The electroretinogram (ERG) is a measure of electrical activity of the retina. It is typically recorded from the cornea with a wire electrode or gold-foil electrode. Generally, the the ERG is considered to reflect the retinal processing of a light stimulus, typically a photoreceptor response followed by secondary neuronal activity and electrical activity evoked by supportive Muller cells.

I am wondering if the pupil response will add signal on top of the ERG? Firstly, there is the direct pupil response, which causes the pupil to constrict when the eye is subjected to bright light. Pupil constriction may, perhaps, induce electromyogram (EMG) activity when the smooth sphincter muscle around the pupil contracts, which in turn could theoretically contaminate the ERG. Secondly, there is neuronal activity involved in constricting the pupil, as light is processed in the retina by ganglion cells that in turn drive the pupil reflex. Furthermore, there is the consensual pupil response as well. This response causes the contralateral pupil to constrict when the ipsilateral eye is subjected to bright light. The consensual response is mediated by efferent input from the brain following stimulation of the other eye and causes both eyes to restrict. Hence, neural input arrives in the contralateral retina from the brain. This activity is not directly related to the processing of the light stimulus and may disturb the ERG proper when ERGs are recorded with a flash and both eyes are open.

So my question is

  1. Does a pupil-mediated EMG affects the ERG signal?
  2. Does neuronal activity involved in the pupil reflex affect the ERG, for example the efferent input involved in the consensual response?

I have had no luck after extensive searches in text books, Scopus, Pubmed, Google, Google Scholar.


1 Answer 1


Answer version1 (read the comments, I leave v1 for the those who can make the same assumptions in the future)

To answer the question we should clarify some general concepts and processes mentioned it the clause:

  1. EMG is performed by means of direct placement of the electrode (the are two types here - surface-EMG and needle-EMG) on/into the skeletal muscles, this is why there is no such thing "iris-sphincter-related EMG": the sphincter is inside the eye and there is no way to contact it without eyewall penetration somewhere. Thus, this part of the question is hypothetical.

  2. ERG is performed after pharmacological mydriasis (pupil dilation to ensure retinal mass response or to perform multifocal ERG), so in the standard settings there is no substantial pupils reaction during the test. Thus, this part of the question is hypothetical too.

  3. ERG measures electrical changes when one of the electrodes is directly attached to the cornea - here we should stress "to one cornea", and "reads" the current from only one side. This is why, the reaction of contralateral pupil, skeletal muscles, heart/defibrillator activity are irrelevant since they are out of the electrical loop existing between the electrodes (the second electrode is on the forehead).

  4. ERG measures electrical activity only - there is no muscle contractions in the process, while EMG involves the time needed to acetylcholine to be released and muscle contraction to begin. This means that the time scopes of these tests are different. In addition, as stated, pupils react as an efferent organ (III nerve, mediator related), while retinal photoreceptors react directly to the light stimulation, so there is a clear delay between these two processes. Theoretically, one can perform full cut of the optic nerve with the resultant loss of afferent signal and consequent loss of efferent pupil reaction, but the ERG will still detect the normal curves if the photoreceptors are intact.

Back to question itself: according to the mentioned facts we should state that the question is more hypothetical and theoretical, thus answering it directly is problematic and the answer should not be assessed without facts mentioned above.

So, the final answer - 1+2: No, pupil reaction does not affect the ERG in the real and standard settings and should not be counted as an interfering factor.


I've found an interesting patent claiming that pupillary responses can contribute to eERG (electrical stimulus instead of light). These influences stated to be artifacts, thus the authors raise the question how to diminish/filter them out.

They say

Eye movements and pupil responses may contribute to the eERG;

Pharmacologic dilation abolished these pupil responses;

Bilateral artifacts, such as pupil responses, in the corneal eERG cannot be sufficiently reduced by using dilation drops; Even after dilation, a residual, but substantial electrical response persisted in the contralateral eye.

So, I should convert my answer to yes/yes. Taking into account the obscure nature of the subject, I hope you will agree that by making mistakes we can learn something new. I personally learnt a lot trying to answer this excellent question.

Patent application number: 20120143080 (if the link will be broken in future, one can find the patent by this number).

Direct link http://www.faqs.org/patents/app/20120143080

  • $\begingroup$ I admire your detective work greatly. In fact, I am one of the authors on that patent. In this patent we are basically making guesses, but we have no answers. I left out any reference to electrical stimulation in this question at Bio.SE to make it more general applicable to light-evoked ERGs. Hence, although I cannot accept your answer, I have started a bounty which is yours to keep in 24 hours - as a token of appreciation (cannot give it away immediately). $\endgroup$
    – AliceD
    Apr 1, 2015 at 11:02
  • $\begingroup$ @aliced I thought that pupullary elecrical stimuli are filtered out from regular ERG $\endgroup$
    – Ilan
    Apr 1, 2015 at 12:05
  • $\begingroup$ @AliceD there is an amplifier and some filtering mechanism which should affect the noise in some way - I've read about that while searching information about ERG, but don't have a source now.... $\endgroup$
    – Ilan
    Apr 1, 2015 at 12:40
  • $\begingroup$ Never encountered any mention of filtering of pupil responses. A normal ERG a and b peak develop prior to the pupil response so timing may be brought forward, but not so on long-latency responses. Perhaps this discussion should be proceeded in chat when necessary. Thanks! $\endgroup$
    – AliceD
    Apr 1, 2015 at 12:43
  • 1
    $\begingroup$ @AliceD If I understand you properly, we don't have sufficient sources to support both answers neither "no", nor "yes". What can we add to "no" answer is the fact that sphincter has constant tonic efferent current, which is why its consequent rise might be less influential on ERG... In addition, the current for dilator should in some way lessen the ERG curve of the sphincter... $\endgroup$
    – Ilan
    Apr 3, 2015 at 10:47

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .