I am thinking where to put optimally depth electrodes in stereo EEG outside sulci in studying the functioning of neural transmission of N neurons over one perivascular space in rodents, mice or rat, where I would like to know how small N can be. The theoretical minimum is one neuron but I would like to understand the Physics how small current you can measure by EEG reliably with not too high Signal Noise Ratio (SNR). I did not find any meta-analysis about EEG electrode locations in DynaMed Plus.

I am not sure about the anatomic therminology of Neurology so I use the terms cerebral cortex, sagittal and coronal planes in describing directions.

Definition of electrode's location somehow in cerebral cortex

The further the electrode is away from the source, the lower SNR. This proposes me that there has has to be at least 20 neurons to minimize the variance. What is the average distance between two neurons?

Here my conjunctures about location of electrodes where the picture is from Virginia's university Nature publication here:

enter image description here

where the electrode is surrounding N neurons and one blood vessel such that one electrode is below the neuron and one above. It is contraindicated to put the electrode too close to the blood vessel.

Definition of electrode's Coronal-Sagital locations

I am interested in

  • how close perivascular space can the electrode be? Stereoelectroencephalography (SEEG) implantation method has a safety distance minimally 2 mm but practically more than 5 mm of depth electrodes. [TODO cite]
  • how superficial can the electrode #2 be? 1.0 - 1.5 mm in microelectrode EEG here i.e. mostly in superficial layers.
  • how deep can the electrode #1 be? I think mostly in superficial layer i.e. 1.0 - 1.5 mm with microelectrodes. SEEG with depth electrodes - minimally 2 mm but practically more than 5 mm.

Stability of electrodes

The stability is affected by many processes discussed here. The array of electrodes is allowed to float with the motions of the cortex due

  • heartbeat
  • respiration
  • ...


  1. Talairach J, Bancaud J, Szikla G, et al. Approche nouvelle de la neurochirugie de l’epilepsie. Méthodologie stérérotaxique et résultats thérapeutiques. Neurochirurgie (1974) 20 (Suppl. 1): 1-240.
  2. Cossu M, Cardinale F, Castana L, et al. Stereo-EEG in the presurgical evaluation of focal epilepsy: a retrospective analysis of 215 procedures. Neurosurgery (2005) 57: 706-718.

How can you define evidence-based locations of stereo EEG electrodes?

  • 2
    $\begingroup$ I am not sure why this question received a down vote. It is usually good to give an indication if it isn't plainly obvious what the problem is. That said, I am assuming that you are asking about rodents, mice or rat. Please clarify with an edit to your question. $\endgroup$
    – AMR
    Dec 28, 2015 at 6:24
  • 2
    $\begingroup$ @AMR - The downvote was probably the result from trying to measure single cells with EEG, that's impossible. This question is on depth electrodes for single-unit recordings. Further, the answer to this question is opinion based, because there is no strict definition of where an electrode needs to be. The further away from the source, the lower the SNR. What is acceptable and what not is pob. $\endgroup$
    – AliceD
    Dec 28, 2015 at 9:52
  • 1
    $\begingroup$ @Christiaan Thank you for your comment! I made the body clearer. The one neuron study is just a theoretical model. I want to understand how you can estimate the amount of neurons required to be optimally within the distance in the cerebral cortex in EEG measurement, minimizing the variance. $\endgroup$ Dec 28, 2015 at 11:36
  • $\begingroup$ The question is still unclear. An EEG is performed with noninvasive scalp electrodes in humans, or, perhaps, with transcranial epidural electrodes in special cases. The figure implies depth electrodes which is per definition not an EEG as far as I know. $\endgroup$
    – AliceD
    Dec 28, 2015 at 14:24
  • 1
    $\begingroup$ That's interesting. I didn't know that. +1 $\endgroup$
    – AliceD
    Dec 28, 2015 at 16:06


You must log in to answer this question.