[FieldTrip] Neighbourhoods and Geodesic Arrays

[Schoffelen, J.M. (Jan Mathijs)]

Hi Matthew,

The definition of neighbours for electrodes is always a bit subject to subjectivity, I’d say.

Ideally, indeed a distance-based heuristic is usually the best thing to do (as opposed to a triangulation heuristic). Yet, this is only really meaningful if the electrode positions are defined in 3D coordinates, rather than defined as a 2D projection on the plane (as in the layout you created from the picture).

One can debate as to whether (in the figure you sent around) (e.g.) LMPf should count as a neighbour to RMFr etc., particularly if the array is really geodesic. Yet, in the grand scheme of things, I don’t think that practically it makes too much of a difference in the final results (i.e. the inferential conclusion that you are drawing based on the statistical test). This despite the fact that the exact neighbours determine the ’shape’ of the clusters, but this counts for both the clusters in the observed data, and in the instances of the permuted data. Thus, the inferential conclusion based on a particular clustering scheme is valid.

Best wishes,
Jan-Mathijs




On 22 Sep 2020, at 16:18, Matthew Wood <matthew.wood at utsa.edu<mailto:matthew.wood at utsa.edu>> wrote:

Hello all,

My name is Matthew Wood and I'm a graduate student at The University of Texas at San Antonio. I'm starting to use cluster analysis for my EEG experiments and wanted to make sure I understood a few things about how neighborhoods in FieldTrip.

1) Our lab uses a 26 EEG electrode setup in a geodesic array (link to diagram: https://www.researchgate.net/figure/Diagram-representing-the-26-electrode-placement-in-geodesic-array_fig2_235354855). I noticed there do not exist any templates on Fieldtrip for my specific array, so I used the feature to use an image to build a layout file. If anyone happens to have a layout file I can use with more accurate measurements, I would appreciate if you could share it so that I can use that instead.

2) Since I'm building the layout and neighborhoods from scratch, I'm a little nervous about their validity. I used the "enableedit" configuration in the ft_neighbourplot function to build my neighbors (I've attached a figure here). One of the features of a geodesic array is that all electrodes are equally spaced apart. Is it correct to then assume that this should be reflected in the neighborhood structure?

3) Finally, for cluster analysis, does the distance between the neighbors matter? If so, does the fact that my layout has my electrodes placed with (seemingly) arbitrary units matter?

I appreciate any help anyone can provide!
Matthew
[https://www.researchgate.net/profile/Nicole_Wicha/publication/235354855/figure/download/fig2/AS:767255897993216@1559939469184/Diagram-representing-the-26-electrode-placement-in-geodesic-array.jpg]<https://www.researchgate.net/figure/Diagram-representing-the-26-electrode-placement-in-geodesic-array_fig2_235354855>

Figure 2. Diagram representing the 26-electrode placement in geodesic...<https://www.researchgate.net/figure/Diagram-representing-the-26-electrode-placement-in-geodesic-array_fig2_235354855>
Download scientific diagram | Diagram representing the 26-electrode placement in geodesic array. from publication: Meaning first: A case for language-independent access to word meaning in the bilingual brain | This study aimed to determine how deeply a word is processed in the bilingual brain before the word's language membership plays a role in lexical selection. In two ERP experiments, balanced Spanish-English bilinguals read lists of words and pseudowords in Spanish and English,... | Bilingual, Lexical Access and Spanish | ResearchGate, the professional network for scientists.
www.researchgate.net<http://www.researchgate.net/>


<neighbor_layout.png>_______________________________________________
fieldtrip mailing list
https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
https://doi.org/10.1371/journal.pcbi.1002202

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20200923/bf478486/attachment.htm>