[Eeglablist] Why most of good 'brain' ICs are 'dipolar' with show 'red'-centerd scalp topos, although 2/3 of the cortex is in sulci?

[Makoto Miyakoshi]

Hi Pål,

Here are some statistics from O'kusky and Colonnier (1982).

"There are thus about 160 neurons in the striate cortex for each
geniculocortical relay cell [...] Each individual axon is not restricted to
a vertical column of 160 neurons: rather, each spreads out tangentially and
overlaps with its neighbors. For each mm^2 of tangential spread there are
potentially thousands of neurons on which it might form synaptic contacts
in layer IVC alone."

Thus 1 relay cell is present for every 160 neurons, hence I reported 0.63%
in my Wiki article.
https://sccn.ucsd.edu/wiki/Makoto%27s_preprocessing_pipeline#A_big_scalp_topography.3F_.28For_170.2C000_page_views.2C_12.2F18.2F2023_updated.29
But as the authors indicate in the last sentence above, much more neurons
are effectively connected to the relay neuron via synaptic contact.

As Mountcastle showed, neurons are functionally organized to form cortical
columns. With the column, the signals are more redundant. This may give us
an insight into the sparsity of the structure: if every cortical column has
at least one thalamo-cortical network connection, that should be
sufficient.

"There are about 200 million (2×108) cortical minicolumns in the human
neocortex with up to about 110 neurons each,[16] and with estimates of
21–26 billion (2.1×1010–2.6×1010) neurons in the neocortex. With 50 to 100
cortical minicolumns per cortical column a human would have 2–4 million
(2×106–4×106) cortical columns." (Wikipedia
https://urldefense.com/v3/__https://en.wikipedia.org/wiki/Cortical_column*:*:text=their*20Nobel*20Prize.-,Number*20of*20cortical*20columns,10)*20neurons*20in*20the*20neocortex__;I34lJSUlJSUlJSU!!Mih3wA!FvrdZluv-Y4_QJI6Gcw1_cwgdWfSsa2T-y6_cwpe508IJvg6s0fnFc5MzsmCoK2MkQSGytyD-wOBB6hhBuQ2-8Hf3fs$ 
)

To sum, in monkey BA17, 1 relay cell is present per 160 neurons in average,
while a minicolumn contains 110 neurons. A cortical column consists of
50-100 minicolumns. I don't know how far this finding can be generalized to
other cases, but it seems thalamo-cortical connection is generally
abundant. And if the authors are only talking about specific (core)
thalamic nuclei, the side of the thalamus, the non-specific (matrix)
thalamic nuclei may need to be considered separately.

I'm still learning these things and could be wrong. I appreciate your input!

Makoto

On Wed, Dec 13, 2023 at 4:21 PM Makoto Miyakoshi <mmiyakoshi at ucsd.edu>
wrote:

