[Neuroscience] Re: The definition of latency?

r norman via neur-sci%40net.bio.net (by r_s_norman from comcast.net)
Fri Nov 27 09:29:37 EST 2009


On Thu, 26 Nov 2009 23:08:58 -0800 (PST), Bill
<connelly.bill from gmail.com> wrote:

>Hi,
>
>If it wasn't explicitly stated, how do you think latency is
>calculated? From the time from the stimulus artifact (or presynaptic
>action potential) to the peak of the synaptic event? To the start of
>the synaptic event? To the rise-10% of the synaptic event?

Since "latency" is vaguely defined as the interval from stimulus to
response, you have to explicitly state the details, probably in
"Materials and Methods".  If changes in latency are what is important,
it doesn't really matter just when you begin or when you end timing
(assuming the changes are due to factors in the middle, not at the
ends) as long as you are consistent in measuring.

There are still ambiguities in the definition even if the questions
you pose are fully answered.  For example, is the synaptic "event" you
are recording a synaptic potential or is it synaptic current?  The
time of peak response differs greatly in these cases as does the
observable time that it starts -- synaptic current being earlier than
potential because of membrane capacitance and longitudinal current
(cable effects).  Where is the stimulus located?  is presynaptic
action potential propagation a factor in the response latency?

So, all in all, if the fine details of latnecy are important you have
to know all the details.  But that is rarely the case and it is
changes in latency that are of interest.



More information about the Neur-sci mailing list