[Eeglablist] p300 convention

[Jasna Martinovic]

Hi,

around two months ago the same topic came up on CVnet and most people 
seemed to argue that the convention originated from clinical 
neurophysiology, as the plotters have been wired that way since the 40s. 
No one seemed to be able to come up with a reference as to why the 
original amplifiers were connected that way to the plotters, but some 
people also speculated that it may be due to the EPSP and IPSP relations 
to negativity/positivity.

For those interested in technical details, here's a quote from  
"Clinical Neurophysiology, Volume 1: EMG, Nerve Conduction and Evoked 
Potential" editeb by Binnie CD, Cooper R, Mauguiere F, Osselton J, Prior 
PF and Tedman BM (2004, Elsevier - I think most med school libraries 
should have this book, in case anyone wants to look it up themselves), 
chapter 1.2 on techniques (p.25-26):

"At this point it would be appropriate to look at the polarity of the 
displayed data. There is considerable ignorance and confusion about the 
way that voltages applied to the input terminals cause the output of the 
recording system (pen, oscilloscope trace, etc.) to move (up or down), 
especially in evoked potential systems. This is vital in the 
interpretation of recordings, so it is prudent to take a little time to 
understand it.

As we have just seen, the differential amplifiers used in 
neurophysiology have two inputs and an earth. These inputs have been 
variously known as grid 1 and grid 2 (dating from the days of thermionic 
valves), "black and white" because of the colour of the wires used, 
"positive and negative" according to the convention of electrical 
engineering, and lead 1 and lead 2. Now imagine lead 1 connected via a 
switch to the negative pole of the battery and lead 2 to the positive 
pole. Let us also imagine that the potential difference applied to the 
two inputs is 100 nmicrovolts. This can be obtained using a potential 
divider as in Fig 1.2.10.

When the switch is closed, the output will be deflected (say) up 
(depending on how the manufacturer has done his wiring) (Fig 1.2.10a). 
If the polarity of the battery is now reversed (positive to input 1; 
negative to input 2), the output will be displaced downwards (Fig 
1.2.10b) - this is how a differential amplifier works. In 
neurophysiology, where we are trying to locate sources from the 
deflections of the display (the inverse problem to that in Fig. 1.2.10), 
we can say that, if the amplifiers are connected as in Fig.1.2.10a, an 
upward deflection of the display means that the electrode connected to 
lead 1 is negative with respect to that connected to lead 2. This is the 
same as describing lead 2 as being positive to lead 1. It is not 
possible to be more specific.

 From 1945 to 1966 there was a generally accepted convention that if a 
potential difference was applied to the two inputs of an amplifier with 
lead 1 connected to the negative pole, then the output display (usually 
pen writers) would be deflected upwards. The same negative signal 
applied to lead 2 would deflect the display downwards. However, with the 
arrival into clinical neurophysiology of other techniques than EEG 
(particularly evoked potentials), this convention was sometimes changed 
to match engineering practice where positivity is usually displayed upwards.

This caused confusion and a straw poll at an international meeting in 
1976 showed that about 60% workers in evoked potentials displayed a 
negative signal applied to input 1 as an upward deflection on the output 
and 30% a downward deflection (the rest did not know!). No one was 
prepared to change their convention, so both were deemed acceptable 
provided that the polarity was clearly indicated on any diagram (Donchin 
et al., 1977). The matter was not improved by some manufacturers 
labelling their input terminals positive and negative, red and green or 
active and passive! It is interesting to note that Grey Walter and 
George Dawson deliberately chose "black" and "white" to avoid any 
connotation with other conventions. The matter remains contentious, 
although in clinical EEG negative at lead 1 (grid 1, black lead) up is 
the accepted convention (BUN: black up negative); in evoked potential 
work it usually depends on the type of EP being recorded. It really does 
not matter which way up the display is - we can all learn to read a map 
with south at the top - so long as we know."


regards,
jasna


Dr Jasna Martinovic
School of Psychology
University of Aberdeen
William Guild Building
Aberdeen
AB24 2UB

tel: 01224 272240
email: j.martinovic @ abdn.ac.uk
web: http://www.abdn.ac.uk/~psy527/dept/ 
<https://mail.abdn.ac.uk/owa/redir.aspx?C=c3a81cef73474a5ebe7379f5a81cb70f&URL=http%3a%2f%2fwww.abdn.ac.uk%2f%7epsy527%2fdept%2f>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20110621/485510ad/attachment.html>