> Thanks Pål for your comments.
> I don't think many people believe the thalamus==pacemaker hypothesis any
> more.
> Lopes da Silva found high coherence between dog's visual cortex and
> pulvinar instead of LGN in his 1980 paper. Saalmann et al. (2012) was a
> nice confirmation of that using monkeys. Basically, they identified the
> pulvinar-V4/TEO (i.e. thalamo-cortical) network of the active visual
> attention. Such a network is 'recruited' upon top-down demand, perhaps as a
> part of a larger network.
>
> Although Nunez insisted on the presence of traveling and standing waves, I
> do not think his claim was to replace the conventional alpha models.
> Instead, such a network can be present on top of something else. I agree
> with you that his claim was 'more or less shot down' except no one dared to
> shoot it but it has been unattended. But you know, Alex Fornito's the
> Nature paper in the early 2023 is nothing but fMRI version of Nunez's
> global field theory with less explanation. They could not explain what
> exactly is traveling in their model. In the case of Nunez's model, the
> medium of the traveling wave is clear: the synaptic action fields mediated
> by white matters.
>
> > 3. The amplitudes of EEG will be linear to the number of synchronous
> firing neurons. If they are firing stochasticly the amplitude will be
> closer to the square root. E.g. 1000000 neurons will in synchrony give an
> unite amplitude of 1000000 as if they have a stochastic firing 1000.
> Coherences gives amplitudes somewhere in between.
>
> I made a humble plot from this simulation. If you are curious, please see
> slide 49 linked below. Basically, after passing some critical point, the
> small-number synchronous firing effect start to outweigh the mass random
> firing effect. It's a replication of Hari et al. (1997).
>
> https://sccn.ucsd.edu/mediawiki/images/1/13/On_EEG%27s_generative_mechanism.pdf
>
> In the actual 3-D implementation of the sheet of (a massive number of)
> parallel pyramidal neurons, there is also a counter-intuitive non-linear
> spatial summation effect. I call it a transducer array effect. This also
> favors the wide EEG sources to be scalp-recorded. The explanation is lined
> below.
>
> https://sccn.ucsd.edu/wiki/Makoto's_preprocessing_pipeline#Transducer_array_effect_.28For_140.2C000_page_views.2C_10.2F09.2F2020_added.29
>
> > 2. The nature is not redundant in the way that many neurons are used for
> the same function. You will not expect close by neurons to be in synchrony.
> Why should they? But there will be coherent elements.
>
> Right. I think the point is sparsity and efficiency.
> For example, we might think 1% of synchrony is trivial. But most of the
> esoteric cognitive neuroscience studies rely on BOLD signal measurement
> that shows changes often less than 1%.
> I think when samples are ample, a small fraction of changes still counts.
>
> Speaking of the global field theory, when I asked Paul Nunez what
> he thought of the view by Buszaki that frequencies of the brain rhythm are
> stable across species, he immediately disagreed. I would love to
> see someone replicate Buszaki's claim there.
>
> Makoto
>
> On Wed, Dec 13, 2023 at 12:10 PM Pål Gunnar Larsson <pall at ous-hf.no>
> wrote:
>
>> Hi Makoto
>>
>> I hope I did not open Pandoras box. First a small history.
>> Andersen and Andersson publish in 1968 that alpha activity came from
>> pacemaker cells in Thalamus. This was a cat study. In 1971 Sturm van Leuwen
>> and Lopez da Silva showed in dog that there were no good coherence between
>> the thalamus activity and the alpha. Even the frequencies didn’t match.
>> Then Nunez suggested alpha was generated by standing waves in cortex with
>> some human data. However, also his hypothesis was more or less shot down.
