover only half of the retina, only the covered half of the eye becomes enlarged and develops myopia (64), and if positive lenses cover only half of the retina, only that covered half exhibits inhibited eye growth.

Taken together, the intraocular mechanisms of eye growth regulation are primarily mediated through a retinal signaling cascade, which subsequently influences the RPE/choroid and ultimately the sclera, acting as an effector to regulates eye growth through scleral remodeling (65).

With the advancement of imaging technologies, our understanding of myopia has been revolutionized, allowing for a detailed investigation of the brain’s involvement in myopia. Table 1 shows the results of the clinical findings between myopia and brain regions. Functional magnetic resonance imaging (fMRI), resting-state fMRI (rs-fMRI), and amplitude of low-frequency fluctuations (ALFF) have been pivotal in elucidating the central neural mechanisms associated with myopia. These advanced imaging techniques provide comprehensive insights into how myopia impacts brain structure and function, thereby enhancing our understanding of its underlying pathophysiology.

Li et al. conducted a voxel-based analysis to compare regional gray matter and white matter concentrations in HM patients versus controls (86). The study found an increased concentration of white matter in HM patients, primarily in the calcarine area, with smaller increases in the prefrontal and parietal lobes. Wu’s study provided evidence of cortical thickness reduction and disconnection in visual centers and processing areas in HM patients. Additionally, an increase in cortical thickness was observed in the left multimodal integration region (87).

Mirzajani et al. demonstrated that blur induced by myopia significantly impacts fMRI experimental outcomes. Their study explored how mild and moderate lens-induced defocus affected the activity of the occipital visual cortex when exposed to a visual stimulus with moderate spatial frequency. They found that even a slight induced myopia of +1D notably influences visual cortex activity (88). Building on this, Mirzajani examined the effects of lens-induced high myopia on occipital visual cortex activity with two different spatial frequency visual stimuli. For a stimulus with 1.84 cycles per degree, lens-induced high myopia significantly reduced visual cortex activity (p = 0.01). However, for a 0.34 cycles per degree stimulus, there was no significant effect [p = 0.17; (89)]. Overall, these studies emphasize the significant effect of myopia on visual cortex activity.

Nelles et al. examined how refractive error affects cortical regulation of an oculomotor task using fMRI. The results showed increased activation in the bilateral frontal and parietal eye fields, supplementary eye fields, and bilateral extrastriate cortex among individuals with refractive errors compared to those with normal vision (90). Cheng et al. compared intrinsic brain activity in individuals with low/moderate myopia versus HM using the ALFF method (91). The findings indicated that individuals with mild to moderate myopia and HM had abnormal intrinsic brain activities in regions associated with the limbic system, default mode network, and thalamo-occipital pathway. Kang et al. assessed neural changes induced by myopic and hyperopic defocus stimuli. They found that myopic defocus significantly increased regional cerebral blood flow in the right precentral gyrus, right superior temporal gyrus, left inferior parietal lobule, and left middle temporal gyrus (66), indicating enhanced blood perfusion in visual attention-related regions.

Clinical evidence strongly supports the association between myopia and central neural changes. Consistent findings across different imaging modalities suggest that myopia involves both functional and structural alterations in the brain. These changes may reflect compensatory mechanisms, neural plasticity, or disruptions in normal brain function due to altered visual input.

Animal experiments that sever the connection between the retina and brain reveal that the regulation of visually-driven eye growth and emmetropization is primarily governed by intraocular mechanisms. For instance, Raviola and Wiesel demonstrated that lid-fusion myopia developed in Macaca mulatta following optic nerve section (ONS) surgery (92). Subsequent studies using the chick model revealed that performing ONS surgery did not inhibit the progression of FDM or lens-induced myopia (93, 94), suggesting that central communication is not essential for the development of myopia. However, there is also evidence suggesting that central mechanisms play a regulatory role in emmetropization. Gong’s research using optic nerve crush (ONC) to block visual input in mice found that recalibrating the refractive set-point in both directions led to significant refractive changes in the majority of animals. Specifically, 54.5% developed significant myopia (<−3 D) and 18.2% exhibited significant hyperopia (> + 3 D), primarily due to changes in ocular AL (10). McFadden discovered that the gain control in response to FD was significantly altered by ONS. Compared to the sham group, ONS markedly enhanced the response to FD. These findings suggest that advanced visual centers are probably involved in the fine-tuning of eye growth and thereby influence refractive status (9). Dillingham conducted experiments on chicks using electrolytic lesions to target the IOTr nucleus, a limited subset of ectopic neurons analogous to the neurons of the mammalian superior colliculus (SC) (95). The findings revealed that chicks with the highest percentage of successful lesions showed notable axial hyperopia in the treated eye relative to the control eye.

