Every time you move your eyes, speak, or reach for a cup of coffee, your brain quietly solves a problem that would stump most robots: it tells the difference between sensations you caused yourself and sensations caused by the outside world. The mechanism behind this feat is called corollary discharge, a copy of a motor command sent from movement-planning areas of the brain to sensory areas, essentially a heads-up that says “this is us, not them.” Without it, every eye movement would make the world appear to lurch, your own voice would startle you, and you would lose track of which actions were yours. The concept has been studied across species from crickets to primates, and when the system breaks down in humans, the consequences can be severe.
Why the World Stays Still When Your Eyes Move
You make rapid eye movements, called saccades, thousands of times a day. Each one drags the entire image on your retina across your photoreceptors at high speed. If you moved a camera that fast, the footage would be an unwatchable blur. Yet the world looks perfectly stable. One reason is that your brain briefly dials down visual sensitivity during each saccade, a phenomenon called saccadic suppression.1PubMed Central. Learning to silence saccadic suppression The timing of that suppression is not a coincidence. It is triggered by the corollary discharge that accompanies the command to move your eyes. Before the eyes even begin to rotate, the motor plan sends a copy of itself to visual processing areas, effectively warning them to expect a burst of retinal motion that should be ignored.
How the brain distinguishes self-generated motion from real-world motion at the neural level has been debated. Early work in monkeys suggested that neurons in the primary visual cortex could not tell the difference at all when given identical retinal input.2PubMed. Inability of rhesus monkey area V1 to discriminate between self-induced and externally induced retinal image slip But more recent experiments found that the same region actually does respond differently to real microsaccades compared to externally simulated ones. Real microsaccades produced a characteristic two-phase response: a quick burst of firing followed by a dip below baseline, while simulated motion produced only the burst.3PubMed Central. V1 neurons respond differently to object motion versus motion from eye movements That suppressive dip is the fingerprint of the corollary discharge arriving in the visual cortex, and it tells downstream brain areas “this motion was us.”
Why You Cannot Tickle Yourself
The visual system is the most intuitive place to see corollary discharge at work, but the same logic runs through every sense you use while acting. When you speak, your auditory cortex does not respond to your own voice with the same intensity it would give to someone else’s voice at the same volume. Neural activity across most auditory processing areas is suppressed during vocalization.4PubMed Central. Speech-induced suppression and vocal feedback sensitivity in human cortex This suppression is not because the sound is familiar or expected in a general sense; it is specifically driven by a prediction generated from the motor command to speak. When the sound that arrives at your ears matches the prediction, the response is dampened. When it does not match, the response stays large, which is exactly how you notice when your voice cracks or when audio feedback is delayed over a conference call.
The same principle explains the old party trick: you cannot tickle yourself. When your hand moves toward your own foot, the corollary discharge from the motor command predicts the exact sensation that will follow. Because the prediction matches the incoming touch, the sensory signal is suppressed. Someone else’s fingers generate the same physical contact on your skin, but without the internal prediction, so the sensation comes through at full strength and triggers the tickle response. This is not a metaphor. It is a measurable difference in neural activity between self-generated and externally generated touch.
The Cerebellum as Prediction Engine
If the brain is sending copies of motor commands to sensory areas, something needs to do the comparison between what was predicted and what actually happened. A large body of evidence points to the cerebellum as the key structure doing this work. The cerebellum sits in a unique position, receiving both copies of the motor command from the motor cortex and sensory feedback from the body, making it well suited to act as a comparator that detects mismatches between plan and outcome.5Frontiers in Systems Neuroscience. The Forward Model: A Unifying Theory for the Role of the Cerebellum in Motor Control and Sense of Agency Many models of cerebellar function borrow from control systems theory, treating the cerebellum as a processor that uses copies of motor commands to compute learned predictions for calibrating movements and adjusting sensory expectations.6PubMed Central. Corollary Discharge Signals in the Cerebellum
This comparison process is often described as a “forward model.” Before you act, the brain generates a prediction of what sensory consequences should follow. When you reach for a mug and your hand closes around it at the expected weight and angle, the forward model’s prediction closely matches reality, and the resulting sensation is dampened. But if the mug turns out to be empty when you expected it to be full, the prediction error is large, the sensory response is not suppressed, and you get that familiar jolt of surprise. The forward model feeds back into motor learning too: each prediction error helps the cerebellum refine future predictions, so the next time you lift a similar mug, the prediction is better calibrated.