>> Now, it seems like a leading hypothesis is that alpha is generated by
>> cortical spreading activity that to some extend is influenced by standing
>> waves. My point here is that model and methods to a large extend influences
>> your results.
>>
>> Let me make some points
>> 1. Propagating takes time, hence you get a phase shifted as a function of
>> distance in cortex. This will give cancellations in the EEG due to the
>> spatial averaging. Therefore our electrodes "sees" patchy activity.
>> 2. The cortico-cortical fibers propagate AP with up to 9 m/s and
>> intracortical propagation is 0.2m/s- 0.5 m/s.
>> 2. The nature is not redundant in the way that many neurons are used for
>> the same function. You will not expect close by neurons to be in synchrony.
>> Why should they? But there will be coherent elements.
>> 3. The amplitudes of EEG will be linear to the number of synchronous
>> firing neurons. If they are firing stochasticly the amplitude will be
>> closer to the square root. E.g. 1000000 neurons will in synchrony give an
>> unite amplitude of 1000000 as if they have a stochastic firing 1000.
>> Coherences gives amplitudes somewhere in between.
>> 4. The EEG we record are to a large extend from not highly correlated
>> neurons with coherences waxing and waning depending on the activity. E.g.
>> Evoked potentials gives higher coherences and hence, higher amplitudes.
>> 5. Thalamus is an important relay station in the brain, but I am not
>> convinced that it has an important pacemaker function, at least not in
>> humans.
>> 6. I think I have to go back to the PL Nunez book to find his references
>> on the human vs rodent brain wiring.
>>
>> To me the EEG is mostly modulated by changing activities in the brain and
>> not by some pacemakers.
>>
>> Best
>> Pål
>>
>> E-mail: pall at ous-hf.no
>>
>> Ikke sensitiv
>>
>> -----Opprinnelig melding-----
>> Fra: eeglablist <eeglablist-bounces at sccn.ucsd.edu> På vegne av Makoto
>> Miyakoshi via eeglablist
>> Sendt: 12. desember 2023 19:35
>> Til: eeglablist <eeglablist at sccn.ucsd.edu>
>> Emne: Re: [Eeglablist] Why most of good 'brain' ICs are 'dipolar' with
>> show 'red'-centerd scalp topos, although 2/3 of the cortex is in sulci?
>>
>> Hi Pal and Ramesh,
>>
>> Thank you for your comments.
>> Let me quote an explanation from Jones (2002) "Thalamic circuitry and
>> thalamocortical synchrony" p.1669. He explains how the two types of
>> thalamocortical projections, core and matrix, interact to create cortical
>> activities, which is probably the direct source of EEG/MEG signals.
>>
>> %%%%%%%%%%%%%%%%%
>> The relay cells of the thalamic core, with their focused projections to
>> an individual cortical area, clearly form the basis for the relay of place-
>> and modality-specific information to the cortex whereas those of the
>> thalamic matrix form a more obvious basis for the dispersion of activity in
>> the thalamocortical network across larger areas of cortex. Within a zone of
>> cortex, the terminations of matrix cell axons on distal dendrites in
>> superficial layers and of matrix cell axons on more proximal dendrites in
>> middle layers should serve as a coincidence detection circuit, providing
>> for a high degree of temporal integration between inputs coming from the
>> two classes of thalamic cells (Llinas&Pare 1997; figure 11). Coincidence of
>> this kind should promote synchronous activity in the cells of individual
>> cortical columns and in any group of columns activated by the same
>> stimulus. Activity in these columns would then be returned via layer VI
>> corticothalamic cells to the thalamic nucleus from which they receive
>> input, serving  to reinforce thalamocortical synchrony. This activity would
>> be spread to other cortical columns in the same cortical area and in
