Every second you are awake, your brain produces a continuous stream of electrical activity. Embedded within that stream are small, reproducible signals that occur in direct response to specific events around you: a click, a flicker of light, a touch on the skin.
These responses are called evoked potentials, and they are among the most precisely timed measurements in neuroscience.
What Is an Evoked Potential?
An evoked potential is a distinct, measurable change in EEG voltage that occurs immediately after an external stimulus. "Evoked" refers to the causal trigger: something happened, and the brain responded. Meanwhile, "potential" refers to the electrical potential difference that develops between recording electrodes on the scalp.
What makes this signal scientifically useful is a property called stimulus-locked timing. The neural response follows the stimulus at a short, reasonably predictable delay and is always measured in reference to the moment of stimulus onset.
In practical terms, a researcher presenting a tone through headphones or flashing a visual pattern on a screen aligns every EEG epoch to the exact instant the stimulus was delivered. Any consistent electrical change that appears shortly after that alignment point is considered part of the evoked response.
This is fundamentally different from the brain's ongoing spontaneous activity. The brain produces electrical fields continuously, regardless of whether a stimulus is present.
An evoked potential is a small, time-locked response that rides on top of the usual background. Spontaneous activity continues whether or not a stimulus occurs. An evoked potential appears only because a stimulus occurred.
Evoked potentials can be triggered in two broad ways:
Natural sensory input: a sound, a visual pattern, a touch.
Direct experimental stimulation of the cortex: For example the magnetic pulses used in transcranial magnetic stimulation.
Where Evoked Potentials Come From
A scalp-recorded evoked potential is the summed electrical activity of large populations of cortical neurons.
The primary source of this summed activity is the postsynaptic potential. When one neuron sends a signal to another, the receiving neuron experiences a small electrical change at its synapses. This postsynaptic potential is the basic currency of neural communication.
In the cortex, the largest and most orderly contributors are pyramidal cells, a class of excitatory neurons that form layered arrangements across the cortical surface. That geometric regularity allows their postsynaptic potentials to sum into coherent electrical fields.
Then, these local fields spread through the surrounding brain tissue, pass through the cerebrospinal fluid that cushions the brain, travel through the skull, and finally reach the scalp.
Types of Evoked Potentials Tests
The type of evoked potentials test selected depends on the sensory system or nerve pathway under examination. Each method uses a different stimulus and records activity from locations chosen to capture the relevant response.
Although the procedures share a common principle, their preparation, waveform patterns, and clinical uses differ. The main categories are visual, auditory, and somatosensory testing.
Visual Evoked Potentials (VEP)
Visual evoked potentials assess electrical activity generated along the visual pathway, from the eyes through the optic nerves and visual structures toward the back of the brain. A common protocol presents a patterned screen, although the exact stimulus can vary according to the testing setting. Electrodes on the scalp record the brain’s response while the individual focuses on the visual target.
VEP testing can help provide functional information when visual symptoms are unexplained or when clinicians are assessing possible impairment of optic or central visual pathways. The result is not a direct measure of visual sharpness, and a normal recording does not exclude every eye or brain disorder. Factors such as attention, visual fixation, refractive correction, and the clarity of the stimulus can influence the waveform.
Brainstem Auditory Evoked Potentials (BAEP) Testing
Brainstem auditory evoked potentials measure electrical responses generated as sound travels through the auditory nerve and brainstem pathways. Clicks or tone-based sounds are commonly presented through earphones, and scalp electrodes record a series of responses occurring within a short period after each sound. The waveform contains several peaks whose timing can provide information about conduction through successive parts of the auditory pathway.
BAEP testing is useful when hearing-pathway function needs to be assessed without relying entirely on a person’s behavioral response. It may be used in selected neurological evaluations, hearing assessments, and situations in which cooperation is limited. Since the recording is sensitive to stimulus level, ear conditions, electrode placement, and electrical noise, results must be considered alongside other hearing and neurological information.
Somatosensory Evoked Potentials (SSEP)
Somatosensory evoked potentials evaluate pathways carrying sensation from a limb toward the spinal cord and brain. Mild electrical pulses are delivered to a peripheral nerve, often at the wrist or ankle, while responses are recorded at several points along the route. The resulting signals can help assess whether conduction is preserved through peripheral nerves, spinal structures, and central sensory pathways.