In addition to physical methods, chemical approaches also support this hypothesis. For instance, the injection of TTX or colchicine, which blocks the generation of RGC action potentials, and thereby prevents the transmission of information to higher visual centers, has been shown to have similar effects (96, 97). Norton’s research on tree shrews demonstrated that intravitreal injection of TTX resulted in a significant shift toward hyperopia (8). In normal visual environments, intravitreal injection of colchicine in chicks led to axial elongation and the development of myopia (97, 98). Liu’s research on developing mice demonstrated that selective ablation and activation of ipRGCs induced opposite refractive shifts, leading to myopia and hyperopia. These changes were achieved by modulating the CRC and AL in mice. These experiments collectively suggest that changes in the integrity of eye-brain neural connections can affect refractive status. However, the broad and non-specific nature of methods prevents the identification of specific brain nuclei involved in this process. The close proximity of the ophthalmic artery to the optic nerve raises concerns about potential impacts on blood supply (99). Meanwhile, ONS surgery not only disrupts RGCs efferents to the primary visual centers of the brain but also midbrain centrifugal visual system axons terminating in the retina. Chemical methods, such as using TTX or colchicine to block action potentials, also affect ON and OFF cone bipolar cells, amacrine cells, and ganglion cells due to their influence on sodium ion channels (100). Moreover, the need for continuous daily intravitreal injections of TTX can exert mechanical pressure on the eye. Table 2 summarizes the results of animal models with interventions and refraction changes.

In avian species, centrifugal visual fibers originating from the brain and projecting to retinal neurons are involved in early eye growth. Although these fibers are very limited in number in mammals (101), they exhibit extensive branching within the retina, encompassing almost the entire retinal region (102). Research has confirmed that these centrifugal visual fibers are dopaminergic and can interact with DACs (103), thereby regulating eye growth by modulating retinal DA release levels. Additionally, central mechanisms may influence eye growth through melanopsin-expressing ipRGCs, which project to the suprachiasmatic nucleus (SCN) and are involved in light-induced circadian rhythms. Current literature indicates that normal refractive development relies on specific daily rhythms of eye growth, and disruption of these rhythms can lead to excessive axial elongation. Meanwhile, the involvement of ipRGCs on ocular growth may be linked to their extensive interaction with retinal blood vessels. Recent studies have shown that some ipRGC processes are near the intermediate capillary plexus and that melanopsin mediates light-induced relaxation in blood vessels (104, 105). Thirdly, degeneration of RGCs could impair the retina’s ability to detect optical defocus and modify the signals generated.

Visual information received by the eyes is transmitted through the visual signal pathways. The brain processes visual information in parallel at three distinct levels: individual neurons, cell types, and neural pathways. Information processing at the cell type level begins in the retina. Visual signals travel via the optic nerve, formed by the axons of RGCs. After partially crossing at the optic chiasm, these fibers form the optic tract, which then transmits the signals to various brain nuclei such as the lateral geniculate nucleus (LGN), SC, SCN, and visual cortex.

The SC is a conserved visual center within the mammalian visual system. It not only receives projections from RGCs but also integrates input from the visual cortex, serving as a crucial hub for information processing and integration (106, 107). Approximately 90% of ganglion cells project directly to the SC (108), which is organized into alternating layers of fibers and cell bodies. The superficial layers receive visual signals, while the intermediate and deep layers process inputs from primary motor, somatosensory, and auditory cortices. The SC mediates various physiological functions, including eye movements, innate fear responses, and sleep regulation (109, 110). In most vertebrates, the extensive inputs and outputs of the SC suggest that the SC can influence almost the entire neural axis. Lee investigated how induced blur affects the response efficiency of midbrain cells. While a small subset of the cells showed minimal changes in their response patterns when subjected to either hyperopic or myopic blur, approximately 86% experienced a significant decline in response efficiency (111). Gehr’s research demonstrated that both excitatory and inhibitory neurons in the superior colliculus receive similarly robust inputs from retinal ganglion cells, and that the same wiring principles govern RGC innervation of both types of SC neurons (112). Li′s study indicated that the superficial layers of the superior colliculus respond to various visual stimuli, such as blue and green flashes (113). Additionally, previous research suggested that flickering blue light may play a role in the progression of myopia (60). Therefore, it is conceivable that the SC may play a role in the development of myopia.