>> adjacent cortical areas via the diffuse projections of matrix cells in the
>> thalamic nucleus through which externally or internally generated activity
>> was first passed to the cortex.
>> %%%%%%%%%%%%%%%%%
>>
>> I guess fMRI-based 'functional mapping' is rather close to the mapping of
>> the projections by 'core thalamic nuclei'.
>>
>> When I analyzed the USCD mismatch negativity (MMN) database, I found MMN
>> is a whole-brain phenomenon and not limited to Fz. One of the coauthors
>> (probably Juan) asked me why the 'visual cortex' showed ERP as well. I
>> could not answer to his question. Now I have a better explanation--From the
>> core/matrix point of view, it is not surprising that auditory stimulus
>> activates cortices of other sensory modalities.
>>
>> Giandomenico Ianetti from U Rome showed me 4 or 5 ERPs evoked by stimuli
>> of different modalities including visual, auditory, tactile, and
>> laser-evoked pain. His point was clear: These ERPs are the same. It was
>> eyes opening.
>>
>> So, this is my proposal: Let us unlearn the fMRI-based functional brain
>> mapping when we do EEG. Instead, let us pay more attention to thalamus.
>>
>> I do not know about the cortico-cortical connection via u fiber very much.
>> If you know a paper detailing that point, please let me know. My initial
>> respose is, is the u-fiber connection fast enough to form a
>> near-simultaneous activity across the cortex? Isn't a region-wide
>> projection from the thalamus more feasible to explain it?
>>
>> Makoto
>>
>>
>> On Tue, Dec 12, 2023 at 12:25 PM Ramesh Srinivasan via eeglablist <
>> eeglablist at sccn.ucsd.edu> wrote:
>>
>> > I've been enjoying this discussion because it taps into one of those
>> > EEG truths/inconsistencies we never talk about.
>> >
>> > 1. We artifact edit EEG data mostly based on the idea it should be
>> > smooth low spatial frequency information.  We don't trust very
>> > (channel, frequency, time)  localized EEG signals.
>> >
>> > 2. Then after we clean the EEG data we want a story for our paper that
>> > is time, frequency, source localized as compact as possible because it
>> > makes a nice narrative.
>> >
>> > Regarding synchrony in adjacent gyri, u-fibers are helpful and yes, I
>> > think most of it is corticocortical rather than thalamocortical but I
>> > think the 2% is just a guess. It's clearly not as thalamocortical as
>> > animal models
>> >
>> > Ramesh Srinivasan
>> > Professor
>> > Cognitive Sciences
>> > Biomedical Engineering
>> >
>> > On Tue, Dec 12, 2023, 6:40 AM Pål Gunnar Larsson via eeglablist <
>> > eeglablist at sccn.ucsd.edu> wrote:
>> >
>> > > Just want to add. In rats about 50% of all fiber going in and out of
>> > > the cortex are connected to the thalamus. In humans connections are
>> > > about 2%, according to Nunez. Hence, we should be very careful when
>> > > you try to extrapolate from animal research to humans.
>> > >
>> > > Pål G. Larsson
>> > >
>> > > Ikke sensitiv
>> > >
>> > >
>> > > -----Opprinnelig melding-----
>> > > Fra: eeglablist <eeglablist-bounces at sccn.ucsd.edu> På vegne av
>> > > Makoto Miyakoshi via eeglablist
>> > > Sendt: 11. desember 2023 19:06
>> > > Til: EEGLAB List <eeglablist at sccn.ucsd.edu>
>> > > Emne: Re: [Eeglablist] Why most of good 'brain' ICs are 'dipolar'
>> > > with show 'red'-centerd scalp topos, although 2/3 of the cortex is in
>> sulci?
>> > >
>> > > Hi Scott,
>> > >
>> > > > "How are LFP signals across each of these gyrii synchronized
>> > > > across the
>> > > dataset?"
>> > >
>> > > The answer is not so special. The synchrony is achieved via
>> > > thalamo-cortical loops.
>> > > In the following Wiki article, I linked to my presentation at an NIH
>> > > summer seminar in which I showed multiple evidence that cortical
>> > synchrony