The Sense of Agency
Corollary discharge does not just filter sensory noise. It also underlies your feeling that you are the one controlling your actions, a phenomenon researchers call the sense of agency. The comparison between predicted and actual outcomes from the forward model produces what is essentially a report card: if the prediction matched well, the brain registers the action as self-caused. If there is a significant mismatch, the brain starts to flag the action or its consequences as potentially external.5Frontiers in Systems Neuroscience. The Forward Model: A Unifying Theory for the Role of the Cerebellum in Motor Control and Sense of Agency This is a low-level, automatic process, not the philosophical musing about free will that “sense of agency” might imply. It is the fast, preconscious determination: “my hand did that.”
When this mechanism is working well, it is invisible. You never think about whether you caused your arm to move. But in neurological and psychiatric conditions where the system malfunctions, the absence of this automatic self-tagging becomes dramatically apparent.
What Bats, Crickets, and Electric Fish Reveal
Corollary discharge is not a luxury upgrade that appeared only in mammalian brains. It is found across the animal kingdom, shaped by the specific survival pressures of each species. Some of the most striking examples come from animals whose own signals would be catastrophically loud or confusing without it.
Echolocating bats face an extreme version of the self-generated-sound problem. They emit intense ultrasonic pulses and listen for faint echoes bouncing off insects or obstacles. If their brains responded to the outgoing pulse with the same sensitivity as the incoming echo, they would be effectively deafened at close range. Bats solve this with a neural attenuation mechanism synchronized to vocalization. In one species, neurons in a relay station between the ear and higher auditory areas reduced their response to self-generated sounds by about 15 decibels.7PubMed. Site of neural attenuation of responses to self-vocalized sounds in echolocating bats In another family of bats, the mean attenuation reached about 25 decibels, despite the auditory nerve itself responding identically to both emitted sounds and playback sounds.8PubMed. Neural attenuation of responses to emitted sounds in echolocating rats The suppression is happening centrally, not at the ear. The brain knows the pulse is coming because it sent the command to produce it.
Crickets face a similar dilemma. Male crickets stridulate loudly to attract mates, and their own chirps would overwhelm the auditory neurons they need for detecting predators and rivals. Recordings from ascending auditory neurons show that a corollary discharge from the chirping motor network actively inhibits these neurons during each chirp. The inhibition was present even during fictive chirps generated without actual sound, confirming it is driven by the motor command itself, not by acoustic feedback.9PubMed Central. Corollary discharge inhibition of ascending auditory neurons in the stridulating cricket
Weakly electric fish offer perhaps the richest natural laboratory for studying corollary discharge. Mormyrid electric fish generate electric organ discharges to sense their surroundings and communicate. Their brain must distinguish the sensory input caused by their own discharge from signals produced by other fish or environmental changes. Research on these animals has revealed at least three distinct corollary discharge functions: inhibition to suppress predictable self-generated input, enhancement to boost sensitivity at key moments, and the generation of a “negative image” prediction that cancels expected reafference.10PubMed Central. A History of Corollary Discharge: Contributions of Mormyrid Weakly Electric Fish In the electrosensory lobe, the output cells responsible for processing environmental signals showed dramatic cancellation of self-generated sensory input during free behavior, regardless of the fish’s position in its environment.11Neuron. State-dependent spatial predictions of self-generated sensory inputs in cerebellum-like circuitry The timing of corollary discharge signals in these fish even adapts across evolutionary timescales, with the onset and duration of the suppression correlating with the delay of returning sensory input across different species.12Current Biology. Shared neural substrates support corollary discharge plasticity and evolution across timescales in mormyrid electric fish
The breadth of species that rely on corollary discharge, from insects to fish to primates, suggests it is one of the most ancient and fundamental solutions to a problem that every nervous system with both motor output and sensory input must solve.
When the System Breaks Down in Schizophrenia
If corollary discharge is the brain’s mechanism for tagging its own actions and silencing the expected sensory consequences, then a failure in this system should produce a specific and disturbing symptom: difficulty telling self-generated experiences from external ones. That is precisely what has been proposed for auditory hallucinations in schizophrenia. The theory holds that when inner speech fails to generate a proper corollary discharge, the resulting auditory cortex activity is not tagged as self-produced. The brain then interprets it as an external voice.13PubMed. Corollary discharge dysfunction in schizophrenia: can it explain auditory hallucinations?
A recent study directly tested this. In healthy participants, when the auditory input matched what was predicted by inner speech preparation, the brain’s early auditory response was reduced, exactly what you would expect from functional corollary discharge. But in patients with schizophrenia who experienced auditory verbal hallucinations, the pattern flipped: the early auditory response was enhanced rather than suppressed when the sound matched the inner speech prediction.14PubMed. Corollary Discharge Dysfunction to Inner Speech and its Relationship to Auditory Verbal Hallucinations in Patients with Schizophrenia Spectrum Disorders Rather than dampening its response to self-generated signals, the brain was amplifying them. This provides direct electrophysiological support for the idea that hallucinations are partly a corollary discharge failure.