SSEP testing is particularly relevant when clinicians need functional information about spinal cord or sensory-pathway integrity. It can also be used during selected surgical procedures to monitor pathway function over time. The test does not measure every aspect of sensation, and an abnormal response must be distinguished from effects caused by temperature, limb position, peripheral nerve disease, medications, or technical interference.
Challenges and Variables in Measuring Evoked Potentials
The population signal that defines an evoked potential is always embedded within the brain's ongoing spontaneous electrical activity, and that background is typically larger than the response it surrounds.
This creates a practical challenge. If the spontaneous EEG is larger than the evoked response, a single trial often cannot reveal the stimulus-locked signal clearly.
The classic solution is signal averaging where the same stimulus is presented many times. Each EEG epoch is aligned to the exact moment of stimulus onset, and the resulting trials are averaged together.
The evoked response, which occurs at the same latency on every trial, survives the averaging process. The spontaneous activity, which is not synchronized to the stimulus, tends to cancel out.
Single-Trial Analysis and Signal Processing
While averaging has been the standard approach for decades, it is not the only route. With appropriate signal-processing methods, single-trial analysis is possible.
A study by Başar et al. about 40 Hz spontaneous activity and auditory evoked potentials reported that the applied signal analysis method enabled single-trial inspection of combined EEG and evoked potential epochs rather than requiring a large averaged waveform. Single-trial approaches matter because they preserve trial-by-trial variability that averaging erases.
The authors of that study also noted that their method could be extended to studies on cognitive processes, where the meaningful signal may change from one trial to the next.
Impact of Pre-Stimulus Brain State
The brain's pre-stimulus state can shape the size of the response that follows. A transcranial magnetic stimulation study of four clinically healthy adult males explored this connection. Researchers recorded EEG and electromyography while delivering magnetic pulses to the left motor cortex at 100% motor threshold, the pulse intensity that reliably triggers a muscle response, and at a rate of 0.1 Hz.
They found:
Significant negative correlation (−0.22) with high-alpha power
Significant positive correlation (+0.17) with gamma power
Strongest correlation (0.27) with low-gamma-to-high-alpha ratio
The authors interpreted these correlations as possible indicators of cortical excitability. A brain running in an activated state, with relatively more gamma activity and less alpha, may respond more vigorously to a stimulation pulse. A brain dominated by slower alpha rhythms may show a smaller response.
But the correlations were modest, and the sample was only four participants. The findings suggest a real relationship between pre-stimulus frequency bands and evoked response size, but they cannot stand alone as a settled model.
What Is Actually Being Evoked?
A second core challenge is establishing what exactly is being evoked in the first place.
Not every potential labeled "evoked" is produced purely by the intended stimulus. Many brain stimulation techniques carry built-in sensory side effects that can themselves trigger brain responses.
Transcranial magnetic stimulation (TMS) is a useful case study. TMS uses brief magnetic pulses to excite populations of neurons in a targeted cortical region. But the magnetic coil also produces a loud click, and the pulse creates a physical sensation on the scalp.
Both of these peripheral effects are sensory events in their own right. The click activates the auditory system, and the scalp sensation activates the somatosensory system.
The significance of this ambiguity was demonstrated in a 2019 experiment involving 17 healthy young individuals. Researchers delivered real TMS over two cortical regions, the left posterior parietal cortex and the superior frontal gyrus, and recorded responses to a realistic sham stimulation that mimicked the click and scalp sensation without producing a transcranial pulse. They applied standard precautions, including a foam layer underneath the coil and auditory noise masking.
However, despite these measures, the cortical potentials evoked by real TMS closely resembled the potentials evoked by the realistic sham TMS. The temporal and spatial features matched for both early and late components of the response.
This does not mean evoked potentials are invalid. It means that a recorded response is not automatically the response to the intended stimulus. The study authors concluded that peripheral multisensory control conditions are required in TMS-EEG designs to separate transcranial from non-transcranial components. The logic extends to any stimulation method with sensory side effects that needs a control that isolates the target pathway.
Why Evoked Potentials Matter
The methodological caution above is not a reason to dismiss evoked potentials. It is part of the reason they remain useful. A signal that can be precisely timed, recorded, and controlled for is a signal that can be applied.
