A patient sits in your chair with 20/20 acuity, clean ocular health findings, and a chief complaint that doesn't match the chart: dizziness in crowded stores, motion sickness on screens, fatigue after a normal day of reading. The eyes test fine. The patient does not feel fine.
This is one of the more common diagnostic puzzles in functional vision care, and it usually resolves once you stop asking "what's wrong with the eyes" and start asking a different question: how is this brain weighting its sensory input right now?
Perception Is Evidence, Not Truth
The visual system does not hand the brain a finished picture of the world. It hands the brain a stream of evidence, and that evidence gets combined with auditory input and proprioceptive input from muscles, joints, and skin before anything resembling "perception" emerges. When all three streams agree, the result feels effortless and stable. When they disagree, the nervous system has to decide which signal to trust.
That decision process has a name: sensory weighting, sometimes called sensory reweighting in the balance literature. It is not a fixed hierarchy where vision always wins. It is a dynamic, context-dependent calculation. A patient relies more heavily on vision in a dark room, more heavily on proprioception when standing on firm ground, and more heavily on vestibular and auditory cues when visual information becomes noisy, delayed, or unreliable. The brain is constantly asking which channel currently deserves the most trust, and the answer changes from one moment to the next.
This reframes a lot of what shows up in a developmental optometry practice. A patient with motion sensitivity, postural instability, or post-concussion symptoms is rarely showing a purely ocular problem. More often, they are showing altered cue weighting: a visual system being asked to carry more than its fair share of the sensory load, or a system that has lost confidence in vision and is now over-relying on inputs that can't fully compensate.
How the Brain Decides Where Something Is
Some of the clearest evidence for this triad comes from how the brain localizes objects in space. Vision and audition answer the same underlying question — where is this relative to me — using completely different cue sets. Vision uses retinotopic position and scene structure. Audition uses interaural timing, intensity, and spectral differences between the two ears. Those signals converge in shared cortical and multisensory regions across the temporal, parietal, and frontal lobes, building a single, unified map of external space rather than two competing ones.
This is why vision can sharpen sound localization when the two cues agree, and why a sudden noise can reflexively pull the eyes and head toward it. The system is built for cooperation, not competition. One of the more elegant demonstrations of this is the ventriloquism effect: when a sound and a visual stimulus are spatially mismatched but temporally linked, the brain doesn't average the two locations. It shifts the perceived sound location toward the more reliable visual cue, attenuating the auditory spatial signal in the process. Repeated exposure to that mismatch can even recalibrate auditory localization afterward, evidence that this isn't a perceptual trick but genuine adaptive plasticity in how the brain weights its inputs over time.
The Midbrain's Coincidence Detector
A lot of this integration happens earlier and faster than conscious perception, in a structure worth knowing well: the superior colliculus. It contains neurons with overlapping visual, auditory, and somatosensory receptive fields, tuned so that a coincident stimulus across modalities produces a stronger orienting response than any single input alone. Its job isn't to interpret the world. It's to answer a faster, more primitive question: is something out there worth turning toward right now?
What makes this clinically relevant is that the colliculus depends on accurate spatial alignment between modalities. If eye position, head posture, or body position shifts, the correspondence between visual and auditory receptive fields can become mismatched, and multisensory integration changes accordingly. This is a structure that needs calibrated inputs to do its job well, which means postural or oculomotor dysfunction doesn't just affect vision in isolation. It can degrade the accuracy of the brain's most basic orienting reflex.
Keeping the World Stable While the Body Moves
The other piece of this triad worth understanding clinically is the posterior parietal cortex, which solves a different problem: how does the brain keep an object's location stable when the eyes, head, and hands are constantly moving relative to it? The PPC converts retinal, head-centered, and limb-centered information into the coordinates needed for action, using gain-modulated neurons whose responses shift with eye position, hand position, and intended movement.
This is the system that lets a patient saccade across a page without losing their place, or reach accurately for an object without watching their hand the whole way there. When it's compromised, the result often isn't a primary sensory loss. It's a mismatch between seeing and acting: trouble maintaining spatial awareness after an eye movement, or difficulty coordinating eye-hand behavior when visual and proprioceptive references stop agreeing with each other.
Learning to Read Is Learning to Match Sounds to Sights
Nowhere is the vision-hearing link more consequential than in learning to read, and nowhere does it become clearer that the pairing has to be built, not assumed. A child does not arrive at print already knowing that letters represent sounds. They first have to develop phonemic awareness, the ability to hear that a spoken word like cat is built from /k/ /a/ /t/, before that auditory skill has anything to attach to. Print mapping comes second. The oral foundation has to be in place first, or the visual symbol stays just that: a shape with no sound behind it.
Once phonemic awareness is established, the brain has to build an actual neural bridge between what the eyes see and what the ears have already learned to isolate. For regular, decodable words, that bridge runs along a left-lateralized dorsal pathway: the visual word form area in the fusiform gyrus identifies the letters, the signal travels through posterior superior temporal and temporo-parietal regions for phonological analysis, and from there into the inferior frontal gyrus for articulatory planning. It is, in effect, a dedicated decoding circuit connecting sight to sound, and it is the circuit doing most of the work for a beginning reader sounding out an unfamiliar word.
English does not let that circuit carry the whole burden. Words like said, laugh, and yacht break the rules the dorsal pathway depends on, and when the letters don't sound the way they look, the brain has to lean on a second route. The angular gyrus, sitting in the ventral pathway, binds orthography directly to stored phonological and semantic representations, allowing a word to be retrieved as a whole rather than sounded out piece by piece. This is the same kind of crossmodal binding seen in the ventriloquism effect or the McGurk effect, just applied to print instead of speech: a visual symbol and an auditory representation, learned together until the pairing becomes automatic enough to bypass step-by-step decoding entirely.
This has a direct clinical implication for anyone evaluating a struggling reader. A child can have flawless letter recognition and still be a poor reader if the auditory side of the mapping, the phonemic awareness, was never solid to begin with. Conversely, a child can hear sounds perfectly well and still struggle with irregular words if the ventral whole-word pathway hasn't had enough repetition to build a stable visual-auditory association. Reading difficulty is rarely a single-system failure. It is, again, a question of where the sight-to-sound link is breaking down, and which side of the pairing needs more support.
What This Means in the Chair
None of this is abstract neuroscience for its own sake. It's a working model for why a patient's symptoms can be multisensory even when every complaint sounds visual. Dizziness, postural instability, and reading difficulty can all reflect a breakdown in cue weighting rather than a single damaged sense. The visual system, the auditory system, and the proprioceptive system are not three separate patients sharing one body. They are three coordinated inputs into one ongoing estimate of where the person is and what's happening around them, an estimate the brain rebuilds and reweights continuously.
Even speech perception and reading follow this same logic. The McGurk effect, where mismatched lip movements change what a person hears, shows that the brain doesn't process audio and visual speech separately and compare notes afterward. It binds them into a single percept before the listener is even aware a conflict existed. Learning to read asks the brain to build that same kind of binding deliberately, pairing a visual symbol with a sound until the two are no longer experienced as separate. If hearing speech and reading print are already multisensory computations, it should be no surprise that balance and spatial orientation are too.
The clinical takeaway is that vision and hearing are not two separate channels that happen to share a brain. They are built to calibrate each other, localizing sound, stabilizing gaze, binding speech into a single percept, and mapping print to sound before any of it reaches conscious awareness. A patient or student whose visual and auditory cues have stopped agreeing, or have never been properly linked, isn't malfunctioning in one sense or the other. They're showing the seams of a partnership that normally runs invisibly in the background.