The implications extend beyond voices. Forward model impairment in schizophrenia has been linked to delusions of control, where patients feel their actions are being directed by an outside force.15JAMA Psychiatry. Functional Magnetic Resonance Imaging of Impaired Sensory Prediction in Schizophrenia If the prediction about your own movement’s consequences is consistently wrong, the mismatch signal that normally confirms “I did that” instead suggests “something else did that.” Behavioral studies have confirmed that patients with delusions of influence show deficits in comparing predicted and actual sensory outcomes of their own actions.16Brain. Misattributions of agency in schizophrenia are based on imprecise predictions about the sensory consequences of one’s actions The corollary discharge framework has opened interest in whether neuromodulation techniques that target these prediction circuits could become a treatment avenue for some of these symptoms.17PubMed Central. Schizophrenia and Corollary Discharge: A Neuroscientific Overview and Translational Implications
The evidence linking corollary discharge dysfunction to hallucinations is strong in concept and growing in empirical support, but researchers have been candid about the difficulty of nailing down the relationship. Earlier work documented clear abnormalities in physiological markers of the corollary discharge system in patients with schizophrenia but found only limited success in linking those abnormalities specifically to patients’ reported hallucination experiences.13PubMed. Corollary discharge dysfunction in schizophrenia: can it explain auditory hallucinations? Hallucinations are complex phenomena, and corollary discharge failure is likely one contributor rather than the whole story.
How the Brain Learns Its Own Timing
Corollary discharge is not a fixed, hardwired signal. The brain actively recalibrates the timing of its predictions based on experience, and the speed at which it does this is remarkable. In experiments where researchers artificially inserted a fixed delay between a participant’s keypress and the resulting sensory feedback, something strange happened after a few minutes of adaptation: when the delay was suddenly removed, the flash seemed to occur before the keypress. The brain had shifted its expectation of “when my action should produce a result” to accommodate the artificial lag, and when reality snapped back to normal, the original timing felt wrong.18PubMed. Motor-sensory recalibration leads to an illusory reversal of action and sensation
This temporal recalibration has been confirmed using brain imaging. After adapting to delayed visual feedback, participants’ thresholds for detecting delay shifted, and the effect was strongest when participants were actively performing the movement themselves rather than having their hand moved passively.19Scientific Reports. Temporal recalibration in response to delayed visual feedback of active versus passive actions: an fMRI study This makes sense under the forward model: active movements generate a corollary discharge that feeds the prediction, so there is more signal to recalibrate. Passive movements lack that internal prediction, giving the brain less to work with.
The recalibration ability is not fully mature in children. When adults were exposed to a 200-millisecond delay and then tested, their perception of the timing boundary between “my action came first” and “the sensation came first” shifted substantially. Eight-year-olds showed virtually no shift, and eleven-year-olds showed a much reduced one compared to adults.20PubMed Central. Children do not recalibrate motor-sensory temporal order after exposure to delayed sensory feedback This suggests that the forward model’s ability to flexibly update its timing predictions continues to develop through childhood, which has implications for understanding developmental conditions where the sense of agency or sensory integration feels off.
Corollary Discharge and Voice Control in Neurological Disease
Schizophrenia is the most-studied psychiatric condition in corollary discharge research, but it is not the only disorder where this system appears to matter. Parkinson’s disease, known primarily for tremor and movement difficulties, also involves characteristic problems with speech. People with Parkinson’s often speak more softly than they intend and have difficulty monitoring their own vocal output. Research into the neural basis of voice production in Parkinson’s disease has noted that self-monitoring of speech is abnormal in these patients.21PubMed Central. The intrinsic resting state voice network in Parkinson’s disease The neurodegeneration that disrupts motor circuits in Parkinson’s may simultaneously degrade the corollary discharge signals that help the brain track what the voice is doing, leaving patients unable to accurately gauge their own volume or articulation.
This connection between motor disease and self-monitoring failures is still being worked out, but it reinforces a broader point: corollary discharge is not just a curiosity of basic neuroscience. It is embedded in everyday abilities like controlling the loudness of your voice in a conversation, adjusting your grip when a bag is heavier than expected, and keeping your visual world anchored while you look around a room. The system runs so seamlessly that you never notice it until something goes wrong, whether that something is a psychiatric disorder, a neurological disease, or simply a poorly timed video call where your own voice echoes back with a half-second lag and suddenly you cannot form a sentence. That last scenario is, in miniature, what life might feel like without functioning corollary discharge: a world where every sensation you cause yourself arrives as a surprise.