The phrase seizure threshold is often used as if the brain has one fixed line: stay below it and no seizure occurs; cross it and a seizure begins. That picture is useful as shorthand, but biologically it is much more complicated.
There is no single clinical number that can be measured and labelled as a person’s seizure threshold. Seizure susceptibility emerges from the state of neuronal networks: ion channels, synapses, inhibitory control, extracellular ions, sleep–wake state, medications, hormones, fever or illness, alcohol or drug withdrawal, structural brain changes and genetics can all influence how easily a network can enter a seizure state.
Threshold is a model of susceptibility
In epilepsy, ‘threshold’ is best understood as a conceptual description of how resistant or susceptible a brain network is to transitioning into a seizure. A high threshold means stronger perturbation may be needed before a seizure develops. A lower threshold means less perturbation may be enough.
This is a population-and-network concept, not the same thing as the membrane threshold at which one neuron fires an action potential. An individual neuron can cross its firing threshold many times during normal brain activity without producing a seizure.
Seizure threshold is not fixed
Seizure susceptibility changes over minutes, hours, days and longer periods. The same person may be much more vulnerable under one set of conditions than another. Sleep deprivation, illness, abrupt withdrawal from certain substances, medication changes and biological rhythms can all alter the probability of a seizure without permanently changing the underlying diagnosis.
This helps explain why seizures can cluster at particular times, disappear for long intervals, or occur during temporary physiological stress without implying that the brain has one permanent threshold value.
How a network approaches the seizure state
Modern seizure science increasingly treats seizure onset as a transition between dynamical brain states. Networks normally fluctuate through many patterns of activity. In a susceptible system, combinations of altered excitation, inhibition, ion concentrations and connectivity can make the seizure state more accessible.
Researchers sometimes describe this using the language of stability. A stable network can absorb perturbations and return to normal activity. A less stable network may be pushed into a self-sustaining seizure by a smaller disturbance.
Excitation and inhibition are part of the story — not the whole story
A common explanation is that seizures occur when excitation exceeds inhibition. That can be useful at a basic level, but it is incomplete. Seizure susceptibility also depends on network synchrony, ion gradients, intrinsic membrane properties, glial regulation and the geometry of connected circuits.
GABAergic inhibition itself can change during intense activity as intracellular chloride accumulates. Extracellular potassium can rise and make nearby neurons easier to depolarise. These mechanisms show why ‘threshold’ is an emergent property of a whole network rather than a simple chemical ratio.
Genetics can influence baseline susceptibility
Genetic variants affecting sodium, potassium, calcium or chloride channels, neurotransmitter receptors, synaptic proteins and other cellular systems can alter neuronal excitability. In some genetic epilepsies, these changes create a long-term predisposition to seizure generation.
Genetic susceptibility does not mean that every carrier will have the same seizure burden. The effect of a variant can depend on developmental stage, other genes, brain region and environmental or physiological factors.
Structural brain changes can alter threshold
Stroke, trauma, cortical malformations, tumours, hippocampal sclerosis and other structural abnormalities can reorganise local circuits. Loss of neurons, gliosis, synaptic sprouting, altered inhibition and changes in connectivity can make some networks more capable of generating recurrent seizures.
This is different from saying that every structural lesion causes epilepsy. Many people with brain lesions never develop recurrent unprovoked seizures. The eventual risk depends on the type, location and biological consequences of the lesion as well as individual susceptibility.
Sleep and sleep deprivation
Sleep has a powerful effect on epileptic networks. Non-REM sleep can increase synchronisation and make interictal epileptiform discharges more visible, while some epilepsies show strong relationships with sleep or awakening.
Sleep deprivation is also associated with increased seizure susceptibility in many people with epilepsy and is used clinically in some EEG protocols because it can increase the chance of revealing epileptiform activity. The mechanisms are not completely settled, but changes in cortical excitability, network synchronisation and GABAergic inhibition have all been implicated.
Importantly, sleep deprivation does not trigger a seizure every time and does not affect every epilepsy equally. It is better understood as a factor that can shift susceptibility in some people.
Fever and temperature
Fever can increase seizure susceptibility, especially in infants and young children. Febrile seizures are the clearest example, but temperature sensitivity is also relevant in specific genetic epilepsies such as Dravet syndrome and other fever-sensitive epilepsy syndromes.
Fever-related mechanisms are complex. Temperature can affect ion-channel kinetics, synaptic transmission, inflammatory signalling, respiration and brain pH. The importance of each mechanism differs by age and syndrome.
Alcohol withdrawal
Long-term heavy alcohol exposure causes the nervous system to adapt to alcohol’s inhibitory and excitatory effects. Abrupt cessation can then produce a rebound state of increased excitability. Alcohol-withdrawal seizures are therefore classified as acute symptomatic seizures rather than proof, by themselves, of epilepsy.
The important point is that the direction of risk changes with brain state. Acute alcohol exposure and alcohol withdrawal do not affect seizure susceptibility in the same way, which is another reason the threshold idea cannot be reduced to one fixed number.
Medicines and other substances
Some medicines can increase seizure risk, particularly at high doses, in overdose, during rapid dose changes or when several risk factors are present together. The degree of risk varies greatly between drugs and individuals.
It is therefore misleading to label every medicine on a ‘lowers seizure threshold’ list as equally dangerous. Dose, metabolism, interacting medicines, age, renal or hepatic function, pre-existing epilepsy and structural brain disease all change the clinical significance.
