[Neuroscience] Re: Minimizing stimulus artifacts

r norman via neur-sci%40net.bio.net (by r_s_norman from _comcast.net)
Thu Mar 1 08:22:09 EST 2007


On 28 Feb 2007 20:12:47 -0800, "Bill" <connelly.bill from gmail.com> wrote:

>
>> The biphasic vs. monophasic is important only if you have problems
>> with electrode polarization from running currents always in one
>> direction through the metal/solution interface.  If you have
>> reasonable chlorided silver wires, you shouldn't have a problem.
>>
>> There are lots of different stimulus artifacts and eliminating them is
>> about as easy (i.e. not at all easy!!!) as eliminating line frequency
>> noise.  Is the artifact from a separate electrode or is it from
>> stimulating through the recording electrode?  Those are two very
>> different situations.  If you have problems with series electrode
>> resistance, you will definitely have problems passing current through
>> those electrodes.  "Automatic" compensation methods only work if the
>> electrode resistance is constant and linear.
>
>It's through a bipolar extracellular electrode.
>Here is an example of what I'm talking about:
>http://www.ilikethings.net/stimartifact.GIF   (4k in size).
>That's a thalamocortical EPSP. The stim electrode must be nearly 1mm
>from recording. The stimulus voltage is probably somewhere between
>"30-60V" (at least that's where the dial is on the Grass). The
>duration is between 40-80microseconds. Yet artifact clearly lasts for
>over 2ms; 20 times longer. It has a classical capacitive discharge
>looking falling phase, making me think a biphasic pulse to "discharge"
>whatever the capacitor is, might help.
>
>I'm recording in current clamp with a Rs of about <10MOhm, and an
>input resistance in the 100-200MOhm range.
>
>(In Voltage clamp, the artifact is shorter in duration, but biphasic
>(even triphasic if I have correction/prediction up high) and up to
>20nA in size)
>
>Any comments appreciated.

The real question is whether you have the same stimulus artifact
without the thalamocortical cells.  If the recording electrode is
extracellular, do you get the same artifact (of course without the
response) when the recording electrode is simple in the solution but
outside the cells?  Do you get the same artifact when the stimulating
electrode is not in contact with whatever cells you are stimulating?
Move the electrodes just a tiny bit as to not disturb the system
geometry but just don't have any biological tissue involved.  If the
artifact remains, and I suspect that it will, then you have a good
chance at eliminating it.  It is quite possible that the artifact is
being capacitatively coupled from stimulus electrode to recording
electrode directly.  Are these well shielded from each other?  I have
sometimes found that a piece of grounded aluminum foil placed between
the electrodes can help greatly.  In fact I would always have aluminum
foil attached to grounded clip leads handy and would wave it around
all the equipment looking for sources of noise.  If necessary, wrap
the foil with plastic or wax wrap so it doesn't short out anything if
it contacts an electrode lead.  WIth high resistance electrodes, like
the 100-200M recording ones you use, the electrode lead  and the
barrel of the electrode itself is extremely sensitive to picking up
all sorts of noise and artifact.  Use the lowest resistance electrodes
you can to minimize the problem. 


More information about the Neur-sci mailing list