>> > > and coupling is controlled by thalamus.
>> > >
>> > >
>> > https://urldefense.com/v3/__https://sccn.ucsd.edu/wiki/Makoto*27s_prep
>> > rocessing_pipeline*Two_presentations_at_a_seminar:_EEG_preprocessing_a
>> > nd_generative_mechanism_.28For_240.2C000_page_views.2C_09.2F21.2F2023_
>> > added.2C_10.2F18.2F2023_updated.29__;JSM!!CzAuKJ42GuquVTTmVmPViYEvSg!I
>> > xfdnbB611_BrP_68EFD1xVZHKoKQu6E2vLO7VJL104Il5HhWcGfwu-K0btGTDMoDcUuo0-
>> > 5NDYMK30iA7NCfDue$
>> > >
>> > > So, when thalamus makes different cortical regions to fire together,
>> > > then you see the synchronous activity. That's it. The distant
>> > > cortical regions do not have to be directly connected via each
>> > > neuron's lateral branches (which does exist, but the conduction
>> > > speed is very slow compared with
>> > that
>> > > of white matter) Note this is not a one-way 'imposing' the rhythm
>> > > from
>> > the
>> > > thalamus to the cortex, like the historical 'pace-maker' hypothesis
>> > > by Andersen and Andersen (1967) but it is a bi-directional
>> interaction.
>> > >
>> > > One fact that might help you see the situation is that only 1% of
>> > > neurons need to be synchronized to form 95% of the amplitude of the
>> > > observed
>> > signal
>> > > according to Hari (1997).
>> > >
>> > > Also, it might also help to remember that there is no general
>> > > guarantee that an EEG source is stationary and localizable. See
>> > > Izhikevich's classical simulation.
>> > >
>> > >
>> > https://urldefense.com/v3/__https://www.izhikevich.org/publications/la
>> > rge-scale_model_of_human_brain.htm__;!!Mih3wA!FLAtrGcNSxCRup1LcnfgWIJk
>> > dfC5HMOr_rPujEIjdmGug69GeOA7PxjXg5NqsRrbtx3VtKZRJhKrQ9HIAFxdjg0n5Fg$
>> > > There is an established principle of functional brain mapping but it
>> > > is a product of statistical processing such as (heavy) averaging.
>> > > ICA model is the same, hence it is stationary across time. The small
>> > > and localizable
>> > EEG
>> > > source is heavily a statistical concept. The actual ongoing EEG is
>> > > stochastic, dynamic, and diffuse. When we see ICA results,
>> > > therefore, we should distinguish properties of the filter from
>> properties of data.
>> > >
>> > > Makoto
>> > >
>> > >
>> > > On Mon, Dec 11, 2023 at 11:42 AM Scott Makeig <smakeig at gmail.com>
>> wrote:
>> > >
>> > > > Makoto -
>> > > >
>> > > > When you repeat the claim that EEG sources 'found' by ICA
>> > > > decomposition must be at least several adjacent gyrii in size, you
>> > > > fail to ask, "How are LFP signals across each of these gyrii
>> > > synchronized across the dataset?"
>> > > > Doesn't this require some physiological basis, and if so, what is
>> it??
>> > > >
>> > > > Scott Makeig
>> > > >
>> > > > On Mon, Dec 11, 2023 at 11:17 AM Makoto Miyakoshi via eeglablist <
>> > > > eeglablist at sccn.ucsd.edu> wrote:
>> > > >
>> > > >> Hello EEGLAB list,
>> > > >>
>> > > >> For those who have wondered so, here are my answers.
>> > > >> I asked two questions:
>> > > >>
>> > > >> (1) Why do good 'brain' ICs show dipolar scalp topos although 2/3
>> > > >> of the cortex is in sulci?
>> > > >> (2) Why do these dipolar IC scalp topos show red (positive)
>> centers?
>> > > >>
>> > > >> The answer was published a few days ago.
>> > > >>
>> > > >>
>> > > >> https://urldefense.com/v3/__https://onlinelibrary.wiley.com/doi/1
>> > > >> 0.10
>> > > >> 02/hbm.26540__;!!Mih3wA!FPOThEiX2hsD7TJBq7WyhlV8v6HSkTe_swsBEoB2R
>> > > >> M-Bh -BGerduzZBnmEtDBamyosThbqv9Xrc1gGPSmdm52LpO7jM$
>> > > >>
>> > > >> The answer to (1): It is because scalp-recorded EEG is
>> > > >> insensitive to sulcal sources compared with gyral sources. This
>> > > >> finding justifies the use of lissencephalic (i.e. no sulci) brain
>> > > >> model proposed in Electric Fields of the Brain (Nunez and
