100%, I think the papers aren't confused in themselves. Just different measures used in different papers...

I had a flick through and see neural coding style papers using 'big s = more sparsity', but stat physics style papers (Brunel et al) using 'f = fraction of active neurons' measure of sparsity

Really?? Maybe I need to revise the lit.

Perhaps yeah. A cultural change in the debate would be welcome. But I remain happy for people to pursue their ideas with conviction

yeah that is a risk of course. And we know there are many social dynamics that can lead to years of wasted effort, see eg Alzheimers research, early genomics studies, etc.

But otoh if everyone was equivocal all the time, humming and hawing and debating, I don't think we would progress much either!

I think a lot of arguments in neuroscience stem from this dynamic and I don't even think it's a problem. Good for people to be bold or even ode-sided about their claims and let the ideas battle out in the field. As long as we can keep the debate civil and respectful and less clickbaity

I also had the same impression Dan.

Sounds like another case of insiders in a topic knowing all the caveats and limitations of their claims, but projecting only the bold novel parts to outsiders, who then interpret the core claims as overhyped

see also "ANNs are good models of the brain"

100%. I'm a fan of the idea of multiplexing here, that different receivers can decode different information from the same broadcasted signal.

Great to see!