The LGN serves as a critical relay station for visual signals entering the visual cortex. It is responsible for further processing of visual information, where the signals are analyzed, encoded, and then transmitted to the primary visual cortex via the optic radiations. The mammalian retina transmits the majority of visual stimulus information to two brain regions: the dorsal lateral geniculate nucleus (dLGN) and the SC. Studies have shown that SC neurons can acquire most of the information sent by the retina to the dLGN, but not vice versa. Animal studies have demonstrated a reduction in the size of ocular dominance columns within the primary visual cortex, as well as cellular atrophy in the layers of the LGN corresponding to the deprived eye (114).

The SCN is closely linked to circadian rhythms and has a well-established connection to visual functions. Previous studies have demonstrated that disruptions in the eye’s daily growth rhythms can lead to excessive axial elongation (115). It has also been reported that the amplitude of diurnal variations in the choroid is correlated with axial length in adults with myopia. This suggests that circadian rhythm disruption may be involved in the pathogenesis of myopia (116). Previous research has been documented that M1 type ipRGCs primarily project to non-image forming visual regions of the brain, including the SCN (117). Importantly, M2 type ipRGCs were observed to project to both non-image forming visual regions, such as the SCN, as well as image-forming areas like the dLGN and SC (118). Additionally, Liu’s research revealed that cell type–specific ablation studies demonstrated that M1 subtype cells, and potentially M2/M3 subtype cells, play a role in ocular development (45). Li proposed a hypothesis that ocular rhythms are regulated locally and indirectly through the SCN, which receives input from ipRGCs (119). In mammals, the central circadian pacemaker is located in the SCN (120). It has been reported that the SCN, upon receiving light and time-of-day information from ipRGCs, initiates the central clock. This provides an anatomical basis and a potential mechanism for the involvement of the ipRGC-SCN circuit in neuronal regulation.

The primary visual cortex (V1) receives input from the dLGN and the SC (121). Additionally, V1 plays a crucial role in spatial frequency selectivity, with studies indicating that varying spatial frequencies can influence the development of myopia (122). Moreover, V1 is involved in the processing of ocular dominance and has been shown to play a role in the formation of amblyopia. Zhao’s study demonstrated that in guinea pigs with concave lens-induced myopia, the levels of GABA and the mRNA of GABA receptors in the visual cortex were elevated (123). Nakadate’s study discovered that monocular deprivation during the critical period of ocular dominance plasticity markedly altered both the quantity and pattern of c-Fos expression in the visual cortex. The most sensitive indicator was the number of c-Fos positive cells in layer IV of the binocular subfields of the primary visual cortex (Oc1B) ipsilateral to the stimulated eye (124). A reduction in the plasticity of neurons in the visual cortex can lead to deficits in contrast sensitivity, stereopsis, eye movement, and visual attention capabilities (125). Current research indicates that contrast sensitivity plays a role in the development of myopia (126). Figure 1 illustrates intraocular structures and potential central pathways involved in myopia.

In conclusion, current research indicates that central mechanisms play a significant role in the development and progression of myopia. Historically, the limitations of available methods have hindered our understanding of these central mechanisms. However, recent advancements in exploring neural circuits provide an opportunity to identify specific brain nuclei and neuron types involved in myopia. Additionally, investigating the potential interactions between central and intraocular mechanisms could offer deeper insights into the complex pathophysiology of myopia. Future studies should focus on utilizing these advanced techniques to unravel the precise contributions of central factors and their interplay with intraocular processes, it is hoped that artificial creation of optical defocus cues could be applied in humans, with the signals transmitted from the retinal ganglion cells to the brain to help control myopia progression. Such approaches would be non-invasive, safe, and convenient. It has been demonstrated that artificially altering defocus images can compensate for changes in eye growth (127–130). Additionally, besides defocus cues, other visual signals, such as contrast sensitivity (126, 131) and spatial frequency (132, 133), could also potentially play a role.