Similarly, abruptly stopping some sedative drugs can provoke seizures because the brain has adapted to their presence. Withdrawal physiology can therefore be as important as the drug’s direct effect.
Antiseizure medicines do not simply ‘raise the threshold’ in one universal way
Antiseizure medicines reduce seizure probability through different mechanisms. Some alter voltage-gated sodium channels, some affect calcium channels, some enhance GABAergic inhibition, some reduce glutamate release, and others act on synaptic vesicle proteins or multiple targets.
Describing all of these effects as ‘raising the threshold’ can be useful shorthand, but it hides important differences in seizure type, syndrome and mechanism. A medicine effective in one epilepsy can be less effective — or occasionally worsen seizures — in another.
Missed doses and falling drug concentrations
For someone whose seizures are controlled by medication, a missed dose can reduce the protective effect of treatment as drug concentrations fall. The size of that effect depends on the medicine’s half-life, dosing schedule, seizure history and individual pharmacology.
This does not mean that one missed dose will inevitably cause a seizure. It means one component of the person’s seizure resistance may be temporarily reduced.
Hormones and biological rhythms
Seizure probability can vary with menstrual hormonal changes in some people, which is the basis of catamenial epilepsy. Circadian and multi-day rhythms in seizure occurrence have also been demonstrated in long-term recordings.
These findings challenge the idea of a static threshold. A person’s susceptibility can oscillate over time even when medication and lifestyle remain relatively stable.
Metabolic disturbance can provoke seizures
Severe disturbances in glucose, sodium, calcium and other metabolic variables can provoke acute symptomatic seizures. In these situations, the seizure is caused by a temporary physiological disruption severe enough to push a brain into seizure activity.
Once the disturbance is corrected, the long-term risk may be very different from that of a person with recurrent unprovoked seizures. This distinction is central to separating provoked seizures from epilepsy.
A low threshold is not the same as epilepsy
Anyone can potentially have a seizure if the provocation is strong enough. Severe hypoglycaemia, major electrolyte disturbance, acute brain injury, drug toxicity or alcohol withdrawal can provoke seizures in people who do not have epilepsy.
Epilepsy, by contrast, is characterised by an enduring predisposition to recurrent unprovoked seizures, or by meeting recognised diagnostic criteria that indicate a sufficiently high risk of recurrence. A temporary reduction in seizure resistance is therefore not equivalent to developing epilepsy.
Why one person has a seizure and another does not
Two people exposed to the same apparent trigger may respond differently because their networks do not start from the same biological state. Genetics, age, brain structure, sleep, medicines, previous injury, illness and many other factors influence susceptibility.
This is why population risk factors cannot predict an individual seizure with certainty. The presence of several threshold-lowering factors can increase risk without making a seizure inevitable.
Threshold and kindling are not identical
Kindling is an experimental phenomenon in which repeated subthreshold electrical or chemical stimulation eventually produces progressively stronger seizure responses. It demonstrated that repeated activity can produce lasting changes in seizure susceptibility.
Human epilepsy is not simply ‘kindling in people’. Some epileptogenic processes share mechanisms of plasticity and increasing network excitability, but the causes and biology of human epilepsies are far more diverse.
Can clinicians measure seizure threshold?
There is no routine blood test, scan or EEG that gives a person’s seizure threshold as a number. EEG can reveal epileptiform activity, imaging can reveal structural risk factors, and clinical history can identify circumstances associated with seizures, but none directly measures an overall threshold.
Threshold can be measured experimentally in more restricted ways — for example, by determining how much electrical or chemical stimulation is required to provoke a seizure in a model — but that is not equivalent to measuring an individual’s everyday seizure susceptibility.
Why the term is still useful
Despite its limitations, ‘seizure threshold’ remains useful because it captures an important clinical idea: seizure probability changes when several protective and provoking influences shift at the same time.
The safest interpretation is therefore not ‘every person has one fixed threshold’, but ‘brain networks have changing susceptibility to seizure transitions’.
What science still does not know
Researchers still cannot calculate in advance exactly when a particular seizure will occur from a list of triggers and protective factors. The transition to seizure is nonlinear, and apparently similar brain states can produce different outcomes.
Long-term intracranial and wearable recordings are improving seizure forecasting research, but prediction remains probabilistic rather than certain. The threshold concept should therefore never be used to imply that a person could have prevented a seizure simply by avoiding one trigger.
What this page does — and does not — explain
This page explains changing susceptibility to seizure generation. It does not repeat the electrical threshold of an individual neuron, the full biology of a seizure once it has started, or the long-term process of epileptogenesis. Those subjects have separate Science & Discovery pages.
Sources
Jasper’s Basic Mechanisms of the Epilepsies, 5th ed. — Transition to Seizure from Cellular, Network, and Dynamical Perspectives
Jasper’s Basic Mechanisms of the Epilepsies, 5th ed. — Role of Ion Concentration Dynamics in Epileptic Seizures
Jasper’s Basic Mechanisms of the Epilepsies, 5th ed. — GABAA-Receptor Signaling and Ionic Plasticity
Sleep deprivation: a risk for epileptic seizures — review
Alcohol Withdrawal Seizures: Neurobiological Mechanisms, Clinical Predictors, and Evidence-Based Management
Epileptogenic drugs: a systematic review
Evidence checked: September 2026.