>> > > >> Srinivasan, 2006) together
>> > with
>> > > Spline Laplacian.
>> > > >> This also supports the view that the major source of
>> > > >> scalp-recordable EEG is pretty broad (minimum 6.45 cm^2) which
>> > > >> requires a continuum of multiple gyral crowns.
>> > > >>
>> > > >> I did not write it in the paper, but the result basically refutes
>> > > >> the claim that ICA is a high-resolution EEG spatial filter
>> > > >> because the result confirms that ICA is mostly blind to 2/3 of
>> > > >> the cortex. In fact, it seems ICA results are always dominated by
>> > > >> high-power, low-frequency, and very broad sources. I will publish
>> > > >> this view in the near future.
>> > > >>
>> > > >> The answer to (2): It is because EEGLAB's ICA sets the initial
>> > > >> topos of all ICs red centered (i.e. positive dominant). Thus,
>> > > >> unless necessary, the algorithm does not flip the polarities.
>> > > >>
>> > > >> Now you wonder--when does the ICA algorithm flip the polarity to
>> > > >> produce 'blue' centered (i.e. negative dominant) ICs? I found
>> > > >> that those blue-centered ICs tend to show poor physiological
>> > > >> validity with large index numbers. A known clear exception for
>> > > >> this rule is ICs localized for the motor cortex.
>> > > >>
>> > > >> People use ICA to clean EEG. I use EEG to glean ICA, which is
>> > > >> more
>> > fun.
>> > > >>
>> > > >> Makoto
>> > > >> _______________________________________________
>> > > >> Eeglablist page:
>> > >
>> > https://urldefense.com/v3/__http://sccn.ucsd.edu/eeglab/eeglabmail.htm
>> > l__;!!CzAuKJ42GuquVTTmVmPViYEvSg!IxfdnbB611_BrP_68EFD1xVZHKoKQu6E2vLO7
>> > VJL104Il5HhWcGfwu-K0btGTDMoDcUuo0-5NDYMK30iA9rRvrgG$
>> > > >> To unsubscribe, send an empty email to
>> > > >> eeglablist-unsubscribe at sccn.ucsd.edu
>> > > >> For digest mode, send an email with the subject "set digest mime"
>> > > >> to eeglablist-request at sccn.ucsd.edu
>> > > >>
>> > > >
>> > > >
>> > > > --
>> > > > Scott Makeig, Research Scientist and Director, Swartz Center for
>> > > > Computational Neuroscience, Institute for Neural Computation,
>> > > > University of California San Diego, La Jolla CA 92093-0559,
>> > > >
>> > >
>> > https://urldefense.com/v3/__http://sccn.ucsd.edu/*scott__;fg!!CzAuKJ42
>> > GuquVTTmVmPViYEvSg!IxfdnbB611_BrP_68EFD1xVZHKoKQu6E2vLO7VJL104Il5HhWcG
>> > fwu-K0btGTDMoDcUuo0-5NDYMK30iA0yMHq0i$
>> > > >
>> > > _______________________________________________
>> > > Eeglablist page:
>> > >
>> > https://urldefense.com/v3/__http://sccn.ucsd.edu/eeglab/eeglabmail.htm
>> > l__;!!CzAuKJ42GuquVTTmVmPViYEvSg!IxfdnbB611_BrP_68EFD1xVZHKoKQu6E2vLO7
>> > VJL104Il5HhWcGfwu-K0btGTDMoDcUuo0-5NDYMK30iA9rRvrgG$
>> > > To unsubscribe, send an empty email to
>> > > eeglablist-unsubscribe at sccn.ucsd.edu
>> > > For digest mode, send an email with the subject "set digest mime" to
>> > > eeglablist-request at sccn.ucsd.edu
>> > > _______________________________________________
>> > > Eeglablist page:
>> > >
>> > https://urldefense.com/v3/__http://sccn.ucsd.edu/eeglab/eeglabmail.htm
>> > l__;!!CzAuKJ42GuquVTTmVmPViYEvSg!IxfdnbB611_BrP_68EFD1xVZHKoKQu6E2vLO7
>> > VJL104Il5HhWcGfwu-K0btGTDMoDcUuo0-5NDYMK30iA9rRvrgG$
>> > > To unsubscribe, send an empty email to
>> > > eeglablist-unsubscribe at sccn.ucsd.edu
>> > > For digest mode, send an email with the subject "set digest mime" to
>> > > eeglablist-request at sccn.ucsd.edu
>> > _______________________________________________
>> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> > To unsubscribe, send an empty email to
>> > eeglablist-unsubscribe at sccn.ucsd.edu
>> > For digest mode, send an email with the subject "set digest mime" to
>> > eeglablist-request at sccn.ucsd.edu
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
>