Across research and clinical fields, evoked potentials are commonly used to:
Test whether sensory pathways are intact
Monitor brain responses during medical procedures
Design brain-computer interfaces
The logic behind sensory pathway testing is that if a stimulus travels along a healthy route from the sense organ to the brain, a time-locked response should appear at the expected latency. A delayed or absent response may point to a disruption somewhere along that route.
Why Stimulus-Locked Signals Shape Practical Neuroscience
In neuroscience, evoked potentials show that the brain’s response to the outside world can be measured through precisely timed electrical changes at the scalp. These signals arise from thousands of cortical neurons acting together, making them a reliable window into population-level brain activity rather than the firing of any single cell.
The same timing property that makes them identifiable also creates the core scientific challenge: spontaneous brain rhythms are usually larger, so separating the locked response requires repeated trials or careful single-trial analysis. Researchers also need control conditions when stimulation brings sensory side effects, because the recorded response may reflect those extras rather than the intended stimulus alone.
Before a stimulus even arrives, the brain’s ongoing state can shape how large the response becomes, with faster frequency bands linked to stronger responses in the studies reviewed. This points to a brain that is not simply passive; its internal rhythm at the moment of input affects what gets measured.
Moreover, practical tools like brain-computer interfaces can use the frequency-locked properties of evoked potentials for real-time monitoring, such as detecting fatigue through changes in EEG power. Still, the strongest evidence supports the basic signal properties, while many broader uses remain promising but not fully proven.
References
Başar, E., Rosen, B., Başar-Eroglu, C., & Greitschus, F. (1987). The associations between 40 Hz-EEG and the middle latency response of the auditory evoked potential. International Journal of Neuroscience, 33(1-2), 103-117. https://doi.org/10.3109/00207458708985933
Zarkowski, P., Shin, C. J., Dang, T., Russo, J., & Avery, D. (2006). EEG and the variance of motor evoked potential amplitude. Clinical EEG and neuroscience, 37(3), 247-251. https://doi.org/10.1177/155005940603700316
Conde, V., Tomasevic, L., Akopian, I., Stanek, K., Saturnino, G. B., Thielscher, A., ... & Siebner, H. R. (2019). The non-transcranial TMS-evoked potential is an inherent source of ambiguity in TMS-EEG studies. Neuroimage, 185, 300-312. https://doi.org/10.1016/j.neuroimage.2018.10.052
Frequently Asked Questions
What is an evoked potential?
An evoked potential is a distinct, measurable change in EEG voltage that occurs immediately after and in response to an external stimulus (such as a light, sound, or touch). It is "stimulus-locked," meaning it has a predictable delay following the stimulus.
How do evoked potentials differ from spontaneous brain activity?
Spontaneous activity is the brain's ongoing, continuous stream of electrical fields that occurs regardless of external stimuli. An evoked potential is a small, precisely timed response to a specific event that rides on top of this background activity.
Why is signal averaging used in EEG research?
Spontaneous brain activity is typically larger than the small evoked response, making it hard to see in a single trial. Presenting the stimulus multiple times and averaging the epochs causes the random, unsynchronized spontaneous activity to cancel out while the consistent, stimulus-locked evoked potential survives.
Is single-trial analysis of evoked potentials possible?
Yes. While signal averaging is traditional, advanced signal-processing methods allow for single-trial analysis. This approach, as discussed in a study by Başar et al., is highly valuable because it preserves the trial-by-trial variability that averaging erases, which is particularly useful for studying changing cognitive processes.
What role does the pre-stimulus brain state play?
The brain’s ongoing activity immediately before a stimulus influences the size of the subsequent response. A transcranial magnetic stimulation study suggests that an "activated" pre-stimulus state (characterized by higher gamma power and lower alpha power) correlates with a stronger, more excitable response.
Emotiv is a neurotechnology leader helping advance neuroscience research through accessible EEG and brain data tools.
Medical Disclaimer: The information provided on this website is for educational and informational purposes only and is not intended as medical or health advice. This content may contain errors and should not be relied upon to make life-altering health, medical, or lifestyle choices. Always seek the advice of your physician or other qualified health provider with any questions you may have regarding a medical condition or treatment.
Christian Burgos
