6 ms·

They don't link to the study, but the image accompanying the article doesn't inspire much confidence. What are the lines? All the dots are almost identical wi

They don't link to the study, but the image accompanying the article doesn't inspire much confidence. What are the lines? All the dots are almost identical with only one or two being slightly bigger or smaller. More fundamentally, what makes them think the results have anything to do with language other than culture and life experience? If they did this in Germany, native Arabic speakers will have (mostly) moved during their lives, presumably traversing a bilingual and bicultural life (that might demand more bicameral connections) relative to the German subjects.

esalman 3y ago
I think the lines indicate highly connected/weighted network edges. The weights are inferred from white matter tracts, measured using diffusion MRI. Dots are nodes on the network, highly connected nodes are indicated by a pair of large dots.
- chaxor 3y ago
  Sure, but this is unfortunately not really novel work in this area. White matter tract and (3T, typically not more) fMRI-based analyses of language tasks have been done for decades, and simply don't provide much information due to the horrendous resolution inherent to the task. What is needed in the area is something more similar to EPFL's in silico rat cortex reconstruction, only for humans. I haven't seen a single reference (ever) for single cell resolution information on connectivity in Wernicke's area (arguably the absolute most important and distinguishing cortex in human brain vs non-human primate). This is what is needed now, rather than extremely ham handed or crude, low resolution correlations across brain structures which are well established in many contexts.
tgv 3y ago
Of course they can't distinguish language from culture, nor nature (heridatary influences). It's a farce to make that claim. And the claim really is "The current study suggests that the structural language connectome develops and is modulated by environmental factors such as the characteristic processing demands of the native language." And I don't know how much diffusion MRI has improved, but around 2010 (when I left this field) it wasn't considered reliable.

esalman 3y ago

I think the lines indicate highly connected/weighted network edges. The weights are inferred from white matter tracts, measured using diffusion MRI. Dots are nodes on the network, highly connected nodes are indicated by a pair of large dots.

chaxor 3y ago
Sure, but this is unfortunately not really novel work in this area. White matter tract and (3T, typically not more) fMRI-based analyses of language tasks have been done for decades, and simply don't provide much information due to the horrendous resolution inherent to the task. What is needed in the area is something more similar to EPFL's in silico rat cortex reconstruction, only for humans. I haven't seen a single reference (ever) for single cell resolution information on connectivity in Wernicke's area (arguably the absolute most important and distinguishing cortex in human brain vs non-human primate). This is what is needed now, rather than extremely ham handed or crude, low resolution correlations across brain structures which are well established in many contexts.

tgv 3y ago

Of course they can't distinguish language from culture, nor nature (heridatary influences). It's a farce to make that claim. And the claim really is "The current study suggests that the structural language connectome develops and is modulated by environmental factors such as the characteristic processing demands of the native language." And I don't know how much diffusion MRI has improved, but around 2010 (when I left this field) it wasn't considered reliable.